JP5526596B2 - Image processing apparatus and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing program.

The digitization of copiers is becoming increasingly popular for the purpose of reusing paper documents.
However, in order to reduce the distribution and accumulation load of electronic documents that are scanned images, reading is performed at a low resolution. Therefore, the image quality of the character image is deteriorated.

  As a technology related to this, for example, Patent Document 1 discloses a character recognition dictionary creation method and character recognition that can automatically generate a large number of sample character data for recognition and can reduce the user's labor. An object of the present invention is to provide a dictionary creation device. In a character recognition dictionary creation unit, an image data input unit captures image data of basic characters from the outside, and a cut-out unit captures image data captured by the image data input unit. The basic image data is obtained by cutting out the character portion from the image, and the disturbance processing unit performs predetermined image processing simulating various disturbances on the basic image data to obtain sample character data for recognition. As a result, a large number of sample character data for recognition are automatically generated from a small number of basic image data. It is disclosed that is.

Japanese Patent Laid-Open No. 10-240869

  In the present invention, when information related to a pixel block is added to an image of a pixel block in an image, the added information is information related to a pixel block having a higher resolution than the image of the pixel block in the image. An object of the present invention is to provide an image processing apparatus and an image processing program.

The gist of the present invention for achieving the object lies in the inventions of the following items.
The invention according to claim 1 is an image of a plurality of pixel blocks, an information generation unit that generates information about the pixel block having a higher resolution than the pixel block image, and an image of the pixel blocks generated by the information generation unit And a storage means for storing information relating to the pixel block, a verification unit for verifying the image of the pixel block in the received image and the image of the pixel block stored in the storage unit, and based on the verification result by the verification unit Information adding means for extracting information related to the pixel block from the storage means and adding the information related to the pixel block to the image of the pixel block in the received image , wherein the information generating unit includes a plurality of types Alternatively, an output unit that outputs an image including a pixel block image having a size, a first reading unit that reads an image output by the output unit, and an image output by the output unit Several times, a second reading means for reading at a lower resolution than the first reading means, a first cutting means for cutting out an image of a pixel block in the image read by the first reading means, and the first A second cutout unit that cuts out an image of the pixel block in the image read by the second reading unit, and the storage unit applies the pixel to the image of the pixel block cut out by the second cutout unit. The image processing apparatus stores an image of a pixel block cut out by the first cutout unit as information on the block .

According to a second aspect of the present invention, there is provided an image of a plurality of pixel blocks, an information generation unit that generates information on the pixel block having a higher resolution than the image of the pixel block, and an image of the pixel block generated by the information generation unit And a storage means for storing information relating to the pixel block, a verification unit for verifying the image of the pixel block in the received image and the image of the pixel block stored in the storage unit, and based on the verification result by the verification unit The information generating means for extracting information relating to the pixel block from the storage means and adding the information relating to the pixel block to the image of the pixel block in the received image, wherein the information generating means reads the image A first reading means; a second reading means for reading the image a plurality of times at a lower resolution than the first reading means; and an image in the image read by the first reading means. A first cutout unit that cuts out an image of an element block, and a second cutout unit that cuts out an image of a pixel block in the image read by the second reading unit; and the storage unit includes the second cutout unit An image processing apparatus that stores an image of a pixel block cut out by the first cutout unit as information on the pixel block for an image of the pixel block cut out by the cutout unit .

According to a third aspect of the present invention, there is provided an image of a plurality of pixel blocks, an information generation unit that generates information about the pixel block having a higher resolution than the pixel block image, and an image of the pixel block generated by the information generation unit And a storage means for storing information relating to the pixel block, a verification unit for verifying the image of the pixel block in the received image and the image of the pixel block stored in the storage unit, and based on the verification result by the verification unit Information adding means for extracting information related to the pixel block from the storage means and adding the information related to the pixel block to the image of the pixel block in the received image, wherein the information generating unit includes a plurality of types Alternatively, an output unit that outputs an image including an image of a pixel block of a size, a reading unit that reads an image output by the output unit, and a solution of the image read by the reading unit Generated by the low-resolution phase processing means for generating an image of a different phase at a low degree, a first clipping means for cutting out an image of a pixel block in the image read by the reading means, and the low-resolution phase processing means A second cutout unit that cuts out an image of the pixel block in the image that has been cut out, and the storage unit stores information about the pixel block as information on the pixel block cut out by the second cutout unit. An image processing apparatus that stores an image of a pixel block cut out by one cut-out means .

According to a fourth aspect of the present invention, there is provided a plurality of pixel block images, information generation means for generating information relating to the pixel blocks having a higher resolution than the pixel block image, and pixel block images generated by the information generation unit. And a storage means for storing information relating to the pixel block, a verification unit for verifying the image of the pixel block in the received image and the image of the pixel block stored in the storage unit, and based on the verification result by the verification unit The information generating means for extracting information relating to the pixel block from the storage means and adding the information relating to the pixel block to the image of the pixel block in the received image, wherein the information generating means reads the image A reading means; a low-resolution phase processing means for reducing the resolution of the image read by the reading means and generating an image of a different phase; and the reading means A first cutout unit that cuts out an image of a pixel block in the image, and a second cutout unit that cuts out an image of the pixel block in the image generated by the low-resolution phase processing unit; An image processing apparatus that stores an image of a pixel block cut out by the first cutout unit as information on the pixel block for an image of the pixel block cut out by a second cutout unit .

According to a fifth aspect of the present invention, the information generating unit further includes a recognizing unit that recognizes an image of a pixel block in the image, and the storage unit is an image of the pixel block cut out by the second cutting unit. 5. The image processing apparatus according to claim 2, wherein a recognition result by the recognition unit is stored as information relating to the pixel block .

The invention according to claim 6 is characterized in that the pixel block is a character and includes a character code, a font type, and a size as information relating to the pixel block. It is a processing device.

According to a seventh aspect of the present invention, there is provided a computer comprising: an information generating unit that generates an image of a plurality of pixel blocks, information on the pixel block having a higher resolution than the pixel block image; and pixels generated by the information generating unit A storage unit for storing a block image and information on the pixel block, a verification unit for verifying a pixel block image in the received image and a pixel block image stored in the storage unit, and a verification by the verification unit Based on the result, information on the pixel block is extracted from the storage unit, and the information generation unit functions as an information adding unit that adds the information on the pixel block to the image of the pixel block in the received image. Output means for outputting an image including a plurality of types or sizes of pixel clusters, first reading means for reading an image output by the output means, and the output means A second reading unit that reads the output image a plurality of times at a resolution lower than that of the first reading unit, and a first image that cuts out an image of a pixel block in the image read by the first reading unit. And a second cutout unit that cuts out an image of a pixel block in the image read by the second reading unit, and the storage unit stores the pixel block cut out by the second cutout unit The image processing program is characterized in that an image of a pixel block cut out by the first cutout unit is stored as information relating to the pixel block with respect to the first image.

According to an eighth aspect of the present invention, there is provided a computer comprising: an information generation unit that generates an image of a plurality of pixel blocks, information about the pixel block having a higher resolution than the pixel block image; A storage unit for storing a block image and information on the pixel block, a verification unit for verifying a pixel block image in the received image and a pixel block image stored in the storage unit, and a verification by the verification unit Based on the result, information on the pixel block is extracted from the storage unit, and the information generation unit functions as an information adding unit that adds the information on the pixel block to the image of the pixel block in the received image. A first reading means for reading an image, a second reading means for reading the image a plurality of times at a lower resolution than the first reading means, and the first reading means. First storage means for cutting out an image of a pixel block in the read image; and second cutting out means for cutting out an image of a pixel block in the image read by the second reading unit; An image processing program for storing an image of a pixel block cut out by the first cutout unit as information on the pixel block for an image of the pixel block cut out by the second cutout unit It is.

According to a ninth aspect of the present invention, there is provided a computer comprising: an information generation unit that generates an image of a plurality of pixel blocks, information about the pixel block having a higher resolution than the pixel block image; and pixels generated by the information generation unit A storage unit for storing a block image and information on the pixel block, a verification unit for verifying a pixel block image in the received image and a pixel block image stored in the storage unit, and a verification by the verification unit Based on the result, information on the pixel block is extracted from the storage unit, and the information generation unit functions as an information adding unit that adds the information on the pixel block to the image of the pixel block in the received image. Output means for outputting an image including images of pixel clusters of a plurality of types or sizes, a reading means for reading an image output by the output means, and a reading means Low resolution phase processing means for reducing the resolution of the read image and generating images of different phases, first cutout means for cutting out an image of a pixel block in the image read by the reading means, and the low resolution A second cutout unit that cuts out an image of the pixel block in the image generated by the phase processing unit; and the storage unit applies the pixel block to the pixel block image cut out by the second cutout unit. An image processing program characterized by storing an image of a pixel block cut out by the first cut-out means as information relating to the above.

According to a tenth aspect of the present invention, there is provided a computer, an information generation unit that generates an image of a plurality of pixel blocks, information on the pixel block having a higher resolution than the image of the pixel blocks, and pixels generated by the information generation unit A storage unit for storing a block image and information on the pixel block, a verification unit for verifying a pixel block image in the received image and a pixel block image stored in the storage unit, and a verification by the verification unit Based on the result, information on the pixel block is extracted from the storage unit, and the information generation unit functions as an information adding unit that adds the information on the pixel block to the image of the pixel block in the received image. A reading means for reading an image; a low-resolution phase processing means for reducing the resolution of the image read by the reading means and generating images of different phases; and the reading First clipping means for cutting out an image of a pixel block in the image read by the means, and second cutting out means for cutting out an image of the pixel block in the image generated by the low-resolution phase processing unit, The storage unit stores an image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. It is a processing program.

According to the image processing device of claim 1, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. The pixel block image output by the output means can be used as information on the pixel block to be applied.

According to the image processing device of claim 2, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, it is possible to generate a pixel block image for collation based on the image actually used by the user.

According to the image processing device of claim 3, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, compared to the case where the present configuration is not provided, it is possible to generate an image of a pixel block for collation with a smaller number of readings.

According to the image processing device of claim 4, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, compared to a case where the present configuration is not provided, it is possible to generate an image of a pixel block for collation with a small number of readings based on an image actually used by the user.

According to the image processing apparatus of the fifth aspect, it is possible to include a character code as information on a pixel block having a high resolution as compared with the case where the present configuration is not provided.

According to the image processing apparatus of the sixth aspect, it is possible to process an image including characters as a target.

According to the image processing program of claim 7, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. The pixel block image output by the output means can be used as information on the pixel block to be applied.

According to the image processing program of claim 8, when information relating to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, it is possible to generate a pixel block image for collation based on the image actually used by the user.

According to the image processing program of claim 9, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, compared to the case where the present configuration is not provided, it is possible to generate an image of a pixel block for collation with a smaller number of readings.

According to the image processing program of claim 10, when information related to a pixel block is added to an image of the pixel block in the image, the added information has a higher resolution than the image of the pixel block in the image. It can be information about a pixel block. Then, compared to a case where the present configuration is not provided, it is possible to generate an image of a pixel block for collation with a small number of readings based on an image actually used by the user.

It is a conceptual module block diagram about the basic structural example of this Embodiment. It is a flowchart which shows the production | generation process example of multi-resolution character image information by this Embodiment. It is a flowchart which shows the example of a high resolution information provision process by this Embodiment. It is a conceptual module block diagram about the structural example of 1st specific embodiment. It is explanatory drawing which shows the data structure example of a multiple resolution information table. It is explanatory drawing which shows the data structure example of a multiple resolution information table. It is a module block diagram about the whole structural example (1) in implement | achieving 1st specific embodiment. It is a module block diagram about the whole structural example (2) in implement | achieving 1st specific embodiment. It is a conceptual module block diagram about the structural example of 2nd specific embodiment. It is explanatory drawing which shows the data structure example of a multiple resolution information table. It is explanatory drawing which shows the data structure example of a multiple resolution information table. It is explanatory drawing which shows the data structure example of a multiple resolution information table. It is a module block diagram about the whole structural example (1) in implement | achieving 2nd specific embodiment. It is a module block diagram about the whole structural example (2) in implement | achieving 2nd specific embodiment. It is a module block diagram about the whole structural example (3) in implement | achieving 2nd specific embodiment. It is a notional module block diagram about the structural example of 3rd concrete embodiment. It is a notional module block diagram about the structural example of 4th specific embodiment. It is a module block diagram about the whole structural example (1) in implement | achieving 4th specific embodiment. It is a module block diagram about the whole structural example (2) in implement | achieving 4th specific embodiment. It is a module block diagram about the whole structural example (3) in implement | achieving 4th specific embodiment. It is a notional module block diagram about the structural example of 5th specific embodiment. It is a notional module block diagram about the structural example of 6th specific embodiment. It is a module block diagram about the whole structural example (1) in implement | achieving 6th specific embodiment. It is a module block diagram about the whole structural example (2) in implement | achieving 6th specific embodiment. It is a module block diagram about the whole structural example (3) in implement | achieving 6th specific embodiment. It is a module block diagram about the whole structural example (4) in implement | achieving 6th specific embodiment. It is explanatory drawing which shows the process example at the time of implement | achieving 1st, 1, 2, 4, 6 specific embodiment as a fax. It is explanatory drawing which shows the process example at the time of implement | achieving 3rd, 5th specific embodiment as a fax. It is explanatory drawing which shows the process example at the time of implement | achieving 1st, 1st, 4th, 6th specific embodiment as a scanner. It is explanatory drawing which shows the process example at the time of implement | achieving 3rd, 5th specific embodiment as a scanner. It is a block diagram which shows the hardware structural example of the computer which implement | achieves this Embodiment.

First, in order to facilitate understanding of the present embodiment, a scene where the present embodiment is used will be described.
As described above, in many cases, an image (electronic document) read by an image reading apparatus (hereinafter also referred to as a scanner) is read at a low resolution in order to circulate or store the image.
On the other hand, a character recognition technology is introduced to enable character search of an electronic document. For example, a character code recognized on a document image using a PDF (Portable Document Format) format is made transparent in a system font. Drawing can be made possible by drawing in color, but if the image quality of characters is poor, the recognition rate decreases.
Furthermore, although the character quality can be improved by drawing using the system font with the character code obtained by character recognition, the font information (size, font type, etc.) cannot be obtained by character recognition. The appearance will change from that of the document.

