Finance and accounting robot and system thereof
Through the image and digital processing module of the accounting robot system, combined with optical character recognition technology, the image quality problem of the accounting robot when scanning bank documents is solved, the data processing accuracy and system security are improved, the bank document processing process is optimized, and the bank document processing process is reduced. The risk is achieved and the efficiency and quality of banking processes are improved.
Patent Information
- Application Number
- CN202411224464.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-03
AI Technical Summary
Existing accounting robots have image quality problems when scanning bank documents, such as blur, overexposed or too dark, which leads to image distortion, which in turn affects the accuracy of extracting key information, resulting in data loss and incompleteness, and insufficient system safety and reliability.
Through accounting robot data input and processing module, bank document image processing module, digital processing module and letter processing module, combined with optical character recognition technology, image processing technology and character segmentation technology, the text and digital information in bank documents are identified and processed, and evaluated Image, numeric and letter processing accuracy, generate comprehensive performance evaluation indexes, optimize robot performance to improve data accuracy and system security.
It improves the data processing accuracy and safety and reliability of accounting robots and their systems, reduces human errors, optimizes bank document processing processes, reduces potential risks, and improves the efficiency and quality of bank business processes.
Smart Images

Figure CN119068316A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of accounting robot data processing, and in particular to an accounting robot and a system thereof. Background Art
[0002] With the rapid development of artificial intelligence and machine learning technologies, accounting robots are able to perform complex financial tasks, such as automatic bookkeeping, tax planning, risk assessment, etc. With the intervention of accounting robots, accounting professionals need to change their skills from traditional data entry to higher-level financial analysis and strategy formulation. The development of big data technology enables accounting robots to process and analyze large amounts of data. When accounting robots have problems scanning bank documents, such as frequent recognition errors, system crashes, or hardware failures, technicians conduct a preliminary inspection of the robot, record the robot's failure phenomenon, occurrence time, and frequency, and determine the nature of the problem. If the problem is complex and cannot be solved, contact the robot manufacturer to explain the situation and arrange for it to be returned to the factory for repair.
[0003] Existing accounting robots and their systems use supervised learning and unsupervised learning algorithms through artificial intelligence and machine learning to perform tasks such as classification, prediction, and anomaly detection. Accounting robots are trained through historical financial data to improve their decision-making and forecasting capabilities; they use optical character technology to recognize characters; they use natural language processing to extract useful information from unstructured text so that accounting robots can understand finance-related terms and queries; they use robotic process automation to automate repetitive financial tasks such as data entry and report generation; and they use system integration to achieve consistency and real-time updating of accounting robots and their system data.
[0004] For example, a patent application with publication number CN111142635A discloses an accounting robot and its system, comprising: a vehicle body, a chassis driving structure arranged on the vehicle body, and a foot-controlled switch installed at the bottom of the front end of the vehicle body, a password cabinet, a shredder, a tool box, a voucher preprocessing box and an integrated computer installed on the vehicle body from bottom to top, wherein a voucher information entry box, a voucher punching box and a voucher stamping box are arranged in the voucher preprocessing box from top to bottom, the voucher information entry box realizes the entry of voucher information through an image acquisition module and an image processing module loaded in the integrated computer; the voucher punching box realizes the voucher punching operation through a voucher fixing mechanism and a voucher punching mechanism; the voucher stamping box realizes the voucher stamping operation through a voucher fixing component and a voucher stamping mechanism, and the foot-controlled switch is used to respectively start the image acquisition module, the voucher punching mechanism and the voucher stamping mechanism.
[0005] For example, the patent application with publication number: CN115760460A discloses a financial robot financial management system, which includes: a U-shield management platform for centralized management of U-shield information; a financial robot operation platform for managing and dispatching financial robots, calling financial information through the financial management platform to carry out corresponding business, and accessing online banking, an integrated platform and local application software to process business processes; a financial management platform for providing an interface for data supplementation and entry and displaying data processing results.
[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems: In the prior art, there are image quality problems when accounting robots scan bank documents, such as blur, overexposure or too dark, resulting in image distortion. At the same time, when extracting key information, such as amount, account number, etc., it may cause extraction errors due to changes in or missing keyword positions, resulting in loss and incomplete bank document data. There is a problem of insufficient security and reliability of accounting robots and their systems. Summary of the invention
[0007] The embodiments of the present application provide an accounting robot and a system thereof, thereby solving the problem of insufficient safety and reliability of accounting robots and systems thereof in the prior art, and improving the safety and reliability of accounting robots and systems thereof.
[0008] The embodiment of the present application provides an accounting robot and its system, including: an accounting robot data input and processing module, an accounting robot bank document image processing module, an accounting robot bank document digital processing module, an accounting robot bank document letter processing module, an accounting robot comprehensive performance evaluation module and an accounting robot evaluation optimization module: wherein the accounting robot data input and processing module is used to convert the text information in the paper document of the bank document into a processable electronic format through optical character recognition technology to obtain the accounting robot bank document data; the accounting robot bank document image processing module is used to process blurred and unclear bank documents to obtain the accounting robot bank document image processing accuracy evaluation index; the accounting robot The bank document digital processing module is used to process the digital information of the bank document to obtain the accounting robot bank document digital processing accuracy evaluation index; the accounting robot bank document letter processing module is used to process the letter information of the bank document to obtain the accounting robot bank document letter processing accuracy evaluation index; the accounting robot comprehensive performance evaluation module is used to obtain the accounting robot comprehensive performance evaluation index based on the accounting robot bank document image processing accuracy evaluation index, the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index; the accounting robot evaluation and optimization module is used to optimize the accounting robot according to the accounting robot evaluation and analysis results.
