Bill information integration method, system and terminal

Abstract

The invention relates to a draft information integration method, system and terminal, relates to the field of draft transaction, and is used for correcting the price of a draft according to the historical transaction condition of the draft so as to obtain a draft transaction price trend conforming to the market and improve the convenience of the draft transaction. The application has the effects of improving the convenience of the draft transaction and obtaining the market information of the draft in time.

Description

Technical Field

[0001] The present invention relates to the field of bill of exchange transactions, and in particular to a bill of exchange information integration method, system and terminal. Background Art

[0002] A bill of exchange is a written instruction issued by the drawee to the payee requiring the payee to pay a certain price unconditionally at a specific time to the payee or holder.

[0003] In the prior art, transactions of bills of exchange generally include the primary market and the secondary market. The primary market refers to the market formed when new securities are first issued to the public. In this market, issuers sell securities to raise funds and investors purchase these newly issued securities. The secondary market refers to the place where issued securities are bought, sold and circulated. In the secondary market, investors buy and sell issued securities from each other without directly affecting the issuers of the securities.

[0004] When trading in the secondary market, the market's trading information relies on manual statistics, which can easily lead to investors obtaining trading information with a lag, resulting in the trading price of the bill not being in line with the market conditions, which in turn causes damage to the interests of investors. Summary of the invention

[0005] In order to improve the convenience of bill transactions and obtain the market situation of bills in a timely manner, the present invention provides a bill information integration method, system and terminal.

[0006] In a first aspect, the present invention provides a method for integrating bill information, which adopts the following technical solution: A method for integrating bill information, comprising: Get transaction information of bill of exchange; Determine the transaction quotation based on the bill image, and retrieve the previous fitting price from the preset fitting price database; Calculate the difference between the transaction quote and the previous fit price and define it as the fit price difference; When the absolute value of the fitted price difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0; When the fitting price difference is greater than 0, the sum of the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment; When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment; When the absolute value of the fitting price difference is not greater than the preset secondary market volatility coefficient, the transaction quote is defined as the fitting price at this moment; Based on the existence of the fitting price at this moment, the fitting price at this moment is entered into a preset fitting price database, and the data in the preset fitting price database is retrieved to form market trading trend parameters.

[0007] By adopting the above technical solution, the existing bills of exchange are fitted and corrected according to the historical records of the bills of exchange, so that when the bill price is too high, the transaction quotation is judged to be falsely high, and the corrected fitting price is obtained, and the obtained fitting price is merged into the historical records of the bills of exchange to obtain the transaction price trend data of the bills of exchange, and then the market fluctuations of the bills of exchange are obtained, thereby improving the convenience of bill transactions.

[0008] Optionally, it also includes an accepting bank correction method, the accepting bank correction method comprising: Determine the accepting bank based on transaction information; Determine the accepting bank parameters based on the accepting bank; Determine the accepting bank correction value based on the transaction quotation and accepting bank parameters; Calculate the difference between the accepting bank's revised value and the previous fitting price and define it as the fitting price difference.

[0009] By adopting the above technical solution, the bill price is corrected according to the credit of the accepting unit of the bill, thereby increasing the price of the bill of exchange of the accepting unit with higher credit, and reducing the price of the bill of exchange of the accepting unit with lower credit, so as to obtain a fitting price that conforms to the market conditions, and then obtain the market fluctuations of the bill of exchange.

[0010] Optionally, a deviation correction method is also included, and the deviation correction method includes: Retrieve historical fitting price data from the preset fitting price database; Determine the average increase value based on the historical data of the fitted price; Calculate the difference between the absolute value of the fitted price difference and the average increase value and define it as the increase difference; When the absolute value of the increase difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0; When the fitted price difference is greater than 0, the sum of the previous fitted price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value; When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value; When the absolute value of the increase difference is not greater than the preset secondary market volatility coefficient, the accepting bank's correction value is defined as the deviation correction value; Calculate the difference between the deviation correction value and the previous fitting price and define it as the fitting price difference.

