JP4157833B2 - Image processing apparatus, image processing method, program, and recording medium - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, a program, and a recording medium, and more particularly to processing for correcting distortion of a read image when a document is a sheet bundle that is stopped with a book or staples.

Many originals read using a scanner are sheet-like originals, and a pressure plate that can be freely opened and closed is provided on a contact glass. After placing the original on the contact glass, the pressure plate is closed to read the original.
However, the original is not limited to a sheet, and a book original (a book, a booklet, a bundle of stapled sheets, etc.) may be handled as an original. In such a case, the book original is placed on the contact glass. The document will be read.
However, when a book document is used as the document, the page binding portion 101 of the book document 100 is lifted from the contact glass 102 as shown in FIG. 7, and the read image of the floating binding portion is distorted. There was a problem that characters and ruled lines were blurred due to shadows and shadows.
A technique for correcting distortion of a read image when a book document is used as a document in this way is disclosed in, for example, Patent Document 1.
According to the technique disclosed in Patent Document 1, a read image is binarized, a circumscribed rectangle of a character is extracted from the binarized image, a change in shape of the extracted character circumscribed rectangle, and a background of the read image The distortion amount of the read image is estimated using the change in density, and the enlargement process is performed on the read image or the binarized image based on the estimated distortion amount.
As a result, even when the page binding portion of the book document is perpendicular to the main scanning direction of the scan, the character circumscribed rectangle changes, so that the read image of the book document can be corrected.

However, the method for correcting the distortion of the binding portion of the book document or the like as described above may require time for processing or may fail in correction.
In addition, in a region identification method (for example, see Patent Document 2) that extracts a region such as a character, a figure, a photo, or a table from a document image when distortion or shadow occurs in the binding portion, the extraction accuracy of the character region or the like is high. There was a problem of getting worse.
In addition, when a sheet bundle stopped with staples (stoppers) is read, the image near the staples (stoppers) is distorted, or characters, figures, photos, tables, etc. in the document with staples (stoppers) There is a problem that is hidden.
A technique for removing staples (stoppers) from such a sheet bundle is disclosed in, for example, Patent Document 3 and the like.
According to the technique disclosed in Patent Document 3, a stapled sheet bundle placed on a sheet placing table is fed, and staples on the sheet bundle are detected, and the detected staple is at a staple removal position. Then, the sheet bundle is moved so as to stop, and the staples on the moved sheet bundle are removed from the sheet bundle.
As a result, the staples can be automatically removed from the stapled sheet bundle, so that the staples can be removed without requiring manual operation.
However, this staple removing method also has problems that it takes time for the processing and that the original is damaged in order to remove the staples.
JP 2002-247348 A Patent No. 2940936 Japanese Patent No. 3236376

  The present invention has been made in consideration of the above-described circumstances. When extracting a region such as a character, a figure, or a photograph of an image read from a book document or a stapled sheet bundle, the region identification target in the read image is determined. An image processing apparatus, an image processing method, and an image processing apparatus that control to perform distortion correction and staple removal only when the area is in the page binding portion or the area identification target area is an image distorted by staples, It is an object of the present invention to provide a program for executing the functions of an image processing apparatus and a computer-readable recording medium on which the program is recorded.

In order to solve the above-described problems, an image processing apparatus according to a first aspect of the present invention corrects a distortion in the shape of an image read by an image reading unit that reads an original image to obtain image information without distortion. In the apparatus, the region identifying means for identifying the character of the image information and the region of the figure / photo , and the distortion magnitude of the region identified by the region identifying means is greater than a predetermined threshold A distortion area detecting means for detecting a distortion area by determining a portion where the magnitude of the distortion is larger than a predetermined threshold as a distortion area, and a character or a character in the distortion area detected by the distortion area detection means. FIG Photography contain regions, and a staple detecting means for detecting whether there is a staple in the strain region, if the presence of the staple is detected by the staple detecting means, the stearate And a staple removing means for removing the pull, characterized after performing staples removed, that performs area identification process for the read again the document image and so as to identify the areas of text and diagrams, photographs And

