Intelligent identification system and method for preventing repetition of bills

Abstract

The invention discloses an intelligent identification system and method for preventing repetition of bills, and relates to the technical field of data identification, and the method comprises the steps: collecting digital signatures and text information of electronic bills, adding timestamps, and constructing historical reimbursement behaviors; extracting writing features of the digital signature to construct a to-be-detected bill; to-be-detected bills in a preset interval are screened according to timestamps, the writing feature similarity is calculated, and suspected repeated recognition bills are marked; sorting the suspected repeated bills to form a to-be-detected sequence, comparing the to-be-detected sequence with a historical reimbursement sequence, marking the to-be-detected sequence successfully if a consistent sequence exists, otherwise, automatically reimbursement the first bill, and transferring the rest to manual auditing; the system comprises a data acquisition and construction module, a data feature extraction module, a screening similarity judgment module, a classification comparison processing module and a result record updating module, the bill repeated recognition accuracy is improved, the invalid manual auditing amount is greatly reduced, and the reimbursement auditing efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data analysis, and particularly relates to an intelligent identification system and method for preventing ticket repetition. BACKGROUND

[0002] Under the dual driving of digital transformation and deepening of enterprise cross-regional operation, as the core voucher of financial reimbursement and compliance audit, the flow scene of tickets is increasingly complex. The popularity of electronic invoices solves the problem of low efficiency of paper ticket delivery, but the same ticket may need to be submitted at the same time in multiple sub-departments, resulting in multiple system judgments of repeated reimbursement and the need for multiple manual audits, which reduces the efficiency of ticket review. SUMMARY

[0003] The present application relates to the technical field of big data analysis, and particularly relates to an intelligent identification system and method for preventing ticket repetition.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical scheme: an intelligent identification method for preventing ticket repetition, the intelligent identification method comprising the following steps: Step S1, collecting the digital signature and text information of the electronic ticket and attaching a time stamp and historical reimbursement behavior, the historical reimbursement behavior being all ticket reimbursement sequences of the same digital signature; Step S2, cleaning and standardizing the collected digital signature and text information, extracting the writing features of the digital signature, and constructing the to-be-detected ticket; Step S3, extracting all to-be-detected tickets within a preset range before and after the time stamp, collecting all writing features, and marking the to-be-detected tickets with a similarity exceeding a preset threshold as suspected repeated identification tickets; Step S4, extracting all suspected repeated identification tickets, classifying them according to the writing features, sorting the suspected repeated identification tickets according to the time stamp to form a to-be-detected sequence, and querying whether there is a same ticket reimbursement sequence in the historical reimbursement behavior, if there is, marking it as passed, if there is not, reimbursing the first suspected repeated identification ticket, and converting the subsequent suspected repeated identification tickets into manual audits; Step S5, collecting the manual audit results and updating the corresponding ticket reimbursement history record.

[0005] Further, the specific steps of step S1 are as follows: Step S1-1, collecting the digital signature and text information of the electronic ticket, the text information including all text information on the electronic ticket; Step S1-2, generating a time stamp according to the real-time time of the electronic ticket submission, binding and associating the time stamp with the digital signature and text information of the corresponding electronic ticket to form a preliminary data set; Step S1-3, extracting all past ticket reimbursement data corresponding to a certain digital signature, arranging the complete ticket reimbursement sequence in chronological order, and constructing historical reimbursement behavior.

[0006] Further, the specific steps of step S2 are as follows: Step S2-1, normalizing the digital signature according to the system preset format, and uniformly converting the text information into a preset format; Step S2-2, extracting writing features from the normalized digital signature, the writing features including stroke logic and structural layout of the signature; Step S2-3, associating and integrating the extracted writing features, the converted text information, and the timestamp, and the historical reimbursement behavior to form a to-be-detected ticket.

