JP2022054710A - Image processing apparatus and image processing method - Google Patents

Abstract

To provide an image processing apparatus and an image processing method that can improve the character recognition rate of handwritten characters without using data indicating the correspondence relationship between an image representing the handwritten characters and characters represented by the image.SOLUTION: An image processing apparatus 10 comprises: a character recognition unit 92 that, based on the match rate between character images included in image data and reference images representing predetermined characters, recognizes the character represented by the character image; a determination processing unit 94 that determines whether predetermined image processing can improve the match rate between a specific character image, of the character images, which represents a predetermined specific character and a specific reference image, of the reference images, which represents the specific character; and a processing execution unit 96 that, when the determination processing unit 94 determines that the image processing can improve the match rate between the specific character image and the specific reference image, executes the image processing on the character image in which the character is recognized.SELECTED DRAWING: Figure 2

Description

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

In an image processing device such as a multifunction device, character recognition processing for recognizing characters represented by a character image included in image data may be executed. For example, in the character recognition process, the character represented by the character image is recognized based on the matching rate between the character image and each of the reference images representing the predetermined characters.

By the way, when the character image included in the image data is an image representing a user's handwritten character, the character image reflects the user's habit, and the difference from the reference image becomes large. Therefore, the literacy rate in the character recognition process is lowered. On the other hand, an image processing device that recognizes the user's handwritten characters included in the image data by using the data showing the correspondence between the image representing the user's handwritten characters and the characters represented by the image is known as a related technology. (See Patent Document 1).

Japanese Unexamined Patent Publication No. 2-81190

However, in the image processing apparatus according to the above-mentioned related technology, it is necessary to prepare in advance data showing the correspondence relationship between the image representing the user's handwritten character and the character represented by the image.

An object of the present invention is to provide an image processing device and an image processing method capable of improving the literacy rate of handwritten characters without using data showing the correspondence between an image representing handwritten characters and the characters represented by the image. It is in.

The image processing apparatus according to one aspect of the present invention includes a character recognition unit, a determination processing unit, and a processing execution unit. The character recognition unit recognizes the characters represented by the character image based on the matching rate between the character image included in the image data and each of the reference images representing the predetermined characters. The determination processing unit can improve the matching rate between the specific character image representing the predetermined specific character in the character image and the specific reference image representing the specific character in the reference image by the predetermined image processing. Judge whether or not. When the determination processing unit determines that the image processing can improve the matching rate between the specific character image and the specific reference image, the processing execution unit recognizes characters by the character recognition unit. The image processing is executed on the character image.

The image processing method according to another aspect of the present invention includes a character recognition step, a determination step, and a processing execution step. In the character recognition step, the character represented by the character image is recognized based on the matching rate between the character image included in the image data and each of the reference images representing the predetermined characters. In the determination step, the predetermined image processing can improve the matching rate between the specific character image representing the predetermined specific character in the character image and the specific reference image representing the specific character in the reference image. It is judged whether or not there is. In the processing execution step, when it is determined by the determination step that the image processing can improve the matching rate between the specific character image and the specific reference image, the character is recognized by the character recognition step. The image processing is executed for the character image.

According to the present invention, it is possible to improve the literacy rate of handwritten characters without using data showing the correspondence between an image showing handwritten characters and the characters represented by the image.

FIG. 1 is a diagram showing a configuration of an image processing apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram showing a system configuration of the image processing apparatus according to the embodiment of the present invention. FIG. 3 is a flowchart showing an example of character recognition processing executed by the image processing apparatus according to the embodiment of the present invention. FIG. 4 is a flowchart showing an example of the first re-recognition process executed by the image processing apparatus according to the embodiment of the present invention. FIG. 5 is a flowchart showing an example of a second re-recognition process executed by the image processing apparatus according to the embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. The following embodiments are examples that embody the present invention, and do not limit the technical scope of the present invention.