[0009] Furthermore, the specific process of obtaining the bank document data of the accounting robot is: converting the paper document of the bank document into a digital image format through an automatic paper-fed scanner to obtain a bank document image; denoising and tilt correction of the bank document image to obtain a corrected bank document image.
[0010] Furthermore, the specific process of obtaining the accounting robot bank document data also includes: using image processing technology to detect the text area in the bank document image, converting the recognized text into an editable electronic text format, and obtaining the accounting robot bank document digital data; cleaning, denoising and storing the accounting robot bank document digital data in a database.
[0011] Furthermore, the specific process of obtaining the bank document data of the accounting robot also includes: segmenting the text area in the bank document image into single characters through character segmentation technology, mapping each character image to a corresponding letter, and obtaining the bank document letter data of the accounting robot.
[0012] Furthermore, the specific steps for processing blurred and unclear bank documents and obtaining the accuracy evaluation index of bank document image processing by the accounting robot are as follows: identifying the area in the bank document image through edge detection and connected component marking to obtain the area of the bank document image of the accounting robot; identifying the shadow edge in the bank document image and marking the shadow area of the bank document image through edge detection and connected component marking, traversing the pixels marked as the shadow of the bank document image, and obtaining the area of the shadow area of the bank document image of the accounting robot; identifying the ink dots in the bank document image through noise detection and connected component marking to obtain the number of ink dots in the bank document image of the accounting robot; obtaining the contrast of the bank document image of the accounting robot through image processing software; and obtaining the accuracy evaluation index of the bank document image processing by the accounting robot through analysis.
[0013] Furthermore, the specific process for processing the digital information of bank documents and obtaining the digital processing accuracy evaluation index of the accounting robot bank documents is as follows: obtaining the number of digits of the accounting robot bank document through a flatbed scanner; marking the number area of the bank document image through edge detection, using optical character recognition technology to recognize the numbers in the bank document image, and obtaining the correct number of digits of the accounting robot bank document; scanning the numbers on the bank document image through optical character recognition technology to obtain the correct currency symbols of the accounting robot bank document; and obtaining the digital processing accuracy evaluation index of the accounting robot bank document through this analysis.
[0014] Furthermore, the specific process for processing the letter information of bank documents and obtaining the letter processing accuracy evaluation index of the accounting robot bank documents is: obtaining the number of names on the accounting robot bank documents through a flatbed scanner; obtaining the number of correct names on the accounting robot bank documents through intelligent character recognition technology; obtaining the number of letters on the accounting robot bank documents through a flatbed scanner; obtaining the number of correct letters on the accounting robot bank documents through optical character technology; obtaining the number of letters actually existing in the accounting robot bank documents through a flatbed scanner and a text analysis tool; and obtaining the letter processing accuracy evaluation index of the accounting robot bank documents through this analysis.
[0015] Furthermore, the specific process of obtaining the comprehensive performance evaluation index of the accounting robot is as follows: obtaining the comprehensive performance evaluation index of the accounting robot through the accuracy evaluation index of bank document image processing of the accounting robot, the accuracy evaluation index of bank document digital processing of the accounting robot and the accuracy evaluation index of bank document letter processing of the accounting robot; ; represents the comprehensive performance evaluation index of accounting robots, It represents the weight factor affecting the accuracy evaluation index of bank document image processing by accounting robots. It represents the accuracy evaluation index of bank document image processing by accounting robot. It represents the weight factor affecting the accuracy evaluation index of bank document digital processing by accounting robots. It represents the accuracy evaluation index of bank document digital processing by accounting robots. It represents the weight factor affecting the accuracy evaluation index of bank document letter processing of accounting robots. It represents the accuracy evaluation index of bank document letter processing by accounting robot.
[0016] Furthermore, the specific process of optimizing the accounting robot according to the evaluation and analysis results of the accounting robot is as follows: the comprehensive error deviation value of the accounting robot scanning bank documents, the first error judgment value of the accounting robot scanning bank documents, the second error judgment value of the accounting robot scanning bank documents and the third error judgment value of the accounting robot scanning bank documents are imported into the accounting robot error detection model, and the error detection evaluation value of the accounting robot scanning bank documents is obtained through comprehensive analysis; the error detection evaluation value of the accounting robot scanning bank documents is compared with the error detection boundary evaluation value of the accounting robot scanning bank documents stored in the database; if the error detection evaluation value of the accounting robot scanning bank documents is higher than the error detection boundary evaluation value of the accounting robot scanning bank documents, then the accounting robot scanning bank documents corresponding to the error detection evaluation value of the accounting robot scanning bank documents are qualified; if the error detection evaluation value of the accounting robot scanning bank documents is equal to the error detection boundary evaluation value of the accounting robot scanning bank documents, then there is a risk in the accounting robot scanning bank documents; if the accounting robot scanning bank documents error detection evaluation value ... If the error detection evaluation value of the robot scanning bank documents is lower than the error detection boundary evaluation value of the accounting robot scanning bank documents, the accounting robot scanning bank documents corresponding to the error detection evaluation value of the accounting robot scanning bank documents are unqualified; if the image processing accuracy evaluation index of the accounting robot bank documents is less than the first error judgment value of the accounting robot scanning bank documents, then by using image editing software, the noise and scattered points in the bank document image are removed, the contrast between the text and the background in the bank document image is enhanced, and the bank document image is sharpened; if the digital processing accuracy evaluation index of the accounting robot bank documents is less than the second error judgment value of the accounting robot scanning bank documents, then by adjusting the scanning environment of the accounting robot to make the light uniform, avoid shadows and reflections, and adjust the brightness of the bank document image through image editing software; if the letter processing accuracy evaluation index of the accounting robot bank documents is less than the third error judgment value of the accounting robot scanning bank documents, then by adjusting the scanning mode of letter recognition of the accounting robot scanning bank documents, the grayscale of the bank document image is adjusted.