In addition, due to the reduction in resolution when reading with a scanner, the same character causes various degraded characters depending on the position due to the relationship between the sampling frequency and the character position. This is one of the causes for lowering the recognition rate of character recognition.
In particular, in the standard mode of fax, the character image quality is poor because it is read at a low resolution of about 200 dpi × 100 dpi, and the recognition rate of character recognition is lowered because of the large variance in image quality degradation.
Furthermore, the existing font recognition algorithm also receives disturbance due to the relationship between the sampling frequency and the character position described above, and the recognition accuracy is reduced.

In this embodiment, a signal degradation pattern generated when a document is read by a scanner is used as a database, so that even a character read at a low resolution can be replaced with a high resolution character. Thus, the recognition rate of character recognition can be improved.
In other words, if you prepare a database that simulates the degradation of character images when scanned by a scanner, you can specify the font size, font type, etc. of characters using existing character pattern recognition technology. The character image quality can be improved by replacing the character image with the high resolution associated with.

Hereinafter, examples of various preferred embodiments for realizing the present invention will be described with reference to the drawings.
<Example of basic configuration of the embodiment>
FIG. 1 shows a conceptual module configuration diagram of a basic configuration example of the present embodiment.
The module generally refers to components such as software (computer program) and hardware that can be logically separated. Therefore, the module in the present embodiment indicates not only a module in a computer program but also a module in a hardware configuration. Therefore, the present embodiment also serves as an explanation of a computer program, a system, and a method. However, for the sake of explanation, the words “store”, “store”, and equivalents thereof are used. However, when the embodiment is a computer program, these words are stored in a storage device or stored in memory. It is the control to be stored in the device. In addition, the modules correspond almost one-to-one with the functions. However, in mounting, one module may be composed of one program, or a plurality of modules may be composed of one program. A plurality of programs may be used. The plurality of modules may be executed by one computer, or one module may be executed by a plurality of computers in a distributed or parallel environment. Note that one module may include other modules. Hereinafter, “connection” is used not only for physical connection but also for logical connection (data exchange, instruction, reference relationship between data, etc.).
In addition, the system or device is configured by connecting a plurality of computers, hardware, devices, and the like by communication means such as a network (including one-to-one correspondence communication connection), etc., and one computer, hardware, device. The case where it implement | achieves by etc. is included. “Apparatus” and “system” are used as synonymous terms. “Predetermined” means that the process is determined before the target process, and not only before the process according to this embodiment starts but also after the process according to this embodiment starts. In addition, if it is before the target processing, it is used in accordance with the situation / state at that time or with the intention to be decided according to the situation / state up to that point.

Further, the pixel block includes at least a pixel region that is continuous in four or eight connections, and includes a set of these pixel regions. The set of these pixel areas means that there are a plurality of continuous pixel areas such as 4-connected, and the plurality of pixel areas are in the vicinity. Here, what is in the vicinity is, for example, a region in which the pixel regions are close to each other in distance, a region that is projected in the vertical or horizontal direction so as to cut out one character at a time from one line as a sentence, and surrounded by a blank. There are an image area cut out, an image area cut out at a predetermined interval, and the like.
In many cases, an image of one character is formed as one pixel block. However, it is not necessary that the pixel area is actually recognizable as a character by humans. There are a part of a character, a pixel region that does not form a character, and the like, and any pixel block may be used.
A character is illustrated as a pixel block, and a character image is illustrated as an image of the pixel block.
When a character is shown as an image, it is called a character image, and when it is shown by a code that uniquely identifies the character, it is called a character. In other words, a character image is obtained by visualizing characters on a medium such as paper. Therefore, a single character has a plurality of character images depending on the font type, font size, resolution, and the like. In addition, the pixel block information includes a higher resolution character image, character code, font type, and size.

  The basic configuration example of the image processing apparatus according to the present embodiment includes a multi-resolution information generation module 110, a multi-resolution information storage module 120, a low-resolution character image collation module 130, and a high-resolution information addition module 140.

The multi-resolution information generation module 110 generates information on a plurality of pixel blocks and information on the pixel blocks having a higher resolution than the pixel block image. Here, the “information about the pixel block having a higher resolution than the image of the pixel block” may be information regarding the pixel block for generating an image of the pixel block having a higher resolution than the image of the pixel block. For example, there are a character code, a font type, a high-resolution character image, and the like.
The multi-resolution information storage module 120 stores an image of a pixel block generated by the multi-resolution information generation module 110 and information related to the pixel block.
The low resolution character image collation module 130 collates the image of the pixel block in the received image with the image of the pixel block stored by the multi-resolution information storage module 120.
The high-resolution information adding module 140 extracts pixel block information from the multi-resolution information storage module 120 based on the collation result by the low-resolution character image collation module 130 and receives the pixel block information in the image. Appends to chunk image. Here, “applying to the image of the pixel block in the image” includes the case of replacing with a high-resolution image.

As a more specific example, each module will be described.
The multi-resolution information generation module 110 and the multi-resolution information storage module 120 are connected, and the multi-resolution information generation module 110 passes the multi-resolution character image information 112 to the multi-resolution information storage module 120. The multi-resolution information generation module 110 is an image when a character such as hiragana, katakana, alphanumeric characters, common kanji is scanned a plurality of times at a predetermined resolution with a font type and font size expected to be generally used. A character image). The characters prepared here are characters used for a target document in a scene where the present embodiment is used. A document on which these characters are printed is prepared and read by a scanner. By reading character images in the same document a plurality of times, character images having different phases can be obtained. It is empirically known that this is because the scanner includes a reading part by a mechanical system, so that even if the same character image is read, the character image has a different phase. Also, the resolution may be read at a plurality of resolutions. An existing method is used as a method for cutting a character image from a document image. For example, the character image may be cut out by projecting in a certain direction by separating with a blank having a predetermined area or more (one side may be a predetermined length or more).
These character images, character font information, and resolution information are generated as multi-resolution character image information 112 and sent to the multi-resolution information storage module 120. Since the character font information and resolution information are specified in advance when printing the document, the specified information is obtained by the multi-resolution information generation module 110 and passed to the multi-resolution information storage module 120.

  The multi-resolution information storage module 120, the low-resolution character image collation module 130, and the high-resolution information addition module 140 are connected, and the multi-resolution information storage module 120 passes the low-resolution character image 122 to the low-resolution character image collation module 130. . The multi-resolution information storage module 120 passes the high-resolution information 124 to the high-resolution information adding module 140, and the high-resolution information adding module 140 passes the verification ID 142 to the multi-resolution information storing module 120. The multi-resolution information storage module 120 stores the multi-resolution character image information 112 sent from the multi-resolution information generation module 110, and the stored low-resolution character images 122 to the low-resolution character image collation module 130. The high resolution information 124 associated with the collation ID 142 received from the high resolution information adding module 140 is sent to the high resolution information adding module 140.

  The low resolution character image collation module 130 and the high resolution information provision module 140 are connected, and the low resolution character image collation module 130 passes the collation ID 132 to the high resolution information provision module 140. The received character image 128 and the low resolution character image 122 stored in the multi-resolution information storage module 120 are collated. If there is a match, the collation ID 132 indicating the matched character is not matched. A collation ID 132 (for example, a predetermined special numerical value, more specifically, 0, a negative numerical value, etc.) is transmitted to the high-resolution information adding module 140.

  The high resolution information adding module 140 outputs an output character image 144. The high-resolution information adding module 140 receives the collation ID 132 that is the collation result by the low-resolution character image collation module 130, and if they match, passes the collation ID 142 (collation ID 132) to the multi-resolution information storage module 120. The high resolution information 124 corresponding to the collation ID 142 (collation ID 132) is received, the high resolution information 124 is added to the low resolution character image 128 (may be replaced), and the received character image 128 is output if they do not match. Output as is.

FIG. 2 is a flowchart showing an example of processing for generating the multi-resolution character image information 112 according to this embodiment.
In step S202, the multi-resolution information generation module 110 generates a multi-resolution character image information 112 that is information about a character image and a plurality of resolutions for one character.
In step S204, the multi-resolution information storage module 120 stores multi-resolution character image information 112 including a character image and high-resolution information 124 related to the character.

FIG. 3 is a flowchart showing an example of high resolution information providing processing according to the present embodiment.
In step S302, the present embodiment accepts an image. A character image is cut out from the image.
In step S304, the low resolution character image collation module 130 collates the character image 128 in the image with the low resolution character image 122 stored in the multi-resolution information storage module 120.
In step S306, the high resolution information adding module 140 acquires the high resolution information 124 for the character image 128 collated in step S304 from the multiple resolution information storage module 120.
In step S308, the high resolution information adding module 140 adds the high resolution information 124 to the character image 128 and outputs it.

<First specific embodiment>
FIG. 4 is a conceptual module configuration diagram of a configuration example of the first specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied.

  The configuration example shown in FIG. 4 includes a dictionary document generation module 410, a low resolution image multiple reading module 420, a character segmentation module 430, and a multiple resolution information storage module 120-4. The dictionary document generation module 410, the low resolution image multiple reading module 420, and the character segmentation module 430 correspond to a specific configuration example of the multiple resolution information generation module 110 illustrated in FIG. 1, and the multiple resolution information storage module 120-4. Corresponds to the multi-resolution information storage module 120 shown in FIG.

The dictionary document generation module 410 generates and outputs an image including a plurality of types or sizes of pixel clusters. The output here corresponds to printing on a medium such as paper.
The low resolution image multiple reading module 420 reads the image output by the dictionary document generation module 410 multiple times. The reading here corresponds to, for example, image reading with a scanner.
The character cutout module 430 cuts out an image of a pixel block in the image read by the low resolution image multiple reading module 420.
The multi-resolution information storage module 120-4 stores information regarding the pixel block with respect to the pixel block image cut out by the character cutout module 430.

As a more specific example, each module will be described.
The dictionary document generation module 410, the low resolution image multiple reading module 420, and the multiple resolution information storage module 120-4 are connected, and the dictionary document generation module 410 passes the dictionary document 412 to the low resolution image multiple reading module 420. In addition, the character code / font information 414 is passed to the multi-resolution information storage module 120-4. The dictionary document generation module 410 generates a dictionary document 412 in which a document including characters such as hiragana, katakana, alphanumeric characters, and common kanji is printed with a font type and font size expected to be generally used. Is sent to the low resolution image multiple reading module 420, and the character code / font information 414 (including font type and font size as font information) of each dictionary document 412 is sent to the multiple resolution information storage module 120-4. . The character code / font information 414 is specified when the dictionary document 412 is generated.

  The low resolution image multiple reading module 420, the character segmentation module 430, and the multiple resolution information storage module 120-4 are connected, and the low resolution image multiple scan module 420 passes the low resolution image 422 to the character segmentation module 430. Also, the resolution information 424 is transferred to the multi-resolution information storage module 120-4. The low-resolution image multiple reading module 420 receives the resolution of the character image 128 received by the low-resolution character image collation module 130 shown in the example of FIG. 1 (for example, about 200 dpi × 100 dpi in the standard mode of fax, 200 dpi × 200 dpi in the scanner). In order to generate a plurality of low-resolution images 422 with different resolutions at a resolution equivalent to the reading resolution possible for the character image 128, a plurality of readings are performed, and the read low-resolution image 422 is converted into a character segmentation module 430. The resolution information 424 at the time of reading is sent to the multi-resolution information storage module 120-4.

  The character segmentation module 430 and the multi-resolution information storage module 120-4 are connected, and the character segmentation module 430 passes the one-character image 432 to the multi-resolution information storage module 120-4. The character cutout module 430 cuts out characters one by one by image processing from characters arranged in a grid pattern, or cuts out a character image based on an operation in which an operator manually specifies a cutout rectangle, and the cutout single character image 432 is sent to the multi-resolution information storage module 120-4.

The multi-resolution information storage module 120-4 links a plurality of low-resolution one-character images 432 and character code / font information 414 to one character, and generates and stores multi-resolution information. The multi-resolution information includes a multi-resolution information table 500 shown in the example of FIG. 5 or a multi-resolution information table 600 shown in the example of FIG.
FIG. 5 is an explanatory diagram showing an example of the data structure of the multi-resolution information table 500. The multi-resolution information table 500 has a collation ID column 510, a low-resolution character image column 520, and a high-resolution information (character code) column 530. The collation ID column 510 stores a code that can uniquely identify a character, and the low resolution character image column 520 is an image obtained by reading the character at a low resolution. The character image of the phase is stored, and the high resolution information (character code) column 530 stores the character code of the character image as high resolution information. The resolution of the character image stored in the low resolution character image column 520 is equivalent to the resolution of the character image 128 received by the low resolution character image collation module 130 shown in the example of FIG.

  FIG. 6 is an explanatory diagram showing an example of the data structure of the multi-resolution information table 600. The multi-resolution information table 600 includes a collation ID column 610, a low resolution character image column 620, a high resolution information (character code) column 630, a high resolution information (font type) column 640, and a high resolution information (font size) column 650. doing. The collation ID column 610 stores a code that can uniquely identify a character (a combination of a character code, a font type, and a font size). The low resolution character image column 620 and the high resolution information (character code) column 630 are equivalent to the low resolution character image column 520 and the high resolution information (character code) column 530 of the multi-resolution information table 500 shown in the example of FIG. The high resolution information (font type) column 640 stores the font type of the character (for example, Gothic), and the high resolution information (font size) column 650 stores the font size of the character. In other words, in addition to the information stored in the high resolution information (character code) column 630 as the high resolution information, there are those stored in the high resolution information (font type) column 640 and the high resolution information (font size) column 650. .

  There are two types of the multi-resolution information table 500 and the multi-resolution information table 600, that is, high-resolution information according to the embodiment according to the purpose. The multi-resolution information table 500 is used in the configuration example shown in the example of FIG. 7, and the multi-resolution information table 600 is used in the configuration example shown in the example of FIG.

  FIG. 7 is a module configuration diagram of an overall configuration example (1) for realizing the first specific embodiment. This receives character document 700-A and generates character document 700-B and character document 700-C having a layer structure by rasterizing character codes with a transparent color. For example, in the PDF format, a character code is assigned as a transparent layer (character document 700-C) using a system font. Therefore, the dictionary document generation module 410-7 passes the character code 414-7 to the multiple resolution information storage module 120-7, and the multiple resolution information storage module 120-7 receives the multiple resolution information table 500 shown in the example of FIG. The high resolution information adding module 140-7 assigns a character code as high resolution information.