According to a second aspect of the present invention, there is provided an image processing apparatus that corrects distortion of a shape of an image read by an image reading unit that reads an original image and obtains image information without distortion. A region identifying unit for identifying a region of a character, a figure or a photo, and a region including a character among the regions identified by the region identifying unit , where the distortion is greater than a predetermined threshold, includes the magnitude of strain and strain region detection means for detecting a decision to strained region and strained region larger portion than a predetermined threshold value, said strained region characters and figures, photos area to the distortion area detected by the detection means And a staple detecting means for detecting whether or not there is a staple in the distorted area, and a stay for removing the staple when the staple detecting means detects the presence of the staple. And a distortion correction unit that corrects distortion of the shape of the image when no staple is present by the staple detection unit. When the staple is detected by the staple detection unit, the staple removal is performed. Then, the area identification process is performed on the re-read original image. On the other hand, if there is no staple, the distortion correction process is performed, and the area identification process is performed again on the image from which the distortion is removed. It is characterized in that it has been made to identify areas of characters and figures / photos .

According to a third aspect of the present invention, there is provided an image processing method for correcting image information read by an image reading means for reading a document image to obtain image information without distortion , wherein the image information A region identification step for identifying a region of a character, a figure or a photo, and a region including a character among the regions identified by the region identification step, where a distortion is greater than a predetermined threshold, A distortion area detection step of detecting a distortion area by determining a distortion area where the magnitude of the distortion is larger than a predetermined threshold as a distortion area, and the distortion area detected in the distortion area detection process includes a character, a figure, or a photograph area. And a staple detection step for detecting whether or not there is a staple in the distorted region, and when the presence of the staple is detected in the staple detection step, the staple is removed. Et al, performs area identification processing on a document image is read again, characterized by being adapted to identify areas of text and diagrams and photographs.
According to a fourth aspect of the present invention, there is provided an image processing method for correcting image shape distortion read by an image reading means for reading a document image to obtain image information without distortion . A region identification step for identifying a region of a character, a figure or a photo, and a region including a character among the regions identified by the region identification step, where a distortion is greater than a predetermined threshold, A distortion area detection step of detecting a distortion area by determining a distortion area where the magnitude of the distortion is larger than a predetermined threshold as a distortion area, and the distortion area detected in the distortion area detection process includes a character, a figure, or a photograph area. And a staple detection step for detecting whether or not there is a staple in the distortion region, and a distortion for correcting distortion of the shape of the image when no staple is present in the staple detection step. A normal process, and when the presence of staples is detected in the staple detection process, the staples are removed, the region image is re-read on the original image, and the staples are not present. Then, the distortion correction processing is performed, and the region identification processing is performed again on the image from which the distortion has been removed, so that the region of the character or the figure / photograph is identified .
According to a fifth aspect of the present invention, there is provided a program for causing a computer to execute each step of the image processing method according to the third or fourth aspect .
A recording medium according to a sixth aspect of the invention is a computer-readable recording medium in which the program according to the fifth aspect is recorded.

According to the present invention, when a book document or a bundle of stapled sheets is read, and a region identification process is performed to extract a character, a figure, a photograph, or the like of the read image, a distortion region of the read image is detected. However, since it is determined whether or not this distortion area includes an area such as a character, figure, or photo that is subject to area identification processing, the subsequent processing is switched. Since correction processing and staple removal processing are not performed when an area such as a photograph is not included, the processing speed of the entire image processing is improved.
In addition, if there is a character, figure, photo, or other area subject to area identification processing in the distortion area, the area identification process is performed again after performing distortion correction processing or staple removal processing. In addition, the accuracy of area identification can be improved.

Hereinafter, preferred embodiments according to an image processing apparatus of the present invention will be described with reference to the drawings.
<Embodiment 1>
FIG. 1 is a block diagram illustrating a functional configuration of the image processing apparatus according to the first embodiment. In the figure, an image processing device reads an image reading unit 10 that reads a book placed on a contact glass or a sheet bundle bound with staples and stores image information in an image information storage unit 20. Area identification means 30 for identifying areas such as characters, figures and photographs, a distortion area detection means 40 for detecting a distortion area using area information identified by the area identification means 30, and a distortion area detection means 40 If an area such as a character, a figure, or a photo overlaps in the distortion area, the image forming apparatus includes a distortion correction unit 50 that corrects and updates the distortion of the image information stored in the image information storage unit 20.
Further, instead of obtaining image information from the original by the image reading unit 10, the image information may be prepared in the image information storage unit 20 in advance.
The area identifying means 30 extracts a circumscribed rectangle constituting a character from the document image stored in the image information storage means 20, and uses information such as the arrangement and size of the rectangle to identify the area such as a character, a figure, and a photograph. To store the region position, range and region type. As a method for area identification, a known technique, for example, a method described in Japanese Patent No. 2940936, Japanese Patent Laid-Open No. 7-322061, or the like is used.