[0017] Furthermore, the specific process of the hardware equipment of the accounting robot is as follows: the central processing unit of the accounting robot is used to analyze the bank financial data of the accounting robot, run preset banking services, identify and process data loss and recognition errors, etc.; the scanner of the accounting robot is used to scan bank documents into digital image format; the image processing unit of the accounting robot is used to process and manage image data; the camera of the accounting robot is used to capture images of bank documents, and by capturing images of bank documents with the camera, the image quality can be adjusted in real time.
[0018] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The present invention obtains and processes bank document data by identifying information in paper documents of bank documents, obtains the accuracy evaluation index of bank document image processing by the accounting robot, the accuracy evaluation index of bank document digital processing by the accounting robot and the accuracy evaluation index of bank document letter processing by the accounting robot through evaluation and analysis, and optimizes by comparative analysis, thereby improving the safety and reliability of the accounting robot and its system, and solving the problem of insufficient safety and reliability of the accounting robot and its system in the prior art.
[0019] 2. The present invention can reduce human errors, reduce the time and cost of manual operations, monitor the bank document processing flow in real time, discover and solve problems, and at the same time, the evaluation results of the bank document processing accuracy of the accounting robot can be used to improve the bank document processing flow and optimize the performance of the accounting robot.
[0020] 3. The present invention can help banks discover problems in document business through abnormal frequency risk assessment of bank business processes by accounting robots, reduce the impact of potential risks on bank business processes through preventive measures, and optimize bank business processes through abnormal frequency risk assessment results of bank business processes by accounting robots, thereby improving the efficiency and quality of bank business processes. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 A schematic diagram of the structure of the accounting robot and its system provided in the embodiment of the present application; Figure 2 A schematic diagram of an index function for comprehensive performance evaluation of an accounting robot provided in an embodiment of the present application; DETAILED DESCRIPTION
[0022] The embodiments of the present application solve the problem of insufficient security and reliability of accounting robots and their systems in the prior art by providing an accounting robot and its system, optimize the accounting robot by comparing and analyzing the results, and improve the security and reliability of the accounting robot and its system.
[0023] The technical solution in the embodiment of the present application is to solve the problem of insufficient security and reliability of the above-mentioned accounting robot and its system. The overall idea is as follows: By identifying information in paper documents of bank receipts, obtaining and processing accounting robot data, planning accounting robot data flow, identifying and handling abnormal situations, and obtaining the accounting robot bank document processing accuracy assessment index and the bank business process abnormal frequency risk assessment index through evaluation and analysis, comparative analysis and optimization are carried out to improve the security and reliability of accounting robots and their systems.
[0024] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0025] like Figure 1 As shown, it is a structural schematic diagram of the accounting robot and its system provided in the embodiment of the present application. The accounting robot and its system provided in the embodiment of the present application include: an accounting robot data input and processing module, an accounting robot bank document image processing module, an accounting robot bank document digital processing module, an accounting robot bank document letter processing module, an accounting robot comprehensive performance evaluation module and an accounting robot evaluation optimization module: wherein the accounting robot data input and processing module is used to convert the text information in the paper document of the bank document into a processable electronic format through optical character recognition technology to obtain the accounting robot bank document data; the accounting robot bank document image processing module is used to process blurred and unclear bank documents to obtain the accounting robot bank document image processing accuracy The accuracy evaluation index is obtained; the accounting robot bank document digital processing module is used to process the digital information of the bank document to obtain the accounting robot bank document digital processing accuracy evaluation index; the accounting robot bank document letter processing module is used to process the letter information of the bank document to obtain the accounting robot bank document letter processing accuracy evaluation index; the accounting robot comprehensive performance evaluation module is used to obtain the accounting robot comprehensive performance evaluation index based on the accounting robot bank document image processing accuracy evaluation index, the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index; the accounting robot evaluation optimization module is used to optimize the accounting robot according to the accounting robot evaluation analysis results.
[0026] In this embodiment, optical character recognition technology is a technology that converts text in an image into an editable and searchable text format. The accounting robot can recognize the text content in pictures, scans, and photos through optical character recognition technology. The optical character recognition software needs to detect the text area in the paper document of the bank document, divide the detected text area into individual characters or words, extract the characteristics of each character or word, such as the stroke shape, relative position, etc., proofread and correct the recognition results, obtain the accounting robot bank document data, convert the accounting robot bank document data into a specific data format, and improve the accuracy of the accounting robot data.
