An image information management system and method based on artificial intelligence

Abstract

The invention discloses an image information management system and method based on artificial intelligence, which relates to the field of image data technology, and comprises an archive digitization management module, an image information acquisition module, an object screening module to be verified and an image data management module. The archive is digitally scanned by the archive digitization management module and the image data generated after scanning is stored and processed. The archive scanning information and the generated image quality information are collected by the image information acquisition module. The necessary verification index of the archive is analyzed according to the archive scanning information and the image quality information by the object screening module to be verified, and the target archive is screened out according to the necessary verification index. The image data management module performs storage planning management on the image data corresponding to the target archive, and the verification efficiency and the reliability of the verification result after the digitization of the archive are improved by the electronic archive image data verification and storage management.

Description

Technical Field

[0001] The present invention relates to the technical field of image data management, and in particular to an image information management system and method based on artificial intelligence. Background Art

[0002] Archives digitization is a new form of archives information derived from the development of computer technology, scanning technology, database technology, multimedia technology and storage technology. It can transform archive resources of various carriers into digital archives information, store them in digital form, connect them in a networked form, and manage them using computer systems to form an orderly archives information library, which is conducive to internal archives management and resource sharing. In the process of archives digitization, it is generally necessary to scan paper archives with a scanner to generate image data. After the archives are digitized, in order to ensure the integrity of the image data information, that is, the integrity of the electronic archives information, relevant personnel are still required to perform archive verification, that is, to compare the information of paper archives and electronic archives to confirm the integrity of the image information generated after digital scanning;

[0003] However, firstly, when performing manual file verification, the verification is generally carried out by randomly selecting files. In this way, since the selected files are random and non-targeted, there may be a problem that the electronic image information generated after the digitization of the files that were not selected is incomplete but has not been verified. The existing technology has not improved the random sampling method, and cannot reduce the probability that the information of the files is incomplete but has not been verified after digitization; secondly, during the file verification process, personnel are required to go to the paper file storage place to take the paper file and then verify it. If the paper files taken at one time are too scattered in step by step, it is not conducive to personnel to quickly take the paper files to further improve the probability of file verification. Summary of the invention

[0004] The purpose of the present invention is to provide an image information management system and method based on artificial intelligence to solve the problems raised in the prior art.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an image information management system based on artificial intelligence, the system comprising an archive digitization management module, an image information acquisition module, an object to be verified screening module and an image data management module; the archive is digitally scanned by the archive digitization management module and the image data generated after scanning is stored and processed; the archive scanning information and the generated image quality information are collected by the image information acquisition module; the necessary verification index of the archive is analyzed by the object to be verified screening module based on the archive scanning information and the image quality information, and the target archive is screened out based on the necessary verification index; the image data management module performs storage planning and management on the image data corresponding to the target archive.

[0006] Preferably, the archive digitalization management module includes an archive digitalization scanning unit, an image data generation unit and an image data storage unit. The archive digitalization scanning unit is used to digitize paper archives: scanning paper archives with a scanner; the image data generation unit is used to generate electronic image data corresponding to the paper archives after scanning; the image data storage unit is used to encrypt the generated image data with an AES encryption algorithm, and store the encrypted image data in a data center;

[0007] Using artificial intelligence technology and the AES encryption algorithm to encrypt electronic archive data before storing it can help ensure the storage security of archive information and avoid information leakage.

[0008] Preferably, the image information acquisition module includes a scanning information acquisition unit and an initial image information acquisition unit. The scanning information acquisition unit is used to collect the number of scans of each paper file placed in the same cabinet after the digital scanning of the paper file is completed. Since the images generated after scanning of some paper files may not meet the image quality requirements, in this case, rescanning is required until the generated image meets the image quality requirements. Therefore, more than one scan of a paper file may occur; the initial image information acquisition unit is used to collect image quality information generated when each paper file placed in the same cabinet is scanned for the first time. The image quality information refers to the image clarity information generated after the initial scan.

[0009] Preferably, the object screening module to be verified includes a verification necessity index analysis unit and a target screening unit. The verification necessity index analysis unit is used to analyze the verification necessity index for verifying the electronic file corresponding to each file placed in the same cabinet based on the image quality information generated during the initial scan and the number of scans. The verification processing operation is that relevant personnel compare the paper file with the electronic file generated after digitizing the paper file to see whether the information in the electronic file is complete; the target screening unit is used to compare the verification necessity indexes for verifying different files, and select the target file as the recommended verification file based on the comparison result, and it is recommended that relevant personnel extract the target file for verification during verification.

