Document data processing method for financial reimbursement

Abstract

The invention relates to the technical field of document processing, in particular to a document data processing method for financial reimbursement, which comprises the following steps: acquiring image information of each financial document, preprocessing each image information, and extracting data features of the preprocessed image information; based on the extracted data features, performing statistics on a to-be-reimbursed amount; through automatic data acquisition, identification and statistical processes, the time and workload of manual operation are greatly reduced, the processing speed of a financial reimbursement process is remarkably improved, whether data checking is qualified or not is determined based on a comparison result of a statistical amount and an expected amount, and when it is judged that data checking is abnormal, the data checking efficiency is improved. According to the method, the parameters for reprocessing the financial documents are determined based on the data difference ratio, the number of the items in each financial document and the item data difference quantity, the accuracy of financial data acquisition is improved, and then the processing efficiency for the document data is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of document processing, and particularly to a method for processing document data for financial reimbursement. Background Art

[0002] In today's digital office environment, the financial reimbursement process involves a large amount of document data processing. The traditional method for processing financial reimbursement data mainly relies on manual input and verification. The manual processing efficiency is low. Facing a large number of reimbursement vouchers, financial personnel need to spend a lot of time on data entry and review, which easily leads to a long reimbursement process cycle, and manual operations are prone to errors.

[0003] With the expansion of enterprise scale and the diversification of business, the types and complexity of reimbursement data are constantly increasing. The traditional method is difficult to comprehensively and accurately analyze and manage the data, and it is impossible to timely discover potential financial problems and abnormal situations. Therefore, there is an urgent need for an efficient, accurate, and intelligent method for processing financial reimbursement document data to improve the quality and efficiency of financial work and reduce financial risks.

[0004] Chinese Patent Publication No.: CN119107659A discloses a method and device for identifying document-based financial reports in a natural scene, including: receiving a financial report file to be identified, converting it into an image format to obtain a financial report image; correcting the orientation of the financial report image and erasing the covering objects on the financial report image; using a first preset deep learning algorithm to detect the position of the table on the financial image and crop out the table area; in the table area, using a second preset deep learning algorithm to detect the unit lines of the table structure, and dividing the cells according to the unit lines; performing centering processing on each cell to extract the information of each cell; standardizing the extracted information according to the subject category to unify the description form to obtain the recognition result; and performing structured output on the recognition result according to the business scenario. It can be seen that the above technical solution has the following problems: it does not consider analyzing the extracted data to adjust the parameters for processing financial documents in a timely manner when abnormal situations occur, which affects the processing efficiency of document data. Summary of the Invention

[0005] Therefore, the present invention provides a method for processing document data for financial reimbursement to overcome the problem in the prior art that it does not consider analyzing the extracted data to adjust the parameters for processing financial documents in a timely manner when abnormal situations occur, which affects the processing efficiency of document data.

[0006] To achieve the above object, the present invention provides a method for processing document data for financial reimbursement, including: S1, obtaining the image information of each financial document; S2, preprocess each image information; S3, extract the data features of the preprocessed image information; S4, based on the extracted data features, calculate the amount to be reimbursed; S5, determine whether the verification of the data is qualified based on the comparison result between the calculated amount and the expected amount, including: Determine that the verification of the data is abnormal, and based on the data difference ratio, the quantity of items in each financial document, and the item data difference quantity, determine the parameters for reprocessing the financial document. The parameters include adjusting the image sharpening parameter for the image information during the preprocessing of the image information to the corresponding value, or adjusting the extraction parameter for the data features; Or, determine that the verification of the data is qualified, and store the calculated amount.

[0007] Further, the process of extracting data features in the S3 includes: Based on matching the preprocessed image information with each preset feature template to identify the corresponding item area of the data feature; Match the text contour information in the extracted item area with several standard preset contours; Obtain the text contour information with a coincidence degree higher than the preset coincidence value, and record the corresponding standard preset contour as the data feature; In the S4, determine whether the verification of the data is qualified based on the comparison result between the calculated amount and the expected amount, including: Compare the calculated amount with the expected amount, calculate the absolute value of the difference between the calculated amount and the expected amount to obtain the data deviation parameter; If the data deviation parameter is less than or equal to the first preset deviation amount, determine that the verification of the data is qualified, and store the calculated amount; If the data deviation parameter is less than or equal to the second preset deviation amount and greater than the first preset deviation amount, determine whether the verification of the data is qualified based on the quota ratio; If the data deviation parameter is greater than the second preset deviation amount, determine that the verification of the data is abnormal, and determine the parameters for reprocessing the financial document based on the data difference ratio.