[0011] By adopting the above technical solution, the price fluctuation of the current bill of exchange is predicted based on the historical fluctuation data of the bill of exchange price, so that when the current bill of exchange price fluctuates too much, it is judged that the bill of exchange price does not conform to the market, and the bill of exchange price is corrected according to the historical record of the bill of exchange price, thereby obtaining the market fluctuation of the bill of exchange.

[0012] Optionally, a fitting correction method is also included, and the fitting correction method includes: Determine the buying and selling data based on the historical data of the fitting price; Determine the lowest purchase price based on the purchase data; Determine the highest selling price based on the selling data; When the highest selling price is not 0, the highest selling price is defined as the fitting correction value; When the highest selling price is 0, the average of the lowest buying price and the deviation correction value is calculated and defined as the fitting correction value; Calculate the difference between the fitted correction value and the previous fitted price and define it as the fitted price difference.

[0013] By adopting the above technical solution, the price is modified according to the situation of bill buying and selling, so that when there is a transaction record of selling bills, the maximum price in the transaction record is selected, and when there is no transaction record of selling bills, the lowest price in the transaction record of purchasing bills is selected, thereby increasing the interests of bill traders.

[0014] Optionally, a transaction information acquisition method is also included, and the transaction information acquisition method includes: Get the money order image; Determine whether the bill of exchange is torn based on the bill of exchange image; When there is a tear in the bill, the tear area is determined based on the bill image; Judging whether there is text in the torn area according to the torn area; When text exists in the tearing area, a tearing contour and a tearing extreme point are determined based on the tearing area; Determine a first tear boundary and a second tear boundary according to the tear contour and the tear extreme point; Extracting tear feature points according to the first tear boundary; Matching restoration feature points from the second tear boundary according to the tear feature points; Determine a restored image according to the tearing feature points, the restoration feature points and the tearing area; Replaces the torn area by restoring the image.

[0015] By adopting the above technical solution, corresponding feature points are selected from the contours on both sides of the crack, so that the torn area is overlapped according to the corresponding feature points to repair the crack, and then the image of the untorn bill of exchange is obtained, thereby improving the accuracy of bill of exchange recognition.

[0016] Optionally, the transaction information acquisition method further includes: When there are words in the torn area, judging whether the words are split based on the torn area; When there is a situation where the text is split, determining the first torn text feature point according to the torn area and the first torn boundary; Determine a second torn text feature point according to the torn area and the second torn boundary; Determining whether the torn text feature points correspond to each other one by one according to the first torn text feature points, the second torn text feature points and the restored image; When the torn text feature points do not correspond, a preset display device is controlled to issue a text missing warning.

[0017] By adopting the above technical solution, the torn text on both sides of the tear is identified, and it is checked in the repaired image whether the torn text is merged, so that when there is a situation where the text is not merged, the staff is notified of the missing characters.

[0018] Optionally, the transaction information acquisition method further includes: When the torn text feature points do not correspond, determining the missing text feature points according to the first torn text feature points, the second torn text feature points and the restored image; Extract missing characters based on missing text feature points and bill of exchange image; Determine whether fuzzy recognition can be performed based on the missing characters; When fuzzy recognition is possible, matching characters are determined based on missing characters; replacing the missing characters in the restored image according to the matching characters and the missing text feature points to form a new restored image; When fuzzy recognition cannot be performed, a preset display device is controlled to issue a text missing warning.

[0019] By adopting the above technical solution, when there are missing characters, an attempt is made to determine the correct characters through fuzzy recognition, thereby repairing the image of the bill of exchange, improving the accuracy of bill of exchange recognition, and reducing the situation where bill of exchange recognition errors are caused by missing characters.

[0020] Optionally, the transaction information acquisition method further includes: When fuzzy recognition is not possible, the missing fields are extracted from the restored image based on the missing text feature points; Determine whether there are synonymous fields in the bill of exchange image based on the missing fields; When there are synonymous fields in the bill of exchange image, the synonymous fields are determined based on the missing fields and the bill of exchange image; Determine synonym characters based on synonym fields and missing fields; The missing characters in the restored image are replaced according to the synonymous characters and the missing text feature points to form a new restored image.

