Relay protection setting value automatic checking system and method based on artificial intelligence

Abstract

The invention relates to the field of electric power systems, and discloses a relay protection setting value automatic checking system and method based on artificial intelligence. The method comprises the following steps of: inputting a constant value sheet into a constant value sheet table, carrying out region correction and pre-segmentation processing on the input constant value sheet picture by using an OCR (Optical Character Recognition) pattern recognition technology and a BMHS (Binary Matrix Hazardous Substances) improved algorithm, recognizing characters in the constant value sheet table one by one by using a TableMaster algorithm, and carrying out correlation comparison analysis on a recognition result and a preset dictionary, so as to correct the recognition result. The method comprises the following steps: recombining a corrected identification result into a complete fixed value sheet document, performing item-by-item comparative analysis on the recombined fixed value sheet document and fixed value items in a fixed value sheet published in a system, so as to automatically check the difference problem of the fixed value sheet, generate a check report, and display the check information of the fixed value sheet to a user, thereby improving the user experience. Checking errors can be reduced, and the efficiency and accuracy of constant value checking can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and more particularly to an automatic checking system and method for relay protection setting values based on artificial intelligence. Background Art

[0002] With the continuous development of power systems, the operating modes of power grids change very frequently. To ensure the safe, stable, and reliable operation of power grids, protection devices at various voltage levels also need to be adjusted and checked as the operating modes change. The traditional method of setting value checking mainly relies on manual operations using paper, telephone, fax, and email to check the differences in relay protection setting values. However, the above process still has the following disadvantages: Firstly, the traditional manual checking method requires a large amount of time and effort for setting value sheets, and is prone to errors. It is difficult to monitor the real-time changes of setting value sheets and record the historical changes of setting value sheets, resulting in low efficiency of setting value checking. Secondly, the traditional checking method may lack a unified standardized checking and analysis process, making it difficult to analyze setting value sheets, unable to accurately identify and check setting value errors in setting value sheets, and resulting in the inability to improve the efficiency and accuracy of checking. Summary of the Invention

[0003] To overcome the above-mentioned defects of the prior art, the present invention provides an automatic checking system and method for relay protection setting values based on artificial intelligence to solve the problems existing in the above background art.

[0004] The present invention provides the following technical solutions: An automatic checking system for relay protection setting values based on artificial intelligence, comprising: A setting value sheet picture acquisition module: Relay protection personnel upload pictures of on-site operating setting value sheets and dispatching setting value sheets to the computer terminal. A picture processing module: Using OCR graphic recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the incoming setting value sheet pictures. The processing process includes grayscale, binarization, erosion, and dilation, and transmits the processed setting value sheet pictures to the picture information recognition module. A picture information recognition module: Using the TableMaster algorithm to recognize the text in the setting value sheet table one by one, then associating and comparing the recognition results with a preset dictionary to obtain a matching difference value. Based on the difference matching value, initially judge whether the text recognition result in the cell is accurate, and correct the result of the cell with incorrect text recognition. At the same time, transmit the recognition and correction results to the setting value information recombination module. Fixed value information recombination module: Based on storing the recognized and corrected recognition results in the database, and according to the format of the fixed value list, recombining the recognition results into a complete fixed value list document; Fixed value list comparison and analysis module: By comparing and analyzing each item of the fixed value in the recombined fixed value list document with the fixed value items in the fixed value list already published in the system, calculating the content difference coefficient, and transmitting the content difference coefficient to the fixed value list verification module; Fixed value list verification module: Based on the fixed value list comparison and analysis results, automatically verifying the difference problems of the fixed value list, confirming whether the fixed value list is accurate, and transmitting the verification results to the result display module; Result display module: Based on the verification results, automatically generating a verification report, and displaying the verification information of the fixed value list to the user through the verification report.

[0005] Preferably, for the upload of the fixed value list picture, the fixed value list picture acquisition module is that the relay protection personnel use a mobile device or a camera to take pictures of the on-site operating fixed value list and the dispatching fixed value list, and then the taken fixed value list picture is uploaded to the server through the application program or the web interface on the computer side.

