A data processing method and device for bill management

Abstract

This invention discloses a data processing method and apparatus for invoice management. The method includes: acquiring invoice image information to be processed; the invoice image information to be processed includes several invoice images to be processed; performing detection processing on the invoice image information to be processed to obtain an image detection result information set; the image detection result information set includes several image detection result information; performing identification and classification processing on the image detection result information set to obtain invoice identification information; the invoice identification information is used to indicate intelligent management of invoices. Therefore, this invention can obtain invoice identification information for indicating intelligent management of invoices through comprehensive processing such as detection processing and identification and classification of invoice image information to be processed. This is beneficial for improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving invoice registration efficiency and reducing invoice review risks.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a data processing method and apparatus for ticket management. Background Technology

[0002] Invoice management is one of the most important tasks in financial management, as all contract payment transactions require invoice registration. However, given the large volume of invoices generated by contract payments, current invoice information management still relies on manual registration of each invoice. This method is not only costly in terms of manpower, resources, and time, but also carries the risk of input errors, resulting in a high error rate and low efficiency. Therefore, it is crucial to provide a data processing method and apparatus for invoice management that improves the intelligent collection, storage, and verification of invoice information, thereby increasing invoice registration efficiency and reducing invoice review risks. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a data processing method and apparatus for invoice management, which can obtain invoice identification information for intelligent management of invoices by comprehensively processing the image information of the invoice to be processed, such as detection processing and identification and classification processing. This is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0004] To address the aforementioned technical problems, a first aspect of the present invention discloses a data processing method for invoice management, the method comprising:

[0005] Obtain image information of the invoice to be processed; the image information of the invoice to be processed includes several images of the invoice to be processed;

[0006] The image information of the ticket to be processed is detected and processed to obtain a set of image detection result information; the set of image detection result information includes several image detection result information.

[0007] The image detection result information set is processed for identification and classification to obtain invoice identification information; the invoice identification information is used to indicate the intelligent management of invoices.

[0008] As an optional implementation, in the first aspect of the present invention, the step of detecting and processing the image information of the document to be processed to obtain an image detection result information set includes:

[0009] For any of the aforementioned invoice images to be processed, an angle detection process is performed on the invoice image to obtain the first processed image information corresponding to the invoice image to be processed; the angle detection process is used to perform angle correction on the invoice image to be processed.

[0010] The first processed image information is subjected to text detection processing to obtain the second processed image information corresponding to the document image to be processed; the text detection processing is used to identify the text line region of the first processed image information;

[0011] The second processed image information is subjected to line and stamp detection to obtain the image detection result information corresponding to the document image to be processed.

[0012] As an optional implementation, in the first aspect of the present invention, the step of identifying and classifying the image detection result information set to obtain ticket identification information includes:

[0013] The image detection result information set is processed to obtain an image recognition result information set; the image recognition result information set includes several image recognition result information sets.

[0014] The image recognition result information set is post-processed to obtain the ticket recognition information.

[0015] As an optional implementation, in a first aspect of the present invention, the step of performing recognition processing on the image detection result information set to obtain an image recognition result information set includes:

[0016] The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets;

[0017] A dedicated engine is invoked to perform region recognition on the image detection result information set to obtain a second recognition result information set; the first recognition result information set includes several second recognition result information sets.

[0018] The first recognition result information set and the second recognition result information set are integrated to obtain the image recognition result information set.

[0019] As an optional implementation, in a first aspect of the present invention, the post-processing of the image recognition result information set to obtain ticket recognition information includes:

[0020] The image recognition result information set is filtered to obtain a filtered image result information set; the filtered image result information set includes several filtered image result information sets; the filtering process is used to remove text lines with low credibility;

[0021] The filtered image result information set is corrected to obtain ticket recognition information; the correction process is used to correct common errors in the filtered image result information.

[0022] As an optional implementation, in the first aspect of the present invention, after the image detection result information set is identified and classified to obtain ticket identification information, the method further includes:

[0023] The ticket identification information is then subjected to type identification to obtain ticket type information;

[0024] Based on the bill type information, execute the preset bill management strategy.

