JP6922365B2 - Document processing equipment and programs - Google Patents

Description

The present invention relates to a document processing apparatus and a program.

According to Patent Document 1, the image input device of the present invention inputs an image from a scanner device, executes optical character recognition (hereinafter, zone OCR) on a partial region on the input image, and obtains it as an OCR result. It is an image input device that uses the text data as the index data of the document and finally saves the document (image and index data) in the document management system. It is an image input device equipped with a process of correcting erroneous recognition and adopting an appropriate OCR output as index data. First, an OCR area is set in advance on a document for OCR processing for a partial area for creating index data. It has a form registration step to register the document as a form, and also has an image input step to scan the actual manuscript, create an index by OCR, and save the document (image and index) in the document management system. In the input step, in the OCR processing for the partial area on the image, in the index creation processing of the scanned image, the partial area of the scanned image is cut out according to the OCR area setting of the form registered in the previous term form registration step. For the partial area image, the first OCR processing and then the partial area is automatically expanded and the second OCR processing is executed, and the analysis result of the first OCR and the analysis result of the second OCR are displayed. First, as a judgment of condition 1, if the results of comparing the "first OCR result" and the "second OCR result" are the same text data, the "first OCR result" is adopted as the index data, and then As a judgment of condition 2, if the "first OCR result" and the "second OCR result" are different text data and the "second OCR result" does not include a line feed code, the "second OCR result" is determined. "OCR result" is adopted as index data, and finally, as a judgment of condition 3, when "first OCR result" and "second OCR result" are different text data, "second OCR result" If contains a line feed code, the character string of the "second OCR result" is divided by the line feed code, the divided text data is compared with the "first OCR result", and the number of characters and the character string are calculated. Partial image cutout by the judgment processing of condition 1, condition 2, and condition 3, such as adopting the one whose constituent character is close to the "first OCR result" as index data. Described is an image input device characterized by correcting erroneous recognition due to character breakage due to a sword and acquiring index data.

Japanese Unexamined Patent Publication No. 2008-40598

In the character recognition process, the present invention is compared with a technique in which a designated area designated to recognize a character is not expanded to an area including the entire object other than the character, and the character other than the character is around the character to be recognized. It is an object of the present invention to provide a document processing device and a program capable of suppressing a decrease in the character recognition rate due to the existence of the object.

The present invention according to claim 1 expands the designated area to an area including the entire non-character object when an object other than the character exists around the designated area designated to perform character recognition processing. It is a document processing device having an area expansion means.

The present invention according to claim 2 is the document processing apparatus according to claim 1, wherein when an object other than the characters is a ruled line, the area expanding means expands a designated area to an area including the entire ruled line. ..

The present invention according to claim 3 claims that when the ruled lines form a table having at least one square, the area expanding means expands the designated range to an area including all the squares of the table. 2. The document processing apparatus according to 2.

According to the fourth aspect of the present invention, when the ruled line is a dividing line that separates the character writing area in which the characters are written and another area adjacent to the character writing area, the area expanding means is the dividing line. The document processing apparatus according to claim 2, wherein the designated area is expanded to include the entire area.

According to the fifth aspect of the present invention, when the designated area includes only a part of a character string composed of a plurality of characters arranged continuously, the expanding means includes at least the entire character string. The document processing apparatus according to any one of claims 1 to 4, wherein the designated area is expanded to.

According to the sixth aspect of the present invention, when the designated area includes only a part of a plurality of character strings, the expansion means expands the designated area to an area including at least all of the plurality of character strings. The document processing apparatus according to any one of claims 1 to 4.

According to the seventh aspect of the present invention, when there are a plurality of objects other than characters in the vicinity of the designated area, the area expanding means extends the designated area to an area including all of the objects other than the plurality of characters. The document processing apparatus according to any one of claims 1 to 6, which is expanded.

The present invention according to claim 8 expands the designated area to an area including the entire non-character object when there is a non-character object around the designated area designated to perform character recognition processing. A program that causes a computer to perform an area expansion step.

According to the first aspect of the present invention, in the character recognition process, an attempt is made to recognize a character as compared with a technique of not expanding a designated area designated to recognize a character to an area including the entire object other than the character. It is possible to provide a document processing device capable of providing a document processing device capable of suppressing a decrease in the recognition rate of a character due to the existence of an object other than the character in the vicinity of the character to be used.