[0010] Preferably, the image data management module includes an archive distribution information acquisition unit, a distribution information mapping unit and a target data storage management unit. The archive distribution information acquisition unit is used to acquire the distribution information of the paper files corresponding to the target archive, and the distribution information includes the number of rows and columns of the cabinets where the corresponding paper files are placed, and the order information of the paper files placed in the cabinets;

[0011] The distribution information mapping unit is used to establish a three-dimensional coordinate system, map the distribution information of the paper files corresponding to the target archive to the three-dimensional coordinate system, and generate the distribution point coordinates of the paper files corresponding to the target archive;

[0012] The paper files are stored vertically, and the order information indicates the number of files in the cabinet from left to right. For example, if a paper file is the second file from left to right in the cabinet of the first row and the first column, the distribution point coordinates of the corresponding paper file are (1, 1, 2).

[0013] The target data storage management unit is used to select a random point in the three-dimensional coordinate system as a reference comparison point, analyze the straight-line distance from the distribution point of the paper document corresponding to the target file to the reference comparison point, and plan the electronic image data corresponding to the target file for separate partition storage according to the comparison result. The electronic image data corresponding to the target file is the electronic file data generated after scanning the target file. The separate partition storage means that in addition to uniformly storing the image data generated after scanning each paper file placed in the same cabinet in the data center, the electronic image data corresponding to the target file is stored in another partition, that is, the electronic image data corresponding to the target file is stored in two copies, and the target file is one of the files placed in the same cabinet.

[0014] An image information management method based on artificial intelligence comprises the following steps:

[0015] S1: Digitally scan the archives and store and process the image data generated after scanning;

[0016] S2: Collecting archive scan information and generated image quality information;

[0017] S3: Analyze the necessary verification index of the archive according to the archive scanning information and image quality information, and select the target archive according to the necessary verification index;

[0018] S4: Perform separate storage planning and management on the image data corresponding to the target file.

[0019] Preferably, S1 includes: scanning the paper file using a scanner, generating electronic image data corresponding to the paper file after the scanning is completed, encrypting the generated image data using an AES encryption algorithm, and storing the encrypted image data in a data center.

[0020] Preferably, S2 includes: after the digital scanning of the paper archive is completed, the number of times each paper archive placed in a random cabinet is scanned is collected as L={L 1 , L 2 , ...L m}, where m represents the number of paper files placed in a random cabinet. If a paper file has more than one page, the maximum number of times the multiple pages are scanned is counted as the number of times the corresponding paper file is scanned. For example, if a paper file has 3 pages in total and the 3 pages are scanned 1 time, 2 times, and 1 time respectively during the scanning process, the number of times the corresponding paper file is scanned is counted as 2. The image quality information generated when each paper file placed in the same cabinet is scanned for the first time is collected: After obtaining a random paper file, a total of n images are generated. The definition set of the detected n images is E={E 1 , E 2 , ...E n}.

[0021] Preferably, S3 includes: calculating the necessary verification index WR for verifying the electronic file corresponding to a random file placed in a random cabinet according to the following formula: i :

[0022] WR i =(L i ) / [∑ m i＝1 (L i )]+1 / [(∑ n j＝1 (E j )) / n];

[0023] Among them, L i represents the number of times a random paper file placed in a random cabinet is scanned, E j It represents the clarity of the jth image detected after scanning the corresponding paper file to generate the image. The necessary verification index set for verifying the electronic file corresponding to m files is {WR 1 , WR 2 , ...WR i , ...WR m}, compare m necessary verification indexes, randomly divide the m necessary verification indexes into a groups in descending order, and obtain the average value set of the necessary verification index in each group in a group after grouping in a random grouping method is {P 1 , P 2 , ...P a}, according to Y = [∑ a e＝1 (P e -∑ a e＝1 (P e ) / a) 2] / aFilter the target grouping results, where Y represents the reference matching value after grouping by a random grouping method, and the result obtained after grouping by the grouping method with the highest reference matching value is selected as the target grouping result. The electronic files corresponding to the necessary verification indexes of the remaining groups except the last group in the target grouping result are selected as the target files, and the target files are used as the recommended verification files. It is recommended that relevant personnel extract the target files for verification during verification;