[0025] As an optional implementation, in the first aspect of the present invention, the step of executing a preset bill management strategy based on the bill type information includes:

[0026] When the invoice type information is a non-VAT invoice, the invoice identification information is stored and updated.

[0027] When the invoice type information is a value-added tax invoice, the authenticity of the invoice identification information is verified to obtain the verification result information.

[0028] When the verification result is yes, the ticket identification information is repeatedly collected and verified, and then stored and updated.

[0029] A second aspect of this invention discloses a data processing apparatus for ticket management, the apparatus comprising:

[0030] The acquisition module is used to acquire image information of the invoices to be processed; the image information of the invoices to be processed includes several images of invoices to be processed;

[0031] The first processing module is used to detect and process the image information of the ticket to be processed, and obtain an image detection result information set; the image detection result information set includes several image detection result information;

[0032] The second processing module is used to identify and classify the image detection result information set to obtain ticket identification information; the ticket identification information is used to indicate the intelligent management of tickets.

[0033] As an optional implementation, in the second aspect of the present invention, the first processing module performs detection processing on the image information of the document to be processed to obtain an image detection result information set in the following specific manner:

[0034] For any of the aforementioned invoice images to be processed, an angle detection process is performed on the invoice image to obtain the first processed image information corresponding to the invoice image to be processed; the angle detection process is used to perform angle correction on the invoice image to be processed.

[0035] The first processed image information is subjected to text detection processing to obtain the second processed image information corresponding to the document image to be processed; the text detection processing is used to identify the text line region of the first processed image information;

[0036] The second processed image information is subjected to line and stamp detection to obtain the image detection result information corresponding to the document image to be processed.

[0037] As an optional implementation, in a second aspect of the present invention, the second processing module performs identification and classification processing on the image detection result information set to obtain the ticket identification information in the following specific manner:

[0038] The image detection result information set is processed to obtain an image recognition result information set; the image recognition result information set includes several image recognition result information sets.

[0039] The image recognition result information set is post-processed to obtain the ticket recognition information.

[0040] As an optional implementation, in a second aspect of the present invention, the second processing module performs recognition processing on the image detection result information set to obtain the image recognition result information set in the following specific manner:

[0041] The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets;

[0042] A dedicated engine is invoked to perform region recognition on the image detection result information set to obtain a second recognition result information set; the first recognition result information set includes several second recognition result information sets.

[0043] The first recognition result information set and the second recognition result information set are integrated to obtain the image recognition result information set.

[0044] As an optional implementation, in a second aspect of the present invention, the second processing module performs post-processing on the image recognition result information set to obtain the invoice recognition information in the following specific manner:

[0045] The image recognition result information set is filtered to obtain a filtered image result information set; the filtered image result information set includes several filtered image result information sets; the filtering process is used to remove text lines with low credibility;

[0046] The filtered image result information set is corrected to obtain ticket recognition information; the correction process is used to correct common errors in the filtered image result information.

[0047] As an optional implementation, in a second aspect of the present invention, after the second processing module performs identification and classification processing on the image detection result information set to obtain ticket identification information, the apparatus further includes:

[0048] The identification module is used to identify the type of the ticket identification information to obtain ticket type information;

[0049] The execution module is used to execute a preset invoice management strategy based on the invoice type information.

[0050] As an optional implementation, in the second aspect of the present invention, the execution module executes the preset bill management strategy according to the bill type information in the following specific manner:

[0051] When the invoice type information is a non-VAT invoice, the invoice identification information is stored and updated.

[0052] When the invoice type information is a value-added tax invoice, the authenticity of the invoice identification information is verified to obtain the verification result information.

[0053] When the verification result is yes, the ticket identification information is repeatedly collected and verified, and then stored and updated.

[0054] A third aspect of the present invention discloses another data processing apparatus for ticket management, the apparatus comprising:

[0055] Memory containing executable program code;

[0056] A processor coupled to the memory;

[0057] The processor calls the executable program code stored in the memory to execute some or all of the steps in the data processing method for ticket management disclosed in the first aspect of the present invention.

[0058] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the data processing method for ticket management disclosed in the first aspect of the present invention.