[Configuration of image processing device 10]
First, the configuration of the image processing apparatus 10 according to the embodiment of the present invention will be described with reference to FIGS. 1 and 2.

The image processing device 10 is a multifunction device having a scan function for reading an image of a document, a print function for forming an image based on image data, and a plurality of functions such as a facsimile function and a copy function. The image processing device 10 may be a scanner, a printer, a facsimile machine, a copier, a personal computer, a notebook computer, a smartphone, a tablet terminal, or the like.

As shown in FIGS. 1 and 2, the image processing device 10 includes an ADF (Auto Document Feeder) 1, an image reading unit 2, an image forming unit 3, a paper feeding unit 4, a control unit 5, an operation display unit 6, and communication. A unit 7 and a storage unit 8 are provided.

The ADF1 conveys a document to be read by the scanning function. For example, the ADF1 includes a document setting unit, a plurality of transport rollers, a document holder, and a paper ejection unit.

The image reading unit 2 realizes the scanning function. For example, the image reading unit 2 includes a platen, a light source, a plurality of mirrors, an optical lens, and a CCD (Charge Coupled Device). The image scanning unit 2 reads the image of the original and outputs the scanned image data including the scanned image.

The image forming unit 3 realizes the printing function. For example, the image forming unit 3 forms an image by an electrophotographic method. For example, the image forming unit 3 includes a photoconductor drum, a charging roller, an optical scanning device, a developing device, a transfer roller, a cleaning device, a fixing device, and a paper ejection tray.

The paper feeding unit 4 supplies the sheet to the image forming unit 3. For example, the paper feed unit 4 includes a paper feed cassette, a manual feed tray, a sheet transport path, and a plurality of transport rollers.

The control unit 5 comprehensively controls the image processing device 10. As shown in FIG. 2, the control unit 5 includes a CPU 51, a ROM 52, and a RAM 53. The CPU 51 is a processor that executes various arithmetic processes. The ROM 52 is a non-volatile storage device in which information such as a control program for causing the CPU 51 to execute various processes is stored in advance. The RAM 53 is a volatile storage device used as a temporary storage memory (working area) for various processes executed by the CPU 51. In the control unit 5, various control programs stored in advance in the ROM 52 are executed by the CPU 51. As a result, the image processing device 10 is collectively controlled by the control unit 5.

The control unit 5 may be composed of an electronic circuit such as an integrated circuit (ASIC). Further, the control unit 5 may be a control unit provided separately from the main control unit that collectively controls the image processing device 10.

The operation display unit 6 is a user interface of the image processing device 10. For example, the operation display unit 6 is a display unit such as a liquid crystal display that displays various information in response to a control instruction from the control unit 5, and an operation key for inputting various information to the control unit 5 in response to a user operation. Alternatively, it is provided with an operation unit such as a touch panel.

The communication unit 7 is a communication interface that executes wired or wireless data communication with an external communication device via a communication network such as the Internet or LAN.

The storage unit 8 is a non-volatile storage device. For example, the storage unit 8 is a storage device such as a flash memory, a non-volatile memory such as EEPROM (registered trademark), an SSD (Solid State Drive), or an HDD (Hard Disk Drive).

The storage unit 8 stores user information about a user who can log in to the image processing device 10. The user information includes authentication information used for user authentication. For example, the authentication information is user identification information such as a user name and a password.

Further, the storage unit 8 stores group information about a group to which a user who can log in to the image processing device 10 belongs. The group information includes group identification information such as a group name and one or more user identification information corresponding to one or more users belonging to the group.

The storage unit 8 may be provided in an information processing device that is communicably connected to the image processing device 10. In this case, the control unit 5 may acquire the data stored in the storage unit 8 by performing data communication with the information processing device.

[Structure of control unit 5]
Next, the configuration of the control unit 5 will be described with reference to FIG. 2.

As shown in FIG. 2, the control unit 5 includes an authentication processing unit 91, a character recognition unit 92, a feature determination unit 93, a determination processing unit 94, a first association unit 95, a processing execution unit 96, and a character acquisition unit 97. And the second mapping unit 98 are included.