[0006] Preferably, the picture processing module detects the straight lines in the fixed value list picture by using OCR graphic recognition technology. According to the detected straight line information, geometric transformation is performed on the picture to correct the tilt or distortion caused by the shooting angle or the uneven placement of the paper. Then, based on the corrected picture, the BMHS improved algorithm is used to perform pre-segmentation processing on the table layout or text distribution in the picture. By identifying the rows and columns of the table, the picture is segmented into multiple cells. The BMHS improved algorithm analyzes the line spacing and word spacing of the text, thereby dividing the image into cells containing single or multiple text blocks.

[0007] Preferably, the picture information recognition module first locates the table area in the image through the TableMaster algorithm, further divides the table area into individual cells, and numbers each of the divided cells, that is, numbered 1, 2,..., n. Then, the text in each of the 1, 2,..., n cells is recognized one by one; the preset dictionary should contain all correct texts and their corresponding codes or identifiers, which can be set according to the specific content and format of the fixed value list; through the TableMaster algorithm, the text recognized in each cell is compared with the entries in the preset dictionary one by one. Then, based on the recognition results of the text in the cell, the matching difference value between it and the dictionary entry is calculated. The matching difference value is used to identify the similarity or difference degree between the recognition result and the dictionary entry; The specific analysis method of the matching difference value is: Step S311: For the text in each cell, use the TF-IDF model to vectorize the text in each cell, which is then converted into a vector representation as A, and each entry in the preset dictionary is also converted into a vector representation as B; Step S312: For the recognition result vector of the i-th cell and the j-th dictionary entry vector , calculate the cosine similarity as , and , , , where represents the value of the recognition result vector of the i-th cell in the k-th dimension, represents the value of the j-th dictionary entry vector in the k-th dimension, and m represents the dimension of the vector; Step S313: Traverse all dictionary entries, analyze the cosine similarity between each entry and the recognition result of the i-th cell, and calculate the matching difference value as .

[0008] Set a matching difference threshold according to the matching difference value, and compare the matching difference value of each cell with the matching difference threshold to determine whether the text recognition result in the cell is accurate, and correct the recognition error of the cell; if the matching difference value of each cell matches the matching difference threshold , it is considered that the text recognition result in the cell is accurate, and the confirmed correct recognition result is reorganized according to the format and structure of the fixed value form; if the matching difference value of each cell matches the matching difference threshold , it is considered that the text recognition result in the cell is incorrect, and then the information in the preset dictionary is used for automatic correction. If the automatic correction is not feasible, the cell is marked and manually reviewed. After correcting the error, the text in the cell is further reorganized.

[0009] Preferably, the fixed value information reorganization module creates a corresponding data table in the database to store the recognized and corrected cell data, then queries the recognized and corrected cell data from the database, creates a document structure according to the format of the fixed value form, combines the cell data into a complete document, and saves the reorganized fixed value form document.

[0010] Preferably, the fixed value list comparison and analysis module preprocesses the recombined fixed value list document to make its data format consistent, and then further compares and analyzes each fixed value item in the fixed value list already published in the system item by item; The specific calculation formula of the content difference coefficient is , where represents the value of the v-th fixed value item in the standard fixed value list already published in the system, represents the value of the v-th fixed value item in the recombined fixed value list document, represents the weight coefficient of the v-th fixed value item, and N represents the total number of fixed value items.

[0011] Preferably, the fixed value list verification module sets a content difference threshold A, compares the content difference coefficient D with the content difference threshold A. If , it means that there is no difference problem with this fixed value list, and it is determined that the fixed value list is accurate. If , it means that there are differences in this fixed value list, automatically marks the difference items, and records the difference types.

[0012] Preferably, the result display module creates a verification report containing the comparison result, and in the report, uses colors and markers to display the difference items, and at the same time sends the verification report to the user terminal.

