Character recognition system and character recognition program

Abstract

To provide a character recognition system that can reduce garbling of characters by detecting garbling of characters occurring due to character recognition, and thereby can improve the quality of a document to be output.SOLUTION: A character recognition system 1 comprises: a document acquisition unit 11 that acquires a document in an image format; a character recognition unit 12 that recognizes character information included in the document; and a written error detection unit 14 that detects written error information included in the character information. The written error detection unit 14 can detect, as the written error information, special characters that are displayed when the character information is not correctly recognized by the character recognition unit 12.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a character recognition system and a character recognition program. [Background technology]

[0002] OCR (Optical Character Recognition) is a technology that converts handwritten or printed characters displayed in an image into digital character codes that a computer can use. OCR may not be able to recognize characters in an image if, for example, the characters are small, unclear, or too complex, such as characters that are not in a standard font or are handwritten. When OCR cannot recognize characters in an image, the characters are displayed as completely different characters, or as special characters such as a white square or a black diamond with a question mark, resulting in garbled characters.

[0003] To solve these problems, a document file detection and analysis system has been developed that is improved to accurately recognize characters in an image (see, for example, Patent Document 1). The document file detection and analysis system in Patent Document 1 is a system for digitally recording invoices of various formats. This document file detection and analysis system reduces image noise to improve image quality, extracts character blocks using OCR, and performs document layout analysis based on image processing technology to separate zones containing character information from zones containing non-character information. This document file detection and analysis system then detects document fields based on the document layout analysis and analyzes the document fields to extract characters and numbers. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Canadian Patent No. 3027038 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the document file detection and analysis system in Patent Document 1 is specialized in detecting and analyzing characters and numbers on invoices, and has the problem that it cannot detect garbled characters that occur during the character recognition process in general document files, reduce garbled characters, and improve the accuracy of character recognition.

[0006] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a character recognition system and a character recognition program that can reduce garbled characters by detecting garbled characters caused by character recognition, thereby improving the quality of output documents. [Means for solving the problem]

[0007] The character recognition system of the present invention includes a document acquisition unit that acquires an image document, a character recognition unit that recognizes character information contained in the document, and an error detection unit that detects typographical errors contained in the character information. The error detection unit is capable of detecting special characters that are displayed when the character recognition unit fails to recognize the character accurately as typographical errors.

[0008] The character recognition program of the present invention causes a computer to execute a process including a document acquisition step of acquiring an image document, a character recognition step of recognizing character information contained in the document, and an error detection step of detecting error information contained in the character information. The error detection step can detect, as error information, special characters that are displayed when the character recognition step fails to accurately recognize the special characters. [Effects of the Invention]

[0009] According to the character recognition system and character recognition program of the present invention, garbled characters displayed as special characters caused by character recognition can be detected, thereby reducing garbled characters and improving the quality of output documents. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a functional block diagram showing functions of a character recognition system according to an embodiment of the present invention. [Figure 2] 1 is a flowchart illustrating a character recognition process of a character recognition system according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 4] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 5] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 6] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 7] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 8] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. [Figure 9] 1 is a flowchart illustrating a character recognition process of a character recognition system according to an embodiment of the present invention. [Figure 10] FIG. 2 is a diagram showing an example of an operation screen of the character recognition system of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference characters and description thereof will not be repeated.

[0012] The character recognition system 1 will be described with reference to Fig. 1. Fig. 1 is a functional block diagram showing the functions of the character recognition system 1 according to this embodiment.

[0013] Character recognition system 1 is a system that recognizes character information contained in image documents. Here, image documents refer to general image files and image files in Portable Document Format (PDF). Character recognition system 1 is constructed in user terminal 2 and / or server terminal 3 by installing a character recognition program in user terminal 2 and / or server terminal 3 connected to user terminal 2 via network 4. All functions of character recognition system 1 may be constructed in either user terminal 2 or server terminal 3, or some functions of character recognition system 1 may be constructed in user terminal 2 and other functions of character recognition system 1 may be constructed in server terminal 3. In the following description, character recognition system 1 will be described assuming that a character recognition program is installed in server terminal 3, all functions are constructed in server terminal 3, and the system operates upon receiving instructions from user terminal 2 connected via network 4.