[0027] Furthermore, the specific process of obtaining the bank document data of the accounting robot is: converting the paper document of the bank document into a digital image format through an automatic paper-fed scanner to obtain a bank document image; denoising and tilt correction are performed on the bank document image to obtain a corrected bank document image.
[0028] Furthermore, the specific process of obtaining the accounting robot bank document data also includes: using image processing technology to detect the text area in the bank document image, converting the recognized text into an editable electronic text format, and obtaining the accounting robot bank document digital data; cleaning, denoising and storing the accounting robot bank document digital data in a database.
[0029] In this embodiment, a color bank document image is converted into a grayscale bank document image to reduce the influence of color information on text detection. The image is converted into a black and white image through threshold processing so that the text area forms a clear contrast with the background. A median filter is used to remove noise in the image to keep the edges of the text area clear. An edge detection algorithm, such as Canny edge detection, is used to detect the edges of text in the image to obtain the text area in the bank document image.
[0030] Furthermore, the specific process of obtaining the bank document data of the accounting robot also includes: segmenting the text area in the bank document image into single characters through character segmentation technology, mapping each character image to a corresponding letter, and obtaining the bank document letter data of the accounting robot.
[0031] Furthermore, the specific steps for processing blurred and unclear bank documents and obtaining the accuracy evaluation index of bank document image processing by the accounting robot are as follows: identifying the area in the bank document image through edge detection and connected component marking to obtain the area of the bank document image of the accounting robot; identifying the shadow edge in the bank document image and marking the shadow area of the bank document image through edge detection and connected component marking, traversing the pixels marked as the shadow of the bank document image, and obtaining the area of the shadow area of the bank document image of the accounting robot; identifying the ink dots in the bank document image through noise detection and connected component marking to obtain the number of ink dots in the bank document image of the accounting robot; obtaining the contrast of the bank document image of the accounting robot through image processing software; and obtaining the accuracy evaluation index of the bank document image processing by the accounting robot through analysis.
[0032] In this embodiment, the specific formula of the accuracy evaluation index of the bank document image processing of the accounting robot is obtained by analysis as follows: ; ; Represents the accuracy evaluation index of bank document image processing by accounting robots.
[0033] Represents the shadow area of the accounting robot bank document image.
[0034] Represents the image area of the accounting robot bank document.
[0035] Indicates the number of ink dots in the bank document image of the accounting robot.
[0036] Represents the contrast of bank document image of accounting robot.
[0037] Represents the contrast threshold of the bank document image obtained from the accounting robot database.
[0038] It is the influence weight factor of the shadow area of the preset accounting robot bank document image obtained from the accounting robot database.
[0039] It is the influence weight factor of the ink dot number of the preset accounting robot bank document image obtained from the accounting robot database.
[0040] It is the preset accounting robot bank document image contrast impact weight factor obtained from the accounting robot database.
[0041] In a specific embodiment, a mapping set of accounting robot bank document image shadow area and accounting robot scanning lighting conditions and their corresponding weight factors is constructed based on the relationship between the historical accounting robot bank document image shadow area and accounting robot scanning lighting conditions and the accounting robot bank document image processing accuracy evaluation index, and the real-time independent variables accounting robot bank document image shadow area and accounting robot scanning lighting conditions are input into the mapping set to obtain the corresponding accounting robot bank document image shadow area influencing weight factors.
[0042] According to the relationship between the historical accounting robot bank document image ink dot number and accounting robot scanning lighting conditions and the accounting robot bank document image processing accuracy evaluation index, a mapping set of the accounting robot bank document image ink dot number and accounting robot scanning lighting conditions and their corresponding weight factors is constructed, and the real-time independent variables accounting robot bank document image ink dot number and accounting robot scanning lighting conditions are input into the mapping set to obtain the corresponding accounting robot bank document image ink dot number influencing weight factors.
[0043] According to the relationship between the historical accounting robot bank document image contrast and accounting robot scanning lighting conditions and the accounting robot bank document image processing accuracy evaluation index, a mapping set of accounting robot bank document image contrast and accounting robot scanning lighting conditions and their corresponding weight factors is constructed, and the real-time independent variables accounting robot bank document image contrast and accounting robot scanning lighting conditions are input into the mapping set to obtain the corresponding accounting robot bank document image contrast influencing weight factors.
[0044] Table 1 is an example table of the accuracy evaluation index of the accounting robot bank document processing. The example parameters in Table 1 only take the parameters of one accounting robot bank document processing accuracy time detection point as an example. The weight factor Set to 0.4, weight factor Set to 0.2, the weight factor Set to 0.4 Table 1 is shown below.
[0045] Table 1 Example of the evaluation index of accuracy of bank document processing by accounting robots Shadow area of accounting robot bank document image (pixels) Accounting robot bank document image area (pixels) Number of ink dots of bank document image produced by accounting robot (pcs) Accounting Robot Bank Document Image Contrast Thresholds obtained from the accounting robot database Accuracy evaluation index of bank document image processing by accounting robot 2 5 3 5 4 2.36316069371 3 4 4 6 5 3.25437420289 2 4 2 5 3 2.37896772991 5 8 1 8 5 1.99371553324 6 9 0 7 4 1.76238264073 The shadow area of the accounting robot bank document image affects the contrast of the accounting robot bank document image because it absorbs or reflects less light, resulting in reduced brightness. The number of ink dots in the accounting robot bank document image interferes with the contrast of the accounting robot bank document image because the brightness value of the accounting robot bank document pixel will change. The smaller the shadow area of the accounting robot bank document image, the fewer the number of ink dots in the accounting robot bank document image, the higher the contrast of the accounting robot bank document image, and the higher the accounting robot bank document image processing accuracy evaluation index.