[0013] To achieve the above object, the present invention provides the following technical solution: An automatic verification method for relay protection fixed values based on artificial intelligence, using the above-mentioned automatic verification system for relay protection fixed values based on artificial intelligence, including the following steps: Step S1: The relay protection personnel upload the pictures of the on-site operating fixed value list and the dispatching fixed value list to the computer terminal; Step S2: Use OCR graphic recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the incoming fixed value list pictures. The processing process includes grayscale, binarization, erosion, and dilation; Step S3: Use the TableMaster algorithm to recognize the text in the fixed value list table one by one, then associate and compare the recognition results with a preset dictionary to obtain a matching difference value, and preliminarily judge whether the text recognition result in the cell is accurate according to the difference matching value, and correct the result of the cell with incorrect text recognition; Step S4: Store the recognized and corrected recognition results in the database, and recombine the recognition results into a complete fixed value list document according to the format of the fixed value list; Step S5: Compare and analyze each fixed value item in the recombined fixed value list document with the fixed value items in the fixed value list already published in the system, and calculate the content difference coefficient; Step S6: Based on the comparison and analysis results of the setting sheet, automatically check the difference problems of the setting sheet and confirm whether the setting sheet is accurate; Step S7: Based on the check results, automatically generate a check report and display the check information of the setting sheet to the user through the check report.

[0014] Technical effects and advantages of the present invention: In the present invention, through the setting sheet picture acquisition module, relay protection personnel upload pictures of the on-site operating setting sheet and the dispatching setting sheet to the computer terminal. Through the picture processing module, the OCR graphic recognition technology and the BMHS improved algorithm are used to perform regional correction and pre-segmentation processing on the incoming setting sheet pictures. Through the picture information recognition module, the TableMaster algorithm is used to recognize the text in the setting sheet table one by one, and then the recognition results are associated and compared with the preset dictionary for analysis to preliminarily judge whether the text recognition results in the cells are accurate. Through the setting information recombination module, according to the format of the setting sheet, the recognition results are recombined into a complete setting sheet document. Through the setting sheet comparison and analysis module, the recombined setting sheet document is compared and analyzed item by item with the setting items in the published setting sheet in the system. Through the setting sheet check module, the difference problems of the setting sheet are automatically checked and it is confirmed whether the setting sheet is accurate. Finally, through the result display module, a check report is automatically generated and the check information of the setting sheet is displayed to the user, which can continuously track and update the setting information and record the modification and check history of each step, reduce check errors, and through the unified check process and standards, can accurately identify and check the setting errors in the setting sheet, which is beneficial to improving the efficiency and accuracy of setting check. Brief Description of the Drawings

[0015] Figure 1 It is a system structure block diagram of the present invention.

[0016] Figure 2 It is a method step diagram of the present invention Detailed Embodiment

[0017] Next, the technical solutions in the present invention will be clearly and completely described in conjunction with the drawings in the present invention. In addition, the forms of each structure described in the following embodiments are only examples, and a relay protection setting automatic check system and method based on artificial intelligence involved in the present invention are not limited to the structures described in the following embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0018] As Figure 1 shown, this embodiment provides a relay protection setting automatic check system based on artificial intelligence, including: Fixed value list picture acquisition module: Relay protection personnel upload pictures of on-site operating fixed value lists and dispatching fixed value lists to the computer terminal.

[0019] In this embodiment, for the upload of fixed value list pictures by the fixed value list picture acquisition module, relay protection personnel use mobile devices or cameras to take pictures of on-site operating fixed value lists and dispatching fixed value lists, and then the application program or web interface on the computer terminal uploads the taken fixed value list pictures to the server.

[0020] Picture processing module: Use OCR graphic recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the incoming fixed value list pictures. The processing process includes grayscale, binarization, erosion, and dilation, and transmits the processed fixed value list pictures to the picture information recognition module.

[0021] In this embodiment, according to the detected straight line information, geometric transformation is performed on the picture to correct the tilt or distortion caused by the shooting angle or uneven paper placement. Then, based on the corrected picture, the BMHS improved algorithm is used to perform pre-segmentation processing on the table layout or text distribution in the picture. By identifying the rows and columns of the table, the picture is segmented into multiple cells. The BMHS improved algorithm analyzes the line spacing and character spacing of the text to segment the image into cells containing single or multiple characters.