[0014] The user terminal 2 is a terminal used by a user who wants to process documents, including character recognition. The user terminal 2 is equipped with a display device, such as a display, on which the user can view information output from the character recognition system 1, and an input device, such as a keyboard, mouse, or touch panel, on which the user can specify the operation of the character recognition system 1. The user terminal 2 also has a storage device, such as a semiconductor memory, for storing various documents. The user terminal 2 is, for example, a personal computer, a tablet computer, or a smartphone.

[0015] The server terminal 3 is a terminal on which a character recognition program is installed to configure the character recognition system 1. The server terminal 3 is, for example, a server computer equipped with a CPU (Central Processing Unit) and storage.

[0016] The character recognition system 1 includes a document acquisition unit 11, a character recognition unit 12, a proofreading confirmation unit 13, an error detection unit 14, a correction proposal proposal unit 15, an information display unit 16, a corrected document creation unit 17, a control unit 21, and a memory unit 22.

[0017] The document acquisition unit 11 acquires a document in image format of one or more pages. Specifically, the user selects a document in image format stored in the storage device of the user terminal 2. The document acquisition unit 11 receives a signal indicating that the user has selected the document, and acquires the document by reading the document in image format selected by the user from the storage device of the user terminal 2.

[0018] Character recognition unit 12 recognizes character information included in the document acquired by document acquisition unit 11. Character recognition unit 12 is, for example, an OCR. Character recognition unit 12 recognizes character information included in an image-format document by converting it into character codes, and generates a text-format document. If the document consists of multiple pages, character recognition unit 12 recognizes character information on each page, and generates a text-format document consisting of multiple pages.

[0019] The proofreading confirmation unit 13 confirms with the user whether proofreading is required. When the user instructs the execution of document output, such as printing or saving the document, the proofreading confirmation unit 13 confirms whether proofreading is required before the document is output. Specifically, when the user instructs the execution of document output, the proofreading confirmation unit 13 receives an output execution signal from the control unit 21 indicating that the user has instructed the execution of document output. Upon receiving the output execution signal, the proofreading confirmation unit 13 returns an output interruption signal to the control unit 21 to interrupt the execution of document output, and displays a screen on the display device of the user terminal 2 to confirm the user whether proofreading is required. When the user instructs that proofreading is required, the proofreading confirmation unit 13 transmits a proofreading execution signal to the control unit 21, and the control unit 21 operates the error detection unit 14, the correction proposal proposal unit 15, the information display unit 16, and the corrected document creation unit 17. On the other hand, when the user instructs that proofreading is not required, the proofreading confirmation unit 13 transmits a proofreading-not required signal to the control unit 21, and the control unit 21 executes document output.

[0020] The error detection unit 14 detects error information contained in the character information. Error information includes characters that differ from the original characters, special characters that indicate garbled characters such as a white square or a black diamond with a question mark, typos, and grammatical errors. That is, the error detection unit 14 can detect, as error information, at least special characters that are displayed when the character recognition unit 12 does not accurately recognize the character. In addition, the error detection unit 14 may be able to detect, as error information, at least one of typos and grammatical errors contained in character information that is accurately recognized by the character recognition unit 12.

[0021] Special characters are detected as follows. Special characters that appear when characters are garbled during character recognition are limited to specific characters, such as the above-mentioned open square or black diamond with a question mark. Special characters that appear when characters are garbled are stored in the storage unit 22 as special character data. The error detection unit 14 reads the special character data from the storage unit 22 and detects special characters by searching whether or not special characters included in the special character data exist in the document converted into text format by the character recognition unit 12. In addition, the error detection unit 14 has a function similar to the proofreading function of Microsoft Word, for example, and uses this function to detect typographical errors and grammatical errors.

[0022] The correction proposal proposal unit 15 proposes correction proposal information for the typographical error information. The correction proposal proposal unit 15 proposes correction proposal information that is estimated to be correct for the typographical errors and grammatical errors in the typographical error information. For the correction proposals for the typographical errors and grammatical errors, the correction proposal proposal unit 15 has a function similar to the proofreading function of Microsoft Word, for example, and proposes correction proposals for the typographical errors and grammatical errors by utilizing this function.

