An archive management system based on OCR recognition

Abstract

The application provides an archive management system based on OCR recognition, which can effectively restore and repair a file and ensure the integrity of character symbol information of the file in electronic form obtained subsequently and improve the reliability of archive management of the file by twice OCR scanning and recognition of the file, repairing the incomplete information area of the file, and converting the file according to the character information obtained by the second scanning.

Description

Technical Field

[0001] This invention relates to the technical field of archival document management, and in particular to an archival management system based on OCR recognition. Background Technology

[0002] Current technology for converting paper documents into electronic documents typically requires OCR technology to scan the entire paper document, followed by the organization, classification, and archiving of the resulting electronic files. In practice, large numbers of paper documents are scanned and recognized using OCR in an assembly line, without any inspection of the resulting electronic files. This means that the electronic files may contain unclear scans due to defects in the paper documents themselves, making it impossible to effectively and comprehensively restore the paper documents. Furthermore, it cannot guarantee that the scanned electronic files accurately reflect the text content recorded in the paper documents, thus reducing the reliability of document archiving. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an OCR-based document management system. The system performs a first OCR scan on a document to determine if there is any missing information, and then restores and repairs the missing information areas. A second OCR scan is then performed to obtain the text information present in the document. Based on the obtained text information, the document is converted into an editable or read-only file, and identification information is generated. The editable and read-only files are then entered into the document storage space. Upon receiving a file retrieval request from a user terminal, the system locates the editable or read-only file from the document storage space and returns it to the user terminal. This system, by performing two OCR scans on the document, first repairing any missing information areas, and then converting the document based on the text information obtained from the second scan, effectively restores and repairs the document, ensuring the integrity of the textual information in the subsequent electronic document and improving the reliability of document management.

[0004] This invention provides an OCR-based document management system, comprising:

[0005] The first OCR scanning and recognition module is used to perform the first OCR scanning and recognition process on the file to determine the first information existence status information of the file.

[0006] The document layout integrity status judgment module is used to determine whether there is any missing information on the entire page of the document based on the first information existence status information.

[0007] The document layout repair module restores and repairs the corresponding missing information areas when there are missing information areas on the entire document layout.

[0008] The second OCR scanning and recognition module is used to perform a second OCR scanning and recognition process on files that have no missing information on the entire page or files that have been successfully restored and repaired, to determine the existence status information of the second information of the file; wherein, the scanning action parameters of the second OCR scanning and recognition process are different from the scanning action parameters of the first OCR scanning and recognition process;

[0009] The file identification entry module is used to convert the file into an editable file and a read-only file according to the second information existence status information, and to generate identification information about the file; according to the identification information, the editable file and the read-only file are entered into the corresponding archive storage space;

[0010] The file search module is used to determine the file type information and file access permissions that the user terminal expects to obtain based on the file access request from the user terminal; then, based on the file type information and the file access permissions, it searches for the corresponding editable or read-only file from the archive storage space and returns it to the user terminal.

[0011] Furthermore, the first OCR scanning and recognition module performs the first OCR scanning and recognition process on the file to determine the file's first information existence status information, which specifically includes:

[0012] The first OCR scanning and recognition module performs line-by-line optical scanning of the document according to the number of lines in the document's layout, thereby obtaining the image information corresponding to each line in the document's layout, which is used as the first information existence status information.

[0013] The second OCR scanning and recognition module performs a second OCR scanning and recognition process to determine the existence status information of the second information of the file, specifically including:

[0014] The second OCR scanning and recognition module performs line-by-line optical scanning of the document according to the number of lines in the document layout, thereby obtaining the image information corresponding to one line in the document layout, which is used as the second information existence status information.

[0015] Wherein, the optical scanning duration for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is less than the optical scanning duration for each line in the line-by-line optical scanning process of the second OCR scanning and recognition module; or, the average scanning illumination intensity for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is less than the average scanning illumination intensity for each line in the line-by-line optical scanning process of the second OCR scanning and recognition module.

[0016] Furthermore, the first OCR scanning and recognition module performs the first OCR scanning and recognition process on the file, and the first information existence status information of the file also includes:

[0017] The first OCR scanning and recognition module performs multiple line-by-line optical scans of the document's layout based on the number of lines, thereby obtaining multiple image information corresponding to each line of the document's layout. It analyzes the clarity of these multiple images corresponding to each line, selects images whose clarity meets certain conditions, and performs weighted image overlay processing to obtain the final image information corresponding to each line of the document's layout. This final image information is used as the first information's existence status information. The specific process is as follows:

[0018] Step S1: Using the following formula (1), analyze the clarity of multiple images corresponding to each row in the document's layout.