[0008] Further, the process of determining whether the verification of the data is qualified based on the quota ratio includes: Calculate the ratio of the data deviation parameter to the expected amount to obtain the quota ratio; If the quota ratio is less than or equal to the preset quota ratio, adjust the first preset deviation amount and the second preset deviation amount based on the quota ratio; If the quota ratio is less than or equal to the preset quota ratio, determine that the verification of the data is abnormal, and determine the parameters for reprocessing the financial document based on the data difference ratio.

[0009] Further, adjust the first preset deviation amount and the second preset deviation amount based on the quota ratio, where the increase rates of the first preset deviation amount and the second preset deviation amount are inversely proportional to the quota ratio.

[0010] Further, the process of determining the parameters for reprocessing financial documents based on the data difference ratio includes: Calculate the ratio of the data deviation parameter to the second preset deviation amount to obtain the data difference ratio; If the data difference ratio is less than or equal to the first preset data difference ratio, it is determined that there are abnormal data, and a document verification notice is issued; If the data difference ratio is less than or equal to the second preset data difference ratio and greater than the first preset data difference ratio, determine the parameters for reprocessing financial documents based on the number of items in each obtained financial document; If the data difference ratio is greater than the second preset data difference ratio, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio.

[0011] Further, determining the parameters for reprocessing financial documents based on the number of items in each obtained financial document includes: If the number of items is less than or equal to the preset number of items, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio; If the number of items is greater than the preset number of items, determine the parameters for reprocessing financial documents based on the project data difference amount.

[0012] Further, determining the parameters for reprocessing financial documents based on the project data difference amount includes: Calculate the variance of the amount data of each project area to obtain the project data difference amount; If the project data difference amount is less than or equal to the preset project data difference amount, increase the preset coincidence value to the corresponding value based on the project data difference amount; If the project data difference amount is greater than the preset project data difference amount, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio.

[0013] Further, increasing the preset coincidence value to the corresponding value based on the project data difference amount, where: the increase rate of the preset coincidence value is proportional to the project data difference amount.

[0014] Further, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio, where The increase amplitude of the image sharpening parameter is proportional to the data difference ratio.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: obtaining the image information of each financial document, preprocessing each image information, and extracting the data features of the preprocessed image information; based on the extracted data features, counting the amount to be reimbursed; through the automated data collection, recognition, and counting processes, the time and workload of manual operations are greatly reduced, and the processing speed of the financial reimbursement process is significantly improved. Determine whether the verification of the data is qualified based on the comparison result between the counted amount and the expected amount. When it is determined in a timely manner that the verification of the data is abnormal, determine the parameters for reprocessing the financial document based on the data difference ratio, the number of items in each financial document, and the item data difference amount, improving the accuracy of financial data acquisition, and further improving the processing efficiency of document data.

[0016] Furthermore, determine whether the verification of the data is qualified based on the data deviation parameter. The data deviation parameter characterizes the actual amount abnormality. When the data deviation parameter is less than or equal to the first preset deviation amount, it is determined to be qualified, that is, the actual reimbursement amount is relatively close to the expected amount, and the extraction of data features is within the normal range, and the currently extracted data is stored. When the data deviation parameter is less than or equal to the second preset deviation amount and greater than the first preset deviation amount, enter the secondary determination process. The quota ratio characterizes the specific situation of the current amount's difference fluctuation with the expected amount as the reference scale. When the quota ratio is less than or equal to the preset quota ratio, lower the judgment standard for the current data to determine whether there is an abnormality in the processing of the document data, and determine whether the verification of the data is qualified according to the actual situation of the data. When it is determined in a timely manner whether there is an abnormality in the processing of the financial document data, re-determine the parameters for processing the financial document, realizing the digital management of financial data. By determining detailed extraction standards and strict data verification mechanisms, intelligent analysis and processing of abnormal situations are carried out, effectively reducing problems such as data entry errors and calculation errors, improving the accuracy and reliability of financial data, and further improving the processing efficiency of document data.