Specifically, the ROM 52 of the control unit 5 stores in advance a character recognition program for causing the CPU 51 of the control unit 5 to execute the character recognition process (see the flowchart of FIG. 3) described later. Then, the CPU 51 of the control unit 5 functions as each of the above-mentioned units by executing the character recognition program stored in the ROM 52.

The character recognition program is recorded on a computer-readable recording medium such as a CD, DVD, or flash memory, and may be read from the recording medium and stored in a storage device such as a storage unit 8.

The authentication processing unit 91 authenticates the user of the image processing device 10 by using the authentication information stored in the storage unit 8.

For example, the authentication processing unit 91 causes the operation display unit 6 to display an authentication screen used for a predetermined authentication operation in response to a predetermined user operation in the operation display unit 6. Then, the authentication processing unit 91 accepts the authentication operation on the authentication screen. For example, the authentication operation is an operation of inputting the user identification information and the password. After that, the authentication processing unit 91 authenticates the user when the information input from the operation display unit 6 matches any of the authentication information stored in the storage unit 8. That is, the user is logged in to the image processing device 10.

The authentication processing unit 91 is a currently logged-in user when a predetermined logout operation is performed in the operation display unit 6, a preset time has elapsed, or another user is logged in. Executes the logout process to log out.

The character recognition unit 92 recognizes the characters represented by the character image based on the matching rate between the character image included in the image data and each of the reference images representing the predetermined characters.

For example, in the image processing device 10, a plurality of the reference images corresponding to a plurality of characters recognizable by the device are stored in the storage unit 8 in advance. Then, the character recognition unit 92 recognizes the characters represented by the character image included in the image data by using the plurality of reference images stored in the storage unit 8.

For example, the character recognition unit 92 acquires the matching rate between the reference image and the character image to be recognized for each of the reference images stored in the storage unit 8. Then, the character recognition unit 92 has one reference image having the highest matching rate with the character image to be recognized, and the matching rate between the reference image and the character image to be recognized is predetermined. When the first threshold value is exceeded, the character represented by the reference image is recognized as the character represented by the character image to be recognized.

On the other hand, in the character recognition unit 92, the reference image having the highest matching rate with the character image to be recognized is one, and the matching rate between the reference image and the character image to be recognized is the first threshold value. In the following cases, it is determined that the recognition of the character represented by the character image to be recognized has failed.

Further, the character recognition unit 92 determines that the recognition of the character represented by the character image to be recognized has failed when there are a plurality of reference images having the highest matching rate with the character image to be recognized.

For example, when the read image data is output from the image reading unit 2, the character recognition unit 92 recognizes the characters represented by the character image included in the read image data.

By the way, when the character image included in the image data is an image representing a user's handwritten character, the character image reflects the user's habit, and as a result, the difference from the reference image becomes large. , The character recognition rate (character recognition rate) by the character recognition unit 92 decreases. On the other hand, an image processing device that recognizes the user's handwritten characters included in the image data by using the data showing the correspondence between the image representing the user's handwritten characters and the characters represented by the image is known as a related technology. There is.

However, in the image processing apparatus according to the above-mentioned related technology, it is necessary to prepare in advance data showing the correspondence relationship between the image representing the user's handwritten character and the character represented by the image.

On the other hand, in the image processing apparatus 10 according to the embodiment of the present invention, as described below, the handwritten characters are represented without using the data indicating the correspondence between the image representing the handwritten characters and the characters represented by the images. It is possible to improve the literacy rate.

The feature determination unit 93 is based on a comparison result between a specific character image representing a predetermined specific character among the character images included in the image data and a specific reference image representing the specific character among the reference images. It is determined whether or not the specific character image has predetermined characteristics.

For example, the specific character includes a character recognized by the character recognition unit 92.