[0046] Furthermore, the specific process for processing the digital information of bank documents and obtaining the digital processing accuracy evaluation index of the accounting robot bank documents is as follows: obtaining the number of digits of the accounting robot bank document through a flatbed scanner; marking the number area of the bank document image through edge detection, using optical character recognition technology to recognize the numbers in the bank document image, and obtaining the correct number of digits of the accounting robot bank document; scanning the numbers on the bank document image through optical character recognition technology to obtain the correct currency symbols of the accounting robot bank document; and obtaining the digital processing accuracy evaluation index of the accounting robot bank document through this analysis.
[0047] In this embodiment, the specific formula of the accuracy evaluation index of bank document digital processing obtained by the accounting robot is as follows: ; ; Represents the accuracy evaluation index of the digital processing of bank documents by the accounting robot.
[0048] Indicates the correct digital quantity of the accounting robot bank document.
[0049] Indicates the number of bank documents of the accounting robot.
[0050] Indicates the correct currency symbol for the accounting robot scanning bank documents.
[0051] Indicates the currency symbol of a bank document scanned by an accounting robot.
[0052] The weighting factor affecting the correct digital quantity of the preset accounting robot bank document obtained from the accounting robot database.
[0053] The weighting factor affecting the correct currency symbol quantity of the preset accounting robot scanning bank documents obtained from the accounting robot database.
[0054] In a specific embodiment, a mapping set of the correct number of digits in accounting robot bank documents and the resolution of accounting robot scanner and their corresponding weight factors is constructed according to the relationship between the historical correct number of digits in accounting robot bank documents and the resolution of accounting robot scanner and the digital processing accuracy evaluation index of accounting robot bank documents, and the real-time independent variables correct number of digits in accounting robot bank documents and the resolution of accounting robot scanner are input into the mapping set to obtain the corresponding weight factors affecting the correct number of digits in accounting robot bank documents.
[0055] According to the relationship between the historical number of correct currency symbols on bank documents scanned by accounting robots, the resolution of accounting robot scanners and the digital processing accuracy evaluation index of accounting robot bank documents, a mapping set of the number of correct currency symbols on bank documents scanned by accounting robots and the resolution of accounting robot scanners and their corresponding weight factors is constructed, and the real-time independent variables, the number of correct currency symbols on bank documents scanned by accounting robots and the resolution of accounting robot scanners, are input into the mapping set to obtain the corresponding weight factors affecting the number of correct currency symbols on bank documents scanned by accounting robots.
[0056] The more correct digits the accounting robot has in bank documents, the more accurate the currency symbols the accounting robot scans in bank documents, and the higher the accuracy assessment index of the accounting robot's bank document digital processing. The more correct digits the accounting robot has in bank documents, the higher the accuracy assessment index of the accounting robot's bank document digital processing.
[0057] Furthermore, the specific process for processing the letter information of bank documents and obtaining the letter processing accuracy evaluation index of the accounting robot bank documents is as follows: obtaining the number of names on the accounting robot bank documents through a flatbed scanner; obtaining the number of correct names on the accounting robot bank documents through intelligent character recognition technology; obtaining the number of letters on the accounting robot bank documents through a flatbed scanner; obtaining the number of correct letters on the accounting robot bank documents through optical character technology; obtaining the number of actual letters on the accounting robot bank documents through a flatbed scanner and text analysis tools; and obtaining the letter processing accuracy evaluation index of the accounting robot bank documents through this analysis.
[0058] In this embodiment, the specific formula of the accuracy evaluation index of bank document letter processing of the accounting robot is obtained through analysis as follows: ; ; ; It represents the accuracy evaluation index of bank document letter processing by accounting robot.
[0059] Indicates the correct name quantity of the accounting robot bank document.
[0060] Indicates the number of names on the accounting robot bank documents.
[0061] Indicates the correct number of letters in the accounting robot bank document.
[0062] Indicates the number of letters in the accounting robot bank document.
[0063] Indicates the actual number of letters that exist in the accounting robot bank document.
[0064] It represents the harmonic mean of the accounting robot bank document letter accuracy and the accounting robot bank document letter recall rate. The accounting robot bank document letter recall rate represents the ratio of the number of correct letters in the accounting robot bank document to the number of letters actually existing in the accounting robot bank document. The harmonic mean of the accounting robot bank document letter accuracy and the accounting robot bank document letter recall rate is used to comprehensively evaluate the accounting robot bank document letter accuracy and the accounting robot bank document letter recall rate.
[0065] The weight factor affecting the number of correct names on the preset accounting robot bank documents obtained from the accounting robot database.
[0066] The weight factor affecting the correct number of letters of the preset accounting robot bank document obtained from the accounting robot database.