The distortion area detection means 40 is an original bound by a book or a staple, so that it is distorted at the periphery of the page binding portion as shown in FIG. 2A, or when an image near the staple is distorted. For this purpose, the position and range of distortion occurring are detected by a known technique, for example, a method described in JP-A-2002-247348.
For this purpose, for example, a circumscribed rectangle constituting the character is cut out from the document image information with respect to the region including the character among the regions identified by the region identifying means 30 (see FIG. 3).
Here, excluding those with extremely small or large circumscribed rectangles, calculate (the length of the side of the circumscribed rectangle) / (length of the vertical side of the circumscribed rectangle) for the circumscribed rectangle of each character. For each line, an average value is calculated and used as a feature value for each line.
A portion where the feature amount of each row becomes larger than a predetermined threshold is set as a boundary, and a portion larger than the predetermined threshold is determined as a distortion region (see FIG. 2A).
Next, as shown in FIG. 2B, the overlap between the strain area obtained by the strain area detection means 40 and the area obtained by the area identification means 30 is checked. If there is no overlap, the area identification is performed first. The identification result obtained by the means 30 is used as it is.
However, as shown in FIG. 2C, when the distortion region and the region obtained by the region identification unit 30 overlap, the distortion correction unit 50 reduces the distortion of the image information stored in the image information storage unit 20. Correction is performed, the image information in the image information storage means 20 is updated, and the region identification process is performed again on the corrected image information.
This distortion correction means 50 corrects by a known technique, for example, a method described in JP-A-2002-247348.

FIG. 4 is a flowchart illustrating a processing flow of the image processing apparatus according to the first embodiment.
The book document placed on the contact glass is read and image information is stored in the image information storage means 20 (step S10).
Based on the image information stored in the image information storage means 20, a known technique, for example, a method described in Japanese Patent No. 2940936, Japanese Patent Laid-Open No. 7-322061, or the like is used. The region is identified, and the region position, range, and region type are stored (step S11).
Based on the image information stored in the image information storage means 20, the position and range of the distortion are detected by a known technique, for example, the method described in Japanese Patent Application Laid-Open No. 2002-247348, etc. If there is no overlap (NO in step S12), the region identification result obtained by the region identification unit 30 is used as it is. .
On the other hand, when there is an area overlapping with the distortion area (YES in step S12), the distortion of the image information stored in the image information storage unit 20 is described in a known technique, for example, JP-A-2002-247348. The data is corrected and updated by the method (step S13).
Based on the distortion-corrected image information, areas such as characters, drawings / photos are identified again, and the position, range, and type of area are stored (step S14).
<Embodiment 2>

In the first embodiment, an original such as a book is targeted. However, in the second embodiment, a sheet bundle bound by staples is also targeted, and when a staple is detected in the distorted area, the staple is removed. Then, the image is read again and the region identification process is re-executed.
FIG. 5 is a block diagram illustrating a functional configuration of the image processing apparatus according to the second embodiment. In the figure, an image processing device reads an image reading unit 10 that reads a book placed on a contact glass or a sheet bundle bound with staples and stores image information in an image information storage unit 20. Area identification means 30 for identifying areas such as characters, figures and photographs, a distortion area detection means 40 for detecting a distortion area using area information identified by the area identification means 30, and a distortion area detection means 40 When the distortion area includes an area such as a character, a figure, a photograph, and the like, and the staple detection means 60 that detects whether or not there is a staple in the distortion area, the staple detection means 60 detects the presence of the staple. If no staples are present in the staple removing means 70 and the staple detecting means 60, the image information storage means 20 stores the staples. Consists distortion correcting unit 50 for updating and correcting distortion of the image information.