[0059] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0060] In this embodiment of the invention, image information of invoices to be processed is acquired; the image information of invoices to be processed includes several images of invoices to be processed; the image information of invoices to be processed is detected and processed to obtain a set of image detection result information; the set of image detection result information includes several sets of image detection result information; the set of image detection result information is identified and classified to obtain invoice identification information; the invoice identification information is used to indicate intelligent management of invoices. It can be seen that this invention, through comprehensive processing such as detection and identification / classification of image information of invoices to be processed, can obtain invoice identification information for indicating intelligent management of invoices, which is beneficial to improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving invoice registration efficiency and reducing invoice review risks. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating a data processing method for invoice management disclosed in an embodiment of the present invention;

[0063] Figure 2 This is a flowchart illustrating another data processing method for bill management disclosed in an embodiment of the present invention;

[0064] Figure 3 This is a schematic diagram of the structure of a data processing device for ticket management disclosed in an embodiment of the present invention;

[0065] Figure 4 This is a schematic diagram of another data processing device for ticket management disclosed in an embodiment of the present invention;

[0066] Figure 5 A schematic diagram of the structure of another data processing device for ticket management disclosed in this embodiment of the invention. Detailed Implementation

[0067] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0069] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0070] This invention discloses a data processing method and apparatus for invoice management. It can obtain invoice identification information for intelligent management of invoices through comprehensive processing of invoice image information, including detection, recognition, and classification. This improves the intelligent collection, storage, and verification of invoice information, thereby increasing invoice registration efficiency and reducing invoice review risks. Detailed descriptions follow.

[0071] Example 1

[0072] Please see Figure 1 , Figure 1 This is a flowchart illustrating a data processing method for invoice management disclosed in an embodiment of the present invention. Figure 1 The described data processing method for invoice management is applied in an invoice management system, such as a local server or cloud server for data processing and management of invoices; this embodiment of the invention is not limited to such applications. Figure 1 As shown, the data processing method for invoice management may include the following operations:

[0073] 101. Obtain the image information of the invoice to be processed.

[0074] In this embodiment of the invention, the above-mentioned document image information to be processed includes several document images to be processed.

[0075] 102. Detect and process the image information of the ticket to be processed to obtain a set of image detection results.

[0076] In this embodiment of the invention, the above-mentioned image detection result information set includes several image detection result information sets.

[0077] 103. The image detection result information set is identified and classified to obtain the invoice recognition information.

[0078] In this embodiment of the invention, the aforementioned invoice identification information is used to indicate intelligent management of invoices.

[0079] Optionally, the aforementioned invoice image information to be processed is obtained by batch uploading scanned copies of invoices or electronic invoice files.

[0080] Optionally, the above-mentioned detection and processing of the invoice image information to be processed is performed using an image recognition model.

[0081] Optionally, the output layer of the above image recognition model is 10 to match the number of ticket categories and improve the data processing efficiency of tickets.

[0082] Optionally, the above-mentioned processing of the information detection of the invoice image to be processed involves inputting the invoice image into a U-shaped network, first passing it through a 5-layer convolutional layer to extract the image features, then upsampling and fusing the convolutional results of the previous layer, finally obtaining a feature map of 128 features at half the size of the image, then outputting different prediction scoresmaps according to different targets, and finally analyzing the corresponding image detection results based on the scoresmaps.

[0083] Optionally, using the U-shaped network described above for the detection and processing of the invoice image information can avoid the problems of large network size and huge storage model, thereby improving the processing efficiency of the invoice image.

[0084] Optionally, the data processing method described above for invoice management can be used to register multiple invoices in batches, controlling the invoice information registration process to 3 to 5 seconds, greatly reducing manual entry time, improving the efficiency of invoice information entry, and also helping to ensure the accuracy and standardization of invoice information.

[0085] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can obtain invoice identification information for intelligent management of invoices by comprehensively processing the image information of the invoice to be processed, such as detection processing and identification and classification processing. This is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0086] In an optional embodiment, step 102 above involves detecting and processing the image information of the ticket to be processed, resulting in a set of image detection result information, including:

[0087] For any document image to be processed, angle detection processing is performed on the document image to obtain the first processed image information corresponding to the document image to be processed; angle detection processing is used to correct the angle of the document image to be processed.