[0024] After the paper archives are digitized, when manual archive verification is required, that is, information comparison between paper archives and electronic archives is performed to determine the information completeness of the electronic archives, considering that there are many archives that need to be extracted for verification, the prior art generally extracts archives for verification in a random manner and does not improve the random extraction method. The present invention obtains the necessary verification index of the electronic archives corresponding to different archives by analyzing the scanned information of the archives and the generated image information. During the archive scanning process, if the quality of the image generated after scanning does not meet the requirements, the image needs to be scanned again. The more scanning times and the lower the clarity of the image generated during the initial scanning, the lower the clarity of the information in the paper archive itself, the higher the probability of incomplete information in the electronic archive after the archive is digitized, and the more necessary it is to perform manual archive verification. Considering that although it is an extraction verification method, most archives must be verified to improve the verification value, the archives with low necessary verification indexes are screened out by grouping and clustering, and the remaining archives are used as recommended verification archives, which improves the method of randomly extracting archives for verification without a target, and reduces the probability that the information of the archives is incomplete but not verified after digitization.

[0025] Preferably, S4 includes: obtaining a total of f copies of the target archive, collecting distribution information of the paper archives corresponding to the target archive, the distribution information including the number of rows and columns of the cabinets where the corresponding paper archives are placed and the order information of the paper archives placed in the cabinets, establishing a three-dimensional coordinate system, mapping the distribution information of the paper archives corresponding to the target archives to the three-dimensional coordinate system, generating the coordinates of the distribution points of the paper archives corresponding to the target archives, selecting a point with coordinates (X, Y, Z) as a reference comparison point in the three-dimensional coordinate system, and calculating the straight-line distance set from the distribution point of the paper archive corresponding to the target archive to (X, Y, Z) as d = {d 1 , d 2 ,...d f}, the paper archives corresponding to the target archives are divided into v categories according to the straight-line distance from the distribution point to the benchmark comparison point. The straight-line distance from the distribution point to the benchmark comparison point of all archives in the first category is smaller than that of the second category. In a random classification result, the average straight-line distance from the distribution point to the benchmark comparison point of each category of v is D = {D 1 , D 2 , ...D v}, according to U=[[∑ v r＝1 (D r -∑ v r＝1 (D r ) / v) 2 ] / v] 2 Calculate the reference contribution value U of a random classification result to the storage plan, where r represents the r-th category of archives, obtain the classification result with the largest reference contribution value, store the electronic image data corresponding to the paper archives in the same category in the classification result with the largest reference contribution value in the same storage area, and store the electronic image data corresponding to the v-category paper archives in different storage areas of the data center;

[0026] When verifying the electronic files corresponding to the target files, relevant personnel can extract the electronic files one by one according to the storage area, and find the paper files corresponding to the electronic files stored in the corresponding storage area for information comparison. Since the paper files corresponding to the electronic image data stored in a storage area found after another storage plan are more concentrated, it is helpful to help relevant personnel quickly find the paper files needed for comparison and verification so as to carry out the file verification work as soon as possible, which further improves the efficiency of file verification.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] In the process of digitizing paper archives, the present invention uses artificial intelligence technology and the AES encryption algorithm to encrypt the electronic image data generated after digitization before storing it, which is beneficial to ensure the storage security of archive information and avoid information leakage problems.

[0029] By analyzing the scanned information of the archives and the generated image information, the necessary verification index of the electronic archives corresponding to different archives is obtained. During the archive scanning process, if the quality of the image generated after scanning does not meet the requirements, the image needs to be scanned again. The more scanning times and the lower the clarity of the image generated during the initial scanning, the lower the clarity of the information in the paper archive itself, and the higher the probability of incomplete information in the electronic archive after the archive is digitized, the more necessary it is to conduct manual archive verification. Considering that although it is an extraction verification method, most archives must be verified to improve the verification value, the archives with low necessary verification index are screened out by grouping and clustering, and the remaining archives are used as recommended verification archives, which improves the method of random sampling of archives for verification without a target and reduces the probability that the information of the archives is incomplete but not verified after digitization.