In FIG. 5, the same functions as those in the first embodiment are denoted by the same reference numerals, and only differences will be described below.
For example, the staple detection unit 60 detects the position of a document on a pressure sensitive conductive member, detects the presence or absence of staples as a convex portion on the document, and identifies the region identified by the region identification unit 30, such as a character, a figure, or a photograph. And whether the detected staple position is included in the strain area detected by the strain area detection means 40.
The position of the staple is detected by a known technique, for example, a method described in Japanese Patent No. 3236376.
In the staple detection means 60, when both the character, figure, photo, etc. area identified by the area identification means 30 and the detected staple position are included in the distortion area detected by the distortion area detection means 40. Then, the staple removing unit 70 removes the staples, the staples are removed, the original without distortion is read again by the image reading unit 10, and the region identification process is performed again.
The staple removing means 70 is removed by a known technique, for example, a method described in Japanese Patent No. 3236376.
As described above, the region identification process is performed again with the staples removed and no distortion, so that the region identification accuracy is improved.
On the other hand, in the staple detection means 60, the area identified by the area identification means 30 is included in the distortion area detected by the distortion area detection means 40, but the detected staple position is the distortion area. If not, the distortion correction means 50 corrects the distortion of the image and updates the image information storage means 20 to perform the area identification process again in the absence of distortion to improve the area identification accuracy.

FIG. 6 is a flowchart illustrating a processing flow of the image processing apparatus according to the second embodiment.
A book placed on the contact glass or a sheet bundle bound with staples is read and image information is stored in the image information storage means 20 (step S20).
Based on the original image stored in the image information storage means 20, a known technique, for example, a method described in Japanese Patent No. 2940936, Japanese Patent Laid-Open No. 7-322061, or the like is used. The region is identified, and the region position, range, and region type are stored (step S21).
Based on the image information stored in the image information storage means 20, the position and range of the distortion are detected by a known technique, for example, the method described in Japanese Patent Application Laid-Open No. 2002-247348, etc. The overlap with the region obtained by the region identification process is checked, and if there is no overlap (NO in step S22), the region identification result obtained earlier is used as it is.
On the other hand, when there is an area overlapping with the distortion area (YES in step S22), the staple position is detected (step S23).
If the detected staple is within the distorted area (YES in step S24), the staple is removed by a known technique, for example, a method described in Japanese Patent No. 3236376 (step S25), and step S20 is used to read the original again. Return to.
On the other hand, when the detected staple is not in the distortion area (NO in step S24), the distortion of the image information stored in the image information storage unit 20 is described in a known technique, for example, Japanese Patent Application Laid-Open No. 2002-247348. The data is corrected and updated by the method (step S26).
Based on the distortion-corrected image information, areas such as characters, drawings and photographs are identified again, and the position, range and area type of the area are stored (step S27).
By configuring as described above, when a character area or the like is included in the distorted area of the read document image, the distortion of the obtained area is corrected or the staple is removed, and the distortion is not generated. The accuracy of area identification can be improved by performing re-reading as an original and performing area identification processing.
In addition, although the distortion area | region detection means 40 of Embodiment 1 and 2 mentioned above used the result of area | region identification by the area | region identification means 30, it was memorize | stored in the image information storage means 20 independently of the area | region identification means 30. For the image information, the position and range of the distortion region may be detected by a method described in JP-A-2002-247348.

The present invention is not limited only to the above-described embodiments. Each function constituting the image processing apparatus of the above-described embodiment is programmed, written in advance in a recording medium such as a ROM, the ROM is installed in the image processing apparatus, and these programs are executed by a microprocessor. Needless to say, the object of the present invention is achieved.
In this case, the state read and executed from the recording medium realizes the functions of the above-described embodiment, and the program and the recording medium on which the program is recorded also constitute the present invention.
Note that a program that realizes such a function includes a semiconductor medium (eg, ROM, nonvolatile memory, etc.), an optical medium (eg, DVD, MO, MD, CD, etc.), a magnetic medium (eg, magnetic tape, flexible disk, etc.). Etc.) may be provided in any form of recording medium.
Alternatively, a program that realizes the functions of the image processing apparatus stored in the storage device via a network or the like may be directly supplied from the server computer. In this case, the storage device of this server computer is also included in the recording medium of the present invention.
When provided by such a recording medium, the function of the above-described embodiment is obtained by reading the program from the recording medium and installing the program in the internal storage device or the external storage device, and executing the installed program by the microprocessor. Is realized. Alternatively, the program recorded on the recording medium may be directly executed.