According to the second aspect of the present invention, as compared with the technique of not expanding the designated area to the area including the entire ruled line, the recognition rate of the character is lowered due to the existence of the ruled line around the character to be recognized. Can be suppressed.

According to the third aspect of the present invention, the recognition rate of characters is lowered due to the presence of the table around the character to be recognized, as compared with the technique of not expanding the designated area to the area including the entire table. Can be suppressed.

According to the fourth aspect of the present invention, the recognition rate of characters due to the existence of a dividing line around the character to be recognized is compared with the technique of not expanding the designated area to the area including the entire dividing line. Can be suppressed.

According to the fifth aspect of the present invention, the number of characters that can be recognized in the character string can be increased as compared with the technique that does not expand the designated area until the entire character string is included.

According to the sixth aspect of the present invention, the number of characters that can be recognized in the character string can be increased as compared with the technique that does not expand the designated area to the area that includes all of the plurality of character strings. ..

According to the seventh aspect of the present invention, as compared with the technique of expanding the designated area to the area including only the entire object other than a part of the character, the object other than the character is formed around the character to be recognized. It is possible to suppress a decrease in the recognition rate of characters due to their existence.

According to the present invention according to claim 7, in the character recognition process, an attempt is made to recognize a character as compared with a technique of not expanding a designated area designated to recognize a character to an area including the entire object other than the character. It is possible to provide a program capable of suppressing a decrease in the recognition rate of a character due to the existence of an object other than the character in the vicinity of the character to be used.

It is a figure which shows the structure of the document processing system used in embodiment of this invention. It is a block diagram which shows the hardware structure of the reading apparatus which the document processing system used in embodiment of this invention has. It is a block diagram which shows the hardware configuration of the server apparatus which the document processing system used in embodiment of this invention has. It is a block diagram which shows the hardware structure of the image forming apparatus which the document processing system used in embodiment of this invention has. It is a block diagram which shows the hardware composition of the terminal apparatus which the document processing system used in embodiment of this invention has. An outline of processing by the document processing system used in the embodiment of the present invention will be described, FIG. 6 (A) is a diagram illustrating a first example of processing, and FIG. 6 (B) is a second processing. 6 (C) is a diagram showing a third example of processing. It is a block diagram which shows the software structure of the document processing system used in embodiment of this invention. It is a flowchart explaining the whole processing in the document processing system used in embodiment of this invention. It is 1st flowchart explaining the process of expanding the designated area in the document processing system used in embodiment of this invention. It is a 2nd flowchart explaining the process which expands the designated area in the document processing system used in embodiment of this invention. It is a flowchart explaining the preprocessing in the document processing system used in embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing an example of the configuration of the document processing system 10 used in the embodiment of the present invention. As shown in FIG. 1, the document processing system 10 has, for example, a plurality of terminal devices 100. As the terminal device 100, for example, a personal computer can be used. Each of the terminal devices 100 has a display unit 102 and an operation unit 104. As the display unit 102, for example, a CRT (Cathode Ray Tube) display device or a liquid crystal display device can be used. As the operation unit 104, for example, a keyboard, a mouse, or the like can be used. The details of the terminal device 100 will be described later.

The document processing system 10 further includes a server device 300. Details of the server device 300 will be described later.

The document processing system 10 further includes a reading device 400. The scanning device 400 is a so-called scanner, which optically reads a document and converts the optically read document into an electronic document (electronic file). The details of the reading device 400 will be described later.

The document processing system 10 further includes an image forming apparatus 500. The image forming apparatus 500 is a so-called multifunction device, and can perform image forming on a recording medium such as copying a document or printing a document file, as well as reading a document and transmitting a facsimile. Here, as the recording medium, for example, plain paper can be used. The image forming apparatus 500 has a document reading unit 502. The document reading unit 502 optically reads the document and converts the optically read document into an electronic document. The details of the image forming apparatus 500 will be described later.

The terminal device 100, the server device 300, the reading device 400, and the image forming device 500 are connected to each other via the network 600, and can communicate with each other via the network 600.