[0017] Further, parameters for reprocessing financial documents are determined based on the data difference ratio, which characterizes the abnormal amount situation of the data deviation compared with the preset difference amount. When the data difference ratio is less than or equal to the first preset data difference ratio, the abnormal amount situation is within a reasonable range, and it is determined that there is no abnormality in the extraction of the document, but there is an abnormality in the reimbursement amount. A document verification notice is issued to prompt relevant personnel to check the accuracy of the document data. When the data difference ratio is less than or equal to the second preset data difference ratio and greater than the first preset data difference ratio, the specific situation is determined in combination with the number of items in the financial document, that is, the number of types of reimbursement amounts. When the number of items is less than or equal to the preset number of items, the number is small and there are large amount errors. In this case, it is determined that the data extraction error is caused by the image being unclear due to mistakes in the preprocessing process of the image information. In this case, the image sharpening parameter of the image information is adjusted to the corresponding value to improve the accuracy of data extraction; when the number of items is greater than the preset number of items, the specific abnormal situation is further determined in combination with the item data difference amount, and the abnormal reason is gradually determined; the item data difference amount characterizes the difference situation of the data of each item. When the item data difference amount is less than or equal to the preset item data difference amount, the amount data differences of the extracted items are small, and it is determined that due to the preset coincidence value being too low, some real data are not accurately extracted, thus affecting the amount statistics. In this case, the preset coincidence value is increased to the corresponding value based on the item data difference amount to improve the data acquisition accuracy. When the item data difference amount is greater than the preset item data difference amount, the amount data differences of the extracted items are large, and it is determined that the image processing parameters are abnormal, and the text contour extraction is inaccurate due to the blurred image. The image sharpening parameter is increased to achieve the image processing and data extraction effects adapted to the current financial document, timely discover problems in data processing, determine the problem reasons through the analysis of different situations, and provide targeted solutions, thereby improving the processing efficiency of document data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of the steps of the method for processing document data for financial reimbursement according to an embodiment of the present invention; Figure 2 It is a logical decision diagram of the process for determining whether the verification of data is qualified based on the quota ratio according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0021] It should be noted that in the description of the present invention, the terms indicating directions or positional relationships such as "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings. This is only for convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0022] In addition, it should also be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0023] Please refer to Figure 1 and Figure 2 as shown, which are respectively the step flow chart of the document data processing method for financial reimbursement in the embodiments of the present invention and the logical decision diagram of the process for determining whether the verification of data is qualified based on the quota ratio; A document data processing method for financial reimbursement in an embodiment of the present invention includes: S1, obtaining the image information of each financial document; S2, preprocessing each image information; S3, extracting the data features of the preprocessed image information; S4, based on the extracted data features, counting the amount to be reimbursed; S5, determining whether the verification of the data is qualified based on the comparison result between the counted amount and the expected amount, including: Determining that the verification of the data is abnormal, and determining the parameters for reprocessing the financial document based on the data difference ratio, the number of items in each financial document, and the item data difference amount. The parameters include adjusting the image sharpening parameter for the image information to the corresponding value during the preprocessing of the image information, or adjusting the parameter for extracting the data features; Or, determining that the verification of the data is qualified and storing the counted amount.

[0024] Specifically, image information of each financial document is obtained, preprocessed for each piece of image information, and data features of the preprocessed image information are extracted; based on the extracted data features, the amount to be reimbursed is statistically calculated; through an automated data collection, recognition, and statistical process, the time and workload of manual operations are significantly reduced, and the processing speed of the financial reimbursement process is notably improved. Based on the comparison result between the statistically calculated amount and the expected amount, it is determined whether the verification of the data is qualified. When it is determined in a timely manner that the verification of the data is abnormal, parameters for reprocessing the financial document are determined based on the data difference ratio, the quantity of items in each financial document, and the item data difference amount, improving the accuracy of financial data acquisition and thereby enhancing the processing efficiency of document data.

[0025] Specifically, there is no limitation on the specific method for S1 to obtain the image information of each financial document. The image information of the financial document can be obtained through a scanning device or taking pictures, etc. It can be understood that as long as the obtained image information is clear, complete, and covers all key contents of the document, this will not be elaborated further here.