[0067] In a specific embodiment, a mapping set of the number of correct names on bank documents of an accounting robot and the brightness of bank document images and their corresponding weight factors is constructed based on the relationship between the historical number of correct names on bank documents of an accounting robot and the brightness of bank document images and the evaluation index of the accuracy of letter processing of bank documents of an accounting robot, and the real-time independent variables the number of correct names on bank documents of an accounting robot and the brightness of bank document images are input into the mapping set to obtain the corresponding weight factors affecting the number of correct names on bank documents of an accounting robot.
[0068] According to the relationship between the historical number of correct letters in accounting robot bank documents and the brightness of bank document images and the accounting robot bank document letter processing accuracy evaluation index, a mapping set of the number of correct letters in accounting robot bank documents and the brightness of bank document images and their corresponding weight factors is constructed, and the real-time independent variables, the number of correct letters in accounting robot bank documents and the brightness of bank document images, are input into the mapping set to obtain the corresponding weight factors affecting the number of correct letters in accounting robot bank documents.
[0069] The more correct letters the accounting robot bank documents have, the more correct names the accounting robot bank documents have, the higher the accounting robot bank document letter processing accuracy assessment index is, the more correct names the accounting robot bank documents have, the accounting robot processes bank documents accurately to identify and extract key personal information, thereby improving the accounting robot bank document letter processing accuracy.
[0070] Furthermore, the specific process of obtaining the comprehensive performance evaluation index of the accounting robot is as follows: the comprehensive performance evaluation index of the accounting robot is obtained through the accuracy evaluation index of the accounting robot bank document image processing, the accuracy evaluation index of the accounting robot bank document digital processing and the accuracy evaluation index of the accounting robot bank document letter processing; ; represents the comprehensive performance evaluation index of accounting robots, It represents the weight factor affecting the accuracy evaluation index of bank document image processing by accounting robots. It represents the accuracy evaluation index of bank document image processing by accounting robot. It represents the weight factor affecting the accuracy evaluation index of bank document digital processing by accounting robots. It represents the accuracy evaluation index of bank document digital processing by accounting robots. It represents the weight factor affecting the accuracy evaluation index of bank document letter processing of accounting robots. It represents the accuracy evaluation index of bank document letter processing by accounting robot.
[0071] In this embodiment, a mapping set of the accounting robot bank document image processing accuracy evaluation index and bank document image noise and their corresponding weight factors is constructed according to the relationship between the historical accounting robot bank document image processing accuracy evaluation index and bank document image noise and the accounting robot comprehensive performance evaluation index, and the real-time independent variables accounting robot bank document image processing accuracy evaluation index and bank document image noise are input into the mapping set to obtain the corresponding accounting robot bank document image processing accuracy evaluation index influencing weight factors.
[0072] According to the relationship between the historical accounting robot bank document digital processing accuracy evaluation index and bank document image noise and the accounting robot comprehensive performance evaluation index, a mapping set of the accounting robot bank document digital processing accuracy evaluation index and bank document image noise and their corresponding weight factors is constructed, and the real-time independent variables accounting robot bank document digital processing accuracy evaluation index and bank document image noise are input into the mapping set to obtain the corresponding accounting robot bank document digital processing accuracy evaluation index influencing weight factors.
[0073] According to the relationship between the historical accounting robot bank document letter processing accuracy evaluation index and bank document image noise and the accounting robot comprehensive performance evaluation index, a mapping set of the accounting robot bank document letter processing accuracy evaluation index and bank document image noise and their corresponding weight factors is constructed, and the real-time independent variables accounting robot bank document letter processing accuracy evaluation index and bank document image noise are input into the mapping set to obtain the corresponding accounting robot bank document letter processing accuracy evaluation index influencing weight factors.
[0074] like Figure 2 As shown, it is a schematic diagram of the comprehensive performance evaluation index function of the accounting robot provided in the embodiment of the present application; x is the positive semi-axis of the horizontal coordinate, y is the positive semi-axis of the vertical coordinate, and the weight factor Set to 0.5, weight factor Set to 0.3, weight factor Set to 0.2.
[0075] Curve a indicates that if the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index are set to a fixed value of 1, the robot bank document image processing accuracy evaluation index is x, and the accounting robot comprehensive performance evaluation index is y, the accounting robot comprehensive performance evaluation index increases with the increase of the accounting robot bank document image processing accuracy evaluation index.
[0076] Curve b indicates that if the accounting robot bank document image processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index are set to a fixed value of 1, the accounting robot bank document digital processing accuracy evaluation index is x, and the accounting robot comprehensive performance evaluation index is y, the accounting robot comprehensive performance evaluation index increases with the increase of the accounting robot bank document digital processing accuracy evaluation index.
[0077] Curve c indicates that if the accounting robot bank document image processing accuracy evaluation index and the accounting robot bank document digital processing accuracy evaluation index are set to a fixed value of 1, the accounting robot bank document letter processing accuracy evaluation index is x, and the accounting robot comprehensive performance evaluation index is y, the accounting robot comprehensive performance evaluation index increases with the increase of the accounting robot bank document letter processing accuracy evaluation index.
[0078] The accounting robot bank document image processing accuracy evaluation index affects the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index. The better the accounting robot bank document image quality, the higher the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index, and the higher the accounting robot comprehensive performance evaluation index.