  7 includes a dictionary document generation module 410-7, a low resolution image multiple reading module 420-7, a character segmentation module 430-7, a multiple resolution information storage module 120-7, an image reading module 710, It has a character segmentation module 720, a low resolution character image collation module 130-7, a high resolution information addition module 140-7, and a formatter module 730.

The dictionary document generation module 410-7, the low resolution image multiple reading module 420-7, and the multiple resolution information storage module 120-7 are connected, and the dictionary document generation module 410-7 includes the low resolution image multiple reading module 420. The dictionary document 412-7 is passed to -7, and the character code 414-7 is passed to the multi-resolution information storage module 120-7.
The low resolution image multiple reading module 420-7, the character cutout module 430-7, and the multiple resolution information storage module 120-7 are connected, and the low resolution image multiple reading module 420-7 is connected to the character cutout module 430-7. The low-resolution image 422-7 is passed to the multi-resolution information storage module 120-7, and the resolution information 424-7 is passed to the multi-resolution information storage module 120-7.
The character segmentation module 430-7 and the multiple resolution information storage module 120-7 are connected, and the character segmentation module 430-7 passes the one-character image 432-7 to the multiple resolution information storage module 120-7.
The multi-resolution information storage module 120-7, the low-resolution character image collation module 130-7, and the high-resolution information adding module 140-7 are connected. The multi-resolution information storage module 120-7 is connected to the low-resolution character image collation module 130. The low resolution character image 122-7 is passed to -7, and the character code 124-7 is passed to the high resolution information adding module 140-7.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 700-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low-resolution character image collation module 130-7, and the formatter module 730 are connected. The character segmentation module 720 segments the character image 712 for each character, and extracts the position of the segment (for example, a character document). The X and Y coordinates in the upper left of the one-character image 722 that is a rectangle in 700-A are extracted. A one-character image 722 is passed to the low-resolution character image collation module 130-7, and a one-character image 722 and character position coordinates 724 are passed to the formatter module 730. Note that one character image 722 and character position coordinates 724 correspond to each other.

  The low resolution character image collation module 130-7 and the high resolution information provision module 140-7 are connected, and the low resolution character image collation module 130-7 passes the collation ID 132-7 to the high resolution information provision module 140-7. . One character image 722 is collated with the low resolution character image (character image stored in the low resolution character image column 520 in the multiple resolution information table 500) from the multiple resolution information storage module 120-7. If the collation ID 132-7 (the code stored in the collation ID column 510 in the multi-resolution information table 500 corresponding to the collated character image) does not match, the collation ID 132-7 is set to 0, and the high resolution information adding module 140- Pass to 7.

  The high resolution information adding module 140-7, the multiple resolution information storage module 120-7, and the formatter module 730 are connected, and the high resolution information adding module 140-7 sends the collation ID 142-7 to the multiple resolution information storage module 120-7. And the character code 144-7 is passed to the formatter module 730. When the collation ID 132-7 matches (when the collation ID 132-7 is other than 0), the collation ID 142-7 as the collation ID 132-7 is sent to the multi-resolution information storage module 120-7, and the corresponding character code 124 is sent. −7 is obtained and sent to the formatter module 730 as a character code 144-7 corresponding to the one-character image 722. If they do not match (when the collation ID 132-7 is 0), the character code 144 corresponding to the one-character image 722 is obtained. As 0, 0 is sent to the formatter module 730.

  The formatter module 730 outputs the character document 700-B and the character document 700-C. If the character code 144-7 from the high resolution information adding module 140-7 is other than 0, the character code is rasterized, and the one character image 722 is arranged based on the character position coordinates 724 received from the character cutout module 720. For example, it is converted into a PDF format, and the character code is attached as a transparent layer (character document 700-C) with a system font and output. If the character code 144-7 is 0, the character image 712 is converted into a character position coordinate. Based on 724, for example, it is converted into a PDF format and output.

  FIG. 8 is a module configuration diagram of an overall configuration example (2) for realizing the first specific embodiment. This accepts the character document 800-A and generates a character document 800-B, which is an electronic document in a format created by a document creation device (such as word processing software). That is, the character document 800-A, which is an image, is converted into a character code to generate a character document 800-B. At that time, the font information is also equivalent to that of the character document 800-A. For this purpose, the dictionary document generation module 410-8 passes the character code / font information 414-8 to the multi-resolution information storage module 120-8, and the multi-resolution information storage module 120-8 receives the multi-resolution information shown in the example of FIG. The information table 600 is stored, and the high resolution information adding module 140-8 assigns a character code and font information as high resolution information.

  The overall configuration example shown in FIG. 8 includes a dictionary document generation module 410-8, a low resolution image multiple reading module 420-8, a character segmentation module 430-8, a multiple resolution information storage module 120-8, an image reading module 710, It has a character segmentation module 720, a low resolution character image collation module 130-8, a high resolution information addition module 140-8, and a documentization module 830.

The dictionary document generation module 410-8 is connected to the low resolution image multiple reading module 420-8 and the multiple resolution information storage module 120-8. The dictionary document generation module 410-8 includes the low resolution image multiple reading module 420. The dictionary document 412-8 is passed to -8, and the character code / font information 414-8 is passed to the multi-resolution information storage module 120-8.
The low resolution image multiple reading module 420-8 is connected to the character cutout module 430-8 and the multiple resolution information storage module 120-8. The low resolution image multiple reading module 420-8 is connected to the character cutout module 430-8. The low-resolution image 422-8 is passed to the multi-resolution information storage module 120-8.
The character segmentation module 430-8 and the multi-resolution information storage module 120-8 are connected, and the character segmentation module 430-8 passes the one-character image 432-8 to the multi-resolution information storage module 120-8.
The multi-resolution information storage module 120-8 is connected to the low-resolution character image collation module 130-8 and the high-resolution information adding module 140-8. The multi-resolution information storage module 120-8 is connected to the low-resolution character image collation module 130. The low resolution character image 122-8 is passed to -8, and the character code / font information 124-8 is passed to the high resolution information adding module 140-8.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 800-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-8, and the documentation module 830 are connected, and the character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-8. A one-character image 722 and character position coordinates 724 are passed to the documentation module 830. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-8 and the high resolution information addition module 140-8 are connected, and the low resolution character image collation module 130-8 passes the collation ID 132-8 to the high resolution information addition module 140-8. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

  The high-resolution information adding module 140-8, the multi-resolution information storing module 120-8, and the documenting module 830 are connected, and the high-resolution information adding module 140-8 sends the collation ID 142- to the multi-resolution information storing module 120-8. 8, and the character code + font information 144-8 is passed to the documentation module 830. When the collation ID 132-8 matches (when the collation ID 132-8 is other than 0), the collation ID 132-8, which is the collation ID 132-8, is sent to the multi-resolution information storage module 120-8, and the corresponding character code / The font information 124-8 is obtained and sent to the documenting module 830 as character code + font information 144-8 corresponding to the one-character image 722. If they do not match (if the collation ID 132-8 is 0), the one-character image 0 is sent to the documenting module 830 as the character code corresponding to 722 + the character code of the font information 144-8.

  The documenting module 830 outputs the character document 800-B. If the character code of the character code + font information 144-8 from the high-resolution information adding module 140-8 is other than 0, the character code is rasterized based on the font information, and a one-character image 722 is received from the character segmentation module 720. If the character code of the character code + font information 144-8 is 0 based on the character position coordinates 724, for example, it is converted into a character document 800-B, which is an electronic document generated by word processing software. Based on the character position coordinates 724, the one-character image 722 of 712 is converted into a font (for example, the one-character image 722 can be specified by a special character code), and converted into an electronic document and output.

<Second specific embodiment>
FIG. 9 is a conceptual module configuration diagram of a configuration example of the second specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied.

  The configuration example shown in FIG. 9 includes a dictionary document generation module 410-9, a high resolution image reading module 920, a low resolution image multiple reading module 420-9, a character segmentation module 930, a character segmentation module 430-9, and multiple resolution information storage. It has a module 120-9. The dictionary document generation module 410-9, the high resolution image reading module 920, the low resolution image multiple reading module 420-9, the character segmentation module 930, and the character segmentation module 430-9 are the multiple resolution information generation module 110 shown in FIG. The multiple resolution information storage module 120-9 corresponds to the multiple resolution information storage module 120 shown in FIG.

The dictionary document generation module 410-9 outputs an image including images of pixel clusters of a plurality of types or sizes. This is equivalent to the dictionary document generation module 410 shown in the example of FIG.
The high resolution image reading module 920 reads the image output by the dictionary document generation module 410-9. The low resolution image multiple reading module 420-9 reads the image output by the dictionary document generation module 410-9 multiple times. Read at a lower resolution than the high resolution image reading module 920.
The images read by the high resolution image reading module 920 and the low resolution image multiple reading module 420-9 are the same image. This image is output by the dictionary document generation module 410-9, and this output image is used as the dictionary document 412-9 for the present embodiment. Further, the high resolution image reading module 920 reads at a higher resolution than the character image 128 received by the low resolution character image matching module 130 shown in the example of FIG.
The character cutout module 930 cuts out an image of a pixel block in the image read by the high resolution image reading module 920.
The character cutout module 430-9 cuts out an image of a pixel block in the image read by the low resolution image multiple reading module 420-9.
The multi-resolution information storage module 120-9 stores the pixel block image cut out by the character cutout module 930 as information related to the pixel block cut out by the character cutout module 430-9.

As a more specific example, each module will be described.
The dictionary document generation module 410-9, the high resolution image reading module 920, the low resolution image multiple reading module 420-9, and the multiple resolution information storage module 120-9 are connected. The dictionary document 412-9 is passed to the resolution image reading module 920 and the low resolution image multiple reading module 420-9, and the character code / font information 414-9 is passed to the multiple resolution information storage module 120-9. This is equivalent to the dictionary document generation module 410 shown in the example of FIG.

  The high resolution image reading module 920 and the character cutout module 930 are connected, and the high resolution image reading module 920 passes the high resolution image 922 to the character cutout module 930. The high resolution image reading module 920 has a predetermined number of times (for example, a high resolution of 600 dpi × 600 dpi) (for example, a high resolution of 600 dpi × 600 dpi) than the resolution of the character image 128 received by the low resolution character image matching module 130 shown in the example of FIG. Once) The image is read by the scanner, and the read high resolution image 922 is sent to the character segmentation module 930.

  The character segmentation module 930 and the multi-resolution information storage module 120-9 are connected, and the character segmentation module 930 passes the one-character image 932 to the multi-resolution information storage module 120-9. This is equivalent to the character segmentation module 430 shown in the example of FIG.

  The low resolution image multiple reading module 420-9, the character cutout module 430-9, and the multiple resolution information storage module 120-9 are connected, and the low resolution image multiple read module 420-9 is connected to the character cutout module 430-9. The low resolution image 422-9 is transferred to the multi-resolution information storage module 120-9. This is equivalent to the low resolution image multiple reading module 420 shown in the example of FIG.

  The character cutout module 430-9 and the multiple resolution information storage module 120-9 are connected, and the character cutout module 430-9 passes the single character image 432-9 to the multiple resolution information storage module 120-9. This is equivalent to the character segmentation module 430 shown in the example of FIG.

  The multi-resolution information storage module 120-9 links the high-resolution one-character image 932, the plurality of low-resolution one-character images 432-9, and the character code / font information 414-9 with respect to one character. Generate and accumulate as. The multi-resolution information includes a multi-resolution information table 1000 shown in the example of FIG. 10, a multi-resolution information table 1100 shown in the example of FIG. 11, or a multi-resolution information table 1200 shown in the example of FIG.

  FIG. 10 is an explanatory diagram showing an example of the data structure of the multi-resolution information table 1000. The multi-resolution information table 1000 includes a collation ID column 1010, a low resolution character image column 1020, and a high resolution information (high resolution image) column 1030. The collation ID column 1010 and the low resolution character image column 1020 are equivalent to the collation ID column 510 and the low resolution character image column 520 of the multi-resolution information table 500 shown in the example of FIG. The high resolution information (high resolution image) column 1030 stores the high resolution single character image 932 received from the character segmentation module 930 as high resolution information.

  FIG. 11 is an explanatory diagram showing an example of the data structure of the multi-resolution information table 1100. The multi-resolution information table 1100 includes a collation ID column 1110, a low resolution character image column 1120, a high resolution information (high resolution image) column 1130, and a high resolution information (character code) column 1140. The collation ID column 1110, the low resolution character image column 1120, and the high resolution information (high resolution image) column 1130 are the collation ID column 1010, the low resolution character image column 1020, and the high resolution of the multi-resolution information table 1000 shown in the example of FIG. This is equivalent to the information (high resolution image) column 1030. The high resolution information (character code) column 1140 stores the character code of the character image as high resolution information. That is, a high resolution character image and character code are stored as high resolution information.

FIG. 12 is an explanatory diagram showing an example of the data structure of the multi-resolution information table 1200. The multi-resolution information table 1200 includes a collation ID column 1210, a low resolution character image column 1220, a high resolution information (high resolution image) column 1230, a high resolution information (character code) column 1240, a high resolution information (font type) column 1250, A high resolution information (font size) column 1260 is provided. The collation ID column 1210, the low resolution character image column 1220, the high resolution information (high resolution image) column 1230, and the high resolution information (character code) column 1240 are the collation ID column 1110 of the multi-resolution information table 1100 shown in the example of FIG. , Low resolution character image column 1120, high resolution information (high resolution image) column 1130, high resolution information (character code) column 1140, high resolution information (font type) column 1250, high resolution information (font size) A column 1260 is equivalent to the high resolution information (font type) column 640 and the high resolution information (font size) column 650 of the multi-resolution information table 600 shown in the example of FIG. That is, a high-resolution character image, character code, font type, and font size are stored as high-resolution information.
In addition, the multi-resolution information stored in the multi-resolution information storage module 120-9 is the multi-resolution information table 600 shown in the example of FIG. ). In particular, this may be used as an alternative to the multi-resolution information table 1200.

  The multi-resolution information table 1000, the multi-resolution information table 1100, and the multi-resolution information table 1200 are selectively used, that is, the high resolution information has the following three types according to the embodiment according to the purpose. The multi-resolution information table 1000 is used for the configuration example shown in the example of FIG. 13, the multi-resolution information table 1100 is used for the configuration example shown in the example of FIG. 14, and the multi-resolution information table 1200 is the configuration example shown in the example of FIG. Used for.