[0026] Specifically, the process of preprocessing each piece of image information in S2 includes: sharpening the image information, performing grayscale processing on the sharpened image information, removing noise interference in the grayscale image information through Gaussian filtering, converting the grayscale image information into a binary image to highlight the text and key information, setting the pixel values greater than the preset binarization threshold to white and those less than or equal to the preset binarization threshold to black to obtain the preprocessed image information.

[0027] Specifically, the preset binarization threshold can be automatically calculated by the Otsu algorithm.

[0028] Specifically, the process of extracting data features from the preprocessed image information includes: Extraction of the amount for each item; Matching the image information with the corresponding preset feature template to identify the area of the amount of the item; For this area, through the contour detection algorithm, the Canny edge detection algorithm combined with the contour discovery function, the contour of the text is obtained; several standard preset contours learned through a large number of samples are pre-stored in the database; When the coincidence degree between the text contour information and the corresponding standard preset contour is higher than the preset coincidence value, it is determined that the area records the reimbursement amount corresponding to the item, and the amount is extracted to obtain the data feature; Specifically, the coincidence degree can be calculated using the contour matching algorithm, or the text contour information can be compared with each standard preset contour for coincidence, and the ratio of the area of the coincidence region to the total area of the text contour information is recorded as the coincidence degree; Specifically, for the extraction of other key information, including the extraction of reimbursement date, reimburser, and reimbursement item, it is carried out by matching with corresponding feature templates and corresponding preset profiles.

[0029] Specifically, in the step S4, the process of counting the amount to be reimbursed based on the extracted data features includes: determining the amount data of each project area based on the data features, traversing all the extracted amount data, and adding their values to obtain the final counted amount to be reimbursed.

[0030] Specifically, the process of extracting data features in the step S3 includes: Based on matching the preprocessed image information with each preset feature template to identify the corresponding project area of the data features; Matching the text contour information in the extracted project area with several standard preset profiles; Obtaining the text contour information with a coincidence degree higher than the preset coincidence value, and recording the corresponding standard preset profile as the data feature; In the step S4, determining whether the check of the data is qualified based on the comparison result of the counted amount and the expected amount includes: Comparing the counted amount with the expected amount, calculating the absolute value of the difference between the counted amount and the expected amount to obtain the data deviation parameter; If the data deviation parameter is less than or equal to the first preset deviation amount, it is determined that the check of the data is qualified, and the counted amount is stored; If the data deviation parameter is less than or equal to the second preset deviation amount and greater than the first preset deviation amount, it is determined whether the check of the data is qualified based on the ratio of the amount; If the data deviation parameter is greater than the second preset deviation amount, it is determined that the check of the data is abnormal, and the parameters for reprocessing the financial document are determined based on the data difference ratio.

[0031] Specifically, the first preset deviation amount is selected within the range of [300, 700], and the second preset deviation amount is selected within the range of [1000, 3000], with the unit being yuan.

[0032] Specifically, the expected amount can be the expected reimbursement amount for this business determined according to the pre-established budget plan. It can also be the expected amount for this business determined based on historical data, and the average reimbursement amount of this business in the counted historical data is determined as the expected amount.

[0033] Specifically, the process of determining whether the check of the data is qualified based on the ratio of the amount includes: Calculating the ratio of the data deviation parameter to the expected amount to obtain the ratio of the amount; If the quota ratio is less than or equal to the preset quota ratio, the first preset deviation amount and the second preset deviation amount are adjusted based on the quota ratio; If the quota ratio is less than or equal to the preset quota ratio, it is determined that the verification of the data is abnormal, and the parameters for reprocessing the financial document are determined based on the data difference ratio.

[0034] Specifically, the preset quota ratio B0 is selected within the range [0.06, 0.1].