For example, the feature includes a first feature in which the specific character represented by the specific character image is rising to the right. For example, the feature determination unit 93 determines that the specific character image has the first feature when only the portion on the right side of the center of the specific character image is shifted upward with respect to the specific reference image. ..

Further, the feature includes a second feature in which the corner portion of the specific character represented by the specific character image is rounded (the specific character is a round character). For example, the feature determination unit 93 has the specific character when the first corner portion included in the specific character image has a rounder shape than the second corner portion corresponding to the first corner portion included in the specific reference image. It is determined that the image has the second feature.

Further, the feature includes a third feature in which the line of the specific character represented by the specific character image is connected (the specific character does not have a "stop"). For example, the feature determination unit 93 determines that the specific character image has the third feature when the two lines separated in the specific reference image are continuous in the specific character image.

For example, the feature determination unit 93 determines whether or not the specific character image has the characteristics based on the comparison result between the specific character image and the specific reference image when the character recognition by the character recognition unit 92 fails. It is possible to judge.

The feature determination unit 93 may display any of the character images included in the image data on the operation display unit 6 and accept input of characters represented by the character image. In this case, the specific character may be a character input by the user.

Further, the feature determination unit 93 determines whether or not the specific character image has the characteristics based on the comparison result between the specific character image and the specific reference image before the character recognition by the character recognition unit 92 fails. It may be possible to judge whether or not.

The determination processing unit 94 determines whether or not the predetermined image processing can improve the matching rate between the specific character image and the specific reference image.

For example, the image processing is a processing associated with the feature.

For example, in the image processing device 10, a rotation process for rotating an image is associated with the first feature. Further, in the image processing device 10, the corner portion enhancement process for emphasizing (sharpening) the corner portion included in the image is associated with the second feature. Further, in the image processing apparatus 10, a thin line erasing process for erasing thin lines whose width included in the image is equal to or less than a predetermined reference value is associated with the third feature.

Then, when the characteristic determination unit 93 determines that the specific character image has the characteristic, the determination processing unit 94 matches the specific character image with the specific reference image in the image processing corresponding to the characteristic. Judge that the rate can be improved.

The control unit 5 does not have to include the feature determination unit 93. In this case, the image processing does not have to be the processing associated with the feature. Further, the determination processing unit 94 determines whether or not the image processing can improve the matching rate between the specific character image and the specific reference image when the character recognition by the character recognition unit 92 fails. You may. For example, the determination processing unit 94 determines whether or not the image processing can improve the matching rate between the specific character image and the specific reference image based on the execution result of the image processing on the specific character image. You may. Further, the determination processing unit 94 determines whether or not the image processing can improve the matching rate between the specific character image and the specific reference image before the character recognition by the character recognition unit 92 fails. You may.

When the feature determination unit 93 determines that the specific character image has the feature, the first association unit 95 associates the feature with the user logged in to the own device.

For example, the first mapping unit 95 adds the feature information indicating the feature to the user information corresponding to the user logged in to the own device.

When the feature determination unit 93 determines that the specific character image has the feature, the first mapping unit 95 associates the feature with the group to which the user logged in to the own device belongs. May be good. For example, the first association unit 95 may add the feature information to the group information corresponding to the group to which the user logged in to the own device belongs. Further, the first mapping unit 95 may switch the mapping destination of the feature according to the operation of the user.

Here, the feature determination unit 93 determines whether or not the specific character image has the feature when the feature associated with the user logged in to the own device or the group to which the user belongs exists. Do not judge. In addition, when the feature associated with the user logged in to the own device or the group to which the user belongs exists, the feature determination unit 93 has another feature whose specific character image is different from the feature. You may decide whether or not to have.

Further, the control unit 5 does not have to include the first association unit 95.

When the determination processing unit 94 determines that the image processing can improve the matching rate between the specific character image and the specific reference image, the processing execution unit 96 recognizes the characters by the character recognition unit 92. The image processing is executed on the character image.