[0007] Further, the specific steps of step S3 are as follows: Step S3-1, extracting a to-be-detected ticket, constructing a preset time interval of the to-be-detected ticket according to its timestamp and the system preset timestamp filtering range, extracting all to-be-detected tickets with timestamps falling within the preset time interval from all to-be-detected tickets to form a temporary to-be-detected ticket set; Step S3-2, extracting the digital signature writing features of each to-be-detected ticket from the temporary to-be-detected ticket set one by one, and forming a temporary writing feature set from all extracted writing features; Step S3-3, using a preset feature similarity calculation method to pairwise compare and calculate all writing features in the temporary writing feature set, obtaining the similarity value between each two groups of writing features, comparing the calculated similarity value with the system preset threshold, if the similarity value of the two groups of writing features exceeds the system preset threshold, then marking the to-be-detected tickets associated with the two groups of features as suspected repeated identification tickets, and summarizing and arranging all information marked as suspected repeated identification tickets to form a suspected repeated ticket list.

[0008] Further, the specific steps of step S4 are as follows: Step S4-1, classifying all suspected repeated identification tickets according to the extracted writing features, and dividing suspected repeated identification tickets with consistent writing features into the same category; Step S4-2, sequentially ordering the suspected repeated identification tickets in each category according to their bound timestamp information, and associating and binding the digital signature, the standardized text information, and the to-be-detected sequence of each suspected repeated identification ticket to form a to-be-detected sequence of each category; Step S4-3, extracting the bill reimbursement sequence in the historical reimbursement behavior associated with the digital signature corresponding to the to-be-detected sequence, comparing the to-be-detected sequence with the extracted historical reimbursement sequence, if the comparison result is that there is a completely consistent bill reimbursement sequence, marking all suspected repeated identification bills in this category as passed; if the comparison result is that there is no completely consistent bill reimbursement sequence, performing an automatic reimbursement process on the suspected repeated identification bill at the top of the sequence, and the remaining bills are transferred to a manual review process, and the determination result and the corresponding bill association information are recorded.

[0009] Further, the specific steps of step S5 are as follows: Step S5-1, collecting the determination results generated by the manual review process, and summarizing the determination results, the digital signature bound to the bill, the standardized text information and the time stamp to form a record document; Step S5-2, updating the bill reimbursement history record corresponding to the digital signature according to the record document.

[0010] Further, the intelligent identification system comprises a data acquisition and construction module, a data feature extraction module, a screening similarity determination module, a classification comparison processing module and a result record updating module; The data acquisition and construction module is used to collect the digital signature, the text information and generate the time stamp of the electronic bill, and construct the historical reimbursement behavior in the form of all bill reimbursement sequences corresponding to the same digital signature; the data feature extraction module is used to clean and standardize the collected digital signature and text information, extract the writing features of the digital signature, and associate the time stamp and the historical reimbursement behavior to construct the to-be-detected bill; the screening similarity determination module is used to screen the bills in a preset time interval according to the time stamp of the to-be-detected bill, and mark the suspected repeated identification bills by calculating the writing feature similarity; the classification comparison processing module is used to classify the suspected repeated identification bills according to the writing features, form a to-be-detected sequence according to the time stamp, and perform corresponding processing after comparing with the historical reimbursement sequence; the result record updating module is used to collect the determination results generated by the manual review process, summarize the results and the bill information, and update the bill reimbursement history record corresponding to the digital signature; The output end of the data acquisition and construction module is electrically connected to the input end of the data feature extraction module; the output end of the data feature extraction module is electrically connected to the input end of the screening similarity determination module; the output end of the screening similarity determination module is electrically connected to the input end of the classification comparison processing module; and the output end of the classification comparison processing module is electrically connected to the input end of the result record updating module.

[0011] Further, the data acquisition and construction module comprises an information acquisition unit and a historical behavior construction unit; The information collection unit is used for collecting the digital signature of the electronic ticket, the text information composed of all the texts on the ticket, and generating a timestamp associated with the ticket according to the real-time time when the ticket is submitted; the historical behavior construction unit is used for extracting all the past ticket reimbursement data corresponding to a specific digital signature, and arranging the data in chronological order to form a complete ticket reimbursement sequence.