  FIG. 13 is a module configuration diagram of an overall configuration example (1) for realizing the second specific embodiment. This accepts the character document 1300-A and generates a character document 1300-B which is a higher resolution document. That is, the character document 1300-B, which is a relatively low resolution image, is generated at a high resolution to generate the character document 1300-B. For this purpose, the multi-resolution information storage module 120-13 stores the multi-resolution information table 1000 shown in the example of FIG. 10, and the high-resolution information adding module 140-13 adds a high-resolution character image as high-resolution information. Yes.

  The overall configuration example shown in FIG. 13 includes a dictionary document generation module 410-13, a high resolution image reading module 920-13, a low resolution image multiple reading module 420-13, a character segmentation module 930-13, and a character segmentation module 430-. 13, a multi-resolution information storage module 120-13, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-13, a high-resolution information adding module 140-13, and a high-resolution documenting module 1330. Yes.

The dictionary document generation module 410-13, the high resolution image reading module 920-13, the low resolution image multiple reading module 420-13, and the multiple resolution information storage module 120-13 are connected, and the dictionary document generation module 410-13 is connected. The dictionary document 412-13 is passed to the high resolution image reading module 920-13 and the low resolution image multiple reading module 420-13, and the character code / font information 414-13 is passed to the multiple resolution information storage module 120-13. .
The high resolution image reading module 920-13 and the character cutout module 930-13 are connected, and the high resolution image reading module 920-13 passes the high resolution image 922-13 to the character cutout module 930-13.
The character segmentation module 930-13 and the multiple resolution information storage module 120-13 are connected, and the character segmentation module 930-13 passes the one-character image 932-13 to the multiple resolution information accumulation module 120-13.
The low resolution image multiple reading module 420-13, the character cutout module 430-13, and the multiple resolution information storage module 120-13 are connected, and the low resolution image multiple reading module 420-13 is connected to the character cutout module 430-13. The low resolution image 422-13 is transferred to the multi-resolution information storage module 120-13.
The character segmentation module 430-13 and the multiple resolution information storage module 120-13 are connected, and the character segmentation module 430-13 passes the one character image 432-13 to the multiple resolution information accumulation module 120-13.
The multi-resolution information storage module 120-13, the low-resolution character image collation module 130-13, and the high-resolution information adding module 140-13 are connected, and the multi-resolution information storage module 120-13 is connected to the low-resolution character image collation module 130. The low-resolution character image 122-13 is passed to -13, and the high-resolution character image 124-13 is passed to the high-resolution information adding module 140-13.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 1300-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-13, and the high resolution document creation module 1330 are connected. The character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-13. In addition, a one-character image 722 and character position coordinates 724 are passed to the high-resolution documenting module 1330. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-13 and the high resolution information provision module 140-13 are connected, and the low resolution character image collation module 130-13 passes the collation ID 132-13 to the high resolution information provision module 140-13. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

  The high resolution information adding module 140-13, the multiple resolution information storage module 120-13, and the high resolution documenting module 1330 are connected, and the high resolution information adding module 140-13 collates with the multiple resolution information storage module 120-13. The ID 142-13 is passed, and the high resolution character image 144-13 is passed to the high resolution documenting module 1330. When the collation ID 132-13 matches (when the collation ID 132-13 is other than 0), the collation ID 132-13, which is the collation ID 132-13, is sent to the multi-resolution information storage module 120-13, and the corresponding high-resolution character is displayed. An image 124-13 is obtained and sent to the high-resolution documenting module 1330 as a high-resolution character image 144-13 corresponding to the one-character image 722. If they do not match (if the collation ID 132-13 is 0), a one-character image 0 is sent to the high resolution documenting module 1330 as the high resolution character image 144-13 corresponding to 722.

  The high resolution documenting module 1330 outputs the character document 1300-B. If the high resolution character image 144-13 from the high resolution information adding module 140-13 is other than 0, the high resolution character image 144-13 is arranged based on the character position coordinates 724 received from the character cutout module 720 (that is, If the high-resolution character image 144-13 is 0) instead of the one-character image 722 and the high-resolution character image 144-13 is 0, the one-character image 722 is based on the character position coordinates 724 received from the character segmentation module 720. Arrangement (that is, the same as the one-character image 722 of the character document 1300-A) is output.

  FIG. 14 is a module configuration diagram of an overall configuration example (2) for realizing the second specific embodiment. This receives character document 1400-A and generates character document 1400-B and character document 1400-C having a layer structure by rasterizing character codes with transparent colors. For example, in the PDF format, a character code is assigned as a transparent layer (character document 1400-C) using a system font. For this purpose, the dictionary document generation module 410-14 passes the character code / font information 414-14 to the multi-resolution information storage module 120-14, and the multi-resolution information storage module 120-14 receives the multi-resolution shown in the example of FIG. The information table 1100 is stored, and the high resolution information adding module 140-14 adds a high resolution character image and a character code as high resolution information.

  The overall configuration example shown in FIG. 14 includes a dictionary document generation module 410-14, a high resolution image reading module 920-14, a low resolution image multiple reading module 420-14, a character segmentation module 930-14, and a character segmentation module 430-. 14, a multi-resolution information storage module 120-14, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-14, a high-resolution information addition module 140-14, and a formatter module 730.

The dictionary document generation module 410-14, the high resolution image reading module 920-14, the low resolution image multiple reading module 420-14, and the multiple resolution information storage module 120-14 are connected, and the dictionary document generation module 410-14 is connected. The dictionary document 412-14 is passed to the high resolution image reading module 920-14 and the low resolution image multiple reading module 420-14, and the character code / font information 414-14 is passed to the multiple resolution information storage module 120-14. .
The high resolution image reading module 920-14 and the character cutout module 930-14 are connected, and the high resolution image reading module 920-14 passes the high resolution image 922-14 to the character cutout module 930-14.
The character segmentation module 930-14 and the multiple resolution information storage module 120-14 are connected, and the character segmentation module 930-14 passes the single character image 932-14 to the multiple resolution information storage module 120-14.
The low resolution image multiple reading module 420-14 is connected to the character segmentation module 430-14 and the multiple resolution information storage module 120-14. The low resolution image multiple scan module 420-14 is connected to the character segmentation module 430-14. The low-resolution image 422-14 is transferred to the multi-resolution information storage module 120-14.
The character cutout module 430-14 and the multiple resolution information storage module 120-14 are connected, and the character cutout module 430-14 passes the single character image 432-14 to the multiple resolution information storage module 120-14.
The multi-resolution information storage module 120-14, the low-resolution character image collation module 130-14, and the high-resolution information addition module 140-14 are connected. The multi-resolution information storage module 120-14 is connected to the low-resolution character image collation module 130. The low resolution character image 122-14 is passed to −14, and the high resolution character image + character code 124-14 is passed to the high resolution information adding module 140-14.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 1400 -A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-14, and the formatter module 730 are connected. The character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-14, and the formatter. A one-character image 722 and character position coordinates 724 are passed to the module 730. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image matching module 130-14 and the high resolution information adding module 140-14 are connected, and the low resolution character image matching module 130-14 sends the matching result 132-14 to the high resolution information adding module 140-14. hand over. This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

  The high resolution information adding module 140-14, the multiple resolution information storage module 120-14, and the formatter module 730 are connected, and the high resolution information adding module 140-14 sends the collation ID 142-14 to the multiple resolution information storage module 120-14. And the high-resolution character image + character code 144-14 is passed to the formatter module 730. If the collation results 132-14 match (if the collation result 132-14 is other than 0), the collation ID 142-14 that is the collation result 132-14 is sent to the multi-resolution information storage module 120-14, and the corresponding result is obtained. A high-resolution character image + character code 124-14 is obtained and sent to the formatter module 730 as a high-resolution character image + character code 144-14 corresponding to one character image 722. If the characters do not match (the collation result 132-14 is 0) In this case, 0 is transmitted to the formatter module 730 as a high resolution character image + character code 144-14 corresponding to the one character image 722.

  The formatter module 730 outputs the character document 1400-B and the character document 1400-C. If the high-resolution character image + character code 144-14 from the high-resolution information adding module 140-14 is other than 0, the character received from the character segmentation module 720 is a high-resolution character image + character code 144-14. Arranged based on the position coordinates 724, converted to PDF format, for example, and the character code is attached and output as a transparent layer (character document 1400-C) in the system font, and the high resolution character image + character code 144-14 is 0 Outputs one character image 722 of the character image 712, for example, in PDF format based on the character position coordinates 724.

  FIG. 15 is a module configuration diagram of an overall configuration example (3) for realizing the second specific embodiment. This accepts the character document 1500-A and generates a character document 1500-B, which is an electronic document in a format created by a document creation device (such as word processing software). That is, a character document 1500-B is generated by character-coding the character document 1500-A, which is an image. At that time, the font information is also equivalent to that of the character document 1500-A. Therefore, the dictionary document generation module 410-15 passes the character code / font information 414-15 to the multi-resolution information storage module 120-15, and the multi-resolution information storage module 120-15 uses the multi-resolution shown in the example of FIG. The information table 1200 (or the multiple resolution information table 600 shown in the example of FIG. 6) is stored, and the high resolution information adding module 140-15 assigns character code and font information as high resolution information.

  15 includes a dictionary document generation module 410-15, a high resolution image reading module 920-15, a low resolution image multiple reading module 420-15, a character segmentation module 930-15, and a character segmentation module 430-. 15, a multi-resolution information storage module 120-15, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-15, a high-resolution information adding module 140-15, and a documenting module 830.

The dictionary document generation module 410-15, the high resolution image reading module 920-15, the low resolution image multiple reading module 420-15, and the multiple resolution information storage module 120-15 are connected, and the dictionary document generation module 410-15 is connected. The dictionary document 412-15 is passed to the high resolution image reading module 920-15 and the low resolution image multiple reading module 420-15, and the character code / font information 414-15 is passed to the multiple resolution information storage module 120-15. .
The high resolution image reading module 920-15 and the character cutout module 930-15 are connected, and the high resolution image reading module 920-15 passes the high resolution image 922-15 to the character cutout module 930-15.
The character segmentation module 930-15 and the multiple resolution information storage module 120-15 are connected, and the character segmentation module 930-15 passes the one-character image 932-15 to the multiple resolution information accumulation module 120-15.
The low resolution image multiple reading module 420-15, the character cutout module 430-15, and the multiple resolution information storage module 120-15 are connected. The low resolution image multiple read module 420-15 is connected to the character cutout module 430-15. The low-resolution image 422-15 is transferred to the multi-resolution information storage module 120-15.
The character segmentation module 430-15 and the multiple resolution information storage module 120-15 are connected, and the character segmentation module 430-15 passes the one character image 432-15 to the multiple resolution information accumulation module 120-15.
The multi-resolution information storage module 120-15, the low-resolution character image collation module 130-15, and the high-resolution information addition module 140-15 are connected. The multi-resolution information storage module 120-15 is connected to the low-resolution character image collation module 130. The low resolution character image 122-15 is passed to -15, and the character code + font information 124-15 is passed to the high resolution information adding module 140-15.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 1500-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-15, and the documentation module 830 are connected, and the character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-15. A one-character image 722 and character position coordinates 724 are passed to the documentation module 830. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-15 and the high resolution information addition module 140-15 are connected, and the low resolution character image collation module 130-15 sends the collation result 132-15 to the high resolution information addition module 140-15. hand over. This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

  The high-resolution information adding module 140-15, the multi-resolution information storage module 120-15, and the documentation module 830 are connected, and the high-resolution information adding module 140-15 sends the collation ID 142- to the multi-resolution information storage module 120-15. 15, and the character code + font information 144-15 is passed to the documentation module 830. When the collation result 132-15 matches (when the collation result 132-15 is other than 0), the collation ID 142-15 that is the collation result 132-15 is sent to the multi-resolution information storage module 120-15, and correspondingly. When the character code + font information 124-15 is obtained and sent to the documenting module 830 as the character code + font information 144-15 corresponding to the one-character image 722, if they do not match (when the collation result 132-15 is 0) Sends 0 to the documenting module 830 as the character code corresponding to the one-character image 722 + the character code of the font information 144-15.

  The documenting module 830 outputs the character document 1500-B. If the character code of the character code + font information 144-15 from the high-resolution information adding module 140-15 is other than 0, the character code is rasterized based on the font information, and a one-character image 722 is received from the character segmentation module 720. If the character code of the character code + font information 144-15 is 0 based on the character position coordinates 724, for example, it is converted into a character document 1500-B which is an electronic document generated by word processing software. Based on the character position coordinates 724, the one-character image 722 of 712 is converted into a font (for example, the one-character image 722 can be specified by a special character code), and converted into an electronic document and output.

<Third specific embodiment>
FIG. 16 is a conceptual module configuration diagram of a configuration example according to the third specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied.

  The configuration example shown in FIG. 16 includes a user document reception module 1610, a high resolution image reading module 1620, a character segmentation module 1630, a low resolution image multiple reading module 1640, a character segmentation module 1650, a character recognition processing module 1660, and multiple resolution information storage. It has a module 120-16. The user document reception module 1610, the high resolution image reading module 1620, the character cutout module 1630, the low resolution image multiple reading module 1640, the character cutout module 1650, and the character recognition processing module 1660 are the multiresolution information generation module 110 shown in FIG. The multi-resolution information storage module 120-16 corresponds to the multi-resolution information storage module 120 shown in FIG.

The user document reception module 1610 passes the user document 1612 to the high resolution image reading module 1620, the low resolution image multiple reading module 1640, and the character recognition processing module 1660.
The high resolution image reading module 1620 reads the user document 1612 passed from the user document reception module 1610.
The character cutout module 1630 cuts out an image of a pixel block in the image read by the high resolution image reading module 1620.
The low resolution image multiple reading module 1640 reads the user document 1612 passed from the user document reception module 1610 multiple times at a lower resolution than the high resolution image reading module 1620.
The character cutout module 1650 cuts out an image of a pixel block in the image read by the low resolution image multiple reading module 1640.