[0088] The first processed image information is subjected to text detection processing to obtain the second processed image information corresponding to the document image to be processed; the text detection processing is used to identify the text line region of the first processed image information;

[0089] Line and stamp detection are performed on the second processed image information to obtain the image detection result information corresponding to the document image to be processed.

[0090] In this optional embodiment, as an optional implementation method, the specific way to perform angle detection processing on the document image to be processed to obtain the first processed image information corresponding to the document image is as follows:

[0091] Identify the main orientation angle of the above-mentioned document image to be processed;

[0092] The image of the invoice to be processed is corrected using the aforementioned main direction angle to obtain the first processed image information corresponding to the image of the invoice to be processed.

[0093] Optionally, by utilizing the aforementioned angle detection and text detection processing, the location information of the ticket data in the ticket image to be processed can be accurately located in a chaotic and complex scene.

[0094] Optionally, the above-mentioned ticket data location information includes the main direction angle, and / or, straight line information, and / or, stamp area, and / or, text area, which are not limited in this embodiment of the invention.

[0095] Optionally, the above-mentioned line detection of the second processed image information is to identify and locate the table lines in the second processed image.

[0096] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can obtain image detection result information by performing angle detection, text detection, line detection and stamp detection on the invoice image to be processed. This is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0097] In another optional embodiment, the above-described image detection result information set is processed for identification and classification to obtain ticket identification information, including:

[0098] The image detection result information set is processed to obtain the image recognition result information set; the image recognition result information set includes several image recognition result information sets.

[0099] Post-processing of the image recognition result set yields the invoice recognition information.

[0100] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can obtain invoice identification information by comprehensively processing the image detection result information set through recognition processing and post-processing, which is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0101] In another optional embodiment, the above-described image detection result information set is processed to obtain an image recognition result information set, including:

[0102] The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets;

[0103] A dedicated engine is invoked to perform region recognition on the image detection result information set, resulting in a second recognition result information set; the first recognition result information set includes several second recognition result information sets.

[0104] The first and second recognition result information sets are integrated to obtain the image recognition result information set.

[0105] Optionally, the above-mentioned whole-line text recognition uses sequence recognition to process the text image, thereby avoiding the extremely difficult single-character segmentation and recognition in traditional algorithms, thus improving recognition accuracy and efficiency.

[0106] Optionally, the above-mentioned recognition processing of text images through sequence recognition is to recognize and process the text regions according to the rules and order of text writing.

[0107] Optionally, the end-to-end average accuracy of the above-mentioned whole-line text recognition can reach over 95%.

[0108] Optionally, the general recognition engine used for recognizing the entire line of text on the image detection result information set includes a CNN network module, and / or an LSTM network module, and / or an Attention network module, and / or a CTC network module. This embodiment of the invention does not limit the specific components.

[0109] In this optional embodiment, as an optional implementation method, the specific way to call the dedicated engine to perform region recognition on the image detection result information set to obtain the second recognition result information set is as follows:

[0110] For any of the above image detection results, obtain table template information; the table template information includes several table templates;

[0111] For any of the above table templates, the table area in the image detection result information is analyzed using the line information in the image detection result information to obtain the first table cell information corresponding to the table template;

[0112] Structural analysis is performed on the first table grid information, and the text information in the image detection result information is fused to obtain the second table grid information corresponding to the table template;

[0113] Feature extraction is performed on the cell information of the second table mentioned above to obtain the table feature information corresponding to the table template;

[0114] The matching score is calculated by performing matching score processing on the above table feature information to obtain the table matching score corresponding to the table template;

[0115] Select the table matching score with the highest value from all the above table matching scores as the target table matching score;

[0116] The table template corresponding to the target table matching score is determined as the target table template;

[0117] The recognition engine corresponding to the target table template is called as the dedicated engine.

[0118] The image detection result information is then used to perform small character set character recognition to obtain the second recognition result information corresponding to the image detection result information.

[0119] Optionally, using the aforementioned dedicated engine to perform small character set text recognition on the image detection result information set can achieve the recognition of special items such as handwritten numbers, handwritten English letters, and handwritten capital amounts, so as to more accurately and quickly recognize small character set text.