Further, when the feature associated with the user logged in to the own device or the group to which the user belongs exists, the processing execution unit 96 processes the image corresponding to the feature with respect to the character image. To execute.

For example, the processing execution unit 96 executes the image processing on the character image in which the character recognition by the character recognition unit 92 fails.

The processing execution unit 96 may execute the image processing on the character image in which the character is not recognized by the character recognition unit 92.

The character acquisition unit 97 acquires the character when the recognition of the character represented by the character image after the image processing fails.

For example, the character acquisition unit 97 causes the operation display unit 6 to display a character input screen used for inputting the character when the recognition of the character represented by the character image after the image processing fails. Then, the character acquisition unit 97 acquires the character represented by the character image whose character recognition has failed in response to the user's operation on the character input screen.

The second mapping unit 98 associates the combination of the character acquired by the character acquisition unit 97 with the character image corresponding to the character to the user logged in to the own device.

For example, in the image processing apparatus 10, a conversion file in which the combination is written is added to the user information in advance.

Then, when the character is acquired by the character acquisition unit 97, the second mapping unit 98 corresponds to the user who is logged in to the own device with the combination of the character and the character image corresponding to the character. Write to the conversion file.

The second mapping unit 98 may map the combination to the group to which the user logged in to the own device belongs. For example, the conversion file may be added to the group information. Further, in the second mapping unit 98, when a character is acquired by the character acquisition unit 97, the user who is logged in to the own device belongs to the combination of the character and the character image corresponding to the character. You may write to the conversion file corresponding to the group. Further, the second mapping unit 98 may switch the mapping destination of the combination according to the operation of the user.

Here, if the combination associated with the user logged in to the own device or the group to which the user belongs exists, the character recognition unit 92 uses the combination to display the characters represented by the character image. It is recognizable. For example, when the character recognition unit 92 fails to recognize the character represented by the character image based on the matching rate between the character image and each of the reference images, the character recognition unit 92 recognizes the character represented by the character image by using the combination. do.

When the unused period of the combination exceeds a predetermined specific time, the control unit 5 cancels the association between the combination and the user logged in to the own device or the group to which the user belongs. You may. For example, the control unit 5 may delete the combination from the conversion file.

Further, the control unit 5 may not include the character acquisition unit 97 and the second association unit 98.

[Character recognition processing]
Hereinafter, the image processing method of the present invention will be described together with an example of the procedure of the character recognition processing executed by the control unit 5 in the image processing apparatus 10 with reference to FIG. Here, steps S11, S12 ... Represent the number of the processing procedure (step) executed by the control unit 5. The character recognition process is executed when the user operates the operation display unit 6 to instruct the execution of the character recognition process.

<Step S11>
First, in step S11, the control unit 5 acquires image data.

For example, the control unit 5 causes the image reading unit 2 to execute the scanning function. Then, the control unit 5 acquires the scanned image data output from the image reading unit 2.

<Step S12>
In step S12, the control unit 5 extracts the character image included in the image data acquired in step S11.

The control unit 5 may execute a normalization process, a binarization process, or the like on each of the extracted character images.

<Step S13>
In step S13, the control unit 5 recognizes the character represented by any of the character images extracted in step S12. Here, the process of step S13 is an example of the character recognition step of the present invention, and is executed by the character recognition unit 92 of the control unit 5.

Specifically, the control unit 5 recognizes the characters represented by the character image based on the matching rate between the character image and each of the reference images.

<Step S14>
In step S14, the control unit 5 determines whether or not the character recognition has failed in step S13.

Here, when the control unit 5 determines that the character recognition has failed (Yes side of S14), the control unit 5 shifts the process to step S15. If the character recognition has not failed (No side of S14), the control unit 5 shifts the process to step S16.

<Step S15>
In step S15, the control unit 5 records the character recognition failure in step S13 and the character image that failed to be recognized.

For example, the control unit 5 records the failure of character recognition in step S13 and the character image that failed in recognition in a predetermined specific storage area in the RAM 53.