[0079] Furthermore, according to the evaluation and analysis results of the accounting robot, the specific process of optimizing the accounting robot is as follows: the comprehensive error deviation value of the accounting robot scanning bank documents, the first error judgment value of the accounting robot scanning bank documents, the second error judgment value of the accounting robot scanning bank documents and the third error judgment value of the accounting robot scanning bank documents are imported into the accounting robot error detection model, and the error detection evaluation value of the accounting robot scanning bank documents is obtained through comprehensive analysis; the error detection evaluation value of the accounting robot scanning bank documents is compared with the error detection boundary evaluation value of the accounting robot scanning bank documents stored in the database; if the error detection evaluation value of the accounting robot scanning bank documents is higher than the error detection boundary evaluation value of the accounting robot scanning bank documents, then the accounting robot scanning bank documents corresponding to the error detection evaluation value of the accounting robot scanning bank documents are qualified; if the error detection evaluation value of the accounting robot scanning bank documents is equal to the error detection boundary evaluation value of the accounting robot scanning bank documents, then there is a risk in the accounting robot scanning bank documents; if the accounting machine If the error detection evaluation value of the bank document scanning robot is lower than the error detection boundary evaluation value of the bank document scanning robot, the bank document scanning by the accounting robot corresponding to the error detection evaluation value of the bank document scanning by the accounting robot is unqualified; if the image processing accuracy evaluation index of the bank document by the accounting robot is less than the first error judgment value of the bank document scanning by the accounting robot, then the noise and spots in the bank document image are removed by using image editing software, the contrast between the text and the background in the bank document image is enhanced, and the bank document image is sharpened; if the digital processing accuracy evaluation index of the bank document by the accounting robot is less than the second error judgment value of the bank document scanning by the accounting robot, then the scanning environment of the accounting robot is adjusted to make the light uniform, avoid shadows and reflections, and the brightness of the bank document image is adjusted by image editing software; if the letter processing accuracy evaluation index of the bank document by the accounting robot is less than the third error judgment value of the bank document scanning by the accounting robot, then the grayscale of the bank document image is adjusted by adjusting the scanning mode of letter recognition of the bank document scanning by the accounting robot.
[0080] Furthermore, the specific process of the hardware equipment of the accounting robot is as follows: the central processing unit of the accounting robot is used to analyze the bank financial data of the accounting robot, run the preset banking business, identify and process data loss and recognition errors, etc.; the scanner of the accounting robot is used to scan bank documents into digital image format; the image processing unit of the accounting robot is used to process and manage image data; the camera of the accounting robot is used to capture images of bank documents, and by capturing images of bank documents with the camera, the image quality can be adjusted in real time.
[0081] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0082] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0085] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0086] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A financial accounting robot system, characterized in that: It includes accounting robot data input and processing module, accounting robot bank document image processing module, accounting robot bank document digital processing module, accounting robot bank document letter processing module, accounting robot comprehensive performance evaluation module and accounting robot evaluation and optimization module: The accounting robot data input and processing module is used to convert the text information in the paper document of the bank receipt into a processable electronic format through optical character recognition technology to obtain the accounting robot bank receipt data; The accounting robot bank document image processing module is used to process blurred bank documents and obtain the accounting robot bank document image processing accuracy evaluation index; The accounting robot bank document digital processing module is used to process the digital information of the bank document and obtain the accuracy evaluation index of the accounting robot bank document digital processing; The accounting robot bank document letter processing module is used to process the letter information of the bank document and obtain the accuracy evaluation index of the accounting robot bank document letter processing; The accounting robot comprehensive performance evaluation module is used to comprehensively evaluate the accounting robot bank document image processing accuracy evaluation index, the accounting robot bank document digital processing accuracy evaluation index and the accounting robot bank document letter processing accuracy evaluation index to obtain the accounting robot comprehensive performance evaluation index; The accounting robot evaluation and optimization module is used to optimize the accounting robot according to the accounting robot evaluation and analysis results.
2. A financial accounting robot system as claimed in claim 1, characterized in that: The specific process of obtaining the bank document data of the accounting robot is as follows: Converting paper documents of bank documents into digital image format by using an automatic sheet-feed scanner to obtain bank document images; The bank document image is denoised and tilt-corrected to obtain a corrected bank document image.
3. The accounting robot system as claimed in claim 1, characterized in that: The specific process of obtaining the bank document data of the accounting robot also includes: Use image processing technology to detect the text area in the bank document image, convert the recognized text into an editable electronic text format, and obtain the digital data of the bank document of the accounting robot; The digital data of bank documents of the accounting robot is cleaned, denoised and stored in the database.
4. The accounting robot system as claimed in claim 1, characterized in that: The specific process of obtaining the bank document data of the accounting robot also includes: The text area in the bank document image is segmented into single characters through character segmentation technology, and each character image is mapped to the corresponding letter to obtain the bank document letter data of the accounting robot.
5. The accounting robot system as claimed in claim 1, characterized in that: The specific steps for processing blurred bank documents and obtaining the accuracy evaluation index of bank document image processing by the accounting robot are as follows: The area in the bank document image is identified by edge detection and connected component labeling, and the bank document image area of the accounting robot is obtained; Through edge detection and connected component marking, the shadow edges in the bank document image are identified and the shadow area of the bank document image is marked. The pixels marked as the shadow of the bank document image are traversed to obtain the shadow area of the bank document image of the accounting robot. The ink dots in the bank document image are identified by noise detection and connected component marking, and the number of ink dots in the bank document image of the accounting robot is obtained; The contrast of bank document images of accounting robots is obtained through image processing software; This analysis resulted in an accuracy evaluation index for bank document image processing by the accounting robot.