[0012] Further, the data feature extraction module comprises a data normalization unit and a feature extraction unit. The data normalization unit is used for normalizing the collected digital signature according to the system preset format, and converting the text information into the system preset format; the feature extraction unit is used for extracting the stroke logic and structural layout writing features from the normalized digital signature, and integrating the features with the converted text information, the timestamp and the historical reimbursement behavior to form a to-be-detected ticket. The screening similarity determination module comprises a time screening unit and a similarity determination unit. The time screening unit is used for constructing a preset time interval according to the timestamp of a to-be-detected ticket and the system preset timestamp screening range, and extracting a ticket whose timestamp falls into the interval from all the to-be-detected tickets to form a temporary to-be-detected ticket set; the similarity determination unit is used for comparing and calculating the writing features of all the tickets in the temporary to-be-detected ticket set by using a preset feature similarity calculation method, and marking and collecting suspected repeatedly identified tickets after comparing the similarity value with a preset threshold.

[0013] Further, the classification comparison processing module comprises a ticket classification and sorting unit and a comparison result processing unit. The ticket classification and sorting unit is used for classifying all the suspected repeatedly identified tickets according to the extracted writing features, dividing the tickets with the same writing features into the same category, and sorting the tickets in the category according to the timestamp to form a to-be-detected sequence; the comparison result processing unit is used for extracting the historical reimbursement sequence of the digital signature corresponding to the to-be-detected sequence, comparing the two sequences, marking the tickets in the category as passed if the two sequences are consistent, and automatically reimbursing the first ticket, transferring the remaining tickets to manual review and recording the results if the two sequences are inconsistent. The result recording and updating module comprises a result collection unit and a record updating unit. The result collection unit is used for collecting the ticket determination results generated in the manual review process, and collecting the results, the digital signature, the standardized text information and the timestamp of the ticket to form a record document; the record updating unit is used for updating the ticket reimbursement history record of the digital signature corresponding to the ticket according to the record document formed by the collection.

[0014] Compared with the prior art, the present application has the following advantages: 1. This invention collects digital signatures and text information of electronic invoices and binds them with timestamps. At the same time, it integrates historical reimbursement sequences of the same digital signature, and after standardized processing, extracts the writing features such as the stroke logic and structural layout of the signature to accurately construct the invoice to be detected, thereby improving the accuracy of invoice duplication identification and providing a reliable data foundation for subsequent screening.

[0015] 2. This invention filters invoices within a preset interval by timestamp, compares the similarity of writing features in pairs to mark suspected duplicate invoices, and then forms a sequence to be detected by classifying writing features and sorting by timestamp. After comparison with historical reimbursement sequences, it performs differentiated processing, which greatly reduces the amount of invalid manual review and solves the problem of low review efficiency caused by repeated submissions of electronic invoices by multiple departments.