<Step S16>
In step S16, the control unit 5 determines whether or not the process of step S13 has been executed for all the character images extracted in step S12.

Here, when the control unit 5 determines that the process of step S13 has been executed for all the character images (Yes side of S16), the control unit 5 shifts the process to step S17. Further, if the process of step S13 is not executed for all the character images (No side of S16), the control unit 5 shifts the process to step S13.

<Step S17>
In step S17, the control unit 5 determines whether or not the character image that fails to recognize the character exists.

For example, the control unit 5 determines whether or not the character image that fails to recognize the character exists based on the storage content of the specific storage area of the RAM 53.

Here, when the control unit 5 determines that the character image that has failed in character recognition exists (Yes side of S17), the control unit 5 shifts the process to step S18. If the character image that failed to recognize the character does not exist (No side of S17), the control unit 5 shifts the process to step S19.

<Step S18>
In step S18, the control unit 5 executes the first re-recognition process described later.

<Step S19>
In step S19, the control unit 5 outputs the recognition result of the character represented by the character image included in the image data acquired in step S11.

For example, the control unit 5 stores data indicating the recognition result in the storage unit 8.

[First re-recognition process]
Next, with reference to FIG. 4, the first re-recognition process executed in step S18 of the character recognition process will be described.

<Step S21>
First, in step S21, the control unit 5 determines whether or not the combination including the character image that failed in character recognition in step S13 of the character recognition process is associated with the user logged in to the own device. ..

For example, when the combination including the character image that fails in character recognition is written in the conversion file corresponding to the logged-in user, the control unit 5 associates the combination with the logged-in user. Judge that it is.

Here, when the control unit 5 determines that the combination including the character image whose character recognition has failed is associated with the logged-in user (Yes side of S21), the control unit 5 shifts the process to step S22. Further, if the combination including the character image for which character recognition has failed is not associated with the logged-in user (No side of S21), the control unit 5 shifts the process to step S24.

<Step S22>
In step S22, the control unit 5 recognizes the character image by using the combination including the character image that failed in character recognition in step S13 of the character recognition process. Here, the process of step S22 is executed by the character recognition unit 92 of the control unit 5.

<Step S23>
In step S23, the control unit 5 determines whether or not the characters have been recognized for all of the character images extracted in step S12 of the character recognition process.

Here, when the control unit 5 determines that the characters have been recognized for all of the character images extracted in step S12 of the character recognition process (Yes side of S23), the control unit 5 ends the first re-recognition process. If the characters are not recognized for all of the character images extracted in step S12 of the character recognition process (No side of S23), the control unit 5 shifts the process to step S24.

<Step S24>
In step S24, the control unit 5 executes the second re-recognition process described below.

[Second re-recognition process]
Next, the second re-recognition process executed in step S24 of the first re-recognition process will be described with reference to FIG.

<Step S31>
First, in step S31, the control unit 5 determines whether or not the feature associated with the user logged in to the own device exists.

For example, the control unit 5 determines that the feature associated with the logged-in user exists when the feature information is added to the user information corresponding to the logged-in user.

Here, when the control unit 5 determines that the feature associated with the logged-in user exists (Yes side of S31), the control unit 5 shifts the process to step S35. Further, if the feature associated with the logged-in user does not exist (No side of S31), the control unit 5 shifts the process to step S32.

<Step S32>
In step S32, the control unit 5 determines that the specific character image included in the image data acquired in step S11 of the character recognition process is compared with the specific reference image corresponding to the specific character image. It is determined whether or not the specific character image has the above-mentioned characteristics. Here, the process of step S32 is executed by the feature determination unit 93 of the control unit 5.

For example, the control unit 5 selects the character image whose matching rate with the reference image exceeds the second threshold value higher than the first threshold value among the character images that have succeeded in character recognition in step S13 of the character recognition process. It is set to the specific character image. Then, the control unit 5 determines whether or not the specific character image has the characteristics based on the comparison result between the set specific character image and the specific reference image corresponding to the specific character image.