FIG. 2 is a block diagram showing a hardware configuration of the reader 400. As shown in FIG. 2, the reading device 400 has at least a CPU 410, a memory 412, a storage device 416, a communication interface 418, a scan driving unit 420, and an operating unit 422, which are via a bus. They are connected to each other and can communicate with each other.

As the storage device 416, for example, a hard disk drive (HDD) or the like can be used. The communication interface 418 transmits and receives data to and from an external device such as the server device 300 described above via the network 600 described above. The CPU 410 executes processing based on a program stored in the memory 412 or the storage device 416 to control the operation of the reading device 400.

FIG. 3 is a block diagram showing a hardware configuration of the server device 300. As shown in FIG. 3, the server device 300 has a CPU 310, a memory 312, a storage device 316, and a communication interface 318, which are connected to each other via a bus and can communicate with each other. It has become.

As the storage device 316, for example, a hard disk drive (HDD) or the like can be used. Further, the storage device 316 receives the electronic document created by the terminal device 100, the electronic document read by the reading device 400, the electronic document read by the image forming device 500, and the image forming device 500 as a facsimile. An electronic document or the like may be stored.

The communication interface 318 transmits and receives data to and from an external device such as the terminal device 100 described above via the network 600 described above. The CPU 310 executes processing based on a program stored in the memory 312 or the storage device 316 to control the operation of the server device 300.

FIG. 4 is a block diagram showing a hardware configuration of the image forming apparatus 500. As shown in FIG. 4, the image forming apparatus 500 includes a CPU 510, a memory 512, a storage device 516, a communication interface 518, a scan driving unit 520, an operation unit 522, and an image forming unit 524. These are connected to each other via a bus so that they can communicate with each other.

As the storage device 516, for example, a hard disk drive (HDD) or the like can be used. The communication interface 518 transmits and receives data to and from an external device such as the server device 300 described above via the network 600 described above. The CPU 510 controls the operation of the image forming apparatus 500 by executing processing based on the program stored in the memory 512 or the storage device 516. The scan drive unit 520 drives the document reading unit 510 described above. The image forming unit 524 uses, for example, an electrophotographic method, an inkjet method, or the like, and forms an image on a recording medium.

FIG. 5 is a block diagram showing a hardware configuration of the terminal device 100. As shown in FIG. 5, the terminal device 100 has at least a CPU 110, a memory 112, a storage device 116, a communication interface 118, the above-mentioned display unit 102, and the above-mentioned operation unit 104, and these are buses. They are connected to each other via and can communicate with each other. As the storage device 116, for example, a hard disk drive (HDD) or the like can be used. The communication interface 118 transmits and receives data to and from an external device such as the server device 300 described above via the network 600 described above.

The program is stored in the memory 112 or the storage device 116. More specifically, the memory 112 or the storage device 116 stores a program 900 (see FIG. 7) according to the present invention, which is a program for document processing and a program for character recognition. The CPU 110 executes processing based on the program 900 stored in the memory 112 or the storage device 116 to control the operation of the terminal device 100. More specifically, the CPU 110 controls the terminal device 100 to perform document processing, controls the terminal device 100 to perform character recognition, and controls the terminal device 100 to execute the program 900 according to the present invention. Let me. As described above, the terminal device 100 corresponds to the document processing device described in the claims. The details of the program 900 will be described later.

In this embodiment, the memory 112 or the storage device 116 stores the program 900 according to the present invention, and the terminal device 100 is used as the document processing device described in the claims, but the memory 412 or the storage device 416 of the present invention is used. The program 900 according to the above may be stored, and the reading device 400 may be used as the document processing device described in the claims. Further, the program 900 of the present invention may be stored in the memory 312 or the storage device 316, and the server device 300 may be used as the document processing device described in the claims. Further, the program 900 of the present invention may be stored in the memory 512 or the storage device 516, and the image forming device 500 may be used as the document processing device described in the claims.

6A and 6B are views for explaining the outline of the processing of the program 900, FIG. 6A illustrates the first example of the processing by the program 900, and FIG. 6B shows the second example of the processing by the program 900. An example is described, and FIG. 6C illustrates a third example by program 900.