[0035] Specifically, it is determined whether the verification of the data is qualified based on the data deviation parameter. The data deviation parameter represents the actual amount abnormality. When the data deviation parameter is less than or equal to the first preset deviation amount, it is determined as qualified, that is, the actual reimbursement amount is close to the expected amount, and the extraction of data features is within the normal range, and the currently extracted data is stored. When the data deviation parameter is less than or equal to the second preset deviation amount and greater than the first preset deviation amount, it enters the secondary determination process. The quota ratio represents the specific situation of the current amount difference fluctuation with the expected amount as the reference scale. When the quota ratio is less than or equal to the preset quota ratio, the judgment standard for the current data is lowered to determine whether there is an abnormality in the processing of the document data. According to the actual situation of the data, it is determined whether the verification of the data is qualified. When it is timely determined whether there is an abnormality in the processing of the financial document data, the parameters for processing the financial document are re-determined, realizing the digital management of financial data. By determining detailed extraction standards and strict data verification mechanisms, intelligent analysis and processing of abnormal situations are carried out, effectively reducing problems such as data entry errors and calculation errors, improving the accuracy and reliability of financial data, and thus improving the processing efficiency of document data.

[0036] Specifically, the first preset deviation amount and the second preset deviation amount are adjusted based on the quota ratio, where The increase amplitudes of the first preset deviation amount and the second preset deviation amount are inversely proportional to the quota ratio.

[0037] In this embodiment, optionally, The quota ratio is compared with the first preset quota comparison threshold and the second preset quota comparison threshold; If the quota ratio is less than or equal to the first preset quota comparison threshold, the first preset deviation amount is adjusted to 1.08 times the initial first preset deviation amount, and the second preset deviation amount is adjusted to 1.08 times the initial second preset deviation amount; If the quota ratio is less than or equal to the second preset quota comparison threshold and greater than the first preset quota comparison threshold, the first preset deviation amount is adjusted to 1.05 times the initial first preset deviation amount, and the second preset deviation amount is adjusted to 1.05 times the initial second preset deviation amount; If the quota ratio is greater than the second preset quota comparison threshold, adjust the first preset deviation amount to 1.03 times the initial first preset deviation amount, and adjust the second preset deviation amount to 1.03 times the initial second preset deviation amount; The first preset quota comparison threshold is taken as 0.5B0, and the second preset quota comparison threshold is taken as 0.7B0.

[0038] Specifically, based on the adjusted first preset deviation amount, re-determine whether the verification of the data is qualified based on the data deviation parameter. If the data deviation parameter is less than or equal to the adjusted first preset deviation amount, it is determined that the verification of the data is qualified, and the statistically calculated amount is stored; if the data deviation parameter is less than or equal to the second preset deviation amount and greater than the adjusted first preset deviation amount of the first preset deviation amount, it is determined that the verification of the data is abnormal, and the parameters for reprocessing the financial document are determined based on the data difference ratio.

[0039] Specifically, the process of determining the parameters for reprocessing the financial document based on the data difference ratio includes: Calculate the ratio of the data deviation parameter to the second preset deviation amount to obtain the data difference ratio; If the data difference ratio is less than or equal to the first preset data difference ratio, it is determined that there is abnormal data, and a document verification notice is issued; If the data difference ratio is less than or equal to the second preset data difference ratio and greater than the first preset data difference ratio, determine the parameters for reprocessing the financial document based on the number of items in each financial document obtained; If the data difference ratio is greater than the second preset data difference ratio, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio.

[0040] Specifically, the first preset data difference ratio Y1 is selected within the range of [0.2, 0.25], and the second preset data difference ratio Y2 is selected within the range of [0.3, 0.42].

[0041] Specifically, determining the parameters for reprocessing the financial document based on the number of items in each financial document obtained includes: If the number of items is less than or equal to the preset number of items, adjust the image sharpening parameter for the image information in the image information preprocessing process to the corresponding value based on the data difference ratio; If the number of items is greater than the preset number of items, determine the parameters for reprocessing the financial document based on the project data difference amount.

[0042] Specifically, the preset number of items is selected within the range of [5, 8].

[0043] Specifically, determining the parameters for reprocessing the financial document based on the project data difference amount includes: Calculate the variance of the amount data for each project area to obtain the project data difference amount; If the project data difference amount is less than or equal to the preset project data difference amount, then increase the preset coincidence value to the corresponding value based on the project data difference amount; If the project data difference amount is greater than the preset project data difference amount, then adjust the image sharpening parameter for the image information during the image information preprocessing process to the corresponding value based on the data difference ratio.

[0044] Specifically, the preset project data difference amount X0 is selected within the interval [1.6Q0, 2.5Q0], where Q0 is the average value of the project data difference amounts of the corresponding services in the historical data.