[0016] 3. This invention collects the results of manual review, summarizes the judgment information and related data of the invoices, and updates the reimbursement history of the corresponding digital signature, forming a complete data closed loop, which is conducive to the traceability and verification of subsequent invoice circulation. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an intelligent identification method for preventing duplicate receipts according to the present invention. Figure 2 This is a schematic diagram of the structure of an intelligent identification system for preventing duplicate receipts according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0019] Example 1: As Figure 1 As shown, the present invention provides a technical solution, an intelligent identification method for preventing duplicate receipts, the intelligent identification method comprising the following steps: Step S1: Collect the digital signature and text information of the electronic invoice and attach a timestamp and historical reimbursement behavior, wherein the historical reimbursement behavior is the reimbursement sequence of all invoices with the same digital signature; The specific steps of step S1 are as follows: Step S1-1: Collect the digital signature and text information that comes with the electronic ticket, wherein the text information includes all text information on the electronic ticket; Step S1-2: Generate a timestamp based on the real-time submission time of the electronic ticket, and bind and associate the timestamp with the digital signature and text information of the corresponding electronic ticket to form a preliminary data set; Step S1-3, extracting all past ticket reimbursement data corresponding to a certain digital signature, arranging the complete ticket reimbursement sequence in chronological order, and constructing historical reimbursement behavior; Step S2, cleaning and standardizing the collected digital signature and text information, extracting the writing features of the digital signature, and constructing the to-be-detected ticket; The specific steps of step S2 are as follows: Step S2-1, regularizing the digital signature according to the system preset format, and uniformly converting the text information into a preset format; Step S2-2, extracting writing features from the regularized digital signature, the writing features including stroke logic and structural layout of the signature; Step S2-3, associating and integrating the extracted writing features, the converted text information, and the timestamp, and the historical reimbursement behavior to form the to-be-detected ticket; Step S3, extracting all to-be-detected tickets within a preset range before and after the timestamp, collecting all writing features, and marking the to-be-detected tickets with a similarity exceeding a preset threshold as suspected repeated identification tickets; The specific steps of step S3 are as follows: Step S3-1, extracting a to-be-detected ticket, constructing a preset time interval of the to-be-detected ticket according to its timestamp and the system preset timestamp filtering range, extracting all to-be-detected tickets with timestamps falling within the preset time interval from all to-be-detected tickets to form a temporary to-be-detected ticket set; Step S3-2, extracting the digital signature writing features of each to-be-detected ticket from the temporary to-be-detected ticket set one by one, and forming a temporary writing feature set from all extracted writing features; Step S3-3, using a preset feature similarity calculation method to pairwise compare and calculate all writing features in the temporary writing feature set, obtaining the similarity values between each two groups of writing features, comparing the calculated similarity values with the system preset threshold, if the similarity values of the two groups of writing features exceed the system preset threshold, marking the to-be-detected tickets associated with the two groups of features as suspected repeated identification tickets, and summarizing and arranging all information marked as suspected repeated identification tickets to form a suspected repeated ticket list; Step S4, extracting all suspected repeated identification tickets, classifying them according to writing features, sorting the suspected repeated identification tickets according to timestamps to form a to-be-detected sequence, querying whether there is a same ticket reimbursement sequence in the historical reimbursement behavior, if there is, marking it as passed, if there is not, reimbursing the first suspected repeated identification ticket, and converting the subsequent suspected repeated identification tickets to manual review; The specific steps of step S4 are as follows: Step S4-1, classify all suspected duplicate identified bills according to the extracted writing features, and divide the suspected duplicate identified bills with consistent writing features into the same category; Step S4-2, for the suspected duplicate identified bills in each category, sort them in chronological order according to their bound timestamp information, and associate the digital signature, standardized text information of each suspected duplicate identified bill with the detection sequence to form the detection sequence of each category; Step S4-3, extract the bill reimbursement sequence in the historical reimbursement behavior associated with the digital signature corresponding to the detection sequence, compare the detection sequence with the extracted historical reimbursement sequence, if the comparison result is that there is a completely consistent bill reimbursement sequence, mark all suspected duplicate identified bills in this category as passed; if the comparison result is that there is no completely consistent bill reimbursement sequence, the first suspected duplicate identified bill is automatically reimbursed, the rest of the bills are transferred to the manual review process, and the judgment result and the corresponding bill association information are recorded; Step S5, collect the manual review results, and update the corresponding bill reimbursement history record; The specific steps of step S5 are as follows: Step S5-1, collect the judgment results generated by the manual review process, and summarize the judgment results, digital signature, standardized text information and timestamp bound to the bill to form a record document; Step S5-2, update the bill reimbursement history record corresponding to the digital signature according to the record document.