The program 900 is a processing program related to optical character recognition, which is also called OCR (Optical Character Recognition / Reader), and as shown in FIG. 6, a designated area designated in advance to perform optical character recognition is set. It is a program that expands under certain conditions. In FIG. 6, the designated area before being expanded by the program 900 (hereinafter referred to as the designated area before expansion R1) is shown by a thick alternate long and short dash line, and the designated area after being expanded by the program 900 (hereinafter referred to as after expansion). The designated area R2) is indicated by a thick dotted line.

More specifically, the program 900 describes a character when an object other than a character exists in the vicinity of the pre-expansion designated area R1, that is, at least one of the inside of the pre-expansion designated area R1 and the outside of the pre-expansion designated area R1. The pre-expansion designated area R1 is expanded to the post-expansion designated area R2 to the area including the entire object other than the above. Here, examples of objects other than characters include images such as arrows and ruled lines. Then, as the ruled line, for example, a table having at least one square or a dividing line can be mentioned. Here, the dividing line means a line segment that separates a character description area in which characters are described and another area adjacent to the above-mentioned character description area. Hereinafter, the case where the object other than the sentence is the ruled line L, the case where the ruled line L forms a table, and the case where the ruled line L is a dividing line will be described as an example.

In the first example shown in FIG. 6A, there is a character "1,000" in the pre-expansion designated area R1 and a character "amount" in the upper part of the pre-expansion designated area R1. There is a character "2,590" at the bottom of the pre-designated area R1. Further, the objects other than the characters around the designated area R1 before enlargement are ruled lines L1, and the ruled lines L1 form a table of 3 horizontal squares and 3 vertical squares. In this first example, the program 900 expands the designated area from the pre-expansion designated area R1 to the post-expansion designated area R2 to the area covered by the designated area (9 squares) in the table. ..

In the second example shown in FIG. 6B, there is a character "1,000" in the pre-expansion designated area R1. Further, the objects other than the characters around the pre-expansion designated area R1 are the ruled line L2, and the ruled line L2 is the dividing line. In this second example, the program 900 expands the designated area from the pre-expansion designated area R1 to the post-expansion designated area R2 to the area where the designated area covers the entire ruled line L2.

In the third example shown in FIG. 6C, all of the character strings "1,000" are arranged in the pre-expansion designated area R1. That is, the pre-expansion designated area R1 includes all of the character strings "1,000" composed of a plurality of characters arranged consecutively. Further, in the third example shown in FIG. 6C, a part of the character string "2,000" is arranged in the pre-expansion designated area R1. That is, the pre-expansion designated area R1 includes only a part of the character string "2,000" composed of a plurality of characters arranged consecutively.

The objects other than the characters around the pre-expansion designated area R1 are the ruled line L3 and the ruled line L4. As described above, in this third example, there are a plurality of objects other than the (two) characters around the pre-expansion designated area R1. The ruled line L3 forms a table composed of one square, and the ruled line L4 also forms a table composed of one square.

In this third example, the program 900 specifies up to an area in which the entire character string "2,000", of which only a part is included in the pre-expansion designated area R1, is included in the post-expansion designated area R2. The area is expanded from the designated area R1 before expansion to the designated area R2 after expansion. Further, in the third example, the program 900 specifies that the designated area includes all of a plurality of objects (a table formed by the ruled line L3 and a table formed by the ruled line L4) which are objects other than characters. The area is expanded from the designated area R1 before expansion to the designated area R2 after expansion.

Furthermore, in this third example, the program 900 is a part of a plurality of character strings of the character string "1,000" and the character string "2,000", "1,000". Only the entire character string with "" is included in the pre-expansion designated area R1, and the entire character string with "2,000", which is another part of the plurality of character strings, is included in the pre-expansion designated area. Designated after expanding the designated area so that it covers all of the plurality of character strings of the character string "1,000" and the character string "2,000" from the state not included in R1. It is expanded to the area R2.

The specific algorithm of the program 900 will be described later.

FIG. 7 is a block diagram showing the software configuration of the document processing system 10, and is a block diagram showing the configuration of the program 900 of the present invention. As shown in FIG. 7, the program 900 has a reception unit 910. The reception unit 910 receives an electronic document from the reading device 400 or the document reading unit 502.