[0120] Optionally, the table feature information mentioned above includes the number of rows and columns, and / or the relative size of the table, and / or the number of rows and columns occupied by the table, and / or the keywords inside the table. This embodiment of the invention does not limit the information.

[0121] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can obtain an image recognition result information set by performing whole-line text recognition, region recognition, and integration processing on the image detection result information set. This is more conducive to improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0122] In another optional embodiment, the post-processing of the image recognition result information set to obtain ticket recognition information includes:

[0123] The image recognition result information set is filtered to obtain a filtered image result information set; the filtered image result information set includes several filtered image result information; the filtering process is used to remove text lines with low credibility;

[0124] The filtered image result information set is corrected to obtain ticket recognition information; the correction process is used to correct common errors in the filtered image result information.

[0125] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can obtain invoice recognition information by filtering and correcting the image recognition result information set, which is more conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0126] Example 2

[0127] Please see Figure 2 , Figure 2 This is a flowchart illustrating another data processing method for invoice management disclosed in an embodiment of the present invention. Figure 2 The described data processing method for invoice management is applied in an invoice management system, such as a local server or cloud server for data processing and management of invoices; this embodiment of the invention is not limited to such applications. Figure 2 As shown, the data processing method for invoice management may include the following operations:

[0128] 201. Obtain the image information of the invoice to be processed.

[0129] 202. Detect and process the image information of the ticket to be processed to obtain a set of image detection results.

[0130] 203. The image detection result information set is identified and classified to obtain the invoice recognition information.

[0131] 204. Perform type identification on the bill identification information to obtain bill type information.

[0132] 205. Execute the preset invoice management strategy based on the invoice type information.

[0133] In this embodiment of the invention, the specific technical details and explanations of technical terms for steps 201-202 can be found in the detailed description of steps 101-102 in Embodiment 1, and will not be repeated here.

[0134] In this optional embodiment, as an optional implementation method, after executing the preset bill management strategy based on the bill type information, the method further includes:

[0135] Retrieve ticket collection information from the database;

[0136] The above-mentioned invoice collection information is categorized and displayed dynamically.

[0137] Optionally, the above-mentioned invoice collection information includes invoice code, and / or, invoice number, date, and / or, amount, and / or, verification code, which is not limited in this embodiment of the invention.

[0138] As can be seen, the data processing method for bill management described in the embodiments of the present invention can obtain bill identification information for intelligent management of bills by comprehensively processing the bill image information to be processed, such as detection processing and identification and classification processing. By processing the bill identification information such as type identification, the bills can be managed intelligently. This is conducive to improving the intelligent collection, intelligent storage and intelligent verification of bill information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0139] In an optional embodiment, step 205 above, which executes a preset bill management strategy based on the bill type information, includes:

[0140] When the invoice type information is a non-VAT invoice, the invoice identification information is stored and updated.

[0141] When the invoice type information is a value-added tax invoice, the authenticity of the invoice identification information is verified to obtain the verification result information.

[0142] When the verification result is positive, the ticket identification information is repeatedly collected and verified, and then stored and updated.

[0143] Optionally, the above-mentioned ticket identification information includes several ticket identification results.

[0144] In this optional embodiment, as an optional implementation method, the specific way to perform duplicate collection and verification processing and storage update processing on the ticket identification information is as follows:

[0145] The ticket identification results from the above ticket identification information are selected sequentially as the tickets to be collected.

[0146] Determine whether any of the archived VAT invoices in the database's archived VAT invoice set exists to match the target VAT invoice mentioned above, and obtain the invoice determination result.

[0147] When the above invoice judgment result is yes, the execution sequence is triggered to select the invoice recognition result in the above invoice recognition information as the invoice to be collected.

[0148] When the above invoice judgment result is negative, the above archived VAT invoice set is updated using the above-mentioned documents to be collected.