The character recognition processing module 1660 recognizes a pixel block image in the user document 1612 delivered from the user document reception module 1610. The character recognition processing module 1660 has a scanner as a reading unit, and recognizes after reading the user document 1612. The character recognition processing module 1660 may receive and recognize the high-resolution image 1622 from the high-resolution image reading module 1620, or may receive and recognize the one-character image 1632 from the character cutout module 1630.
The image read by the high resolution image reading module 1620 and the low resolution image multiple reading module 1640 and the image to be recognized by the character recognition processing module 1660 are the same image. This image is a document actually used by the user.
The multi-resolution information storage module 120-16 stores the pixel block image cut out by the character cutout module 1630 as information related to the pixel block cut out by the character cutout module 1650. Further, the multi-resolution information storage module 120-16 may store the recognition result by the character recognition processing module 1660 as information regarding the pixel block for the pixel block image cut out by the character cutout module 1650.

As a more specific example, each module will be described.
The user document reception module 1610, the high resolution image reading module 1620, the low resolution image multiple reading module 1640, and the character recognition processing module 1660 are connected. The user document reception module 1610 includes the high resolution image reading module 1620, the low resolution image. The user document 1612 is passed to the multiple reading module 1640 and the character recognition processing module 1660. For example, the user document 1612 frequently used by the user is sent to the high resolution image reading module 1620, the low resolution image multiple reading module 1640, and the character recognition processing module 1660.

  The high resolution image reading module 1620 and the character cutout module 1630 are connected, and the high resolution image reading module 1620 passes the high resolution image 1622 to the character cutout module 1630. This is equivalent to the high resolution image reading module 920 shown in the example of FIG.

  The character cutout module 1630 and the multiple resolution information storage module 120-16 are connected, and the character cutout module 1630 passes the single character image 1632 to the multiple resolution information storage module 120-16. This is equivalent to the character segmentation module 930 shown in the example of FIG.

  The low resolution image multiple reading module 1640 is connected to the character segmentation module 1650 and the multiple resolution information storage module 120-16, and the low resolution image multiple scan module 1640 passes the low resolution image 1642 to the character segmentation module 1650, Also, the resolution information 1644 is passed to the multi-resolution information storage module 120-16. This is equivalent to the low-resolution image multiple reading module 420-9 shown in the example of FIG.

  The character cutout module 1650 and the multiple resolution information storage module 120-16 are connected, and the character cutout module 1650 passes the single character image 1652 to the multiple resolution information storage module 120-16. This is equivalent to the character segmentation module 430-9 shown in the example of FIG.

  The character recognition processing module 1660 and the multi-resolution information storage module 120-16 are connected, and the character recognition processing module 1660 passes the character code 1662 to the multi-resolution information storage module 120-16. The character recognition processing module 1660 recognizes a user document 1612 (which may be a high-resolution image 1622, or a single character image 1632) as a target image. The existing method is used as the character recognition method. Also, resolution conversion processing or the like may be performed as preprocessing in accordance with the character recognition processing. Further, the character that cannot be recognized may not be stored in the multi-resolution information storage module 120-16. In that case, the multi-resolution information about the character is lost.

The multi-resolution information storage module 120-16 links a high-resolution single character image 1632, a plurality of low-resolution single character images 1652, and a character code 1662 to generate and store multi-resolution information. The multi-resolution information includes a multi-resolution information table 1000 shown in the example of FIG. 10, a multi-resolution information table 1100 shown in the example of FIG. 11, or a multi-resolution information table 1200 shown in the example of FIG.
Further, the entire configuration example for realizing the third specific embodiment is equivalent to that shown in FIGS. 13, 14, and 15. However, dictionary document generation modules 410-13, 14, and 15, high-resolution image reading modules 920-13, 14, and 15, character segmentation modules 930-13, 14, and 15, low-resolution image multiple-reading modules 420-13 and 14 15, the character segmentation modules 430-13, 14, 15 and the multi-resolution information storage modules 120-13, 14, 15 are replaced with the configuration shown in FIG. 16.

<Fourth Specific Embodiment>
FIG. 17 is a conceptual module configuration diagram of a configuration example of the fourth specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied.

  The configuration example shown in FIG. 17 includes a dictionary document generation module 410-17, a high resolution image reading module 920-17, a low resolution multi-phase scan processing module 1710, a character segmentation module 930-17, a character segmentation module 430-17, and multiple resolutions. An information storage module 120-17 is included. The dictionary document generation module 410-17, the high resolution image reading module 920-17, the low resolution multiphase scan processing module 1710, the character segmentation module 930-17, and the character segmentation module 430-17 generate the multi-resolution information shown in FIG. This corresponds to a specific configuration example of the module 110, and the multi-resolution information storage module 120-17 corresponds to the multi-resolution information storage module 120 illustrated in FIG.

The dictionary document generation module 410-17 outputs an image including images of pixel clusters of a plurality of types or sizes.
The high resolution image reading module 920-17 reads the image output by the dictionary document generation module 410-17. The resolution read by the high resolution image reading module 920-17 is higher than that of the character image 128 received by the low resolution character image collation module 130 shown in the example of FIG.
The low resolution multi-phase scan processing module 1710 lowers the resolution of the image read by the high resolution image reading module 920-17 and generates an image having a different phase. Here, lowering the resolution of the image means that the resolution is equivalent to that of the character image 128 received by the low-resolution character image collation module 130 shown in the example of FIG.
The character cutout module 930-17 cuts out an image of a pixel block in the image read by the high resolution image reading module 920-17.
The character cutout module 430-17 cuts out an image of a pixel block in the image generated by the low resolution multiphase scan processing module 1710.
The multi-resolution information storage module 120-17 stores the pixel block image cut out by the character cutout module 930-17 as information on the pixel block cut out by the character cutout module 430-17. .

As a more specific example, each module will be described.
The dictionary document generation module 410-17, the high-resolution image reading module 920-17, and the multi-resolution information storage module 120-17 are connected, and the dictionary document generation module 410-17 adds the dictionary to the high-resolution image reading module 920-17. The document 412-17 is handed over, and the character code / font information 414-17 is handed over to the multi-resolution information storage module 120-17. This is equivalent to the dictionary document generation module 410-9 shown in the example of FIG.

  The high resolution image reading module 920-17, the character segmentation module 930-17, and the low resolution multi-phase scan processing module 1710 are connected. The high resolution image readout module 920-17 includes the character segmentation module 930-17 and the low resolution multi-level scan processing module 1710. The high resolution image 922-17 is passed to the phase scan processing module 1710. This is equivalent to the high resolution image reading module 920 shown in the example of FIG.

  The character cutout module 930-17 and the multiple resolution information storage module 120-17 are connected, and the character cutout module 930-17 passes the single character image 932-17 to the multiple resolution information storage module 120-17. This is equivalent to the character segmentation module 930 shown in the example of FIG.

  The low-resolution multi-phase scan processing module 1710, the character segmentation module 430-17, and the multi-resolution information storage module 120-17 are connected, and the low-resolution multi-phase scan processing module 1710 adds a low-resolution image to the character segmentation module 430-17. 422-17 is passed, and resolution information 1712 is passed to the multi-resolution information storage module 120-17. A low-resolution image 422-17 corresponding to an image read multiple times to generate a plurality of character images with different phases at a low resolution (for example, about 200 dpi × 100 dpi in the fax standard mode and 200 dpi × 200 dpi for the scanner) It is generated and sent to the character segmentation module 430-17, and the resolution information 1712 is sent to the multi-resolution information storage module 120-17. That is, the low resolution multi-phase scan processing module 1710 does not actually read by the scanner like the low resolution image multiple reading module 420 shown in the example of FIG. A low resolution image 422-17, which is a character image, is generated. As a more specific example, a plurality of low-resolution images 422-17 are generated by performing multiple reduction processing and phase conversion processing (for example, movement processing) on the high-resolution image 922-17 multiple times. To do.

The character segmentation module 430-17 and the multiple resolution information accumulation module 120-17 are connected, and the character segmentation module 430-17 passes the one character image 432-17 to the multiple resolution information accumulation module 120-17. This is equivalent to the character segmentation module 430-9 shown in the example of FIG.
The multi-resolution information storage module 120-17 is equivalent to the multi-resolution information storage module 120-9 shown in the example of FIG. That is, as the multiple resolution information stored in the multiple resolution information storage module 120-17, the multiple resolution information table 1000 shown in the example of FIG. 10, the multiple resolution information table 1100 shown in the example of FIG. 11, or the multiple shown in the example of FIG. There is a resolution information table 1200.

  FIG. 18 is a module configuration diagram of an overall configuration example (1) for realizing the fourth specific embodiment. The character document 1800-A is received, and a character document 1800-B, which is a higher resolution document, is generated. For this purpose, the multi-resolution information storage module 120-18 stores the multi-resolution information table 1000 shown in the example of FIG. 10, and the high-resolution information adding module 140-18 adds a high-resolution character image as high-resolution information. Yes. The result equivalent to the configuration example shown in the example of FIG. 13 is to be obtained.

  18 includes a dictionary document generation module 410-18, a high resolution image reading module 920-18, a low resolution multi-phase scan processing module 1710-18, a character segmentation module 930-18, and a character segmentation module 430-. 18, a multi-resolution information storage module 120-18, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-18, a high-resolution information adding module 140-18, and a high-resolution documenting module 1330. Yes.

The dictionary document generation module 410-18, the high-resolution image reading module 920-18, and the multi-resolution information storage module 120-18 are connected, and the dictionary document generation module 410-18 includes a dictionary in the high-resolution image reading module 920-18. The document 412-18 is delivered, and the character code / font information 414-18 is delivered to the multi-resolution information storage module 120-18.
The high resolution image reading module 920-18, the character segmentation module 930-18, and the low resolution multi-phase scan processing module 1710-18 are connected, and the high resolution image readout module 920-18 is the character segmentation module 930-18, low The high resolution image 922-18 is passed to the resolution multiphase scan processing module 1710-18.
The character segmentation module 930-18 and the multiple resolution information storage module 120-18 are connected, and the character segmentation module 930-18 passes the one-character image 932-18 to the multiple resolution information accumulation module 120-18.
The low resolution multi-phase scan processing module 1710-18, the character segmentation module 430-18, and the multiple resolution information storage module 120-18 are connected, and the low resolution multiphase scan processing module 1710-18 is the character segmentation module 430-18. The low resolution image 422-18 is transferred to the multi-resolution information storage module 120-18, and the resolution information 1712-18 is transferred to the multi-resolution information storage module 120-18.
The character segmentation module 430-18 and the multiple resolution information accumulation module 120-18 are connected, and the character segmentation module 430-18 passes the one character image 432-18 to the multiple resolution information accumulation module 120-18.
The multi-resolution information storage module 120-18 is connected to the low-resolution character image collation module 130-18 and the high-resolution information addition module 140-18. The multi-resolution information storage module 120-18 is connected to the low-resolution character image collation module 130. The low resolution character image 122-18 is passed to −18, and the high resolution character image 124-18 is passed to the high resolution information adding module 140-18.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 1800 -A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-18, and the high resolution document creation module 1330 are connected, and the character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-18, In addition, a one-character image 722 and character position coordinates 724 are passed to the high-resolution documenting module 1330. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-18 and the high resolution information provision module 140-18 are connected, and the low resolution character image collation module 130-18 passes the collation ID 132-18 to the high resolution information provision module 140-18. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high resolution information adding module 140-18, the multiple resolution information storage module 120-18, and the high resolution documenting module 1330 are connected, and the high resolution information adding module 140-18 collates with the multiple resolution information storage module 120-18. ID 142-18 is passed, and high resolution character image 144-18 is passed to high resolution documentation module 1330. This is equivalent to the high-resolution information adding module 140-13 shown in the example of FIG.
The high resolution documenting module 1330 outputs the character document 1800-B. This is equivalent to the high resolution documenting module 1330 shown in the example of FIG.

  FIG. 19 is a module configuration diagram of an overall configuration example (2) for realizing the fourth specific embodiment. This receives character document 1900-A and generates character document 1900-B and character document 1900-C having a layer structure by rasterizing character codes with transparent colors. Therefore, the dictionary document generation module 410-19 passes the character code / font information 414-19 to the multi-resolution information storage module 120-19, and the multi-resolution information storage module 120-19 receives the multi-resolution shown in the example of FIG. The information table 1100 is stored, and the high resolution information adding module 140-19 adds a high resolution character image and a character code as high resolution information. The result equivalent to the configuration example shown in the example of FIG. 14 is to be obtained.

  19 includes a dictionary document generation module 410-19, a high resolution image reading module 920-19, a low resolution multi-phase scan processing module 1710-19, a character segmentation module 930-19, and a character segmentation module 430-. 19, a multi-resolution information storage module 120-19, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-19, a high-resolution information adding module 140-19, and a formatter module 730.

The dictionary document generation module 410-19, the high-resolution image reading module 920-19, and the multi-resolution information storage module 120-19 are connected. The document 412-19 is transferred, and the character code / font information 414-19 is transferred to the multi-resolution information storage module 120-19.
The high resolution image reading module 920-19, the character segmentation module 930-19, and the low resolution multi-phase scan processing module 1710-19 are connected. The high resolution image readout module 920-19 includes the character segmentation module 930-19, The high resolution image 922-19 is passed to the resolution multiphase scan processing module 1710-19.
The character segmentation module 930-19 and the multiple resolution information storage module 120-19 are connected, and the character segmentation module 930-19 passes the one-character image 932-19 to the multiple resolution information accumulation module 120-19.
The low resolution multi-phase scan processing module 1710-19, the character segmentation module 430-19, and the multi-resolution information storage module 120-19 are connected. The low resolution image 422-19 is transferred to the multi-resolution information storage module 120-19, and the resolution information 1712-19 is transferred to the multi-resolution information storage module 120-19.
The character cutout module 430-19 and the multiple resolution information storage module 120-19 are connected, and the character cutout module 430-19 passes the single character image 432-19 to the multiple resolution information storage module 120-19.
The multi-resolution information storage module 120-19, the low-resolution character image collation module 130-19, and the high-resolution information adding module 140-19 are connected. The multi-resolution information storage module 120-19 is connected to the low-resolution character image collation module 130. The low resolution character image 122-19 is passed to -19, and the high resolution character image + character code 124-19 is passed to the high resolution information adding module 140-19.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 1900 -A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-19, and the formatter module 730 are connected. The character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-19, and the formatter. A one-character image 722 and character position coordinates 724 are passed to the module 730. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-19 and the high resolution information addition module 140-19 are connected, and the low resolution character image collation module 130-19 passes the collation ID 132-19 to the high resolution information addition module 140-19. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high resolution information adding module 140-19, the multiple resolution information storage module 120-19, and the formatter module 730 are connected, and the high resolution information adding module 140-19 sends the collation ID 142-19 to the multiple resolution information storage module 120-19. Also, the high-resolution character image + character code 144-19 is passed to the formatter module 730. This is equivalent to the high-resolution information adding module 140-14 shown in the example of FIG.
The formatter module 730 outputs the character document 1900-B and the character document 1900-C. This is equivalent to the formatter module 730 shown in the example of FIG.