[0021] By adopting the above technical solution, when the characters cannot be determined by fuzzy recognition, the field where the missing characters are located is extracted, and a synonym search is performed on the bill image to determine the missing characters, thereby repairing the torn bill image.

[0022] In the second aspect, the present application provides a method for integrating bill information, which adopts the following technical solution: A bill of exchange information integration system, comprising: An acquisition module, used to acquire transaction information and bill images; A memory for storing a program for any of the above-mentioned methods for integrating bill of exchange information; The program in the processor and the memory can be loaded and executed by the processor and implement any of the above-mentioned bill information integration methods.

[0023] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution: An intelligent terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any of the above-mentioned bill information integration methods.

[0024] By adopting the above technical solution, the existing bills of exchange are fitted and corrected according to the historical records of the bills of exchange, so that when the bill price is too high, the transaction quotation is judged to be falsely high, and the corrected fitting price is obtained, and the obtained fitting price is merged into the historical records of the bills of exchange to obtain the transaction price trend data of the bills of exchange, and then the market fluctuations of the bills of exchange are obtained, thereby improving the convenience of bill transactions.

[0025] In summary, the present application includes at least one of the following beneficial technical effects: 1. According to the historical records of the bills, the existing bills are fitted and corrected, so that when the bill price is too high, the transaction quotation is judged to be falsely high, and then the corrected fitting price is obtained, and the obtained fitting price is merged into the historical records of the bills to obtain the transaction price trend data of the bills, and then the market fluctuation of the bills is obtained, which improves the convenience of bill transactions; 2. The bill price is adjusted according to the credit of the acceptor of the bill, so as to increase the bill price of the acceptor with higher credit and reduce the bill price of the acceptor with lower credit, so as to obtain the fitting price that conforms to the market, and then obtain the market fluctuation of the bill; 3. Predict the current bill price fluctuations based on the historical fluctuation data of the bill price, so as to determine that the bill price does not conform to the market when the current bill price fluctuates too much, and revise the bill price based on the historical records of the bill price, and then obtain the market fluctuations of the bill. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1It is a flow chart of a method for integrating bill of exchange information; Figure 2 It is a flow chart of the accepting bank's correction method; Figure 3 is a flow chart of the deviation correction method; Figure 4 is a flow chart of the fitting correction method; Figure 5 This is the process of obtaining transaction information Figure 1 ; Figure 6 This is the process of obtaining transaction information Figure 2 ; Figure 7 This is the process of obtaining transaction information Figure 3 ; Figure 8 This is the process of obtaining transaction information Figure 4 . DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0028] The embodiments of the present application disclose a method, system and terminal for integrating bill information. The present invention is used to correct the price of a bill according to the historical transaction situation of the bill, so as to obtain a bill transaction price trend that conforms to the market conditions and improve the convenience of bill transactions.

[0029] Reference Figure 1 , a method for integrating bill information, comprising: Step 100: Obtain transaction information of the bill of exchange.

[0030] Transaction information refers to a collection of information such as the price, transaction parties, and accepting bank of a bill of exchange transaction. The bill of exchange information can be obtained from a bill of exchange transaction website. The bill of exchange information can also be obtained by first obtaining a bill of exchange image and then identifying it from the bill of exchange image. The method for obtaining transaction information is selected by the staff based on actual conditions and will not be elaborated here.

[0031] Step 101: Determine a transaction quotation based on transaction information, and retrieve the previous fitting price from a preset fitting price database.

[0032] A transaction quotation refers to the amount of currency paid or settled recorded in the transaction information. The transaction quotation can be identified from the transaction information through a transaction quotation recognition model. The transaction quotation recognition model refers to a neural network model obtained in advance through sample training.

[0033] The fitting price database refers to a database that records the historical transaction records of bills of exchange. The previous fitting price is the transaction quotation of the latest bill of exchange transaction record retrieved from the fitting price database.

[0034] Step 102: Calculate the difference between the transaction quotation and the previous fitting price, and define it as the fitting price difference.