[0030] An additional storage plan is made for the image data generated after scanning the target file. When verifying the electronic file corresponding to the target file, relevant personnel can extract the electronic files one by one according to the storage area, and find the paper files corresponding to the electronic files stored in the corresponding storage area for information comparison. Since the paper files corresponding to the electronic image data stored in a storage area found after the additional storage plan are more concentrated, it is helpful to help relevant personnel quickly find the paper files needed for comparison and verification so as to carry out the file verification work as soon as possible, which further improves the efficiency of file verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a structural schematic diagram of an image information management system based on artificial intelligence of the present invention;

[0032] Figure 2 The figure is a flow chart of an image information management method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Embodiment 1:

[0035] like Figure 1 As shown, this embodiment provides an image information management system based on artificial intelligence, the system includes: an archive digitization management module, an image information acquisition module, a to-be-verified object screening module and an image data management module;

[0036] The output end of the archive digitization management module is connected to the input end of the image information acquisition module, the output end of the image information acquisition module is connected to the input end of the to-be-verified object screening module, and the output end of the to-be-verified object screening module is connected to the input end of the image data management module;

[0037] Digitally scan the archives through the archive digital management module and store and process the image data generated after scanning;

[0038] Collecting archive scanning information and generated image quality information through an image information acquisition module;

[0039] The verification object screening module analyzes the necessary verification index of the archives according to the archive scanning information and image quality information, and screens out the target archives according to the necessary verification index;

[0040] The image data management module is used to manage the storage planning of the image data corresponding to the target file.

[0041] The archive digitization management module includes an archive digitization scanning unit, an image data generation unit and an image data storage unit;

[0042] The output end of the archive digitization scanning unit is connected to the input end of the image data generation unit, and the output end of the image data generation unit is connected to the input end of the image data storage unit;

[0043] The archive digitization scanning unit is used to digitize paper archives: use a scanner to scan paper archives;

[0044] The image data generation unit is used to generate electronic image data corresponding to the paper file after scanning is completed;

[0045] The image data storage unit is used to encrypt the generated image data using the AES encryption algorithm, and store the encrypted image data in the data center.

[0046] The image information acquisition module includes a scanning information acquisition unit and an initial image information acquisition unit;

[0047] The input ends of the scanning information acquisition unit and the initial image information acquisition unit are connected to the output end of the image data storage unit;

[0048] The scanning information acquisition unit is used to collect the number of scans of each paper file placed in the same cabinet after the digital scanning of the paper file is completed. Since the images generated after scanning of some paper files may not meet the image quality requirements, in this case, it is necessary to re-scan until the generated images meet the image quality requirements, so it may happen that a paper file is scanned more than once;

[0049] The initial image information acquisition unit is used to acquire image quality information generated when each paper file placed in the same cabinet is scanned for the first time. The image quality information refers to the image clarity information generated after the initial scan.

[0050] The object screening module to be verified includes a verification necessary index analysis unit and a target screening unit;

[0051] The input end of the verification necessary index analysis unit is connected to the output ends of the scanning information acquisition unit and the initial image information acquisition unit, and the output end of the verification necessary index analysis unit is connected to the input end of the target screening unit;

[0052] The verification necessity index analysis unit is used to analyze the verification necessity index for verifying the electronic files corresponding to each file placed in the same cabinet based on the image quality information generated during the initial scanning and the scanning number information. The verification processing operation is that the relevant personnel compare the paper files with the electronic files generated after the paper files are digitized to check whether the information in the electronic files is complete;

[0053] The target screening unit is used to compare the necessary verification indexes for verification processing of different files, and screen out the target files as recommended verification files based on the comparison results, and recommends that relevant personnel extract the target files for verification during verification.

[0054] The image data management module includes an archive distribution information acquisition unit, a distribution information mapping unit, and a target data storage management unit;

[0055] The input end of the archive distribution information collection unit is connected to the output end of the target screening unit, the output end of the archive distribution information collection unit is connected to the input end of the distribution information mapping unit, and the output end of the distribution information mapping unit is connected to the input end of the target data storage management unit;

[0056] The file distribution information collection unit is used to collect the distribution information of the paper files corresponding to the target files, the distribution information including the number of rows and columns of the cabinets where the corresponding paper files are placed and the order information of the paper files placed in the cabinets;

[0057] The distribution information mapping unit is used to establish a three-dimensional coordinate system, map the distribution information of the paper file corresponding to the target file into the three-dimensional coordinate system, and generate the distribution point coordinates of the paper file corresponding to the target file;

[0058] The paper files are stored vertically, and the order information indicates the number of files in the cabinet from left to right. For example, if a paper file is the second file from left to right in the cabinet of the first row and the first column, the distribution point coordinates of the corresponding paper file are (1, 1, 2).