[0019]

[0020] In formula (1) above, Q(a) represents the resolution value of the a-th image corresponding to each line in the document layout; W a (i, j) represents the pixel value of the i-th row and j-th column pixel of the a-th image corresponding to each row in the layout of the file; W a (i-1, j) represents the pixel value of the (i-1)th row and jth column of the image corresponding to each row in the file layout; | represents the absolute value; m represents the total number of pixels in any row of any image corresponding to each row in the file layout; n represents the total number of pixels in any column of any image corresponding to each row in the file layout; G represents the total number of scanned images corresponding to each row in the file layout; ∈ represents the membership symbol; min a∈[1，G] {} indicates that the value of a is taken from 1 to G and substituted into the parentheses to obtain the minimum value inside the parentheses;

[0021] Step S2: Using formula (2) below, obtain the superimposed transparency weight value of each image based on the clarity of the multiple images corresponding to each row.

[0022]

[0023] In the above formula (2), B(a) represents the superimposed transparency weight value of the a-th image corresponding to each row in the layout of the document;

[0024] Step S3: Using the formula (3) below, select the images with higher clarity based on the clarity of multiple images corresponding to each row, and perform image weighted superposition processing on the multiple images to obtain the final image information corresponding to each row in the document layout.

[0025]

[0026] In the above formula (3), W represents the final image corresponding to each row in the layout of the document obtained by performing image weighted superposition processing on multiple images corresponding to each row; Indicates in and In this case, the transparency of the a-th image is adjusted to B(a), and the images are superimposed in order from 1 to G to obtain the final image.

[0027] Furthermore, the document layout integrity judgment module, which is used to determine whether there is any missing information on the entire document layout based on the first information existence status information, specifically includes:

[0028] The document layout integrity judgment module extracts the outline information of each line containing text symbols from the image information corresponding to each line in the document layout.

[0029] The outline information of each text symbol is processed by fuzzy recognition to determine the actual text symbol content corresponding to each text symbol; based on the text symbol content, the outline information of the corresponding text symbol is extracted from a preset text symbol database; the outline information of the text symbol extracted from the image information is compared with the outline information of the text symbol extracted from the preset text symbol database to determine whether there is any outline incompleteness of the text symbol corresponding to each line in the layout of the document.

[0030] Furthermore, when information is missing from the entire document layout, the document layout repair module restores and repairs the corresponding missing information areas, specifically including:

[0031] The document layout repair module determines the area ratio between the missing outline area and the entire text symbol area in each line of the document layout; if the area ratio is greater than or equal to a preset area, it is determined that the text symbol with missing outline cannot be restored and repaired; otherwise, it is determined that the text symbol with missing outline can be restored and repaired.

[0032] Determine the size and font type information of the text symbols that can be restored and repaired; based on the size and font type information, redraw and fill the missing parts of the text symbols that can be restored and repaired, thereby achieving the corresponding restoration and repair.

[0033] Furthermore, the file identification input module, based on the second information existence status information, converts the file into an editable file and a read-only file, and generates identification information about the file, specifically including:

[0034] The file identification input module directly treats the image information corresponding to the second information existence status information as a read-only file;

[0035] The file recognition and input module extracts the corresponding text symbols from the image information corresponding to the second information existence status information, and converts all the extracted text symbols into editable text symbols; then, based on all the editable text symbols, it generates an editable file corresponding to the file.

[0036] The document recognition and input module also extracts the document header information or document page number information corresponding to the document from all editable text symbols, and generates identification information about the document based on the document header information or document page number information.

[0037] Furthermore, the file identification input module, based on the identification information, inputs the editable file and the read-only file into the corresponding archive storage space, specifically including:

[0038] The file identification entry module uses the identification information as the sector index information of one of the storage sectors in the file storage space, and enters and stores the editable files and read-only files corresponding to the same file into the storage sector.

[0039] Furthermore, the file search module determines, based on the user terminal's file retrieval request, the file type information the user terminal expects to obtain and the user terminal's file retrieval permissions, specifically including:

[0040] The file search module extracts the file type information and the user terminal's identity information from the file retrieval request from the user terminal.