[0020] For example: Collect the electronic bill submitted by a certain business unit, extract the unique digital signature and all text information on the bill, including bill type, transaction object, transaction content, amount related description, bill date and other core information; According to the real-time time of submitting the bill to the organization reimbursement system, generate a special timestamp, and bind the timestamp with the digital signature and text information to form a preliminary data set containing the core elements of the bill. Then, extract all past reimbursement data corresponding to the above digital signature, and arrange them into a complete bill reimbursement sequence according to the order of reimbursement submission, so as to build the historical reimbursement behavior of the subject and provide a general historical reference basis for subsequent duplicate identification.

[0021] The digital signature is regularized according to the format requirements of the organization reimbursement system, and the text information of the bill is converted into the unified coding format preset by the system, eliminating the recognition errors caused by the filling habits and data format differences of each business unit. Then, the general writing features are extracted from the regularized digital signature, including the stroke logic and structure layout of the signature.

[0022] Finally, the extracted writing features, the converted text information, and the previously bound timestamps are associated and integrated with the constructed historical reimbursement behaviors to form the to-be-detected bills that can be used for subsequent repeated detection.

[0023] Taking the timestamp of a to-be-detected bill as a reference, a preset time interval is determined in combination with the system-preset time range of "a certain number of regular business cycles before and after". From all the to-be-detected bills in the organization reimbursement system, all bills with timestamps falling within the preset time interval are selected to form a temporary to-be-detected bill set. This step can exclude irrelevant bills that are not of the same period, reduce invalid comparisons, and improve identification efficiency. Subsequently, the system-preset feature similarity calculation method is used to compare and calculate the writing features of all bills in the temporary to-be-detected bill set in pairs to obtain the similarity values between each two sets of writing features. The calculation results are compared with the system-preset threshold value. If the similarity of the two sets of writing features exceeds the threshold value, the two bills are jointly marked as suspected repeated identification bills. Finally, all information marked as suspected repeated identification bills is summarized to form a suspected repeated bill list.

[0024] According to the extracted writing features, all bills in the suspected repeated bill list are classified, and bills with consistent associated writing features are divided into the same category, for example, the same service consumption bill submitted by the same submitter in two different business units is classified into one category. For suspected repeated bills in each category, the order of bill submission is determined according to the bound timestamp information, and the digital signature, standardized text information, and sorting result of each bill are associated to form a to-be-detected sequence for each category. Subsequently, the bill reimbursement sequence in the historical reimbursement behavior associated with the to-be-detected sequence corresponding digital signature is extracted, and the to-be-detected sequence and the historical reimbursement sequence are compared. If there is a completely consistent reimbursement sequence, all suspected repeated bills in the category are marked as "passed". If there is no completely consistent reimbursement sequence, the first suspected repeated bill in the order is executed with the automatic reimbursement process, and the remaining bills are transferred to the manual audit process, and the determination results and associated information of the bills are recorded in detail.

[0025] The determination results generated by the manual audit process are collected. If the bill confirmed to be "compliant report required for cross-business unit cooperation" is partially transferred to the audit, it is determined to be "passed". These determination results, the corresponding bill-bound digital signature, standardized text information, and timestamp are summarized to form a complete record document, ensuring that the determination process of each bill is traceable and meets the organization reimbursement compliance management requirements. Finally, according to the information in the record document, the historical reimbursement record of the digital signature corresponding to the bill is updated, and the determination results of this reimbursement and the business scenario information associated with the bill are supplemented to the historical sequence to provide more business-actual historical data support for the subsequent repeated identification of similar cross-business unit bills.