[0059] The target data storage management unit is used to select a random point in the three-dimensional coordinate system as a benchmark comparison point, analyze the straight-line distance from the distribution point of the paper document corresponding to the target file to the benchmark comparison point, and plan the electronic image data corresponding to the target file for partition storage according to the comparison result. The electronic image data corresponding to the target file is the electronic file data generated after scanning the target file. The partition storage means that in addition to uniformly storing the image data generated after scanning each paper file placed in the same cabinet in the data center, the electronic image data corresponding to the target file is stored in another partition, that is, the electronic image data corresponding to the target file is stored in two copies, and the target file is one of the files placed in the same cabinet.

[0060] Embodiment 2:

[0061] like Figure 2 As shown, this embodiment provides an image information management method based on artificial intelligence, which is implemented based on the image information management system in the embodiment and specifically includes the following steps:

[0062] S1: Scan the archive digitally and store the image data generated after scanning: Scan the paper archive with a scanner, generate electronic image data corresponding to the paper archive after scanning, encrypt the generated image data with the AES encryption algorithm, and store the encrypted image data in the data center;

[0063] S2: Collecting file scanning information and generated image quality information: After the digital scanning of paper files is completed, the number of times each paper file placed in a random cabinet is scanned is L = {L 1 , L 2 , ...L m}, where m represents the number of paper files placed in a random cabinet. If a paper file has more than one page, the maximum number of times the multiple pages are scanned is counted as the number of times the corresponding paper file is scanned. For example, if a paper file has 3 pages in total and the 3 pages are scanned 1 time, 2 times, and 1 time respectively during the scanning process, the number of times the corresponding paper file is scanned is counted as 2. The image quality information generated when each paper file placed in the same cabinet is scanned for the first time is collected: After obtaining a random paper file, a total of n images are generated. The definition set of the detected n images is E={E 1 , E 2 , ...E n};

[0064] S3: Analyze the necessary verification index of the file based on the file scanning information and image quality information, and select the target file based on the necessary verification index: Calculate the necessary verification index WR for the electronic file corresponding to a random file placed in a random cabinet according to the following formula: i :

[0065] WR i =(L i ) / [∑ m i＝1 (L i )]+1 / [(∑ n j＝1 (E j )) / n];

[0066] Among them, Li represents the number of times a random paper file placed in a random cabinet is scanned, E j It represents the clarity of the jth image detected after scanning the corresponding paper file to generate the image. The necessary verification index set for verifying the electronic file corresponding to m files is {WR 1 , WR 2 , ...WR i , ...WR m}, compare m necessary verification indexes, randomly divide the m necessary verification indexes into a groups in descending order, and obtain the average value set of the necessary verification index in each group in a group after grouping in a random grouping method is {P 1 , P 2 , ...P a}, according to Y = [∑ a e＝1 (P e -∑ a e＝1 (P e ) / a) 2 ] / aFilter the target grouping results, where Y represents the reference matching value after grouping by a random grouping method, and the result obtained after grouping by the grouping method with the highest reference matching value is selected as the target grouping result. The electronic files corresponding to the necessary verification indexes of the remaining groups except the last group in the target grouping result are selected as the target files, and the target files are used as the recommended verification files. It is recommended that relevant personnel extract the target files for verification during verification;

[0067] For example, the set of necessary verification indexes for verifying the electronic files corresponding to the 7 files placed in a random cabinet is {WR 1 , WR 2 , WR 3 , WR 4 , WR 5 , WR 6 , WR 7}={0.214, 0.143, 0.070, 0.069, 0.140, 0.283, 0.142}, the 7 necessary verification indexes are in the order from large to small {0.283, 0.214, 0.143, 0.142, 0.140, 0.070, 0.069}, the 7 indexes are divided into 3 groups, and the necessary verification indexes of each group after grouping by a random grouping method are {0.283, 0.214, 0.143}, {0.142, 0.140, 0.070} and {0.069}, and the average value set of the necessary verification indexes in each of the 3 groups after grouping by the corresponding grouping method is {P 1 , P 2, P 3}={0.213, 0.117, 0.069}, and the reference matching value after grouping according to the corresponding grouping method is 0.004. The result obtained after grouping according to the grouping method with the highest reference matching value is selected as the target grouping result. The target grouping result is: each group of necessary verification indexes is {0.283, 0.214}, {0.143, 0.142, 0.140} and {0.070, 0.069} respectively. The electronic archives corresponding to the necessary verification indexes of the remaining groups except the archives with necessary verification indexes of 0.070 and 0.069 are selected as the target archives. The target archives are used as recommended verification archives, and it is recommended that relevant personnel extract the target archives for verification during verification;