  FIG. 20 is a module configuration diagram of an overall configuration example (3) for realizing the fourth specific embodiment. This accepts the character document 2000-A and generates a character document 2000-B which is an electronic document in a format created by a document creation device (such as word processing software). Therefore, the dictionary document generation module 410-20 passes the character code / font information 414-20 to the multi-resolution information storage module 120-20, and the multi-resolution information storage module 120-20 receives the multi-resolution shown in the example of FIG. The information table 1200 (or the multiple resolution information table 600 shown in the example of FIG. 6) is stored, and the high resolution information adding module 140-20 assigns character code and font information as high resolution information. The result equivalent to the configuration example shown in the example of FIG. 15 is to be obtained.

  20 includes a dictionary document generation module 410-20, a high resolution image reading module 920-20, a low resolution multi-phase scan processing module 1710-20, a character segmentation module 930-20, and a character segmentation module 430-. 20, a multi-resolution information storage module 120-20, an image reading module 710, a character segmentation module 720, a low-resolution character image collation module 130-20, a high-resolution information addition module 140-20, and a documenting module 830.

The dictionary document generation module 410-20, the high-resolution image reading module 920-20, and the multi-resolution information storage module 120-20 are connected, and the dictionary document generation module 410-20 adds a dictionary to the high-resolution image reading module 920-20. The document 412-20 is delivered, and the character code / font information 414-20 is delivered to the multi-resolution information storage module 120-20.
The high resolution image reading module 920-20, the character segmentation module 930-20, and the low resolution multi-phase scan processing module 1710-20 are connected, and the high resolution image readout module 920-20 is the character segmentation module 930-20, low The high resolution image 922-20 is passed to the resolution multiphase scan processing module 1710-20.
The character segmentation module 930-20 and the multiple resolution information accumulation module 120-20 are connected, and the character segmentation module 930-20 passes the single character image 932-20 to the multiple resolution information accumulation module 120-20.
The low resolution multi-phase scan processing module 1710-20, the character segmentation module 430-20, and the multi-resolution information storage module 120-20 are connected, and the low resolution multi-phase scan processing module 1710-20 is the character segmentation module 430-20. The low resolution image 422-20 is transferred to the multi-resolution information storage module 120-20, and the resolution information 1712-20 is transferred to the multi-resolution information storage module 120-20.
The character segmentation module 430-20 and the multiple resolution information storage module 120-20 are connected, and the character segmentation module 430-20 passes the single character image 432-20 to the multiple resolution information storage module 120-20.
The multi-resolution information storage module 120-20, the low-resolution character image collation module 130-20, and the high-resolution information adding module 140-20 are connected, and the multi-resolution information storage module 120-20 is connected to the low-resolution character image collation module 130. The low resolution character image 122-20 is passed to −20, and the character code + font information 124-20 is passed to the high resolution information adding module 140-20.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 2000-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low-resolution character image collation module 130-20, and the documentation module 830 are connected, and the character segmentation module 720 passes the one-character image 722 to the low-resolution character image collation module 130-20. A one-character image 722 and character position coordinates 724 are passed to the documentation module 830. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-20 and the high resolution information provision module 140-20 are connected, and the low resolution character image collation module 130-20 passes the collation ID 132-20 to the high resolution information provision module 140-20. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high-resolution information adding module 140-20, the multi-resolution information accumulating module 120-20, and the documenting module 830 are connected, and the high-resolution information adding module 140-20 sends the collation ID 142- 20, and the character code + font information 144-20 is passed to the documentation module 830. This is equivalent to the high resolution information adding module 140-15 shown in the example of FIG.
The documenting module 830 outputs the character document 2000-B. This is equivalent to the documentation module 830 shown in the example of FIG.

<Fifth Specific Embodiment>
FIG. 21 is a conceptual module configuration diagram of a configuration example of the fifth specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied.

  The configuration example shown in FIG. 21 includes a user document reception module 1610-21, a high resolution image reading module 1620-21, a low resolution multi-phase scan processing module 1710-21, a character recognition processing module 2110, a character segmentation module 1630-21, A clipping module 430-21 and a multi-resolution information storage module 120-21 are included. The user document reception module 1610-21, the high resolution image reading module 1620-21, the low resolution multi-phase scan processing module 1710-21, the character recognition processing module 2110, the character segmentation module 1630-21, and the character segmentation module 430-21 are shown in FIG. 1 corresponds to a specific configuration example of the multi-resolution information generation module 110 shown in FIG. 1, and the multi-resolution information storage module 120-21 corresponds to the multi-resolution information storage module 120 shown in FIG.

The user document reception module 1610-21 passes the user document 1612-21 to the high resolution image reading module 1620-21.
The high resolution image reading module 1620-21 reads an image.
The low resolution multi-phase scan processing module 1710-21 lowers the resolution of the image read by the high resolution image reading module 1620-21 and generates an image having a different phase.
The character recognition processing module 2110 recognizes an image of a pixel block in the image read by the high resolution image reading module 1620-21. The character recognition processing module 2110 may receive and recognize a single character image 1632-21 from the high resolution image reading module 1620-21.
The image read by the high resolution image reading module 1620-21 is a document actually used by the user.
The character cutout module 1630-21 cuts out an image of a pixel block in the image read by the high resolution image reading module 1620-21.
The character cutout module 430-21 cuts out an image of a pixel block in the image generated by the low resolution multiphase scan processing module 1710-21.

  The multi-resolution information storage module 120-21 stores the pixel block image cut out by the character cutout module 1630-21 as information on the pixel block cut out by the character cutout module 430-21. . Further, the multi-resolution information storage module 120-21 may store the recognition result by the character recognition processing module 2110 as information regarding the pixel block for the pixel block image cut out by the character cutout module 430-21. Good.

As a more specific example, each module will be described.
The user document reception module 1610-21 and the high resolution image reading module 1620-21 are connected, and the user document reception module 1610-21 passes the user document 1612-21 to the high resolution image reading module 1620-21. This is equivalent to the user document reception module 1610 shown in the example of FIG.
The high resolution image reading module 1620-21, the character segmentation module 1630-21, the low resolution multi-phase scan processing module 1710-21, and the character recognition processing module 2110 are connected, and the high resolution image reading module 1620-21 is character segmentation. The high-resolution image 1622-21 is transferred to the module 1630-21, the low-resolution multiphase scan processing module 1710-21, and the character recognition processing module 2110. This is equivalent to the high resolution image reading module 1620 shown in the example of FIG.
The character cutout module 1630-21 and the multiple resolution information storage module 120-21 are connected, and the character cutout module 1630-21 passes the single character image 1632-21 to the multiple resolution information storage module 120-21. This is equivalent to the character segmentation module 1630 shown in the example of FIG.

The low resolution multiphase scan processing module 1710-21, the character segmentation module 430-21, and the multiple resolution information storage module 120-21 are connected, and the low resolution multiphase scan processing module 1710-21 is the character segmentation module 430-21. The low-resolution image 422-21 is transferred to the multi-resolution information storage module 120-21. This is equivalent to the low resolution multi-phase scan processing module 1710 shown in the example of FIG.
The character segmentation module 430-21 and the multi-resolution information storage module 120-21 are connected, and the character segmentation module 430-21 passes the one-character image 432-21 to the multi-resolution information storage module 120-21. This is equivalent to the character segmentation module 430-17 shown in the example of FIG.

  The character recognition processing module 2110 and the multi-resolution information storage module 120-21 are connected, and the character recognition processing module 2110 passes the character code 2112 to the multi-resolution information storage module 120-21. The character recognition processing module 2110 recognizes characters using the high-resolution image 1622-21 (or a single character image 1632-21) as a target image. The existing method is used as the character recognition method. Also, resolution conversion processing or the like may be performed as preprocessing in accordance with the character recognition processing. Further, the character that could not be recognized may not be stored in the multi-resolution information storage module 120-21. In that case, the multi-resolution information about the character is lost.

The multi-resolution information storage module 120-21 links a high-resolution one-character image 1632-21, a plurality of low-resolution one-character images 432-21, and a character code 2112 with respect to one character to generate multi-resolution information. ,accumulate. The multi-resolution information includes a multi-resolution information table 1000 shown in the example of FIG. 10, a multi-resolution information table 1100 shown in the example of FIG. 11, or a multi-resolution information table 1200 shown in the example of FIG.
Further, the entire configuration example for realizing the fifth specific embodiment is the same as that shown in FIGS. 18, 19, and 20. However, the dictionary document generation modules 410-18, 19, and 20, the high-resolution image reading modules 920-18, 19, and 20, the character segmentation modules 930-18, 19, and 20, and the low-resolution multiphase scan processing modules 1710-18 and 19 , 20, character segmentation modules 430-18, 19, 20 and multi-resolution information storage modules 120-18, 19, 20 are replaced with the configuration shown in FIG. 21.

<Sixth specific embodiment>
FIG. 22 is a conceptual module configuration diagram of a configuration example of the sixth specific embodiment. That is, the multi-resolution information generation module 110 and the multi-resolution information storage module 120, which are a part of the basic configuration example of the embodiment shown in FIG. 1, are embodied. In the embodiments so far, a character image is read by a scanner to generate multi-resolution information. In this embodiment, a high-resolution character image is generated by raster processing and is used to generate multiple resolution information. Information is generated.

  The configuration example shown in FIG. 22 includes a dictionary document character selection module 2210, a font drawing module 2220, a low resolution multi-phase scan processing module 2230, and a multi-resolution information storage module 120-22. The dictionary document character selection module 2210, the font drawing module 2220, and the low resolution multi-phase scan processing module 2230 correspond to a specific configuration example of the multi-resolution information generation module 110 shown in FIG. −22 corresponds to the multi-resolution information storage module 120 illustrated in FIG. 1.

The dictionary document character selection module 2210 selects a pixel block and information regarding the pixel block in accordance with a user operation.
The font drawing module 2220 generates an image of the pixel block based on the pixel block selected by the dictionary document character selection module 2210 and information on the pixel block.
The low resolution multi-phase scan processing module 2230 lowers the resolution of the pixel block image generated by the font drawing module 2220 and generates an image having a different phase.
The multi-resolution information storage module 120-22 stores information on the pixel block selected by the dictionary document character selection module 2210 as information on the pixel block with respect to the pixel block image processed by the low-resolution multiphase scan processing module 2230. Remember.

As a more specific example, each module will be described.
The dictionary document character selection module 2210, the font drawing module 2220, and the multi-resolution information storage module 120-22 are connected. The dictionary document character selection module 2210 passes the character code / font information 2212 to the font drawing module 2220. The character code / font information 2214 is passed to the multi-resolution information storage module 120-22. In response to the user's operation, a character image to be generated with multiple resolution information is selected. That is, the target character (character code) and font information (font type, font size, etc.) are selected.

  The font drawing module 2220, the multi-resolution information storage module 120-22, and the low-resolution multi-phase scan processing module 2230 are connected, and the font drawing module 2220 sends the high-resolution one-character image 2222 to the multi-resolution information storage module 120-22. hand over. Based on the character code / font information 2212 selected by the dictionary document character selection module 2210, a character is drawn at a high resolution to generate a high resolution single character image 2222, a low resolution multi-phase scan processing module 2230, a plurality of resolutions The information is sent to the information storage module 120-22. Since the character image generated by the font drawing module 2220 is one character at a time, it is not necessary to perform character cut-out processing after processing by the font drawing module 2220 and before passing to the multi-resolution information storage module 120-22. .

  The low-resolution multi-phase scan processing module 2230 and the multi-resolution information storage module 120-22 are connected, and the low-resolution multi-phase scan processing module 2230 is connected to the multi-resolution information storage module 120-22 with a low-resolution multi-character image 2232 and resolution. Pass information 2234. This is equivalent to the low resolution multi-phase scan processing module 1710 shown in the example of FIG. Since the character image generated by the font drawing module 2220 is one character at a time, the character segmentation process is performed after the processing by the low-resolution multi-phase scan processing module 2230 and before passing to the multi-resolution information storage module 120-22. There is no need to do it.

  The multi-resolution information storage module 120-22 links the high-resolution single character image 2222, the low-resolution multi-character image 2232, and the character code / font information 2214 to generate and store the multi-resolution information. The multi-resolution information includes a multi-resolution information table 1000 shown in the example of FIG. 10, a multi-resolution information table 1100 shown in the example of FIG. 11, or a multi-resolution information table 1200 shown in the example of FIG.

  FIG. 23 is a module configuration diagram of an overall configuration example (1) for realizing the sixth specific embodiment. The character document 2300-A is received, and a character document 2300-B, which is a higher resolution document, is generated. For this purpose, the multi-resolution information storage module 120-23 stores the multi-resolution information table 1000 shown in the example of FIG. 10, and the high-resolution information adding module 140-23 adds a high-resolution character image as high-resolution information. Yes. The result equivalent to the configuration example shown in the example of FIG. 13 is to be obtained.

  23 includes a dictionary document character selection module 2210-23, a font drawing module 2220-23, a low resolution multi-phase scan processing module 2230-23, a multi-resolution information storage module 120-23, and an image reading module 710. , A character segmentation module 720, a low resolution character image collation module 130-23, a high resolution information addition module 140-23, and a high resolution documenting module 1330.