[0023] The information display unit 16 displays information related to the erroneous information on the display device of the user terminal 2. The information display unit 16 displays the number of erroneous information detected by the erroneous information detection unit 14. The information display unit 16 also displays the correction proposal information proposed by the correction proposal proposing unit 15 together with the erroneous information. At this time, for erroneous information for which no correction proposal information exists, the information display unit 16 leaves the correction proposal information column blank and displays only the erroneous information.

[0024] The correction proposal information displayed on the display device of the user terminal 2 by the information display unit 16 can be revised by the user. That is, the user can check the error information and the correction proposal information and revise the correction proposal information as necessary. The user can also add correction proposals that they consider to be correct to the blank correction proposal information.

[0025] The correction document creation unit 17 creates a correction document in which the erroneous information has been rewritten into correction proposal information. Specifically, the correction document creation unit 17 acquires the correction proposal information created by the correction proposal proposing unit 15, and also acquires the correction proposal information added or revised by the user on the screen displayed by the information display unit 16. Furthermore, if the user adds or corrects something, the correction document creation unit 17 updates the correction proposal information created by the correction proposal proposing unit 15 to the correction proposal information corrected or added by the user. Then, the correction document creation unit 17 converts the erroneous information into correction proposal information for the document converted into text format by the character recognition unit 12, and creates a correction document.

[0026] The control unit 21 is configured with a processor such as a CPU (Central Processing Unit). The control unit 21 reads out and executes a character recognition program from the storage unit 22, thereby controlling the operations of the document acquisition unit 11, the character recognition unit 12, the proofreading confirmation unit 13, the error detection unit 14, the correction proposal proposal unit 15, the information display unit 16, and the corrected document creation unit 17.

[0027] The memory unit 22 stores various data, programs, etc. The memory unit 22 is configured by a storage device and a semiconductor memory. The storage device is, for example, a hard disk drive (HDD) or a solid state drive (SSD). The semiconductor memory is, for example, a read only memory (ROM) or a random access memory (RAM). The memory unit 22 stores a character recognition program and special character data.

[0028] Next, the flow of character recognition in the character recognition system 1 according to the embodiment of the present invention will be described using a flowchart with reference to Figs. 2 to 10. Figs. 2 and 9 are flowcharts for explaining the character recognition process in the character recognition system 1 according to the present embodiment. Fig. 2 is a flowchart for the case where the user proofreads independently, and Fig. 9 is a flowchart for the case where the user does not proofread independently. Figs. 3 to 8 and 10 are diagrams showing an example of the operation screen 30 of the character recognition system 1 according to the present embodiment. The character recognition process is executed by a character recognition program stored in the server terminal 3.

[0029] First, the flow of character recognition processing when a user independently performs proofreading will be described with reference to FIGS.

[0030] When a user wishes to perform character recognition on an image document, the user starts a character recognition program on the user terminal 2, thereby operating the character recognition system 1. When the character recognition system 1 starts, the control unit 21 displays an operation screen 30 on the display device of the user terminal 2. As shown in FIG. 3 , the operation screen 30 displays a document selection screen 31 for selecting an image document 42.

[0031] The user uses the input device of the user terminal 2 to select, on the document selection screen 31, a folder 41 in which a document 42 in an image format that the user wishes to character recognize is stored, and selects a document 42 stored in the folder 41. When the user selects the document 42, the document acquisition unit 11 executes a document acquisition step to acquire the document 42 selected by the user (step S10). Then, the document acquisition unit 11 sends the acquired document 42 to the character recognition unit 12.

[0032] Upon receiving document 42, character recognition unit 12 executes a character recognition step of recognizing character information contained in acquired document 42 (step S12). Specifically, character recognition unit 12 recognizes character information by converting character information contained in one or more pages of document 42 into character code for each page. Then, character recognition unit 12 generates one or more pages of text-format document 42.

[0033] When the character recognition is complete, control unit 21 displays operation screen 30 on the display device of user terminal 2. As shown in Fig. 4, operation screen 30 displays process selection screen 32 for selecting a process to be performed on document 42 that has been character recognized and converted into text format. Process selection screen 32 displays the top page of document 42 that has been character recognized and converted into text format, and icons 51 for specifying an operation to be performed on document 42.