[0041] The terminal identity information is compared with a preset terminal identity information database to determine whether the user terminal has the permission to access editable files or read-only files.

[0042] Furthermore, the file search module, based on the file type information and the file access permissions, searches for corresponding editable or read-only files in the archive storage space and returns them to the user terminal, specifically including:

[0043] When the user terminal has the permission to access editable files, the file search module matches and compares the file type information with the sector index information of each storage sector of the archive storage space, and returns the editable file of the finally matched storage sector to the user terminal.

[0044] When the user terminal has read-only file access permission, the file search module matches and compares the file type information with the sector index information of each storage sector of the file storage space, and returns the read-only file of the finally matched storage sector to the user terminal.

[0045] Compared to existing technologies, this OCR-based document management system performs a first OCR scan to identify any missing information and repairs the damaged areas. A second OCR scan then retrieves the text information. Based on this text, the system converts the document into editable and read-only files, generates identification information, and stores these files in the document storage space. Upon receiving a file retrieval request from a user terminal, the system locates the editable or read-only file in the storage space and returns it to the user terminal. By performing two OCR scans, first repairing any missing information, and then converting the document based on the second scan's text information, the system effectively restores and repairs the document, ensuring the integrity of the text information in the subsequent electronic document and improving the reliability of document management.

[0046] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0047] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the structure of an OCR-based document management system provided by the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] See Figure 1 This is a schematic diagram of a document management system based on OCR recognition, provided as an embodiment of the present invention. The document management system based on OCR recognition includes:

[0052] The first OCR scanning and recognition module is used to perform the first OCR scanning and recognition process on the file to determine the first information existence status information of the file.

[0053] The document layout integrity status judgment module is used to determine whether there is any missing information on the entire page of the document based on the first information existence status information.

[0054] The document layout repair module restores and repairs the corresponding missing information areas when there are missing information areas on the entire document layout.

[0055] The second OCR scanning and recognition module is used to perform a second OCR scanning and recognition process on files that do not have any missing information on the entire page or files that have been successfully restored and repaired, to determine the existence status information of the second information of the file; wherein, the scanning action parameters of the second OCR scanning and recognition process are different from the scanning action parameters of the first OCR scanning and recognition process.

[0056] The file identification entry module is used to convert the file into an editable file and a read-only file according to the existence status information of the second information, and to generate identification information about the file; according to the identification information, the editable file and the read-only file are entered into the corresponding archive storage space;

[0057] The file search module is used to determine the file type information and file access permissions that the user terminal expects to obtain based on the file access request from the user terminal; then, based on the file type information and file access permissions, it searches for the corresponding editable or read-only file in the file storage space and returns it to the user terminal.

[0058] The beneficial effects of the above technical solution are as follows: This OCR-based document management system performs a first OCR scan on the document to determine if there is any missing information, and then restores and repairs the missing information areas. A second OCR scan is then performed to obtain the text information present in the document. Based on the obtained text information, the document is converted into an editable file and a read-only file, and identification information about the document is generated. The editable and read-only files are then entered into the document storage space. Based on a file retrieval request from a user terminal, the editable or read-only file is located from the document storage space and returned to the user terminal. By performing two OCR scans on the document, first repairing any missing information areas, and then converting the document based on the text information obtained from the second scan, the system effectively restores and repairs the document, ensuring the integrity of the textual symbol information in the subsequent electronic document and improving the reliability of document management.

[0059] Preferably, the first OCR scanning and recognition module performs a first OCR scanning and recognition process on the file to determine the first information existence status information of the file, specifically including:

[0060] The first OCR scanning and recognition module performs line-by-line optical scanning of the document based on the number of lines in the document's layout, thereby obtaining the image information corresponding to each line in the document's layout, which is used as the existence status information of the first information.

[0061] The second OCR scanning and recognition module performs a second OCR scanning and recognition process to determine the existence status information of the second information of the file, specifically including:

[0062] The second OCR scanning and recognition module performs line-by-line optical scanning of the document according to the number of lines in the document's layout, thereby obtaining the image information corresponding to one line in the document's layout, which is used as the existence status information of the second information.

[0063] Wherein, the optical scanning duration of each line in the progressive optical scanning process of the first OCR scanning and recognition module is less than the optical scanning duration of each line in the progressive optical scanning process of the second OCR scanning and recognition module; or, the average scanning illumination intensity of each line in the progressive optical scanning process of the first OCR scanning and recognition module is less than the average scanning illumination intensity of each line in the progressive optical scanning process of the second OCR scanning and recognition module.