The control unit 5 may determine that the specific character image has the plurality of features.

<Step S33>
In step S33, the control unit 5 determines whether or not the image processing can improve the matching rate between the specific character image and the specific reference image. Here, the process of step S33 is an example of the determination step of the present invention, and is executed by the determination processing unit 94 of the control unit 5.

Specifically, when the control unit 5 determines in step S32 that the specific character image has the characteristic, the image processing corresponding to the characteristic has a matching rate between the specific character image and the specific reference image. Is judged to be possible to improve.

Here, when the control unit 5 determines that the image processing can improve the matching rate between the specific character image and the specific reference image (Yes side of S33), the control unit 5 shifts the processing to step S34. Further, if the image processing cannot improve the matching rate between the specific character image and the specific reference image (No side of S33), the control unit 5 shifts the processing to step S38.

<Step S34>
In step S34, the control unit 5 executes the first associating process of associating the feature determined to be possessed by the specific character image with the user logged in to the own device in step S32. Here, the process of step S34 is executed by the first mapping unit 95 of the control unit 5.

<Step S35>
In step S35, the control unit 5 executes the image processing for each of the character images for which the characters have not been recognized yet. Here, the process of step S35 is an example of the process execution step of the present invention, and is executed by the process execution unit 96 of the control unit 5.

Specifically, when it is determined in step S31 that the feature associated with the logged-in user exists, the control unit 5 executes the image processing corresponding to the feature. Further, when the control unit 5 determines in step S33 that the image processing can improve the matching rate between the specific character image and the specific reference image, the control unit 5 determines in step S32 that the specific character image has. The image processing corresponding to the said feature is executed.

For example, when the feature determined to be possessed by the specific character image in step S32 is the first feature, the control unit 5 displays the character image for each of the character images for which the character has not been recognized yet. The rotation process of rotating a character by a predetermined reference angle in the clockwise direction is executed.

Further, when the feature determined to be possessed by the specific character image in step S32 is the second feature, the control unit 5 emphasizes the corner portion for each of the character images for which the character has not been recognized yet. Execute the process.

Further, when the feature determined to be possessed by the specific character image in step S32 is the third feature, the control unit 5 performs the fine line erasing process for each of the character images for which characters have not yet been recognized. To execute.

<Step S36>
In step S36, the control unit 5 recognizes the characters represented by each of the character images for which the image processing was executed in step S35. Here, the process of step S36 is executed by the character recognition unit 92 of the control unit 5.

<Step S37>
In step S37, the control unit 5 determines whether or not the character image whose character recognition has failed in step S36 exists.

Here, when the control unit 5 determines that the character image that has failed in character recognition exists (Yes side of S37), the control unit 5 shifts the process to step S38. If the character image that failed to recognize the character does not exist (No side of S37), the control unit 5 ends the second re-recognition process.

When the character image that fails to recognize the character exists, the control unit 5 shifts the processing to step S35 and executes the image processing on the character image under execution conditions different from the previous time. You may. In this case, the control unit 5 may shift the process to step S38 when the number of character recognition failures in the process of step S35 reaches a predetermined number of conditions in step S37.

<Step S38>
In step S38, the control unit 5 acquires the character represented by the character image that failed to recognize the character in step S36. Here, the process of step S38 is executed by the character acquisition unit 97 of the control unit 5.

Specifically, the control unit 5 causes the operation display unit 6 to display the character input screen. Then, the control unit 5 acquires the character represented by the character image whose character recognition has failed in response to the user's operation on the character input screen.

<Step S39>
In step S39, the control unit 5 associates the combination of the character acquired in step S38 with the character image corresponding to the character with the user logged in to the own device. Here, the process of step S39 is executed by the second mapping unit 98 of the control unit 5.

Specifically, the control unit 5 writes the combination in the conversion file corresponding to the logged-in user.