[0034] To proofread the document 42 converted into text format, the user selects the "proofread" icon 51. When the "proofread" icon 51 is selected, the error detection unit 14 executes an error detection step of detecting error information contained in the character information of the document 42 (step S14).

[0035] Specifically, when the "Proofread" icon 51 is selected, the control unit 21 displays a first error detection setting screen 33 on the operation screen 30 of the display device of the user terminal 2, as shown in FIG. 5. The first error detection setting screen 33 is a screen for selecting the number of pages. The user uses the input device of the user terminal 2 to set the page setting section 331 of the first error detection setting screen 33 to the page on which they want to perform error detection. Specifically, the page setting section 331 of the first error detection setting screen 33 displays the total number of pages of the selected document as an initial value. That is, if the total number of pages of the document is 20, as shown in FIG. 5, the page setting section 331 displays "Page 1 to Page 20." For example, if the user wants to perform error detection on pages 5 to 10, the user sets the page setting section 331 to the pages on which they want to perform error detection, such as "Page 5 to Page 10." Then, the user selects "Yes" in the process selection section 332 of the first error detection setting screen 33 to continue the proofreading process. If "No" is selected in the process selection section 332 of the first error detection setting screen 33, the error detection step is not executed and the screen returns to the process selection screen 32.

[0036] When "Yes" is selected in the process selection section 332 of the first error detection setting screen 33, the control unit 21 displays the second error detection setting screen 34 on the operation screen 30 of the display device of the user terminal 2, as shown in FIG. 6. The second error detection setting screen 34 is a screen for selecting the type of error. The user uses the input device of the user terminal 2 to set the type of error to be detected in the type setting section 341 of the second error detection setting screen 34. Specifically, the type setting section 341 of the second error detection setting screen 34 displays the types of error that can be detected and check boxes. In this embodiment, the types of error that can be detected are "special characters," "typographical errors," and "grammatical errors." The user selects the check boxes for the types of error that they want to detect. For example, if all types of error are to be detected, all check boxes are selected. In FIG. 6, it is selected that all types of error need to be detected. The user then selects "OK" in the process selection section 342 of the second error detection setting screen 34 to continue the proofreading process. If "Back" is selected in the process selection section 342 of the second error detection setting screen 34, the first error detection setting screen 33 is displayed again, and the page on which error information is to be detected can be set again.

[0037] When the error detection conditions are set on the first error detection setting screen 33 and the second error detection setting screen 34, the error detection unit 14 detects the specified type of error information contained in a selected page from a multi-page document. For example, when the error detection unit 14 is set to detect all types of error information contained in pages 1 to 20 of the document, it detects error information such as special characters, clerical errors, and grammatical errors contained in pages 1 to 20 of the document.

[0038] When the detection of clerical errors is completed, the error detection unit 14 sends the number of detected clerical errors to the information display unit 16, and also sends the detected clerical errors to the correction proposal proposing unit 15. When the information display unit 16 receives the number of clerical errors, it executes an error detection result display step of displaying an error detection result screen 35 on the operation screen 30 of the display device of the user terminal 2, as shown in FIG. 7 (step S16). The error detection result screen 35 displays the page number on which the error was detected and the number of the error. Also, an icon 52 is displayed for selecting a process for the result of the error information. FIG. 7 shows that four clerical errors were detected on the fifth page.

[0039] The error detection result screen 35 is displayed for each page on which an error was detected in the document 42. That is, when the "Skip" icon 52 is selected on the error detection result screen 35, the page number of the next page on which an error was detected and the number of errors are displayed.

[0040] If the user wishes to check the details of the error information of the page displayed on the error detection result screen 35, the user selects the "Details" icon 52 (Details in step S18). If the user does not need to check the details of the error information of the page displayed on the error detection result screen 35, the user selects the "Skip" icon 52 (Skip in step S18). When the "Skip" icon 52 is selected, the control unit 21 checks whether the page number of the last page among the pages on which error information was detected and the number of pieces of error information are displayed (step S20), and if it is not the last page (Yes in step S20), it executes the error detection result display step again (step S16). If it is the last page (No in step S20), the proofreading of the character-recognized document 42 is completed, and the character recognition process is terminated.