[0149] As can be seen, the data processing method for invoice management described in the embodiments of the present invention can repeatedly collect, verify, and update invoice identification information based on the verification results, which is more conducive to improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0150] Example 3

[0151] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a data processing device for ticket management disclosed in an embodiment of the present invention. Figure 3 The described apparatus can be applied to a ticket management system, such as a local server or cloud server for data processing and management of ticket management, etc., and the embodiments of the present invention are not limited thereto. Figure 3 As shown, the device may include:

[0152] The acquisition module 301 is used to acquire image information of the invoice to be processed; the image information of the invoice to be processed includes several images of the invoice to be processed;

[0153] The first processing module 302 is used for detecting and processing the image information of the ticket to be processed, and obtaining a set of image detection result information; the set of image detection result information includes several image detection result information.

[0154] The second processing module 303 is used to identify and classify the image detection result information set to obtain ticket identification information; the ticket identification information is used to indicate the intelligent management of tickets.

[0155] It is evident that implementation Figure 3The described data processing device for invoice management can obtain invoice recognition information for intelligent management of invoices by comprehensively processing the image information of the invoices to be processed, such as detection, recognition and classification. This is conducive to improving the intelligent collection, storage and verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0156] In another alternative embodiment, such as Figure 4 As shown, the first processing module 302 performs image detection and processing on the document image to obtain the image detection result information set in the following specific way:

[0157] For any document image to be processed, angle detection processing is performed on the document image to obtain the first processed image information corresponding to the document image to be processed; angle detection processing is used to correct the angle of the document image to be processed.

[0158] The first processed image information is subjected to text detection processing to obtain the second processed image information corresponding to the document image to be processed; the text detection processing is used to identify the text line region of the first processed image information;

[0159] Line and stamp detection are performed on the second processed image information to obtain the image detection result information corresponding to the document image to be processed.

[0160] It is evident that implementation Figure 4 The described data processing device for invoice management can obtain image detection results by performing angle detection, text detection, line detection, and stamp detection on the invoice image to be processed. This is beneficial for improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0161] In yet another alternative embodiment, such as Figure 4 As shown, the second processing module 303 performs recognition and classification processing on the image detection result information set to obtain the ticket recognition information in the following specific way:

[0162] The image detection result information set is processed to obtain the image recognition result information set; the image recognition result information set includes several image recognition result information sets.

[0163] Post-processing of the image recognition result set yields the invoice recognition information.

[0164] It is evident that implementation Figure 4The described data processing device for invoice management can perform comprehensive processing such as recognition processing and post-processing on the image detection result information set to obtain invoice recognition information. This is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0165] In yet another alternative embodiment, such as Figure 4 As shown, the second processing module 303 performs recognition processing on the image detection result information set to obtain the image recognition result information set in the following specific way:

[0166] The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets;

[0167] A dedicated engine is invoked to perform region recognition on the image detection result information set, resulting in a second recognition result information set; the first recognition result information set includes several second recognition result information sets.

[0168] The first and second recognition result information sets are integrated to obtain the image recognition result information set.

[0169] It is evident that implementation Figure 4 The data processing device for invoice management described herein can obtain an image recognition result information set by performing whole-line text recognition, region recognition, and integration processing on the image detection result information set. This is more conducive to improving the intelligent collection, intelligent storage, and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0170] In yet another alternative embodiment, such as Figure 4 As shown, the second processing module 303 performs post-processing on the image recognition result information set to obtain the ticket recognition information in the following specific way:

[0171] The image recognition result information set is filtered to obtain a filtered image result information set; the filtered image result information set includes several filtered image result information; the filtering process is used to remove text lines with low credibility;

[0172] The filtered image result information set is corrected to obtain ticket recognition information; the correction process is used to correct common errors in the filtered image result information.

[0173] It is evident that implementation Figure 4The data processing device for invoice management described herein can obtain invoice recognition information by filtering and correcting the image recognition result information set. This is more conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0174] In yet another alternative embodiment, such as Figure 4 As shown, after the second processing module 303 performs recognition and classification processing on the image detection result information set to obtain the ticket recognition information, the device further includes:

[0175] The identification module 304 is used to identify the type of the ticket identification information to obtain the ticket type information.

[0176] The execution module 305 is used to execute a preset invoice management strategy based on the invoice type information.

[0177] It is evident that implementation Figure 4 The described data processing device for invoice management can obtain invoice identification information for intelligent management of invoices by comprehensively processing the image information of the invoices to be processed, such as detection, identification and classification. By processing the invoice identification information by type identification, the device can achieve intelligent management of invoices, which is conducive to improving the intelligent collection, intelligent storage and intelligent verification of invoice information, thereby improving the efficiency of invoice registration and reducing the risk of invoice review.