[0064] The beneficial effects of the above technical solution are as follows: the first OCR scanning and recognition module and the second OCR scanning and recognition module can be, but are not limited to, optical scanners. The first OCR scanning and recognition module is only used to perform a preliminary optical scan of the document to identify missing text outlines from the scanned image. The second OCR scanning and recognition module, on the other hand, is used to perform a secondary optical scan of the document to identify every single character on the document's layout. It is clear that the two OCR scans have different purposes. The first OCR scan does not require detailed recognition of every single character on the entire document, while the second OCR scan does. Therefore, the optical scanning duration for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is set to be shorter than that of the line-by-line optical scanning process of the second OCR scanning and recognition module; or, the average scanning illumination intensity for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is set to be lower than that of the line-by-line optical scanning process of the second OCR scanning and recognition module. This reduces the optical scanning workload of the first OCR scanning and recognition module, effectively shortening its working time and improving its efficiency.

[0065] Preferably, the first OCR scanning and recognition module performs a first OCR scanning and recognition process on the file, and the determination of the file's first information existence status information further includes:

[0066] The first OCR scanning and recognition module performs multiple line-by-line optical scans of the document's layout based on the number of lines, thereby obtaining multiple image information corresponding to each line of the document's layout. It analyzes the clarity of these multiple images corresponding to each line, selects images whose clarity meets certain conditions, and performs weighted image overlay processing to obtain the final image information corresponding to each line of the document's layout. This final image information is used as the first information's existence status information. The specific process is as follows:

[0067] Step S1: Using the formula (1) below, analyze the clarity of the multiple images corresponding to each row in the document's layout.

[0068]

[0069] In formula (1) above, Q(a) represents the resolution value of the a-th image corresponding to each line in the layout of the file; W a (i, j) represents the pixel value of the i-th row and j-th column of the a-th image corresponding to each row in the file layout; W a(i-1, j) represents the pixel value of the (i-1)th row and jth column of the image corresponding to each row in the file layout; || represents the absolute value; m represents the total number of pixels in any row of any image corresponding to each row in the file layout; n represents the total number of pixels in any column of any image corresponding to each row in the file layout; G represents the total number of scanned images corresponding to each row in the file layout; ∈ represents the membership symbol; min a∈[1，G] {} indicates that the value of a is taken from 1 to G and substituted into the parentheses to obtain the minimum value inside the parentheses;

[0070] Step S2: Using formula (2) below, obtain the superimposed transparency weight value of each image based on the clarity of the multiple images corresponding to each row.

[0071]

[0072] In the above formula (2), B(a) represents the superimposed transparency weight value of the a-th image corresponding to each row in the layout of the file;

[0073] Step S3: Using the formula (3) below, select the images with higher clarity based on the clarity of multiple images corresponding to each row, and perform image weighted overlay processing on the multiple images to obtain the final image information corresponding to each row in the layout of the file.

[0074]

[0075] In the above formula (3), W represents the final image corresponding to each row in the layout of the file obtained by performing image weighted superposition processing on multiple images corresponding to each row; Indicates in and In this case, the transparency of the a-th image is adjusted to B(a), and the images are superimposed in order from 1 to G to obtain the final image.

[0076] The beneficial effects of the above technical solution are as follows: Using the above formula (1), the clarity of multiple images corresponding to each row in the document layout is analyzed, thereby removing images with low clarity to prevent subsequent recognition from being affected and improve the efficiency of the system; then using the above formula (2), the superposition transparency weight value of each image is obtained according to the clarity of multiple images corresponding to each row, thereby quantifying the clarity of each image to facilitate subsequent image superposition; finally using the above formula (3), the images with higher clarity are selected according to the clarity of multiple images corresponding to each row, and the images are weighted and superimposed to obtain the final image information corresponding to each row in the document layout, thereby obtaining the latest image. The latest image removes unclear elements and deepens the display in the image in the form of superposition, ensuring that the result of the first OCR scan recognition process is more accurate and the scanned image is clearer.