[0041] When the “Details” icon 52 is selected, the correction proposal proposing unit 15 receives the error information from the error detection unit 14 and executes a correction proposal proposing step of proposing correction information that is estimated to be correct for the error information (step S22). The correction proposed by the correction proposal proposing unit 15 is sent to the information display unit 16.

[0042] When the information display unit 16 receives the correction proposal information for the error, it displays a proofreading result screen 36 on the operation screen 30 of the display device of the user terminal 2 (step S24). As shown in Fig. 8, the proofreading result screen 36 displays the page on which the error was detected, along with the error (original characters) and correction proposal information (correction proposal characters) written side by side. In Fig. 8, an error was detected on the fifth page of the document 42, so the proofreading result screen 36 displays "page 5" and four pieces of error information written side by side with the correction proposal information. At this time, for error information for which the correction proposal proposing unit 15 could not propose a correction proposal, no correction proposal information is displayed and a blank space is displayed.

[0043] The proofreading result screen 36 has a column for correction proposal information that can be added or revised by the user, and a check box is provided next to the column for correction proposal information. The user checks the proofreading result screen 36, and if they determine that the correction proposal information is incorrect or if the correction proposal information is blank, they use the input device of the user terminal 2 to correct or add to the correction proposal information. Furthermore, if they determine that the correction proposal information is correct, or if they add or revise the correction proposal information and it becomes correct, the user selects the check box.

[0044] The proofreading result screen 36 displays an icon 53 for selecting an action to take on the proofreading results. After the user has finished checking the proofreading result screen 36, the user selects the "Print," "Display," or "Back" icon 53, which indicates the action to take on the proofreading results (step S26). When the "Print" or "Display" icon 53 is selected, the correction document creation unit 17 executes a correction document creation step in which it creates a correction document in which the error information has been converted into correction proposal information (step S28). If "Print" is selected, the correction document is printed, the proofreading of the character-recognized document 42 is completed, and the character recognition process is terminated. If "Display" is selected, the correction document is temporarily displayed on the display device of the user terminal 2, and then the displayed correction document is saved in the storage device of the user terminal 2, the proofreading of the character-recognized document 42 is completed, and the character recognition process is terminated. If the "Back" icon 53 is selected, the user returns to the error detection result screen 35. If the user wishes to check the error information on the next page, the user selects the "Skip" icon 52. Then, the process of step S20 is executed.

[0045] Next, the flow of character recognition processing when the user does not proofread independently will be described with reference to Figures 9 and 10. The basic flow is the same as the flow of character recognition processing when the user proofreads independently. Below, we will explain the processing that differs from the flow of character recognition processing when the user proofreads independently.

[0046] As in the case of self-proofreading, the user operates the character recognition system 1 to perform character recognition on an image document and selects the document 42 to be character recognized. When the document 42 is selected, a document acquisition step is executed (step S10), and the character recognition unit 12 executes a character recognition step (step S12). The character recognition unit 12 then generates the document 42 in text format.

[0047] When the character recognition is complete, the control unit 21 displays a process selection screen 32, as shown in Fig. 4, on the operation screen 30 of the display device of the user terminal 2. The process selection screen 32 displays the top page of the document 42 that has been character recognized and converted into text format, and icons 51 for specifying operations for the document 42. On this process selection screen 32, the user selects the "print" or "save" icon 51 to print or save the document 42 that has been converted into text format.

[0048] When the "Print" or "Save" icon 51 is selected, the proofreading confirmation unit 13 executes a proofreading confirmation step to confirm with the user whether proofreading of the document 42 is necessary (step S30).

[0049] Specifically, when the "Print" or "Save" icon 51 is selected, the control unit 21 displays a proofreading confirmation screen 37 on the operation screen 30 of the display device of the user terminal 2, as shown in FIG. 10. The proofreading confirmation screen 37 is a screen for confirming whether or not to enable proofreading before printing or saving. The proofreading confirmation screen 37 has a check box in front of "Enable proofreading." The user selects whether or not proofreading is required on the proofreading confirmation screen 37 (step S32). Specifically, if the user wants to enable proofreading for the document 42, the user uses the input device of the user terminal 2 to select the check box on the proofreading confirmation screen 37. If the document 42 does not require configuration, the user does not select the check box. Thereafter, the user selects the "Yes" icon 54 displayed on the proofreading confirmation screen 37 to confirm the continuation of the process and continues the process. Note that if the "No" icon 54 is selected, the user returns to the process selection screen 32 again.