[0178] In yet another alternative embodiment, such as Figure 4 As shown, the execution module 305 executes the preset invoice management strategy according to the invoice type information in the following specific way:

[0179] When the invoice type information is a non-VAT invoice, the invoice identification information is stored and updated.

[0180] When the invoice type information is a value-added tax invoice, the authenticity of the invoice identification information is verified to obtain the verification result information.

[0181] When the verification result is positive, the ticket identification information is repeatedly collected and verified, and then stored and updated.

[0182] It is evident that implementation Figure 4 The data processing device for invoice management described herein can repeatedly collect, verify, and update invoice identification information based on the verification results. This is more conducive to improving the intelligent collection, storage, and verification of invoice information, thereby increasing the efficiency of invoice registration and reducing the risk of invoice review.

[0183] Example 4

[0184] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another data processing device for ticket management disclosed in an embodiment of the present invention. Figure 5 The described apparatus can be applied to a ticket management system, such as a local server or cloud server for data processing and management of ticket management, etc., and the embodiments of the present invention are not limited thereto. Figure 5 As shown, the device may include:

[0185] Memory 401 storing executable program code;

[0186] Processor 402 coupled to memory 401;

[0187] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the data processing method for ticket management described in Embodiment 1 or Embodiment 2.

[0188] Example 5

[0189] This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps in the data processing method for ticket management described in Embodiment 1 or Embodiment 2.

[0190] Example 6

[0191] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the data processing method for ticket management described in Embodiment 1 or Embodiment 2.

[0192] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0193] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0194] Finally, it should be noted that the data processing method and apparatus for ticket management disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method for invoice management, characterized in that, The method includes: Obtain image information of the invoice to be processed; the image information of the invoice to be processed includes several images of the invoice to be processed; The image information of the ticket to be processed is detected and processed to obtain a set of image detection result information; the set of image detection result information includes several image detection result information. The image detection result information set is identified and classified to obtain invoice identification information; the invoice identification information is used to indicate the intelligent management of invoices. The process of identifying and classifying the image detection result information set to obtain invoice recognition information includes: The image detection result information set is processed to obtain an image recognition result information set; the image recognition result information set includes several image recognition result information sets. Post-processing the image recognition result information set yields invoice recognition information; The step of performing recognition processing on the image detection result information set to obtain the image recognition result information set includes: The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets; A dedicated engine is invoked to perform region recognition on the image detection result information set to obtain a second recognition result information set; the first recognition result information set includes several second recognition result information sets. The first recognition result information set and the second recognition result information set are integrated to obtain the image recognition result information set; The entire line of text recognition is processed by sequence recognition to identify the text image. The process of recognizing the text image by sequence recognition is to identify the text region according to the rules and order of text writing. The specific method for calling the dedicated engine to perform region recognition on the image detection result information set to obtain the second recognition result information set is as follows: For any of the above image detection results, obtain table template information; the table template information includes several table templates; For any of the above table templates, the table area in the image detection result information is analyzed using the line information in the image detection result information to obtain the first table cell information corresponding to the table template; Structural analysis is performed on the first table grid information, and the text information in the image detection result information is fused to obtain the second table grid information corresponding to the table template; Feature extraction is performed on the cell information of the second table mentioned above to obtain the table feature information corresponding to the table template; The matching score is calculated by performing matching score processing on the above table feature information to obtain the table matching score corresponding to the table template; Select the table matching score with the highest value from all the above table matching scores as the target table matching score; The table template corresponding to the target table matching score is determined as the target table template; The recognition engine corresponding to the target table template is called as the dedicated engine. The image detection result information is used to perform small character set character recognition to obtain the second recognition result information corresponding to the image detection result information; The table feature information includes the number of rows and columns, and / or the relative size of the table, and / or the number of rows and columns occupied by the table, and / or the internal keywords of the table.