The dictionary document character selection module 2210-23, the font drawing module 2220-23, and the multi-resolution information storage module 120-23 are connected, and the dictionary document character selection module 2210-23 receives the character code / The font information 2212-23 is passed, and the character code / font information 2214-23 is passed to the multi-resolution information storage module 120-23.
The font drawing module 2220-23 is connected to the multi-resolution information storage module 120-23 and the low-resolution multi-phase scan processing module 2230-23. The font drawing module 2220-23 is connected to the multi-resolution information storage module 120-23. The high resolution single character image 2222-23 is passed to the resolution multi-phase scan processing module 2230-23.
The low-resolution multi-phase scan processing module 2230-23 and the multi-resolution information storage module 120-23 are connected, and the low-resolution multi-phase scan processing module 2230-23 is connected to the multi-resolution information storage module 120-23 with a low-resolution multi-character. The image 2232-23 and resolution information 2234-23 are transferred.
The multi-resolution information storage module 120-23, the low-resolution character image collation module 130-23, and the high-resolution information addition module 140-23 are connected. The multi-resolution information storage module 120-23 is connected to the low-resolution character image collation module 130. The low-resolution character image 122-23 is passed to -23, and the high-resolution character image 124-23 is passed to the high-resolution information adding module 140-23.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 2300-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-23, and the high resolution documentization module 1330 are connected. In addition, a one-character image 722 and character position coordinates 724 are passed to the high-resolution documenting module 1330. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-23 and the high resolution information provision module 140-23 are connected, and the low resolution character image collation module 130-23 passes the collation ID 132-23 to the high resolution information provision module 140-23. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high-resolution information adding module 140-23, the multi-resolution information storage module 120-23, and the high-resolution documenting module 1330 are connected, and the high-resolution information adding module 140-23 collates with the multi-resolution information storage module 120-23. ID 142-23 is passed, and high resolution character image 144-23 is passed to high resolution documentation module 1330. This is equivalent to the high-resolution information adding module 140-13 shown in the example of FIG.
The high resolution documenting module 1330 outputs the character document 2300-B. This is equivalent to the high resolution documenting module 1330 shown in the example of FIG.

  FIG. 24 is a module configuration diagram of an overall configuration example (2) for realizing the sixth specific embodiment. This receives character document 2400-A and generates character document 2400-B and character document 2400-C having a layer structure by rasterizing character codes with transparent colors. For this purpose, the dictionary document character selection module 2210-24 passes the character code / font information 2212-24 to the multi-resolution information storage module 120-24, and the multi-resolution information storage module 120-24 uses the plurality of information shown in the example of FIG. The resolution information table 1100 is stored, and the high resolution information adding module 140-24 adds a high resolution character image and character code as high resolution information. The result equivalent to the configuration example shown in the example of FIG. 14 is to be obtained.

  24 includes a dictionary document character selection module 2210-24, a font drawing module 2220-24, a low resolution multi-phase scan processing module 2230-24, a multi-resolution information storage module 120-24, and an image reading module 710. A character segmentation module 720, a low resolution character image collation module 130-24, a high resolution information addition module 140-24, and a formatter module 730.

The dictionary document character selection module 2210-24, the font drawing module 2220-24, and the multi-resolution information storage module 120-24 are connected, and the dictionary document character selection module 2210-24 sends the character code / code to the font drawing module 2220-24. The font information 2212-24 is transferred, and the character code / font information 2214-24 is transferred to the multi-resolution information storage module 120-24.
The font drawing module 2220-24 is connected to the multi-resolution information storage module 120-24 and the low-resolution multi-phase scan processing module 2230-24. The font drawing module 2220-24 is connected to the multi-resolution information storage module 120-24. The high resolution single character image 2222-24 is passed to the resolution multi-phase scan processing module 2230-24.
The low-resolution multi-phase scan processing module 2230-24 and the multi-resolution information storage module 120-24 are connected, and the low-resolution multi-phase scan processing module 2230-24 is added to the multi-resolution information storage module 120-24. The image 2232-24 and resolution information 2234-24 are passed.
The multi-resolution information storage module 120-24, the low-resolution character image collation module 130-24, and the high-resolution information addition module 140-24 are connected. The multi-resolution information storage module 120-24 is connected to the low-resolution character image collation module 130. The low resolution character image 122-24 is passed to -24, and the high resolution character image + character code 124-24 is passed to the high resolution information adding module 140-24.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 2400-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-24, and the formatter module 730 are connected, and the character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-24, A one-character image 722 and character position coordinates 724 are passed to the module 730. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-24 and the high resolution information provision module 140-24 are connected, and the low resolution character image collation module 130-24 passes the collation ID 132-24 to the high resolution information provision module 140-24. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high resolution information providing module 140-24, the multiple resolution information storage module 120-24, and the formatter module 730 are connected. In addition, the high resolution character image + character code 144-24 is passed to the formatter module 730. This is equivalent to the high-resolution information adding module 140-14 shown in the example of FIG.
The formatter module 730 outputs the character document 2400-B and the character document 2400-C. This is equivalent to the formatter module 730 shown in the example of FIG.

  FIG. 25 is a module configuration diagram of an overall configuration example (3) for realizing the sixth specific embodiment. This receives character document 2500-A and generates character document 2500-B, which is an electronic document in a format created by a document creation device (such as word processing software). For this purpose, the dictionary document character selection module 2210-25 passes the character code / font information 2212-25 to the multi-resolution information storage module 120-25, and the multi-resolution information storage module 120-25 has a plurality of information shown in the example of FIG. The resolution information table 1200 (or the multiple resolution information table 600 shown in the example of FIG. 6) is stored, and the high resolution information adding module 140-25 assigns character code and font information as high resolution information. The result equivalent to the configuration example shown in the example of FIG. 15 is to be obtained.

  The overall configuration example shown in FIG. 25 includes a dictionary document character selection module 2210-25, a font drawing module 2220-25, a low resolution multi-phase scan processing module 2230-25, a multi-resolution information storage module 120-25, and an image reading module 710. , A character segmentation module 720, a low resolution character image collation module 130-25, a high resolution information addition module 140-25, and a documentization module 830.

The dictionary document character selection module 2210-25, the font drawing module 2220-25, and the multi-resolution information storage module 120-25 are connected, and the dictionary document character selection module 2210-25 receives the character code / The font information 2212-25 is transferred, and the character code / font information 2214-25 is transferred to the multi-resolution information storage module 120-25.
The font drawing module 2220-25 is connected to the multi-resolution information storage module 120-25 and the low-resolution multi-phase scan processing module 2230-25. The font drawing module 2220-25 is connected to the multi-resolution information storage module 120-25. The high resolution single character image 2222-25 is passed to the resolution multiphase scan processing module 2230-25.
The low-resolution multi-phase scan processing module 2230-25 and the multi-resolution information storage module 120-25 are connected, and the low-resolution multi-phase scan processing module 2230-25 is connected to the multi-resolution information storage module 120-25 with a low-resolution multi-character. The image 2232-25 and resolution information 2234-25 are passed.
The multi-resolution information storage module 120-25 is connected to the low-resolution character image collation module 130-25 and the high-resolution information adding module 140-25. The multi-resolution information storage module 120-25 is connected to the low-resolution character image collation module 130. The low resolution character image 122-25 is passed to −25, and the character code + font information 124-25 is passed to the high resolution information adding module 140-25.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 2500-A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720, the low resolution character image collation module 130-25, and the documenting module 830 are connected, and the character segmentation module 720 passes the one character image 722 to the low resolution character image collation module 130-25, A one-character image 722 and character position coordinates 724 are passed to the documentation module 830. This is equivalent to the character segmentation module 720 shown in the example of FIG.

  The low resolution character image collation module 130-25 and the high resolution information provision module 140-25 are connected, and the low resolution character image collation module 130-25 passes the collation ID 132-25 to the high resolution information provision module 140-25. . This is equivalent to the low resolution character image collation module 130-7 shown in the example of FIG.

The high resolution information adding module 140-25, the multiple resolution information storage module 120-25, and the documenting module 830 are connected, and the high resolution information adding module 140-25 sends the collation ID 142- to the multiple resolution information storage module 120-25. 25, and the character code + font information 144-25 is passed to the documentation module 830. This is equivalent to the high resolution information adding module 140-15 shown in the example of FIG.
The documenting module 830 outputs the character document 2500-B. This is equivalent to the documentation module 830 shown in the example of FIG.

  FIG. 26 is a module configuration diagram of an overall configuration example (4) for realizing the sixth specific embodiment. This accepts the character document 2600-A and generates a character document 2600-B, which is an electronic document in a format created by a document creation device (such as word processing software). For this purpose, the high resolution information providing module 140-26 assigns a character code and font information as high resolution information. The result equivalent to the configuration example shown in the example of FIG. 15 is to be obtained.

  26 includes a font information determination module 2610, a character recognition processing module 2620, a font drawing module 2630, a low resolution multi-phase scan processing module 2640, a collation control module 2690, an image reading module 710, and a character segmentation module 720. , A low-resolution character image collation module 130-26, a high-resolution information adding module 140-26, and a documenting module 830.

The image reading module 710 reads an image.
The character cutout module 720 cuts out an image of a pixel block in the image read by the image reading module 710.
The font information determination module 2610 determines information related to the pixel block image cut out by the character cutout module 720.
The character recognition processing module 2620 recognizes the pixel block image cut out by the character cut-out module 720.
The font drawing module 2630 generates a pixel block image based on the recognition result by the character recognition processing module 2620 and information on the pixel block image determined by the font information determination module 2610. The font drawing module 2630 may generate a pixel block image having a higher resolution than the pixel block image cut out by the character cutout module 720.
The low-resolution multiphase scan processing module 2640 lowers the resolution of the pixel block image generated by the font drawing module 2630 and generates an image having a different phase.
The collation control module 2690 controls the font information determination module 2610, the character recognition processing module 2620, the font drawing module 2630, and the low resolution multi-phase scan processing module 2640 based on the collation result of the low resolution character image collation module 130-26. Let the process run again.
The low resolution character image collation module 130-26 collates the pixel block image cut out by the character cutout module 720 and the pixel block image generated by the low resolution multiphase scan processing module 2640.
The high-resolution information adding module 140-26 is based on the collation result by the low-resolution character image collation module 130-26, and information on the pixel block image determined by the font information determination module 2610 and the recognition result by the character recognition processing module 2620. Is added to the pixel block cut out by the character cut-out module 720.
The documenting module 830 generates document information based on the information related to the image of the pixel block given by the high resolution information giving module 140-26 and the recognition result.

The image reading module 710 and the character cutout module 720 are connected, and the image reading module 710 reads the character document 2600 -A and passes the character image 712 to the character cutout module 720.
The character segmentation module 720 is connected to the low resolution character image collation module 130-26, the documenting module 830, the font information determination module 2610, and the character recognition processing module 2620. The character segmentation module 720 is connected to the low resolution character image collation module 130. -26, the one character image 722 is passed to the font information determination module 2610 and the character recognition processing module 2620, and the one character image 722 and the character position coordinates 724 are passed to the documenting module 830. This is equivalent to the character segmentation module 720 shown in the example of FIG.

The font information determination module 2610, the font drawing module 2630, and the high resolution information addition module 140-26 are connected. The font information determination module 2610 sends the font information 2612 to the font drawing module 2630 and the high resolution information addition module 140-26. hand over. The font size and font type of the one-character image 722 passed from the character cutout module 720 are recognized and determined, and are sent as font information 2612 to the font drawing module 2630 and the high resolution information adding module 140-26. The existing method is used to recognize the font size and font type.
The character recognition processing module 2620, the font drawing module 2630, and the high-resolution information adding module 140-26 are connected. The character recognition processing module 2620 recognizes the one-character image 722 passed from the character cut-out module 720 to recognize the font. The character code 2622 of the recognition result is passed to the drawing module 2630 and the high resolution information adding module 140-26. This is equivalent to the character recognition processing module 1660 shown in the example of FIG.

The font drawing module 2630 and the low resolution multi-phase scan processing module 2640 are connected, and the font drawing module 2630 receives the font information 2612 from the font information determination module 2610 and the character code 2622 from the character recognition processing module 2620 and receives characters. And the character image is passed to the low resolution multi-phase scan processing module 2640. This is equivalent to the font drawing module 2220 shown in the example of FIG.
The low resolution multiphase scan processing module 2640 and the low resolution character image collation module 130-26 are connected, and the low resolution multiphase scan processing module 2640 sends the low resolution character image 2642 to the low resolution character image collation module 130-26. hand over. This is equivalent to the low-resolution multiphase scan processing module 2230 shown in the example of FIG.

  The low resolution character image collation module 130-26 and the high resolution information addition module 140-26 are connected, and the low resolution character image collation module 130-26 sends the collation result 132-26 to the high resolution information addition module 140-26. hand over. One character image 722 from the character segmentation module 720 and the low resolution character image 2642 from the low resolution multi-phase scan processing module 2640 are collated, and if they match, the collation result indicating that they match is a high resolution information addition module 140- 26. If they do not match, the collation control module 2690 is notified of the collation result to that effect.

  The collation control module 2690 controls the font information determination module 2610, the character recognition processing module 2620, the font drawing module 2630, the low resolution multiphase scan processing module 2640, and the low resolution character image collation module 130-26 using the control signal 2699A. To do. A control signal 2699A is sent to the font information determination module 2610, the character recognition processing module 2620, the font drawing module 2630, and the low resolution multi-phase scan processing module 2640 to change the respective parameters to perform the processing again. The resolution character image collation module 130-26 performs collation. The parameter change includes, for example, a change of a recognition dictionary in the font information determination module 2610 and the character recognition processing module 2620. These processes are repeated until a collation result indicating that they match is obtained. Alternatively, when a predetermined number of repetitions have been performed, the low-resolution character image collation module 130-26 passes to the high-resolution information addition module 140-26 as a collation result 132-26 indicating that the collation could not be performed.

The high resolution information adding module 140-26 and the documenting module 830 are connected, and the high resolution information adding module 140-26 passes the character code + font information 144-26 to the documenting module 830. When the collation result 132-26 received from the low resolution character image collation module 130-26 indicates a match, the font information 2612 from the font information determination module 2610 and the character code from the character recognition processing module 2620 are displayed. 2622 is assigned to the corresponding one-character image 722 cut out by the character cut-out module 720. When the collation result 132-26 indicates that the collation could not be performed, it indicates that the character document 2600-B is generated by using the one-character image 722 cut out by the character cut-out module 720 and the character position coordinates 724 as they are. Pass the code to the documentation module 830.
The documenting module 830 outputs the character document 2600-B. This is equivalent to the documentation module 830 shown in the example of FIG.

<Example 1 as a fax>
FIG. 27 is an explanatory diagram showing a processing example when the first, second, fourth, and sixth specific embodiments are realized as a fax.
FIG. 27A shows an example of processing at the preparation stage, that is, processing for generating and storing multi-resolution information. The fax machine 2700-A transmits a dictionary document 2710 to the fax machine 2700-B in advance, and the receiving fax machine 2700-B generates a plurality of pieces of resolution information based on the dictionary document 2710 transmitted. The information is stored in the resolution information storage module 120. However, this preparation process may be executed first and incorporated into the product fax machine 2700-B. For example, if a fax machine made by the same company is used on the transmission side and the reception side, and a dictionary document 2710 to be sent by the user is a printer product made by the same company and printed using a printer font, a known font is used on the reception side. It is effective because only the target is targeted.