[0068] S4: Perform storage planning and management on the image data corresponding to the target archive: obtain a total of f copies of the target archive, collect the distribution information of the paper archive corresponding to the target archive, the distribution information includes the number of rows and columns of the cabinet where the corresponding paper archive is placed, and the order information of the paper archive placed in the cabinet, establish a three-dimensional coordinate system, map the distribution information of the paper archive corresponding to the target archive to the three-dimensional coordinate system, generate the coordinates of the distribution point of the paper archive corresponding to the target archive, select a point with coordinates (X, Y, Z) as the reference comparison point in the three-dimensional coordinate system, and calculate the straight-line distance set from the distribution point of the paper archive corresponding to the target archive to (X, Y, Z) as d = {d 1 , d 2 ,...d f}, the paper archives corresponding to the target archives are divided into v categories according to the straight-line distance from the distribution point to the benchmark comparison point. The straight-line distance from the distribution point to the benchmark comparison point of all archives in the first category is smaller than that of the second category. In a random classification result, the average straight-line distance from the distribution point to the benchmark comparison point of each category of v is D = {D 1 , D 2 , ...D v}, according to U=[[∑ v r＝1 (D r -∑ v r＝1 (D r ) / v) 2 ] / v] 2 Calculate the reference contribution value U of a random classification result to the storage plan, where r represents the rth category of archives. Obtain the classification result with the largest reference contribution value, and store the electronic image data corresponding to the paper archives in the same category in the classification result with the largest reference contribution value in the same storage area. Store the electronic image data corresponding to the vth category of paper archives in different storage areas of the data center.

[0069] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An image information management method based on artificial intelligence, characterized in that: The following steps are involved: S1: Digitally scan the archives and store and process the image data generated after scanning; S2: Collecting archive scan information and generated image quality information; S3: Analyze the necessary verification index of the archive according to the archive scanning information and image quality information, and select the target archive according to the necessary verification index; S4: Perform storage planning and management on the image data corresponding to the target file; S2 includes: after the digital scanning of the paper archive is completed, the number of times each paper archive placed in a random cabinet is scanned is collected as L = {L1, L2, ...L m }, where m represents the number of paper files placed in a random cabinet, and the image quality information generated when each paper file placed in the same cabinet is scanned for the first time is collected: after scanning a random paper file, a total of n images are generated, and the definition set of the n images detected is E = {E1, E2, ...E n }; S3 includes: calculating the necessary verification index WR for verifying the electronic file corresponding to a random file placed in a random cabinet according to the following formula: i : WR i =(L i ) / [∑ m i＝1 (L i )]+1 / [(∑ n j＝1 (E j )) / n]; Among them, L i represents the number of times a random paper file placed in a random cabinet is scanned, E j It represents the definition of the jth image detected after scanning the corresponding paper file to generate the image, and the necessary verification index set for verifying the electronic file corresponding to m files is {WR1, WR2, ...WR i , ...WR m }, compare m necessary verification indexes, randomly divide the m necessary verification indexes into a groups in descending order, and obtain the average value set of the necessary verification index in each group in a group after grouping in a random grouping method as {P1, P2, ...P a }, according to Y = [∑ a e＝1 (P e -∑ a e＝1 (P e ) / a) 2 ] / aFilter the target grouping results, Y represents the reference matching value after grouping by a random grouping method, and the results obtained after grouping by the grouping method with the highest reference matching value are selected as the target grouping results. The electronic files corresponding to the necessary verification indexes of the remaining groups except the last group in the target grouping results are selected as the target files. The target files are used as recommended verification files, and relevant personnel are recommended to extract the target files for verification during verification.

2. The method for image information management based on artificial intelligence according to claim 1, characterized in that: The S1 includes: scanning the paper file with a scanner, generating electronic image data corresponding to the paper file after the scanning is completed, encrypting the generated image data with an AES encryption algorithm, and storing the encrypted image data in a data center.