[0077] Preferably, the document layout integrity judgment module, which is used to determine whether there is any missing information on the entire page of the document based on the first information existence status information, specifically includes:

[0078] The file layout integrity judgment module extracts the outline information of the text symbols in each line from the image information corresponding to each line in the file layout;

[0079] The outline information of each text symbol is processed by fuzzy recognition to determine the actual text symbol content corresponding to each text symbol; based on the text symbol content, the outline information of the corresponding text symbol is extracted from the preset text symbol database; the outline information of the text symbol extracted from the image information is compared with the outline information of the text symbol extracted from the preset text symbol database to determine whether there is any outline incompleteness of the text symbol corresponding to each line in the layout of the document.

[0080] The beneficial effects of the above technical solution are as follows: The document layout integrity judgment module extracts the outline information of the text symbols in each line from the image information of each line of the document layout obtained by the first OCR scanning and recognition module. Then, it performs fuzzy recognition processing on the extracted outline information, which can estimate the actual text symbol content represented by each text symbol (such as Chinese characters, English letters, or other numerical symbols). The extracted outline information of the text symbols is compared with the outline information of the text symbols extracted from the preset text symbol database (the outline information of each text symbol in the preset text symbol database is complete outline information). This can obtain the difference between the two outline information. If the difference between the two outline information is greater than the preset difference threshold, it is determined that the corresponding text symbol has an outline incompleteness. Otherwise, it is determined that the corresponding text symbol does not have an outline incompleteness. This can quickly and accurately locate the outline incompleteness of the text symbols on the layout.

[0081] Preferably, when the document layout repair module has missing information on the entire document layout, the specific steps for restoring and repairing the corresponding missing information areas include:

[0082] The document layout repair module determines the area ratio between the missing outline area and the entire text symbol area in each line of the document layout. If the area ratio is greater than or equal to the preset area, it is determined that the text symbol with missing outline cannot be restored and repaired; otherwise, it is determined that the text symbol with missing outline can be restored and repaired.

[0083] Determine the size and font type information of the text symbols that can be restored and repaired; based on the size and font type information, redraw and fill the missing parts of the text symbols that can be restored and repaired, thereby achieving the corresponding restoration and repair.

[0084] The beneficial effects of the above technical solution are as follows: Using the above method, the ratio of the area of ​​the missing outline in each line of text symbols with incomplete outlines to the entire text symbol area of ​​each line is used as a benchmark. When this ratio is greater than or equal to a preset area, it indicates that the missing outline area in each line of text symbols with incomplete outlines is too large, and the corresponding text symbol cannot be completely restored and repaired. Otherwise, the missing outline area in each line of text symbols with incomplete outlines is small, and the corresponding text symbol can be completely restored and repaired. Next, the size and font type information of the text symbols that can be restored and repaired are used as reference benchmarks for restoration and repair. This allows for the re-outlining and filling of the missing outline of the corresponding text symbols, thereby achieving complete restoration and repair of the text symbols.

[0085] Preferably, the file identification entry module converts the file into an editable file and a read-only file based on the existence status information of the second information, and generates identification information about the file, specifically including:

[0086] The file identification entry module directly treats the image information corresponding to the second information existence status information as a read-only file;

[0087] The document recognition and input module extracts the corresponding text symbols from the image information corresponding to the second information existence status information, and converts all the extracted text symbols into editable text symbols; then, based on all the editable text symbols, it generates an editable file corresponding to the document.

[0088] The document recognition and input module also extracts the corresponding document header information or document page number information from all editable text symbols, and generates identification information about the document based on the document header information or document page number information.

[0089] The beneficial effects of the above technical solution are as follows: The file identification and input module extracts the intact text symbols from the image information corresponding to the second information existence status information, and combines them with the restored text symbols to generate a read-only file and an editable file that match the original file. This allows for the generation of two files with different file editing permissions, facilitating the subsequent allocation of appropriate versions of the file to different user terminals. Furthermore, based on the file's header information or page number information, identification information about the file is generated, enabling quick and accurate file retrieval in the future.

[0090] Preferably, the file identification entry module, based on the identification information, enters the editable file and the read-only file into the corresponding archive storage space, specifically including:

[0091] The file identification entry module uses this identification information as the sector index information of one of the storage sectors in the file storage space, and also enters and stores the editable and read-only files corresponding to the same file into this storage sector.

[0092] The beneficial effects of the above technical solution are as follows: through the above method, editable files and read-only files obtained by conversion can be stored separately, thereby improving the storage efficiency and reliability of the corresponding files.

[0093] Preferably, the file search module determines, based on the user terminal's file retrieval request, the file type information the user terminal expects to obtain and the user terminal's file retrieval permissions, specifically including:

[0094] The file search module extracts the file type information and the user terminal's identity information from the file retrieval request from the user terminal.