The program 900 further includes an area expansion processing unit 912. The area expansion processing unit 912 is designated to perform character recognition as described above by showing a specific example (see FIGS. 6 (A), 6 (B), and 6 (C)). When an object other than characters exists around the designated area, the designated area is expanded from the pre-expansion designated area R1 to the post-expansion designated area R2 to the area where the entire non-character object is included in the designated area.

The program 900 further includes a preprocessing unit 914. The pre-processing unit 914 performs pre-processing before the character recognition processing unit 916, which will be described later, recognizes a character. More specifically, the reception unit 910 performs a process of removing objects other than characters such as ruled lines from the electronic document received by the reception unit 910 and the electronic document whose designated area is expanded by the area expansion processing unit 912. .. As described above, the preprocessing unit 914 is used as a removing means for removing objects other than characters from the designated area expanded by the area expansion processing unit 912.

In addition to removing objects other than characters, the preprocessing unit 914 removes noise that is multiplied when reading the document, and eliminates the tilt that occurs because the document is read in a tilted state. Skew correction (tilt elimination processing) for this is performed.

The program 900 further includes a character recognition processing unit 916. The character recognition processing unit 916 performs processing for recognizing characters from the electronic document received from the preprocessing unit 914.

The program 900 further includes an output unit 918. The output unit 918 outputs the electronic document received from the character recognition processing unit 916 to an external device such as the server device 300 or the image forming device 500.

FIG. 8 is a flowchart showing an example of the operation of the document processing system 10. As shown in FIG. 8, when a series of operations is started, in step S100, the CPU 110 controls the reception unit 910 and causes the reception unit 910 to accept the electronic document.

In step S200, which is the next step, the CPU 110 controls the area expansion processing unit 912, and the area expansion processing unit 912 has an object other than characters around the designated area designated to perform character recognition. In this case, the designated area is expanded from the pre-expansion designated area R1 to the post-expansion designated area R2 to the area where the entire object other than the characters is included in the designated area.

In step S300, which is the next step, the CPU 110 controls the preprocessing unit 914 to cause the preprocessing unit 914 to perform preprocessing. That is, the CPU 110 controls the preprocessing unit 914 to cause the preprocessing unit 914 to remove objects other than characters from the electronic document, to remove noise, and to perform skew correction.

In step S500, which is the next step, the CPU 110 controls the character recognition processing unit 916 to cause the character recognition processing unit 916 to recognize characters from the electronic document preprocessed in step S300.

In step S600, which is the next step, the CPU 110 controls the output unit 918, and sends the electronic document whose characters are recognized in step S500 to the output unit 918 to an external device such as the server device 300 or the image forming device 500. Is output.

FIG. 9 is a flowchart showing an example of details of the area expansion process in step S200, and is a case where there is a ruled line around the pre-expansion designated area R1 shown in FIGS. 6 (A) and 6 (B). It is a flowchart which shows the detail of a process. When the area expansion process in step S200 is started, the CPU 110 tells the area expansion processing unit 912 in step S202 from the center of the pre-expansion designated area R1 in both the left and right directions, for example, an object other than characters such as a ruled line (hereinafter, simply). The search for "ruled lines" (also simply referred to as "ruled lines" in FIGS. 9 and 10) is performed.

In the next step S204, as a result of the search to the left and right in step S202, it is determined whether or not there is a ruled line, and if there is a ruled line, the process proceeds to step S206, and if there is no ruled line, the process proceeds to step S250.

In step S206, the CPU 110 causes the area expansion processing unit 912 to investigate the rectangle of the ruled line searched in step S202. That is, the CPU 110 causes the area expansion processing unit 912 to extract the coordinates of the ruled line.

In step S208, which is the next step, it is determined whether or not the ruled line searched in step S202 includes the entire pre-expansion designated area R1 by using the coordinates of the ruled line extracted in step S206, and step S208. If it is determined that the ruled line searched in step 2 includes the entire pre-expansion designated area R1, the process proceeds to step S218, and the ruled line searched in step S208 includes the entire pre-expansion designated area R1. If it is determined that the case is not present, the process proceeds to step S210.