[0050] When the checkbox is selected and the "Yes" icon 54 is selected, the "print" or "save" process is interrupted, and the character recognition process is executed from the error detection step (step S14) to the corrected document creation step (step S28), just as when the user proofreads independently. When the "Yes" icon 54 is selected without selecting the checkbox, the character recognition process is terminated, and the "print" or "save" process selected by the user is executed.

[0051] As described above, the character recognition system 1 of the present invention can detect special characters that appear when the character recognition process fails to accurately recognize them. When special characters are detected, the user can edit the special characters before outputting the document 42 that has been converted into text format by character recognition. By editing special characters before outputting the document 42, it is possible to reduce garbled characters caused by character recognition and improve the quality of the output document 42.

[0052] Furthermore, if garbled characters are discovered after document 42 has been output, document 42 must be corrected and output again, wasting paper, toner, ink, time, etc. However, character recognition system 1 of the present invention can detect garbled characters and make corrections before document 42 is output, eliminating the waste of paper, toner, ink, time, etc.

[0053] Furthermore, the character recognition system 1 of the present invention is configured to check whether proofreading is necessary when attempting to output the document 42, such as by printing or saving. Therefore, even if the user forgets to proofread the document 42 after character recognition, the document 42 can be proofread before outputting, thereby more reliably improving the quality of the output document 42.

[0054] The embodiments of the present invention have been described above with reference to the drawings. However, the present invention is not limited to the above embodiments and can be implemented in various forms without departing from the spirit of the present invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Furthermore, the components shown in the above embodiments are merely examples and are not particularly limited, and various modifications are possible within a range that does not substantially deviate from the configuration of the present invention.

[0055] For example, in the above embodiment, character recognition system 1 includes proofreading confirmation unit 13, but the present invention is not limited to this. For example, when a user instructs output of document 42, if there is no problem in outputting document 42 according to the user's instruction, character recognition system 1 does not need to include proofreading confirmation unit 13. [Industrial Applicability]

[0056] The present invention can be used in the field of character recognition systems. [Explanation of symbols]

[0057] 1. Character Recognition System 2. User terminal 11 Document Acquisition Section 12 Character recognition section 13 Calibration Verification Section 14 Error detection section 15 Correction Proposal Department 16 Information display section 17 Correction Document Creation Department

Claims

1. a document acquisition unit for acquiring a document in an image format; a character recognition unit that recognizes character information included in the document; an error detection unit that detects error information included in the character information; Equipped with The character recognition system, wherein the error detection unit is capable of detecting, as the error information, a special character that is displayed when the character recognition unit has not correctly recognized the special character.

2. 2. The character recognition system according to claim 1, wherein the error detection unit is capable of detecting, in addition to the special characters, at least one of a spelling error and a grammatical error contained in the character information accurately recognized by the character recognition unit as the error information.

3. Further provided is a calibration confirmation unit that confirms with the user whether calibration is required, The character recognition system according to claim 1 , wherein the proofreading confirmation unit confirms whether proofreading is required before the document is output when a user instructs the document to be output.

4. Further, an information display unit is provided for displaying information about the error information on a user terminal, The character recognition system according to claim 3 , wherein the information display unit displays the number of pieces of error information.

5. a correction proposal proposing unit that proposes correction proposal information for the error information; The character recognition system according to claim 4 , wherein the information display unit displays the correction suggestion information together with the error information.

6. The character recognition system according to claim 5 , further comprising a correction document creation unit that creates a correction document in which the error information is rewritten with the correction proposal information.

7. a document acquisition step of acquiring a document in an image format; a character recognition step of recognizing character information contained in the document; an error detection step of detecting error information included in the character information; causing a computer to execute a process including The character recognition program, wherein the error detection step is capable of detecting, as the error information, a special character that is displayed when the character is not accurately recognized in the character recognition step.

8. causing the computer to further execute a calibration confirmation step for confirming with the user whether calibration is required; 8. The character recognition program according to claim 7, wherein the proofreading confirmation step, when a user instructs execution of output of the document, confirms whether proofreading is necessary before the document is output.

Images