FIG. 2 is a block diagram showing a functional configuration of the image processing apparatus according to the first embodiment. The figure for demonstrating the relationship between a distortion area | region and a character, figure, photograph area | region, etc. FIG. The figure for demonstrating the distortion area | region detection using the external shape rectangle of a character. 3 is a flowchart illustrating a processing flow of the image processing apparatus according to the first embodiment. FIG. 4 is a block diagram illustrating a functional configuration of an image processing apparatus according to a second embodiment. 9 is a flowchart illustrating a processing flow of the image processing apparatus according to the second embodiment. The figure which shows the state of an original when a book original is mounted on the contact glass.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image reading means, 20 Image information storage means, 30 Area identification means, 40 Distortion area detection means, 50 Distortion correction means, 60 Staple detection means, 70 Staple removal means, 100 Book original, 101 Page binding part, 102 Contact glass

Claims (6)

In an image processing apparatus for obtaining distortion-free image information by correcting distortion of the shape of an image read by an image reading means for reading a document image, an area identifying means for identifying a character or a figure / photograph area of the image information And with respect to the region including the character among the regions identified by the region identifying means , the boundary is a portion where the distortion magnitude is larger than a predetermined threshold value, and the distortion is larger than the predetermined threshold value. A distortion area detecting means for determining a distortion area and detecting a distortion area ; and whether the distortion area detected by the distortion area detection means includes a character, figure or photo area , and whether or not the distortion area has a staple. A staple detecting means for detecting the staple and a staple removing means for removing the staple when the staple detecting means detects the presence of the staple. Since I, an image processing apparatus characterized by performing area discrimination processing on the reread document image and so as to identify the areas of text and diagrams and photographs. In an image processing apparatus for obtaining distortion-free image information by correcting distortion of the shape of an image read by an image reading means for reading a document image, an area identifying means for identifying a character or a figure / photograph area of the image information And with respect to the region including the character among the regions identified by the region identifying means , the boundary is a portion where the distortion magnitude is larger than a predetermined threshold value, and the distortion is larger than the predetermined threshold value. A distortion area detecting means for determining a distortion area and detecting a distortion area ; and whether the distortion area detected by the distortion area detection means includes a character, figure or photo area , and whether or not the distortion area has a staple. A staple detecting means for detecting a staple, a staple removing means for removing the staple when the staple detecting means detects the presence of a staple, and the staple detecting means Distortion correction means for correcting distortion of the shape of the image when there is no staple, and when the staple is detected by the staple detection means, the staple is removed and the original image read again is removed. On the other hand, if there is no staple, the distortion correction process is performed, and the area identification process is performed again on the image from which the distortion has been removed. An image processing apparatus characterized by identifying the above. In an image processing method for obtaining image information without distortion by correcting distortion of the shape of an image read by an image reading means for reading an original image, an area identifying step for identifying a character, a figure / photograph area of the image information And a region including characters among the regions identified by the region identifying step, where a distortion is larger than a predetermined threshold and a portion where the distortion is larger than the predetermined threshold is distorted. A distortion region detection step for determining a region and detecting a distortion region, and whether the distortion region detected in the distortion region detection step includes a character, a figure, or a photograph region, and whether the distortion region has a staple. A staple detection step for detecting the presence of staples in the staple detection step, and removing the staples and then reading the original image again. Performing frequency identification process, the image processing method is characterized in that so as to identify the areas of text and diagrams and photographs. In an image processing method for obtaining image information without distortion by correcting distortion of the shape of an image read by an image reading means for reading an original image, an area identifying step for identifying a character, a figure / photograph area of the image information And a region including characters among the regions identified by the region identifying step, where a distortion is larger than a predetermined threshold and a portion where the distortion is larger than the predetermined threshold is distorted. A distortion region detection step for determining a region and detecting a distortion region, and whether the distortion region detected in the distortion region detection step includes a character, a figure, or a photograph region, and whether the distortion region has a staple. A staple detection step for detecting the image, and a distortion correction step for correcting distortion of the shape of the image when no staple is present in the staple detection step, If the presence of staples is detected, the staple removal is performed, and the region identification process is performed on the re-read document image. If the staples are not present, the distortion correction process is performed to correct the distortion. An image processing method, wherein a region identification process is performed again on a removed image to identify a region of a character, a figure, or a photograph . A program for causing a computer to execute each step of the image processing method according to claim 3 or 4 . A computer-readable recording medium on which the program according to claim 5 is recorded .

Images