[0095] The terminal's identity information is compared with a preset terminal identity information database to determine whether the user terminal has the permission to access editable files or read-only files.

[0096] The beneficial effects of the above technical solution are as follows: by using the user terminal's identity information as a basis, it is determined whether the user terminal has the permission to access editable files or read-only files, thereby preventing files stored in the archive storage space from being stolen by unauthorized user terminals.

[0097] Preferably, the file search module searches for the corresponding editable or read-only file in the archive storage space and returns it to the user terminal based on the file type information and file access permissions. Specifically, this includes:

[0098] When the user terminal has the permission to access editable files, the file search module matches and compares the file type information with the sector index information of each storage sector in the file storage space, and returns the editable file of the finally matched storage sector to the user terminal.

[0099] When a user terminal has read-only file access permission, the file search module matches and compares the file type information with the sector index information of each storage sector in the file storage space, and returns the read-only file of the finally matched storage sector to the user terminal.

[0100] The beneficial effects of the above technical solution are as follows: through the above method, editable or read-only files can be returned to the user terminal in a targeted manner, which can ensure the security and reliability of file archive management.

[0101] As can be seen from the above embodiments, this OCR-based document management system performs a first OCR scan on the document to determine if there is any missing information, and then restores and repairs the missing information areas. A second OCR scan is then performed to obtain the text information present in the document. Based on the obtained text information, the document is converted into an editable file and a read-only file, and identification information about the document is generated. The editable and read-only files are then entered into the document storage space. Based on a file retrieval request from a user terminal, the editable or read-only file is located from the document storage space and returned to the user terminal. By performing two OCR scans on the document, first repairing any missing information areas, and then converting the document based on the text information obtained from the second scan, the system effectively restores and repairs the document, ensuring the integrity of the textual symbol information in the subsequent electronic document and improving the reliability of document management.

[0102] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A document management system based on OCR recognition, characterized in that, It includes: The first OCR scanning and recognition module is used to perform the first OCR scanning and recognition process on the file to determine the first information existence status information of the file. The document layout integrity status judgment module is used to determine whether there is any missing information on the entire page of the document based on the first information existence status information. The document layout repair module restores and repairs the corresponding missing information areas when there are missing information areas on the entire document layout. The second OCR scanning and recognition module is used to perform a second OCR scanning and recognition process on files that have no missing information on the entire page or files that have been successfully restored and repaired, to determine the existence status information of the second information of the file; wherein, the scanning action parameters of the second OCR scanning and recognition process are different from the scanning action parameters of the first OCR scanning and recognition process; The file identification entry module is used to convert the file into an editable file and a read-only file according to the second information existence status information, and to generate identification information about the file; according to the identification information, the editable file and the read-only file are entered into the corresponding archive storage space; The file search module is used to determine the file type information and file access permissions that the user terminal expects to obtain based on the file access request from the user terminal; and then, based on the file type information and the file access permissions, to search for the corresponding editable or read-only file from the archive storage space and return it to the user terminal. The first OCR scanning and recognition module performs the first OCR scanning and recognition process on the file, and the first information existence status information of the file also includes: Step S1: Using the following formula (1), analyze the clarity of multiple images corresponding to each row in the document's layout. In formula (1) above, Q(a) represents the resolution value of the a-th image corresponding to each line in the document layout; W a (i,j) represents the pixel value of the i-th row and j-th column pixel of the a-th image corresponding to each row in the layout of the file; W a (i-1,j) represents the pixel value of the (i-1)th row and jth column of the image corresponding to each row in the file layout; | represents the absolute value; m represents the total number of pixels in any row of any image corresponding to each row in the file layout; n represents the total number of pixels in any column of any image corresponding to each row in the file layout; G represents the total number of scanned images corresponding to each row in the file layout; ∈ represents the membership symbol; min a∈[1,G] {} indicates that the value of a is taken from 1 to G and substituted into the parentheses to obtain the minimum value inside the parentheses; Step S2: Using formula (2) below, obtain the superimposed transparency weight value of each image based on the clarity of the multiple images corresponding to each row. In the above formula (2), B(a) represents the superimposed transparency weight value of the a-th image corresponding to each row in the layout of the document; Step S3: Using the formula (3) below, select the images with higher clarity based on the clarity of multiple images corresponding to each row, and perform image weighted superposition processing on the multiple images to obtain the final image information corresponding to each row in the document layout. In the above formula (3), W represents the final image corresponding to each row in the layout of the document obtained by performing image weighted superposition processing on multiple images corresponding to each row; Indicates in and In this case, the transparency of the a-th image is adjusted to B(a), and the images are superimposed in order from 1 to G to obtain the final image.