In step S210, it is determined whether or not the ruled line searched in step S202 includes a part of the pre-expansion designated area R1 by using the coordinates of the ruled line extracted in step S206, and the ruled line searched in step S208 is determined. If it is determined that a part of the pre-expansion designated area R1 is included, the process proceeds to step S212, and it is determined that the ruled line searched in step S208 does not include a part of the pre-expansion designated area R1. If is done, the process proceeds to step S220.

In step S212, the ruled line determined in step S210 to include a part of the pre-expansion designated area R1 is registered as a primary candidate for subsequent processing.

In step S214, which is the next step, it is determined whether or not there is a ruled line searched in step S202 in addition to the ruled line registered as a temporary candidate in step S212, and the ruled line registered as a temporary candidate in step S212 is determined. If it is determined that there is another ruled line, the process returns to step S206, and the process described above is repeated for the other ruled line. On the other hand, if it is determined that there is no ruled line searched in step S202 other than the ruled line registered as a temporary candidate in step S212, the process proceeds to step S216.

In step S216, the ruled lines registered so far are selected as temporary candidates in step S212 for subsequent processing.

In step S250, the CPU 110 causes the area expansion processing unit 912 to search for ruled lines from the center of the pre-expansion designated area R1 in both the upper and lower directions. The details of searching for ruled lines in the vertical direction will be described later (see FIG. 10).

In step S280, which is the next step, the search result up and down in step S250 determines whether or not there is a ruled line, and if it is determined that there is a ruled line, the process proceeds to step S218, and the up and down in step S250. As a result of the search for, if it is determined that there is no ruled line, the process proceeds to step S220.

In step S218, the designated area is expanded to an area including the entire ruled line, and the designated area is expanded from the pre-expansion designated area R1 to the post-expansion designated area R2. That is, the designated area is expanded to the size of the ruled line determined to include the entire designated area in step S208, or the designated area is expanded to the area including all the ruled lines selected in step S216. Or, the designated area is expanded to the area including the entire ruled line searched in step S250.

In step S220, the designated area is not expanded. That is, in the subsequent step, the pre-expansion designated area R1 is preprocessed (see step S300 in FIG. 8), and further, the character recognition process is performed (see step S500 in FIG. 8).

FIG. 10 is a flowchart showing the details of the search for the vertical ruled lines in step S250 and the discrimination in step S280. When the search for the upper and lower ruled lines in step S250 is started, the CPU 110 causes the area expansion processing unit 912 to search the ruled lines from the center of the designated area to the upper side in step S252.

In step S254, which is the next step, as a result of the upward search in step S252, it is determined whether or not there is a ruled line, and if there is a ruled line, the process proceeds to step S256, and if there is no ruled line, step S258. Proceed to.

In step S256, the CPU 110 causes the area expansion processing unit 912 to determine the upper end side when expanding the designated area so as to expand the designated area to the area including the ruled line searched in step S253.

In step S258, the CPU 110 causes the area expansion processing unit 912 to search for ruled lines from the center of the designated area to the lower side.

In step S260, which is the next step, as a result of the downward search in step S258, it is determined whether or not there is a ruled line, and if there is a ruled line, the process proceeds to step S262, and if there is no ruled line, step S284. Proceed to.

In step S262, the CPU 110 causes the area expansion processing unit 912 to determine the lower end side when expanding the designated area so as to expand the designated area to the area including the ruled line searched in step S258.

In step S282, which is the next step, it is assumed that at least one of the upper side and the lower side of the designated area has a ruled line, and the process proceeds to the next step S218 (see FIG. 9). At this time, in step S218, the pre-expansion designated area R1 is set based on the upper end portion when expanding the designated area determined in step S256 and the lower end portion when expanding the designated area determined in step S262. After expansion, it is expanded to the designated area R2.

In step S284, it is assumed that there is no ruled line on either the upper side or the lower side of the designated area, and the process proceeds to the next step S220 (see FIG. 9).

In the above description, the expansion of the designated area has been described by taking as an example the expansion of the designated area when there is a ruled line around the designated area R1 before expansion, but a character string consisting of a plurality of characters arranged continuously has been described. When the pre-expansion designated area R1 includes only a part of the above, or when the pre-expansion designated area R1 includes only a part of a plurality of character strings, it is the same as when there is a ruled line around the pre-expansion designated area R1. The designated area is expanded.