[0045] Specifically, increase the preset coincidence value to the corresponding value based on the project data difference amount, where: The increase amplitude of the preset coincidence value is proportional to the project data difference amount.

[0046] In this embodiment, optionally, Compare the project data difference amount with the first preset project comparison threshold and the second preset project comparison threshold; If the project data difference amount is less than or equal to the first preset project comparison threshold, then adjust the preset coincidence value to 1.11 times the initial preset coincidence value; If the project data difference amount is less than or equal to the second preset project comparison threshold and greater than the first preset project comparison threshold, then adjust the preset coincidence value to 1.19 times the initial preset coincidence value; If the project data difference amount is greater than the second preset project comparison threshold, then adjust the preset coincidence value to 1.25 times the initial preset coincidence value; The first preset project comparison threshold is taken as 0.8X0, and the second preset project comparison threshold is taken as 0.65X0.

[0047] Specifically, adjust the image sharpening parameter for the image information during the image information preprocessing process to the corresponding value based on the data difference ratio, where, The increase amplitude of the image sharpening parameter is proportional to the data difference ratio.

[0048] In this embodiment, optionally, Compare the data difference ratio with the first preset data comparison threshold and the second preset data comparison threshold; If the data difference ratio is less than or equal to the first preset data comparison threshold, then adjust the image sharpening parameter to 1.1 times the initial image sharpening parameter; If the data difference ratio is less than or equal to the second preset data comparison threshold and greater than the first preset data comparison threshold, then adjust the image sharpening parameter to 1.2 times the initial image sharpening parameter; If the data difference ratio is greater than the second preset data comparison threshold, adjust the image sharpening parameter to 1.3 times the initial image sharpening parameter; The first preset data comparison threshold is taken as 1.25Y2, and the second preset data comparison threshold is taken as 1.47Y2.

[0049] Specifically, determine the parameters for reprocessing the financial document based on the data difference ratio. The data difference ratio characterizes the abnormal amount situation of the data deviation compared with the preset difference amount. When the data difference ratio is less than or equal to the first preset data difference ratio, the abnormal amount situation is within a reasonable range, and it is determined that there is no abnormality in the extraction of the document, but there is an abnormality in the reimbursement amount. A document verification notice is issued to prompt relevant personnel to check the accuracy of the document data. When the data difference ratio is less than or equal to the second preset data difference ratio and greater than the first preset data difference ratio, combine the number of items in the financial document, that is, the number of types of reimbursement amounts, to determine the specific situation. When the number of items is less than or equal to the preset number of items, the number is small and there are large amount errors. In this case, it is determined that due to mistakes in the preprocessing process of the image information, the image is not clear, resulting in incorrect data extraction. In this case, adjust the image sharpening parameter of the image information to the corresponding value to improve the accuracy of data extraction; when the number of items is greater than the preset number of items, further combine the item data difference amount to determine the specific abnormal situation and gradually determine the cause of the abnormality; the item data difference amount characterizes the difference situation of the data of each item. When the item data difference amount is less than or equal to the preset item data difference amount, the difference in the amount data of each item extracted is small. It is determined that due to the too low preset coincidence value, some real data is not accurately extracted, thus affecting the amount statistics. In this case, based on the item data difference amount, adjust the preset coincidence value to the corresponding value to improve the data acquisition accuracy. When the item data difference amount is greater than the preset item data difference amount, the difference in the amount data of each item extracted is large. It is determined that the image processing parameters are abnormal, and due to the blurred image, the text contour extraction is inaccurate. Increase the image sharpening parameter to achieve the image processing and data extraction effects suitable for the current financial document, timely discover problems in data processing, determine the cause of the problem through the analysis of different situations, and provide targeted solutions, thereby improving the processing efficiency of the document data.

[0050] So far, the technical solution of the present invention has been described in combination with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

[0051] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention; for those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A document data processing method for financial reimbursement, characterized in that: include: S1, obtaining image information of each financial document; S2, preprocessing each image information; S3, extracting data features of the preprocessed image information; S4, based on the extracted data features, calculate the amount to be reimbursed; S5, based on the comparison result between the statistical amount and the expected amount, determining whether the data verification is qualified, including: Determine the verification anomaly for the data, and determine the parameters for reprocessing the financial documents based on the data difference ratio, the number of items in each financial document, and the amount of item data difference, the parameters including adjusting the image sharpening parameters for the image information in the image information preprocessing process to the corresponding values, or adjusting the extraction parameters for the data features; Or, determine whether the data verification is qualified and store the statistical amount.