[0026] Example 2, as Figure 2 As shown, the present invention provides an intelligent identification system for preventing duplicate receipts. The intelligent identification system includes a data acquisition and construction module, a data feature extraction module, a similarity judgment module, a classification and comparison processing module, and a result record update module. The data acquisition and construction module is used to collect the digital signatures and text information of electronic invoices and generate timestamps, while constructing a historical reimbursement sequence for all invoices corresponding to the same digital signature. The data feature extraction module is used to clean and standardize the collected digital signatures and text information, extract the writing features of the digital signatures, and associate them with timestamps and historical reimbursement behaviors to construct invoices to be detected. The similarity judgment module is used to filter invoices within a preset time interval based on the timestamps of the invoices to be detected, and to mark suspected duplicate invoices by calculating the similarity of writing features. The classification and comparison processing module is used to classify suspected duplicate invoices according to writing features, sort them according to timestamps to form a sequence to be detected, and perform corresponding processing after comparing them with historical reimbursement sequences. The result recording and updating module is used to collect the judgment results generated by the manual review process, summarize the results with the invoice information, and update the historical reimbursement records of the corresponding digital signatures. The output of the data acquisition and construction module is electrically connected to the input of the data feature extraction module; the output of the data feature extraction module is electrically connected to the input of the screening similarity determination module; the output of the screening similarity determination module is electrically connected to the input of the classification comparison processing module; and the output of the classification comparison processing module is electrically connected to the input of the result recording and updating module.

[0027] The data acquisition and construction module includes an information acquisition unit and a historical behavior construction unit; The information collection unit is used to collect the digital signature of the electronic invoice and the text information consisting of all the text on the invoice, and generate a timestamp associated with the invoice based on the real-time submission time of the invoice; the historical behavior construction unit is used to extract all past invoice reimbursement data corresponding to a specific digital signature, and organize them into a complete invoice reimbursement sequence according to the reimbursement time.

[0028] The data feature extraction module includes a data regularization unit and a feature extraction unit; The data normalization unit is used to normalize the collected digital signatures according to the system's preset format, and at the same time convert the text information into the system's preset format; the feature extraction unit is used to extract the stroke logic and structural layout writing features from the normalized digital signatures, and integrate them with the converted text information, timestamps and historical reimbursement behavior to form the invoice to be tested. The similarity determination module includes a time-based filtering unit and a similarity determination unit; The time screening unit is configured to construct a preset time interval according to a time stamp of a certain to-be-detected bill and a system preset time stamp screening range, and then extract a bill group with time stamps falling into the interval from all to-be-detected bills to form a temporary to-be-detected bill set; the similarity determination unit is configured to adopt a preset feature similarity calculation method to compare and calculate the writing features of all bills in the temporary to-be-detected bill set two by two, and mark suspected repeatedly recognized bills and collect them after comparing the similarity values with a preset threshold.

[0029] The classification comparison processing module comprises a bill classification sorting unit and a comparison result processing unit. The bill classification sorting unit is configured to classify all suspected repeatedly recognized bills according to the extracted writing features, divide bills with consistent writing features into the same category, sort the bills in the category according to the time stamps, and form a to-be-detected sequence by associating information; the comparison result processing unit is configured to extract a history reimbursement sequence of a digital signature corresponding to the to-be-detected sequence, compare the two sequences, mark the bills in the category as passing if the two sequences are consistent, automatically reimburse the first bill if the two sequences are inconsistent, transfer the remaining bills to manual audit, and record the results. The result recording updating module comprises a result collection unit and a record updating unit. The result collection unit is configured to collect bill determination results generated in the manual audit process, collect the results, a digital signature bound to the bill, standardized text information, and a time stamp to form a record document; and the record updating unit is configured to update a bill reimbursement history record of a digital signature corresponding to the bill according to the record document formed by the collection.

[0030] It is apparent for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present application is defined by the appended claims rather than the above description, and all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A method for intelligent identification of ticket duplication prevention, characterized in that: The intelligent identification method comprises the following steps: Step S1, collecting the digital signature and text information of the electronic bill and attaching a time stamp and historical reimbursement behavior, the historical reimbursement behavior being all bill reimbursement sequences of the same digital signature; Step S2, cleaning and standardizing the collected digital signature and text information, extracting the writing features of the digital signature, and constructing the to-be-detected bill; Step S3, extracting all to-be-detected bills within a preset range before and after the time stamp, collecting all writing features, and marking the to-be-detected bills with a similarity exceeding a preset threshold as suspected repeated identification bills; Step S4, extracting all suspected repeated identification bills, classifying them according to the writing features, sorting the suspected repeated identification bills according to the time stamp to form a to-be-detected sequence, and querying whether there is a bill reimbursement sequence identical to the to-be-detected sequence in the historical reimbursement behavior, if there is, marking it as passed, if not, reimbursing the first suspected repeated identification bill and converting the subsequent suspected repeated identification bills into manual review; Step S5, collecting the manual review results and updating the corresponding bill reimbursement history record.