[0022] Picture information recognition module: Use the TableMaster algorithm to identify the text in the fixed value list table one by one, and then perform correlation comparison analysis between the recognition results and the preset dictionary to obtain the matching difference value. According to the difference matching value, initially judge whether the text recognition result in the cell is accurate, and correct the result of the cell with incorrect text recognition. At the same time, transmit the recognition and correction results to the fixed value information recombination module.

[0023] In this embodiment, the picture information recognition module first locates the table area in the image through the TableMaster algorithm, further segments the table area into single cells, and numbers each of the segmented cells, that is, numbered 1, 2,..., n. Then, the text in the 1, 2,..., n cells is identified one by one; the preset dictionary should contain all correct texts and their corresponding codes or identifiers, which can be set according to the specific content and format of the fixed value list; through the TableMaster algorithm, the text recognized in each cell is compared with the entries in the preset dictionary one by one, and based on the recognition result of the text in the cell, calculate the matching difference value between it and the dictionary entry. The matching difference value is used to represent the similarity or difference degree between the recognition result and the dictionary entry; The specific analysis method of the matching difference value is: Step S311: For the text in each cell, use the TF-IDF model to vectorize the text in each cell, which is then converted into a vector representation as A, and each entry in the preset dictionary is also converted into a vector representation as B; Step S312: For the recognition result vector of the i-th cell and the j-th dictionary entry vector , calculate the cosine similarity as , and , , , where represents the value of the recognition result vector of the i-th cell in the k-th dimension, represents the value of the j-th dictionary entry vector in the k-th dimension, and m represents the dimension of the vector; Step S313: Traverse all dictionary entries, analyze the cosine similarity between each entry and the recognition result of the i-th cell, and calculate the matching difference value as .

[0024] Set a matching difference threshold according to the matching difference value, and compare the matching difference value of each cell with the matching difference threshold to determine whether the text recognition result in the cell is accurate, and correct the recognition error of the cell; if the matching difference value of each cell is less than or equal to the matching difference threshold, it is considered that the text recognition result in the cell is accurate, and the confirmed correct recognition result is reorganized according to the format and structure of the fixed value form; if the matching difference value of each cell is greater than the matching difference threshold, it is considered that the text recognition result in the cell is incorrect, and then the information in the preset dictionary is used for automatic correction. If the automatic correction is not feasible, the cell is marked and manually reviewed. After correcting the error, the text in the cell is further reorganized.

[0025] Fixed value information reorganization module: Based on storing the recognized and corrected recognition results in the database, and according to the format of the fixed value form, the recognition results are recombined into a complete fixed value form document.

[0026] In this embodiment, the fixed value information reorganization module creates a corresponding data table in the database to store the recognized and corrected cell data, then queries the recognized and corrected cell data from the database, creates a document structure according to the format of the fixed value form, combines the cell data into a complete document, and saves the reorganized fixed value form document.

[0027] Fixed value order comparison and analysis module: By comparing and analyzing the reorganized fixed value order document with the fixed value items in the fixed value order published in the system item by item, the content difference coefficient is calculated, and the content difference coefficient is transmitted to the fixed value order verification module.

[0028] In this embodiment, the value order comparison and analysis module pre-processes the reorganized value order document to keep its data format consistent, and then further compares and analyzes the value items in the value order published in the system item by item; The specific calculation formula of the content difference coefficient is: ,in, Indicates the value of the vth fixed value item in the standard fixed value list published in the system. Indicates the value of the vth fixed value item in the reorganized fixed value single document. represents the weight coefficient of the vth fixed-value item, and N represents the total number of fixed-value items.

[0029] Fixed value list verification module: Based on the comparison and analysis results of the fixed value list, the difference problems of the fixed value list are automatically checked, and whether the fixed value list is accurate is confirmed, and the verification results are transmitted to the result display module.

[0030] In this embodiment, the fixed value single check module sets a content difference threshold A, compares the content difference coefficient D with the content difference threshold A, and if , it means that there is no discrepancy in the fixed value order, and the fixed value order is determined to be accurate. , it means that there are differences in the fixed value order, and the difference items are automatically marked and the difference types are recorded.

[0031] Result display module: Based on the verification results, a verification report is automatically generated, and the verification information of the fixed value order is displayed to the user through the verification report.