Further, the expansion of the designated area when there is a ruled line around the designated area R1 before expansion, and the designation when the designated area R1 before expansion includes only a part of the character string composed of a plurality of characters arranged continuously. The expansion of the area and the expansion of the designated area when the pre-expansion designated area R1 includes only a part of the plurality of character strings are superimposed, for example, performed at the same time.

FIG. 11 is a flowchart showing the details of the preprocessing in step S300 described above. When the preprocessing in step S300 is started, the CPU 110 causes the preprocessing unit 914 to remove the noise multiplied by, for example, when reading the original in step S302.

In step S304, which is the next step, the CPU 110 causes the preprocessing unit 914 to perform skew correction (tilt elimination processing) for eliminating the inclination caused by reasons such as the original being read in an inclined state.

In step S306, which is the next step, it is determined whether or not there is a vertical ruled line in the designated area R2 after enlargement, and if it is determined that there is a vertical ruled line, the process proceeds to step S310 and the vertical ruled line is determined. If it is determined that there is no such thing, the process proceeds to step S308.

In step S308, it is determined whether or not there is a horizontal ruled line in the designated area R2 after enlargement, and if it is determined that there is a horizontal ruled line, the process proceeds to step S310 and it is determined that there is no horizontal ruled line. If is done, a series of preprocessing is completed.

In step S310, the box structure analysis in the enlarged designated area R2 is performed by using at least one of the information such as the coordinates of the vertical ruled line in the enlarged designated area R2 and the information such as the coordinates of the horizontal ruled line. The character recognition process of step S500 is performed for each area (for example, each square in the table) separated by the ruled line. In this way, by making the character recognition performed for each area delimited by the ruled line, the area expansion processing unit 912 enlarges the character recognition process prior to the character recognition processing performed by the character recognition processing unit 916 in step S500. The ruled line, which is an example of an object other than characters, is removed from the specified area.

10: Document processing system 100: Terminal device 100: Terminal device 300: Server device 400: Reading device 500: Image forming device 900: Program 910: Reception unit 912: Area expansion processing unit 914: Preprocessing unit 916: Character recognition processing unit 918: Output unit L: Ruled line R1: Designated area before enlargement R2: Designated area after enlargement

Claims (8)

When there are non-character objects around the specified area specified for character recognition processing, an area expansion means that expands the specified area to a processing area that includes the entire non-character object, and an area expansion means .
A removing means for removing objects other than characters from the image of the processing area enlarged by the area expanding means,
A character recognition means that performs character recognition processing of an image in which objects other than characters have been removed by the removal means, and
Document processing device with. When the object other than the above characters is a ruled line,
The document processing apparatus according to claim 1, wherein the area expanding means expands a designated area to an area including the entire ruled line. If the ruled lines form a table with at least one square,
The document processing apparatus according to claim 2 , wherein the area expansion means expands a designated area to an area including all the squares in the table. When the ruled line is a delimiter line that separates the character description area in which characters are described from another area adjacent to the character description area.
The document processing apparatus according to claim 2, wherein the area expanding means expands a designated area to an area including the entire dividing line. When the specified area contains only a part of a character string consisting of a plurality of characters arranged consecutively.
The document processing apparatus according to any one of claims 1 to 4, wherein the area expanding means expands a designated area to at least an area including the entire character string. When the specified area contains only a part of multiple character strings
The document processing apparatus according to any one of claims 1 to 4, wherein the area expanding means expands a designated area to an area including all of at least a plurality of character strings. When there are multiple non-character objects around the specified area
The document processing apparatus according to any one of claims 1 to 6, wherein the area expanding means expands a designated area to an area including the entire object other than a plurality of characters. When there are non-character objects around the specified area specified for character recognition processing, the area expansion step to expand the specified area to the area that includes the entire non-character object, and the area expansion step.
A removal step that removes objects other than characters from the image of the processing area enlarged by the area expansion step, and
A character recognition step for performing character recognition processing on an image in which objects other than characters have been removed by the removal step, and
A program that causes a computer to run.

Images