2. The intelligent identification method for ticket anti-repetition according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1, collecting the digital signature and text information of the electronic bill, the text information including all the text information on the electronic bill; Step S1-2, generating a time stamp according to the real-time time of the submission of the electronic bill, binding and associating the time stamp with the digital signature and text information of the corresponding electronic bill to form a preliminary data set; Step S1-3, extracting all past bill reimbursement data corresponding to a certain digital signature, arranging them in chronological order to form a complete bill reimbursement sequence, and constructing a historical reimbursement behavior.

3. The method for intelligent identification of ticket duplication prevention according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1, regularizing the digital signature according to the system preset format, and uniformly converting the text information into a preset format; Step S2-2, extracting writing features from the regularized digital signature, the writing features including the stroke logic and structural layout of the signature; Step S2-3, integrating the extracted writing features, the converted text information, the time stamp, and the historical reimbursement behavior to form a to-be-detected bill.

4. The intelligent identification method for ticket anti-repetition according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1, extracting a to-be-detected bill, constructing a preset time interval of the to-be-detected bill according to its time stamp and the system preset time stamp filtering range, extracting all to-be-detected bills with time stamps falling within the preset time interval from all to-be-detected bills to form a temporary to-be-detected bill set; Step S3-2, extracting the digital signature writing features of each to-be-detected bill from the temporary to-be-detected bill set one by one, and forming a temporary writing feature set from all extracted writing features; Step S3-3, using a preset feature similarity calculation method, all writing features in the temporary writing feature set are compared and calculated, the similarity value between each two groups of writing features is obtained, the calculated similarity value is compared with the system preset threshold, if the similarity value of the two groups of writing features exceeds the system preset threshold, the two groups of features associated with the to-be-detected bill are jointly marked as suspected repeated identification bills, all information marked as suspected repeated identification bills is summarized and arranged to form a suspected repeated bill list.

5. The intelligent identification method for preventing duplicate receipts according to claim 4, characterized in that: The specific steps of the step S4 are as follows: Step S4-1, according to the extracted writing features, all suspected repeated identification bills are classified, and the suspected repeated identification bills with consistent associated writing features are divided into the same category; Step S4-2, for the suspected repeated identification bills in each category, the time stamp information bound thereto is sorted in order, and the digital signature, the standardized text information and the to-be-detected sequence of each suspected repeated identification bill are associated and bound to form a to-be-detected sequence of each category; Step S4-3, the digital signature corresponding to the to-be-detected sequence is extracted, and the to-be-detected sequence is compared with the extracted historical reimbursement sequence, if the comparison result is that there is a completely consistent bill reimbursement sequence, all suspected repeated identification bills in the category are marked as passed; if the comparison result is that there is no completely consistent bill reimbursement sequence, the first suspected repeated identification bill in the order is automatically reimbursed, and the remaining bills are transferred to the manual review process, and the determination result and the corresponding bill associated information are recorded.

6. The intelligent identification method for ticket anti-repetition according to claim 5, characterized in that: The specific steps of the step S5 are as follows: Step S5-1, collecting the determination results generated by the manual review process, and collecting the determination results, the digital signature bound to the bill, the standardized text information and the time stamp to form a record document; Step S5-2, according to the record document, the bill reimbursement history record corresponding to the digital signature is updated.