[0032] In this embodiment, the result display module creates a verification report including the comparison results, and uses colors and markers to display the difference items in the report, and sends the verification report to the user terminal.

[0033] This embodiment provides an automatic checking method for relay protection setting values based on artificial intelligence, comprising the following steps: Step S1: The relay protection personnel upload the pictures of the setting sheet and the dispatching setting sheet of the on-site operation to the computer; Step S2: using OCR pattern recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the input fixed-value single image, the processing process includes grayscale, binarization, corrosion, and expansion; Step S3: Use the TableMaster algorithm to recognize the text in the fixed-value list table one by one, then perform an associated comparison analysis between the recognition results and a preset dictionary to obtain a matching difference value, preliminarily judge whether the text recognition result in the cell is accurate based on the difference matching value, and correct the result of the cell with incorrect text recognition; Step S4: Store the recognized and corrected recognition results in the database, and recombine the recognition results into a complete fixed-value list document according to the format of the fixed-value list; Step S5: Compare and analyze each item of the recombined fixed-value list document with the fixed-value items in the fixed-value lists already published in the system, and calculate the content difference coefficient; Step S6: Automatically check the difference problems of the fixed-value list based on the comparison analysis result of the fixed-value list, and confirm whether the fixed-value list is accurate; Step S7: Automatically generate a check report based on the check result, and display the check information of the fixed-value list to the user through the check report.

[0034] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

[0035] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claimed rights.

Claims

1. An automatic checking system for relay protection setting values based on artificial intelligence, characterized in that, include: Setting sheet picture acquisition module: Relay protection personnel upload pictures of setting sheets and dispatching setting sheets in field operation to the computer; Image processing module: Use OCR pattern recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the incoming fixed-value single image. The processing process includes grayscale, binarization, corrosion, and expansion, and the processed fixed-value single image is transmitted to the image information recognition module; Image information recognition module: Use the TableMaster algorithm to recognize the characters in the fixed value table one by one, then compare and analyze the recognition results with the preset dictionary to obtain the matching difference value, and make a preliminary judgment on whether the text recognition results in the cell are accurate based on the difference matching value, and correct the results of the cells with wrong text recognition. At the same time, transmit the recognition and correction results to the fixed value information reorganization module; Value information reorganization module: based on storing the recognition results after identification and correction into the database, and recombining the recognition results into a complete value sheet document according to the format of the value sheet; Value sheet comparison and analysis module: compares and analyzes the reorganized value sheet document with the value items in the value sheets published in the system one by one, calculates the content difference coefficient, and transmits the content difference coefficient to the value sheet verification module; Fixed value list verification module: Based on the comparison and analysis results of the fixed value list, the module automatically verifies the differences in the fixed value list, confirms whether the fixed value list is accurate, and transmits the verification results to the result display module; Result display module: Based on the verification results, a verification report is automatically generated, and the verification information of the fixed value order is displayed to the user through the verification report.

2. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that The setting order picture acquisition module uploads the setting order picture by having the relay protection personnel use a mobile device or camera to take pictures of the setting order and the dispatching setting order running on site, and then the computer application or web interface uploads the taken setting order pictures to the server.

3. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that The image processing module detects straight lines in a fixed-value single image by using OCR graphic recognition technology, performs geometric transformation on the image based on the detected straight line information, corrects the tilt or distortion of the image caused by the shooting angle or uneven paper placement, and then based on the corrected image, pre-segmentates the table layout or text distribution in the image by using the improved BMHS algorithm, identifies the rows and columns of the table, and segments the image into multiple cells. The improved BMHS algorithm segments the image into cells containing single or multiple text blocks by analyzing the line spacing and character spacing of the text.

4. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that The image information recognition module first locates the table area in the image through the TableMaster algorithm, and further divides the table area into individual cells, and numbers each of the divided cells, that is, numbering them as 1, 2, ..., n, and then recognizes the text in 1, 2, ..., n cells one by one; the preset dictionary should contain all correct texts and their corresponding codes or identifiers, which can be set according to the specific content and format of the setting sheet; the recognized text in each cell is compared with the entries in the preset dictionary one by one through the TableMaster algorithm, and then based on the recognition result of the text in the cell, the matching difference value between it and the preset dictionary entry is calculated, and the matching difference value is used to identify the similarity or difference between the result and the dictionary entry; The specific analysis method of the matching difference value is as follows: Step S311: for the text in each cell, use the TF-IDF model to vectorize the text in each cell, convert it into a vector represented as A, and convert each entry in the preset dictionary into a vector represented as B; Step S312: For the recognition result vector of the i-th cell and the j-th dictionary entry vector , calculate the cosine similarity as , and , , , where represents the value of the recognition result vector of the i-th cell in the k-th dimension, represents the value of the j-th dictionary entry vector in the k-th dimension, and m represents the dimension of the vector; Step S313: Traverse all dictionary entries, analyze the cosine similarity between each entry and the recognition result of the i-th cell, and calculate the matching difference value as .

5. Set a matching difference threshold according to the matching difference value , and compare the matching difference value of each cell with the matching difference threshold to determine whether the text recognition result in the cell is accurate, and correct the recognition error of the cell; if the matching difference value of each cell is less than or equal to the matching difference threshold , it is considered that the text recognition result in the cell is accurate, and the confirmed recognition result is reorganized according to the format and structure of the fixed-value form; if the matching difference value of each cell is greater than the matching difference threshold , it is considered that the text recognition result in the cell is incorrect, and the information in the preset dictionary is used for automatic correction. If the automatic correction is not feasible, the cell is marked and manually reviewed. After correcting the error, the text in the cell is further reorganized.

6. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that, The fixed value information reorganization module creates a corresponding data table in the database to store the identified and corrected cell data, then queries the identified and corrected cell data from the database, creates a document structure according to the format of the fixed value sheet, combines the cell data into a complete document, and saves the reorganized fixed value sheet document.

7. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that The value order comparison and analysis module pre-processes the reorganized value order document to keep its data format consistent, and then further compares and analyzes the value items in the value order published in the system item by item; The specific calculation formula for the content difference coefficient is , where represents the value of the v-th fixed value item in the released standard fixed value list in the system, represents the value of the v-th fixed value item in the reorganized fixed value list document, represents the weight coefficient of the v-th fixed value item, and N represents the total number of fixed value items.

8. The automatic relay protection setting value verification system based on artificial intelligence according to claim 1, characterized in that, The fixed-value list verification module compares the content difference coefficient D with a content difference threshold A by setting a content difference threshold A. If , it means that there are no differences in the fixed-value list, and it is determined that the fixed-value list is accurate. If , it means that there are differences in the fixed-value list, automatically marks the different items, and records the difference type.

9. The automatic checking system for relay protection setting values based on artificial intelligence according to claim 1, characterized in that The result display module creates a verification report including the comparison results, and uses colors and markers to display the difference items in the report, and sends the verification report to the user terminal.

10. A method for automatically checking relay protection setting values based on artificial intelligence, using an automatic relay protection setting value checking system based on artificial intelligence as described in any one of claims 1-8, characterized in that: The following steps are involved: Step S1: The relay protection personnel upload the pictures of the setting sheet and the dispatching setting sheet of the on-site operation to the computer; Step S2: using OCR pattern recognition technology and BMHS improved algorithm to perform regional correction and pre-segmentation processing on the input fixed-value single image, the processing process includes grayscale, binarization, corrosion, and expansion; Step S3: Use the TableMaster algorithm to identify the characters in the value list table one by one, then compare and analyze the recognition results with the preset dictionary to obtain the matching difference value, and preliminarily judge whether the character recognition result in the cell is accurate according to the difference matching value, and correct the result of the cell with wrong character recognition; Step S4: Based on storing the recognized and corrected recognition results in a database, and according to the format of the value order, recombining the recognition results into a complete value order document; Step S5: By comparing and analyzing the reorganized value order document with the value items in the value order that has been published in the system one by one, and calculating the content difference coefficient; Step S6: Based on the comparison and analysis results of the fixed value list, the difference problems of the fixed value list are automatically checked, and whether the fixed value list is accurate is confirmed; Step S7: Based on the verification result, automatically generate a verification report, and display the verification information of the setting list to the user through the verification report.

Images