2. The document data processing method for financial reimbursement according to claim 1, characterized in that: The process of extracting data features in S3 includes: Based on matching the pre-processed image information with each preset feature template to identify the corresponding project area of ​​the data feature; Matching the extracted text contour information within the project area with a number of standard preset contours; Acquire text contour information with a degree of overlap higher than a preset overlap value, and record the corresponding standard preset contour as a data feature; In S4, determining whether the data verification is qualified based on the comparison result of the statistical amount and the expected amount includes: Compare the statistical amount with the expected amount, calculate the absolute value of the difference between the statistical amount and the expected amount, and obtain the data deviation parameter; If the data deviation parameter is less than or equal to the first preset deviation, the data verification is determined to be qualified, and the statistical amount is stored; If the data deviation parameter is less than or equal to the second preset deviation and greater than the first preset deviation, then determining whether the data verification is qualified based on the quota ratio; If the data deviation parameter is greater than the second preset deviation, it is determined that the verification of the data is abnormal, and the parameters for reprocessing the financial document are determined based on the data difference ratio.

3. The document data processing method for financial reimbursement according to claim 2, characterized in that: The process of determining whether the data verification is qualified based on the quota ratio includes: Calculate the ratio of the data deviation parameter to the expected amount to obtain the amount ratio; If the credit ratio is less than or equal to the preset credit ratio, adjusting the first preset deviation amount and the second preset deviation amount based on the credit ratio; If the amount ratio is less than or equal to the preset amount ratio, the data verification is determined to be abnormal, and the parameters for reprocessing the financial document are determined based on the data difference ratio.

4. The document data processing method for financial reimbursement according to claim 3, characterized in that: The first preset deviation amount and the second preset deviation amount are adjusted based on the quota ratio, wherein: The increase in the first preset deviation amount and the second preset deviation amount is inversely proportional to the quota ratio.

5. The document data processing method for financial reimbursement according to claim 4, characterized in that: The process of determining parameters for reprocessing financial documents based on data discrepancy ratios, including: Calculating the ratio of the data deviation parameter to the second preset deviation to obtain a data difference ratio; If the data difference ratio is less than or equal to the first preset data difference ratio, it is determined that abnormal data exists and a document verification notification is issued; If the data difference ratio is less than or equal to the second preset data difference ratio and greater than the first preset data difference ratio, determining parameters for reprocessing the financial documents based on the number of items in each of the acquired financial documents; If the data difference ratio is greater than the second preset data difference ratio, the image sharpening parameter for the image information in the image information preprocessing process is adjusted to a corresponding value based on the data difference ratio.

6. The document data processing method for financial reimbursement according to claim 5, characterized in that: Parameters for reprocessing financial documents based on the number of items in each acquired financial document include: If the number of items is less than or equal to the preset number of items, adjusting the image sharpening parameter for the image information in the image information preprocessing process to a corresponding value based on the data difference ratio; If the number of items is greater than the preset number of items, the parameters for reprocessing the financial document are determined based on the amount of item data difference.

7. The document data processing method for financial reimbursement according to claim 6, characterized in that: The parameters that determine the reprocessing of financial documents based on the amount of project data variance include: Calculate the variance of the amount data of each project area to obtain the project data difference; If the project data difference amount is less than or equal to the preset project data difference amount, the preset overlap value is increased to a corresponding value based on the project data difference amount; If the project data difference amount is greater than a preset project data difference amount, the image sharpening parameter for the image information in the image information preprocessing process is adjusted to a corresponding value based on the data difference ratio.

8. The document data processing method for financial reimbursement according to claim 7, characterized in that: Based on the amount of project data difference, the preset overlap value is adjusted up to the corresponding value, where: The preset overlap value increases in direct proportion to the amount of difference in the project data.

9. The document data processing method for financial reimbursement according to claim 8, characterized in that: Based on the data difference ratio, the image sharpening parameters for the image information in the image information preprocessing process are adjusted to corresponding values, wherein: The increase in image sharpening parameters is proportional to the data difference ratio.

Images