2. The OCR-based document management system as described in claim 1, characterized in that: The second OCR scanning and recognition module performs a second OCR scanning and recognition process to determine the existence status information of the second information of the file, specifically including: The second OCR scanning and recognition module performs line-by-line optical scanning of the document according to the number of lines in the document layout, thereby obtaining the image information corresponding to one line in the document layout, which is used as the second information existence status information. Wherein, the optical scanning duration for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is less than the optical scanning duration for each line in the line-by-line optical scanning process of the second OCR scanning and recognition module; or, the average scanning illumination intensity for each line in the line-by-line optical scanning process of the first OCR scanning and recognition module is less than the average scanning illumination intensity for each line in the line-by-line optical scanning process of the second OCR scanning and recognition module.

3. The OCR-based document management system as described in claim 2, characterized in that: The document layout integrity determination module is used to determine whether there is any missing information on the entire document layout based on the first information existence status information. Specifically, it includes: The document layout integrity judgment module extracts the outline information of each line containing text symbols from the image information corresponding to each line in the document layout. The outline information of each text symbol is processed by fuzzy recognition to determine the actual text symbol content corresponding to each text symbol; based on the text symbol content, the outline information of the corresponding text symbol is extracted from a preset text symbol database; the outline information of the text symbol extracted from the image information is compared with the outline information of the text symbol extracted from the preset text symbol database to determine whether there is any outline incompleteness of the text symbol corresponding to each line in the layout of the document.

4. The OCR-based document management system as described in claim 3, characterized in that: When the file layout repair module has missing information on the entire file layout, it restores and repairs the corresponding missing information areas. Specifically, this includes: The document layout repair module determines the area ratio between the missing outline area and the entire text symbol area in each line of the document layout; if the area ratio is greater than or equal to a preset area, it is determined that the text symbol with missing outline cannot be restored and repaired; otherwise, it is determined that the text symbol with missing outline can be restored and repaired. Determine the size and font type information of the text symbols that can be restored and repaired; based on the size and font type information, redraw and fill the missing parts of the text symbols that can be restored and repaired, thereby achieving the corresponding restoration and repair.

5. The OCR-based document management system as described in claim 4, characterized in that: The file identification input module converts the file into an editable file and a read-only file based on the second information existence status information, and generates identification information about the file, specifically including: The file identification input module directly treats the image information corresponding to the second information existence status information as a read-only file; The file recognition and input module extracts the corresponding text symbols from the image information corresponding to the second information existence status information, and converts all the extracted text symbols into editable text symbols; then, based on all the editable text symbols, it generates an editable file corresponding to the file. The document recognition and input module also extracts the document header information or document page number information corresponding to the document from all editable text symbols, and generates identification information about the document based on the document header information or document page number information.

6. The OCR-based document management system as described in claim 5, characterized in that: The file identification input module, based on the identification information, inputs the editable file and the read-only file into the corresponding archive storage space, specifically including: The file identification entry module uses the identification information as the sector index information of one of the storage sectors in the file storage space, and enters and stores the editable files and read-only files corresponding to the same file into the storage sector.

7. The OCR-based document management system as described in claim 6, characterized in that: The file search module determines, based on the user terminal's file retrieval request, the file type information the user terminal expects to obtain and the user terminal's file retrieval permissions, specifically including: The file search module extracts the file type information and the user terminal's identity information from the file retrieval request from the user terminal. The terminal identity information is compared with a preset terminal identity information database to determine whether the user terminal has the permission to access editable files or read-only files.

8. The OCR-based document management system as described in claim 7, characterized in that: The file search module searches for corresponding editable or read-only files from the archive storage space and returns them to the user terminal based on the file type information and file access permissions. Specifically, this includes: When the user terminal has the permission to access editable files, the file search module matches and compares the file type information with the sector index information of each storage sector of the archive storage space, and returns the editable file of the finally matched storage sector to the user terminal. When the user terminal has read-only file access permission, the file search module matches and compares the file type information with the sector index information of each storage sector of the file storage space, and returns the read-only file of the finally matched storage sector to the user terminal.

Images