  FIG. 27B shows normal processing, that is, processing for receiving a document and converting it into a high-resolution document. When the user transmits the user document 2720 using the fax 2700-C, the user document 2720 is output as a document in which high-resolution information is added to the fax document 2730 deteriorated by the receiving fax 2700-B. The form of the output document varies depending on the high resolution information to be given. For example, FIG. 27 (c1) shows that character search is possible in the form of PDF, and FIG. 27 (c2) shows that character encoding is done in the form of an electronic document or PDF. 27 (c3) indicates that the character image is replaced with a high-resolution character image.

<Realization example 2 as fax>
FIG. 28 is an explanatory diagram showing a processing example when the third and fifth specific embodiments are realized as a fax.
FIG. 28A shows an example of processing at the preparation stage, that is, processing for generating and storing multi-resolution information. The fax 2800-A transmits a user document 2810 (for example, a document already used by the user) to the fax 2800-B in advance, and the user who has been transmitted to the fax 2800-B on the receiving side. Multi-resolution information is generated based on the document 2810 and stored in the multi-resolution information storage module 120. In particular, when a dictionary document cannot be generated separately, such as when it is printed in a special font used for a form document, the user document is scanned in advance at a high resolution and sent to the receiving fax machine 2800-B. The multi-resolution information is generated and stored by the fax machine 2800-B on the receiving side. As a result, the more user documents, the more characters to which high resolution information is assigned in the normal process.

  FIG. 28B shows normal processing, that is, processing for receiving a document and converting it into a high-resolution document. When the user transmits the user document 2820 using the fax 2800-C, the user document 2820 is output as a document in which high-resolution information is added to the fax document 2830 deteriorated by the receiving fax 2800-B. The form of the output document varies depending on the high resolution information to be given. 28 (c1), (c2), and (c3) are equivalent to FIGS. 27 (c1), (c2), and (c3).

<Example 1 as Scanner>
FIG. 29 is an explanatory diagram showing a processing example when the first, second, fourth, and sixth specific embodiments are realized as a scanner.
FIG. 29A shows an example of processing at the preparation stage, that is, processing for generating and storing multi-resolution information. The fax 2900 reads the dictionary document 2910 a plurality of times at a low resolution in advance, reads it once at a higher resolution, generates multi-resolution information, and stores it in the multi-resolution information storage module 120. However, it is the same as described above that this preparation process may be executed first and incorporated in the fax machine 2900 as a product.

  FIG. 29B shows normal processing, that is, processing for receiving a document and converting it into a high-resolution document. The user causes the user document 2920 to be read using the fax 2900. The fax 2900 outputs the deteriorated user document 2920 as a document with high resolution information. The form of the output document varies depending on the high resolution information to be given. 29 (c1), (c2), and (c3) are equivalent to FIGS. 27 (c1), (c2), and (c3).

<Example 2 as a scanner>
FIG. 30 is an explanatory diagram showing a processing example when the third and fifth specific embodiments are realized as a scanner.
FIG. 30A shows an example of processing at the preparation stage, that is, processing for generating and storing multi-resolution information. The fax 3000 reads a user document 3010 (for example, a document already used by a user) in advance at a low resolution a plurality of times, reads it at a higher resolution once, generates a plurality of resolution information, and generates a multi-resolution information storage module 120. Accumulate in. In particular, when a dictionary document cannot be generated separately, such as when it is printed in a special font used for a form-based document, a user document is scanned at a high resolution in advance, and multi-resolution information is generated with a fax 3000. Accumulate. As a result, the more user documents, the more characters to which high resolution information is assigned in the normal process.

  FIG. 30B shows normal processing, that is, processing for receiving a document and converting it into a high-resolution document. The user reads the user document 3020 using the fax 3000. The fax 3000 outputs the deteriorated user document 3020 as a document with high resolution information. The form of the output document varies depending on the high resolution information to be given. 30 (c1), (c2), and (c3) are equivalent to FIGS. 27 (c1), (c2), and (c3).

  A hardware configuration example of the image processing apparatus according to the present embodiment will be described with reference to FIG. The configuration illustrated in FIG. 31 is configured by, for example, a personal computer (PC), and illustrates a hardware configuration example including a data reading unit 3117 such as a scanner and a data output unit 3118 such as a printer.

  A CPU (Central Processing Unit) 3101 includes various modules described in the above-described embodiments, that is, a multi-resolution information generation module 110, a multi-resolution information storage module 120, a low-resolution character image collation module 130, a character segmentation module 720, It is a control part which performs the process according to the computer program which described the execution sequence of each module, such as the formatter module 730.

  A ROM (Read Only Memory) 3102 stores programs used by the CPU 3101, calculation parameters, and the like. A RAM (Random Access Memory) 3103 stores programs used in the execution of the CPU 3101, parameters that change as appropriate during the execution, and the like. These are connected to each other by a host bus 3104 including a CPU bus.

  The host bus 3104 is connected via a bridge 3105 to an external bus 3106 such as a PCI (Peripheral Component Interconnect / Interface) bus.

  A keyboard 3108 and a pointing device 3109 such as a mouse are input devices operated by an operator. The display 3110 includes a liquid crystal display device or a CRT (Cathode Ray Tube), and displays various types of information as text or image information.

  An HDD (Hard Disk Drive) 3111 includes a hard disk, drives the hard disk, and records or reproduces a program executed by the CPU 3101 and information. The hard disk stores received images, multi-resolution information, and the like. Further, various computer programs such as various other data processing programs are stored.

  The drive 3112 reads out data or a program recorded in a removable recording medium 3113 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read as an interface 3107 and an external bus 3106. , The bridge 3105, and the RAM 3103 connected via the host bus 3104. The removable recording medium 3113 can also be used as a data recording area similar to a hard disk.

  The connection port 3114 is a port for connecting the external connection device 3115 and has a connection unit such as USB or IEEE1394. The connection port 3114 is connected to the CPU 3101 and the like via the interface 3107, the external bus 3106, the bridge 3105, the host bus 3104, and the like. The communication unit 3116 is connected to a network and executes data communication processing with the outside. The data reading unit 3117 is a scanner, for example, and executes document reading processing. The data output unit 3118 is a printer, for example, and executes document data output processing.

  Note that the hardware configuration of the image processing apparatus shown in FIG. 31 shows one configuration example, and the present embodiment is not limited to the configuration shown in FIG. 31, and the modules described in the present embodiment are executed. Any configuration is possible. For example, some modules may be configured with dedicated hardware (for example, Application Specific Integrated Circuit (ASIC), etc.), and some modules are in an external system and connected via a communication line Alternatively, a plurality of systems shown in FIG. 31 may be connected to each other via a communication line and cooperate with each other. Further, it may be incorporated in a copying machine, a fax machine, a scanner, a printer, a multifunction machine (an image processing apparatus having any two or more functions of a scanner, a printer, a copying machine, a fax machine, etc.).

  Note that the various embodiments or modules described above may be combined, and the technology described in the background art may be adopted as the processing content of each module.

The program described above may be provided by being stored in a recording medium, or the program may be provided by communication means. In that case, for example, the above-described program may be regarded as an invention of a “computer-readable recording medium recording the program”.
The “computer-readable recording medium on which a program is recorded” refers to a computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution, and the like.
The recording medium is, for example, a digital versatile disc (DVD), which is a standard established by the DVD Forum, such as “DVD-R, DVD-RW, DVD-RAM,” and DVD + RW. Standard “DVD + R, DVD + RW, etc.”, compact disc (CD), read-only memory (CD-ROM), CD recordable (CD-R), CD rewritable (CD-RW), Blu-ray disc ( Blu-ray Disc (registered trademark), magneto-optical disk (MO), flexible disk (FD), magnetic tape, hard disk, read-only memory (ROM), electrically erasable and rewritable read-only memory (EEPROM), flash Includes memory, random access memory (RAM), etc. .
The program or a part of the program may be recorded on the recording medium for storage or distribution. Also, by communication, for example, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a wired network used for the Internet, an intranet, an extranet, etc., or wireless communication It may be transmitted using a transmission medium such as a network or a combination of these, or may be carried on a carrier wave.
Furthermore, the program may be a part of another program, or may be recorded on a recording medium together with a separate program. Moreover, it may be divided and recorded on a plurality of recording media. Further, it may be recorded in any manner as long as it can be restored, such as compression or encryption.

DESCRIPTION OF SYMBOLS 110 ... Multi-resolution information generation module 120 ... Multi-resolution information storage module 130 ... Low resolution character image collation module 140 ... High resolution information provision module 410 ... Dictionary document generation module 420 ... Low resolution image multiple reading module 430 ... Character segmentation module 730 Formatter module 830 Documenting module 920 High resolution image reading module 930 Character cutting module 1330 High resolution documenting module 1610 User document receiving module 1620 High resolution image reading module 1630 Character cutting module 1640 Low resolution image Multiple reading module 1650 ... Character segmentation module 1660 ... Character recognition processing module 1710 ... Low resolution multi-phase CAN processing module 2110 ... Character recognition processing module 2210 ... Dictionary document character selection module 2220 ... Font drawing module 2230 ... Low resolution multi-phase scan processing module 2610 ... Font information determination module 2620 ... Character recognition processing module 2630 ... Font drawing module 2640 ... Low Resolution multi-phase scan processing module 2690 ... collation control module

Claims (10)

An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collating means, comprising information providing means for extracting information relating to the pixel block from the storage means and applying the information relating to the pixel block to the image of the pixel block in the received image ;
The information generating means
Output means for outputting an image including images of pixel clusters of a plurality of types or sizes;
First reading means for reading an image output by the output means;
A second reading unit that reads the image output by the output unit a plurality of times at a resolution lower than that of the first reading unit;
First cutout means for cutting out an image of a pixel block in the image read by the first reading means;
Second cutout means for cutting out an image of a pixel block in the image read by the second reading means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing device. An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collating means, comprising information providing means for extracting information relating to the pixel block from the storage means and applying the information relating to the pixel block to the image of the pixel block in the received image ;
The information generating means
First reading means for reading an image;
Second reading means for reading the image a plurality of times at a lower resolution than the first reading means;
First cutout means for cutting out an image of a pixel block in the image read by the first reading means;
Second cutout means for cutting out an image of a pixel block in the image read by the second reading means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing device. An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collating means, comprising information providing means for extracting information relating to the pixel block from the storage means and applying the information relating to the pixel block to the image of the pixel block in the received image ;
The information generating means
Output means for outputting an image including images of pixel clusters of a plurality of types or sizes;
Reading means for reading an image output by the output means;
Low resolution phase processing means for reducing the resolution of the image read by the reading means and generating images of different phases;
First cutout means for cutting out an image of a pixel block in the image read by the reading means;
Second cutout means for cutting out an image of a pixel block in the image generated by the low resolution phase processing means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing device. An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collating means, comprising information providing means for extracting information relating to the pixel block from the storage means and applying the information relating to the pixel block to the image of the pixel block in the received image ;
The information generating means
Reading means for reading an image;
Low resolution phase processing means for reducing the resolution of the image read by the reading means and generating images of different phases;
First cutout means for cutting out an image of a pixel block in the image read by the reading means;
Second cutout means for cutting out an image of a pixel block in the image generated by the low resolution phase processing means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing device. The information generating means
Recognizing means for recognizing an image of a pixel block in the image;
Said storage means according to claim 2 or 4, characterized in that storing a recognition result by the recognition unit as the information regarding該画pixel block with respect to the image of the pixel block cut out by the second cut-out means Image processing device. Wherein the pixel block is a character, an image processing apparatus according to any one of claims 1 5, characterized in that it comprises a character code, font type, size as information about the pixel block. Computer
An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collation means, information relating to the pixel block is extracted from the storage unit, and the information relating to the pixel block is provided as an information adding unit that is provided to the image of the pixel block in the received image .
The information generating means
Output means for outputting an image including images of pixel clusters of a plurality of types or sizes;
First reading means for reading an image output by the output means;
A second reading unit that reads the image output by the output unit a plurality of times at a resolution lower than that of the first reading unit;
First cutout means for cutting out an image of a pixel block in the image read by the first reading means;
Second cutout means for cutting out an image of a pixel block in the image read by the second reading means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing program. Computer
An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collation means, information relating to the pixel block is extracted from the storage unit, and the information relating to the pixel block is provided as an information adding unit that is provided to the image of the pixel block in the received image .
The information generating means
First reading means for reading an image;
Second reading means for reading the image a plurality of times at a lower resolution than the first reading means;
First cutout means for cutting out an image of a pixel block in the image read by the first reading means;
Second cutout means for cutting out an image of a pixel block in the image read by the second reading means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing program. Computer
An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collation means, information relating to the pixel block is extracted from the storage unit, and the information relating to the pixel block is provided as an information adding unit that is provided to the image of the pixel block in the received image .
The information generating means
Output means for outputting an image including images of pixel clusters of a plurality of types or sizes;
Reading means for reading an image output by the output means;
Low resolution phase processing means for reducing the resolution of the image read by the reading means and generating images of different phases;
First cutout means for cutting out an image of a pixel block in the image read by the reading means;
Second cutout means for cutting out an image of a pixel block in the image generated by the low resolution phase processing means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing program. Computer
An information generating means for generating an image of a plurality of pixel blocks and information on the pixel block having a higher resolution than the image of the pixel blocks;
Storage means for storing an image of the pixel block generated by the information generation unit and information about the pixel block;
Collating means for collating the image of the pixel block in the received image with the image of the pixel block stored by the storage unit;
Based on the collation result by the collation means, information relating to the pixel block is extracted from the storage unit, and the information relating to the pixel block is provided as an information adding unit that is provided to the image of the pixel block in the received image .
The information generating means
Reading means for reading an image;
Low resolution phase processing means for reducing the resolution of the image read by the reading means and generating images of different phases;
First cutout means for cutting out an image of a pixel block in the image read by the reading means;
Second cutout means for cutting out an image of a pixel block in the image generated by the low resolution phase processing means
Have
The storage unit stores the image of the pixel block cut out by the first cutout unit as information on the pixel block for the pixel block image cut out by the second cutout unit. Image processing program.

Images