As described above, in the image processing apparatus 10, whether or not the image processing can improve the matching rate between the specific character image representing the specific character included in the image data and the specific reference image representing the specific character. When it is determined that the image can be improved, the image processing is executed on the character image to be recognized as the character included in the image data. This makes it possible to improve the literacy rate of handwritten characters without using data indicating the correspondence between the image representing the handwritten characters and the characters represented by the image.

1 ADF
2 Image reading unit 3 Image forming unit 4 Paper feeding unit 5 Control unit 6 Operation display unit 7 Communication unit 8 Storage unit 10 Image processing device 91 Authentication processing unit 92 Character recognition unit 93 Feature judgment unit 94 Judgment processing unit 95 First correspondence Part 96 Processing execution part 97 Character acquisition part 98 Second mapping part

Claims (8)

A character recognition unit that recognizes the characters represented by the character image based on the matching rate between the character image included in the image data and each of the reference images representing the predetermined characters.
Whether or not the predetermined image processing can improve the matching rate between the specific character image representing the predetermined specific character in the character image and the specific reference image representing the specific character in the reference image. Judgment processing unit to judge and
When the image processing determines that the matching rate between the specific character image and the specific reference image can be improved by the determination processing unit, the character recognition unit recognizes the character with respect to the character image. The processing execution unit that executes the image processing and
An image processing device comprising. The specific character includes a character recognized by the character recognition unit.
The image processing apparatus according to claim 1. The determination processing unit determines whether or not the image processing can improve the matching rate between the specific character image and the specific reference image when the character recognition by the character recognition unit fails.
The processing execution unit executes the image processing on the character image in which the character recognition by the character recognition unit fails.
The image processing apparatus according to claim 1 or 2. When the character recognition by the character recognition unit fails, it is possible to determine whether or not the specific character image has predetermined characteristics based on the comparison result between the specific character image and the specific reference image. Equipped with a feature judgment unit
The image processing is a process associated with the feature in advance.
When the characteristic determination unit determines that the specific character image has the characteristic, the determination processing unit determines that the image processing corresponding to the characteristic determines the matching rate between the specific character image and the specific reference image. Judging that it can be improved,
The image processing apparatus according to claim 3. When the feature determination unit determines that the specific character image has the feature, the feature determination unit is provided with a first mapping unit that associates the feature with the user logged in to the own device or the group to which the user belongs.
The feature determination unit determines whether or not the specific character image has the feature when the feature associated with the user logged in to the own device or the group to which the user belongs exists. figure,
When the feature associated with the user logged in to the own device or the group to which the user belongs exists, the processing execution unit performs the image processing corresponding to the feature on the character image. Run,
The image processing apparatus according to claim 4. When the recognition of the character represented by the character image after the image processing fails, the character acquisition unit for acquiring the character and the character acquisition unit.
A second mapping unit that associates a combination of a character acquired by the character acquisition unit with the character image corresponding to the character to a user logged in to the own device or a group to which the user belongs.
Equipped with
When the combination associated with the user logged in to the own device or the group to which the user belongs exists, the character recognition unit can recognize the character represented by the character image using the combination. be,
The image processing apparatus according to any one of claims 1 to 5. It is equipped with an image reading unit that reads the image of the original and outputs the scanned image data including the scanned image.
The image data includes the read image data.
The image processing apparatus according to any one of claims 1 to 6. A character recognition step for recognizing a character represented by the character image based on a matching rate between the character image included in the image data and each of the reference images representing the predetermined characters, and a character recognition step.
Whether or not the predetermined image processing can improve the matching rate between the specific character image representing the predetermined specific character in the character image and the specific reference image representing the specific character in the reference image. Judgment steps to judge and
When it is determined by the determination step that the image processing can improve the matching rate between the specific character image and the specific reference image, the character image is recognized by the character recognition step. Processing execution step to execute image processing and
Image processing methods including.

Images