7. A smart identification system for ticket anti-duplication, which is applied to the smart identification method for ticket anti-duplication in any one of claims 1-6, characterized in that: The intelligent identification system comprises a data acquisition and construction module, a data feature extraction module, a screening similarity determination module, a classification comparison processing module and a result record updating module; The data collection and construction module is used for collecting the digital signature and text information of the electronic ticket and generating a time stamp, and constructing a history reimbursement behavior in the form of a sequence of all tickets corresponding to the same digital signature; the data feature extraction module is used for cleaning and standardizing the collected digital signature and text information, extracting the writing features of the digital signature, and associating the time stamp and the history reimbursement behavior to construct a to-be-detected ticket; the screening and similarity determination module is used for screening the tickets within a preset time interval according to the time stamp of the to-be-detected ticket, and marking suspected repeatedly identified tickets by calculating the similarity of the writing features; the classification and comparison processing module is used for classifying the suspected repeatedly identified tickets according to the writing features, sequencing the suspected repeatedly identified tickets according to the time stamp, and performing corresponding processing after comparing with the history reimbursement sequence; and the result recording and updating module is used for collecting the determination results generated in the manual review process, integrating the results and the ticket information, and updating the history reimbursement record corresponding to the digital signature. The output end of the data collection and construction module is electrically connected to the input end of the data feature extraction module; the output end of the data feature extraction module is electrically connected to the input end of the screening and similarity determination module; the output end of the screening and similarity determination module is electrically connected to the input end of the classification and comparison processing module; and the output end of the classification and comparison processing module is electrically connected to the input end of the result recording and updating module.

8. The intelligent identification system for ticket anti-repetition according to claim 7, characterized in that: The data collection and construction module comprises an information collection unit and a history behavior construction unit; The information collection unit is used for collecting the digital signature and text information composed of all the characters on the ticket, and generating a time stamp associated with the ticket according to the real-time time of the ticket submission; and the history behavior construction unit is used for extracting all the past ticket reimbursement data corresponding to a specific digital signature, and arranging the complete ticket reimbursement sequence according to the reimbursement time sequence.

9. The intelligent identification system for ticket anti-repetition according to claim 7, characterized in that: The data feature extraction module comprises a data normalization unit and a feature extraction unit; The data normalization unit is used for normalizing the collected digital signature according to the system preset format, and converting the text information into the system preset format; and the feature extraction unit is used for extracting the stroke logic and structural layout writing features from the normalized digital signature, and integrating the extracted features with the converted text information, the time stamp and the history reimbursement behavior to form a to-be-detected ticket; The screening and similarity determination module comprises a time screening unit and a similarity determination unit; The time screening unit is used for constructing a preset time interval according to the time stamp of a to-be-detected ticket and the system preset time stamp screening range, and extracting the tickets with the time stamp falling into the interval from all the to-be-detected tickets to form a temporary to-be-detected ticket set; and the similarity determination unit is used for adopting a preset feature similarity calculation method to compare and calculate the writing features of all the tickets in the temporary to-be-detected ticket set, marking suspected repeatedly identified tickets after comparing the similarity value with a preset threshold, and integrating the suspected repeatedly identified tickets.

10. The intelligent identification system for ticket duplication prevention according to claim 7, characterized in that: the classification comparison processing module comprises a ticket classification sorting unit and a comparison result processing unit; the ticket classification sorting unit is used for classifying all suspected duplicate identification tickets according to the extracted writing features, dividing the tickets with consistent writing features into the same category, sorting the tickets in the category according to the time stamp, and associating information to form a detection sequence; the comparison result processing unit is used for extracting the historical reimbursement sequence of the digital signature corresponding to the detection sequence, comparing the two sequences, marking the tickets in the category as passed if the sequences are consistent, and automatically reimbursing the first ticket, transferring the remaining tickets to manual review, and recording the results if the sequences are inconsistent; the result record updating module comprises a result collection unit and a record updating unit; the result collection unit is used for collecting the ticket determination results generated by the manual review process, and summarizing the results, the digital signature bound to the ticket, the standardized text information, and the time stamp to form a record document; the record updating unit is used for updating the ticket reimbursement history record of the digital signature corresponding to the ticket according to the record document formed by the summary.

Images