3. The image information management method based on artificial intelligence according to claim 1, characterized in that: The S4 includes: obtaining a total of f copies of the target archive, collecting distribution information of the paper archives corresponding to the target archive, the distribution information including the number of rows and columns of the cabinets where the corresponding paper archives are placed and the order information of the paper archives placed in the cabinets, establishing a three-dimensional coordinate system, mapping the distribution information of the paper archives corresponding to the target archives to the three-dimensional coordinate system, generating the coordinates of the distribution points of the paper archives corresponding to the target archives, selecting a point with coordinates (X, Y, Z) as a reference comparison point in the three-dimensional coordinate system, and calculating the straight-line distance set from the distribution point of the paper archive corresponding to the target archive to (X, Y, Z) as d={d1, d2, ...d f }, the paper archives corresponding to the target archives are divided into v categories according to the straight-line distance from the distribution point to the benchmark comparison point. The straight-line distance from the distribution point to the benchmark comparison point of all archives in the former category is smaller than that of the latter category. In a random classification result, the average straight-line distance from the distribution point to the benchmark comparison point of each category of v categories is D = {D1, D2, ...D v }, according to U=[[∑ v r＝1 (D r -∑ v r＝1 (D r ) / v) 2 ] / v] 2 Calculate the reference contribution value U of a random classification result to the storage plan, where r represents the rth category of archives. Obtain the classification result with the largest reference contribution value, and store the electronic image data corresponding to the paper archives in the same category in the classification result with the largest reference contribution value in the same storage area. Store the electronic image data corresponding to the vth category of paper archives in different storage areas of the data center.

4. An image information management system based on artificial intelligence, applied to the image information management method based on artificial intelligence as claimed in claim 1, characterized in that: The system includes an archive digitization management module, an image information acquisition module, a to-be-verified object screening module, and an image data management module; Digitally scan the archives through the archive digital management module and store and process the image data generated after scanning; Collecting archive scanning information and generated image quality information through the image information acquisition module; The object screening module to be verified analyzes the necessary verification index of the file according to the file scanning information and the image quality information, and screens out the target file according to the necessary verification index; The image data management module is used to perform storage planning and management on the image data corresponding to the target file.

5. The image information management system based on artificial intelligence according to claim 4, characterized in that: The archive digitization management module includes an archive digitization scanning unit, an image data generation unit and an image data storage unit; The file digitization scanning unit is used to digitize paper files: scanning paper files with a scanner; The image data generating unit is used to generate electronic image data corresponding to the paper file after scanning is completed; The image data storage unit is used to encrypt the generated image data using the AES encryption algorithm, and store the encrypted image data in the data center.

6. The image information management system based on artificial intelligence according to claim 5, characterized in that: The image information acquisition module includes a scanning information acquisition unit and an initial image information acquisition unit; The scanning information collection unit is used to collect the number of scans of each paper file placed in the same cabinet after the digital scanning of the paper file is completed; The initial image information acquisition unit is used to acquire image quality information generated when each paper file placed in the same cabinet is scanned for the first time, and the image quality information refers to image clarity information generated after the initial scan.

7. The image information management system based on artificial intelligence according to claim 6, characterized in that: The object screening module to be verified includes a verification necessary index analysis unit and a target screening unit; The verification necessity index analysis unit is used to analyze the verification necessity index for performing verification processing on the electronic files corresponding to each file placed in the same cabinet based on the image quality information generated during the initial scanning and the scanning number information; The target screening unit is used to compare the necessary verification indexes for verification processing of different files, screen out the target files as recommended verification files based on the comparison results, and recommend relevant personnel to extract the target files for verification during verification.

8. The image information management system based on artificial intelligence according to claim 7, characterized in that: The image data management module includes an archive distribution information acquisition unit, a distribution information mapping unit and a target data storage management unit; The archive distribution information collection unit is used to collect the distribution information of the paper files corresponding to the target archive, wherein the distribution information includes the number of rows and columns of the cabinets where the corresponding paper files are placed, and the order information of the paper files placed in the cabinets; The distribution information mapping unit is used to establish a three-dimensional coordinate system, map the distribution information of the paper files corresponding to the target archive to the three-dimensional coordinate system, and generate the distribution point coordinates of the paper files corresponding to the target archive; The target data storage management unit is used to select a random point in the three-dimensional coordinate system as a reference comparison point, analyze the straight-line distance from the distribution point of the paper document corresponding to the target file to the reference comparison point, and plan the electronic image data corresponding to the target file for partition storage according to the comparison result.

Images