6. The accounting robot system as claimed in claim 1, characterized in that: The specific process for processing the digital information of bank documents and obtaining the digital processing accuracy evaluation index of bank documents by the accounting robot is as follows: Get the digital quantity of bank documents from accounting robots through flatbed scanner; Mark the number area of the bank document image through edge detection, use optical character recognition technology to recognize the numbers in the bank document image, and obtain the correct number of numbers in the bank document of the accounting robot; Scan the currency symbol on the bank document image through optical character recognition technology to obtain the correct currency symbol of the accounting robot bank document; This analysis resulted in an evaluation index of the accuracy of the accounting robot’s digital processing of bank documents.
7. The accounting robot system as claimed in claim 1, characterized in that: The specific process for processing the letter information of bank documents and obtaining the accuracy evaluation index of bank document letter processing by the accounting robot is as follows: Get the name and number of the bank document of the accounting robot through the flatbed scanner; Get the correct name and number of bank documents from accounting robots through intelligent character recognition technology; Get the number of letters in the bank document of the accounting robot through the flatbed scanner; Get the correct number of letters in bank documents of accounting robots through optical character technology; The actual number of letters in the accounting robot bank document is obtained through a flatbed scanner and text analysis tools; This analysis resulted in an accuracy evaluation index for the accounting robot’s letter processing of bank documents.
8. The accounting robot system as claimed in claim 1, characterized in that: The specific process of obtaining the comprehensive performance evaluation index of the accounting robot is as follows: The comprehensive performance evaluation index of the accounting robot is obtained through the accuracy evaluation index of the accounting robot bank document image processing, the accuracy evaluation index of the accounting robot bank document digital processing and the accuracy evaluation index of the accounting robot bank document letter processing; ; represents the comprehensive performance evaluation index of accounting robots, It represents the weight factor affecting the accuracy evaluation index of bank document image processing by accounting robots. It represents the accuracy evaluation index of bank document image processing by accounting robot. It represents the weight factor affecting the accuracy evaluation index of bank document digital processing by accounting robots. It represents the accuracy evaluation index of bank document digital processing by accounting robots. It represents the weight factor affecting the accuracy evaluation index of bank document letter processing of accounting robots. It represents the accuracy evaluation index of bank document letter processing by accounting robot.
9. The accounting robot system as claimed in claim 1, characterized in that: The specific process of optimizing the accounting robot according to the evaluation and analysis results of the accounting robot is as follows: The comprehensive error deviation value of the accounting robot scanning bank documents, the first error judgment value of the accounting robot scanning bank documents, the second error judgment value of the accounting robot scanning bank documents and the third error judgment value of the accounting robot scanning bank documents are introduced into the accounting robot error detection model, and a comprehensive analysis is performed to obtain an error detection evaluation value of the accounting robot scanning bank documents; the error detection evaluation value of the accounting robot scanning bank documents is compared with the error detection boundary evaluation value of the accounting robot scanning bank documents stored in the database; If the error detection evaluation value of the accounting robot scanning bank documents is higher than the limit evaluation value of the error detection of the accounting robot scanning bank documents, the accounting robot scanning bank documents corresponding to the error detection evaluation value of the accounting robot scanning bank documents are qualified; If the error detection assessment value of the accounting robot scanning bank documents is equal to the limit assessment value of the error detection of the accounting robot scanning bank documents, then there is a risk in the accounting robot scanning bank documents; If the error detection evaluation value of the accounting robot scanning bank documents is lower than the defined error detection evaluation value of the accounting robot scanning bank documents, the accounting robot scanning bank documents corresponding to the error detection evaluation value of the accounting robot scanning bank documents are unqualified; If the accuracy evaluation index of the bank document image processing by the accounting robot is less than the first error judgment value of the bank document scanning by the accounting robot, then by using image editing software, the noise and miscellaneous points in the bank document image are removed, the contrast between the text and the background in the bank document image is enhanced, and the bank document image is sharpened; If the accuracy evaluation index of the digital processing of bank documents by the accounting robot is less than the second error judgment value of the bank documents scanned by the accounting robot, the scanning environment of the accounting robot is adjusted to make the light uniform, avoid shadows and reflections, and the brightness of the bank document image is adjusted through the image editing software; If the accuracy evaluation index of the accounting robot's bank document letter processing is less than the third error judgment value of the accounting robot scanning the bank document, the grayscale of the bank document image is adjusted by adjusting the scanning mode of the accounting robot's bank document letter recognition.
10. Use of an accounting robot as described in any one of claims 1 to 9, characterized in that: The specific process of the hardware equipment of the accounting robot is as follows: The central processing unit of the accounting robot is used to analyze the accounting robot's bank financial data, run preset banking operations, identify and process data loss and recognition errors, etc.; The accounting robot's scanner is used to scan bank documents into digital image format; The image processing unit of the accounting robot is used to process and manage image data; The camera of the accounting robot is used to capture images of bank documents. By capturing images of bank documents through the camera, the image quality can be adjusted in real time.
Citation Information
Patent Citations
Financial and accounting robot and system thereof
CN107203762A
Model training method and device and image question and answer method and device
CN113360699A
Road element identification method and related device
CN117132958A