2. The data processing method for invoice management according to claim 1, characterized in that, The detection and processing of the image information of the document to be processed yields a set of image detection result information, including: For any of the aforementioned invoice images to be processed, an angle detection process is performed on the invoice image to obtain the first processed image information corresponding to the invoice image to be processed; the angle detection process is used to perform angle correction on the invoice image to be processed. The first processed image information is subjected to text detection processing to obtain the second processed image information corresponding to the document image to be processed; the text detection processing is used to identify the text line region of the first processed image information; The second processed image information is subjected to line and stamp detection to obtain the image detection result information corresponding to the document image to be processed.

3. The data processing method for invoice management according to claim 1, characterized in that, The post-processing of the image recognition result information set to obtain ticket recognition information includes: The image recognition result information set is filtered to obtain a filtered image result information set; the filtered image result information set includes several filtered image result information sets; the filtering process is used to remove text lines with low credibility; The filtered image result information set is corrected to obtain ticket recognition information; the correction process is used to correct common errors in the filtered image result information.

4. The data processing method for invoice management according to claim 1, characterized in that, After performing identification and classification processing on the image detection result information set to obtain ticket identification information, the method further includes: The ticket identification information is then subjected to type identification to obtain ticket type information; Based on the bill type information, execute the preset bill management strategy.

5. The data processing method for ticket management according to claim 4, characterized in that, The step of executing a preset invoice management strategy based on the invoice type information includes: When the invoice type information is a non-VAT invoice, the invoice identification information is stored and updated. When the invoice type information is a value-added tax invoice, the authenticity of the invoice identification information is verified to obtain the verification result information. When the verification result is yes, the ticket identification information is repeatedly collected and verified, and then stored and updated.

6. A data processing device for ticket management, characterized in that, The device includes: The acquisition module is used to acquire image information of the invoices to be processed; the image information of the invoices to be processed includes several images of invoices to be processed; The first processing module is used to detect and process the image information of the ticket to be processed, and obtain an image detection result information set; the image detection result information set includes several image detection result information; The second processing module is used to identify and classify the image detection result information set to obtain ticket identification information; the ticket identification information is used to indicate the intelligent management of tickets. The second processing module performs identification and classification processing on the image detection result information set to obtain the ticket recognition information in the following specific way: The image detection result information set is processed to obtain an image recognition result information set; the image recognition result information set includes several image recognition result information sets. Post-processing the image recognition result information set yields invoice recognition information; The second processing module performs recognition processing on the image detection result information set to obtain the image recognition result information set in the following specific way: The image detection result information set is subjected to whole-line text recognition to obtain a first recognition result information set; the first recognition result information set includes several first recognition result information sets; A dedicated engine is invoked to perform region recognition on the image detection result information set to obtain a second recognition result information set; the first recognition result information set includes several second recognition result information sets. The first recognition result information set and the second recognition result information set are integrated to obtain the image recognition result information set; The entire line of text recognition is processed by sequence recognition to identify the text image. The process of recognizing the text image by sequence recognition is to identify the text region according to the rules and order of text writing. The second processing module calls a dedicated engine to perform region recognition on the image detection result information set, and obtains the second recognition result information set in the following specific way: For any of the above image detection results, obtain table template information; the table template information includes several table templates; For any of the above table templates, the table area in the image detection result information is analyzed using the line information in the image detection result information to obtain the first table cell information corresponding to the table template; Structural analysis is performed on the first table grid information, and the text information in the image detection result information is fused to obtain the second table grid information corresponding to the table template; Feature extraction is performed on the cell information of the second table mentioned above to obtain the table feature information corresponding to the table template; The matching score is calculated by performing matching score processing on the above table feature information to obtain the table matching score corresponding to the table template; Select the table matching score with the highest value from all the above table matching scores as the target table matching score; The table template corresponding to the target table matching score is determined as the target table template; The recognition engine corresponding to the target table template is called as the dedicated engine. The image detection result information is used to perform small character set character recognition to obtain the second recognition result information corresponding to the image detection result information; The table feature information includes the number of rows and columns, and / or the relative size of the table, and / or the number of rows and columns occupied by the table, and / or the internal keywords of the table.

7. A data processing device for ticket management, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the data processing method for ticket management as described in any one of claims 1-5.

8. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the data processing method for ticket management as described in any one of claims 1-5.

Images