[0035] The fitted spread refers to the difference between the current bill transaction quotation and the fitted spread of the previous moment. When the currency units of the current bill quotation and the fitted spread of the previous moment are different, the currency unit of the current bill quotation should be converted into the currency unit of the fitted spread of the previous moment according to the real-time exchange rate before calculation.

[0036] Step 103: When the absolute value of the fitted price difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0.

[0037] The secondary market volatility coefficient refers to an indicator used to measure the degree of market price volatility, usually used to assess the market's risk level. The secondary market volatility coefficient is selected by the staff based on actual conditions and will not be elaborated here.

[0038] If the absolute value of the fitted price difference is greater than the secondary market volatility coefficient, it means that the current bill quotation does not conform to the market's rise and fall. The transaction quotation is either too high or too low, and the bill quotation needs to be corrected.

[0039] Step 104: When the fitting price difference is greater than 0, the sum of the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment.

[0040] The fluctuation unit value refers to the unit value of the transaction quotation fluctuation. The fluctuation unit value is used to correct the inflated or inflated transaction quotation to match the market's rise and fall. The fluctuation unit value is selected by the staff according to the actual situation and will not be elaborated here. If the fitting price difference is greater than 0, it means that the transaction quotation is on an upward trend. At this time, the sum of the previous fitting price and the fluctuation unit value is used as the fitting price at this moment.

[0041] Step 105: When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment.

[0042] If the fitting price difference is not greater than 0, it means that the transaction quotation shows a downward trend. At this time, the difference between the previous fitting price and the fluctuation unit value is used as the current fitting price.

[0043] Step 106: When the absolute value of the fitted price difference is not greater than the preset secondary market volatility coefficient, the transaction quotation is defined as the fitting price at this moment.

[0044] The absolute value of the fitted price difference is not greater than the secondary market volatility coefficient, which means that the current bill quotation is in line with the market's ups and downs, and the transaction quotation can be used as the fitting price at this moment.

[0045] The general price of bill transactions in the market is obtained through the historical transaction records of bills of exchange, so that the transaction quotation can be corrected when the transaction quotation deviates too much from the general price of bill transactions in the market, thereby making the bill price consistent with market conditions.

[0046] Step 107: Based on the existence of the current fitting price, the current fitting price is entered into a preset fitting price database, and data in the preset fitting price database is retrieved to form market transaction trend parameters.

[0047] The historical records of the bill of exchange are retrieved from the fitted price database to obtain the quotation fluctuations of the bill of exchange, and then the trend of the market conditions. The trading trend parameters are the market trends obtained from the quotation fluctuations. The trading trend parameters are generally displayed in the form of a line graph of the quotation fluctuations. The display method of the trading trend parameters is selected by the staff according to the actual situation and will not be elaborated here.

[0048] Reference Figure 2 , the accepting bank's correction method includes the following steps: Step 200: Determine the accepting bank based on the transaction information.

[0049] The accepting bank refers to the bank designated to accept the forward bill of exchange issued by the beneficiary in a forward letter of credit. The accepting bank is usually the issuing bank or other banks designated by the issuing bank. The accepting banks generally include state-owned banks, the Big Five, F4, Dashang and Danguo. The accepting bank can be identified from the transaction information through the accepting bank identification model. The accepting bank identification model refers to a neural network model obtained through sample training in advance.

[0050] Step 201: Determine the accepting bank parameters according to the accepting bank.

[0051] Accepting bank parameters refer to parameters used to modify transaction quotations based on the credit of the accepting bank. Accepting bank parameters can be obtained by querying a preset accepting bank parameter database. The accepting bank parameter database refers to a database that records the accepting banks and their corresponding accepting bank parameters.

[0052] Step 202: Determine the accepting bank correction value according to the transaction quotation and the accepting bank parameters.

[0053] The accepting bank's correction value refers to the amount of currency after the transaction quotation is corrected according to the accepting bank's correction value. Generally, the sum of the transaction quotation and the accepting bank's parameters is calculated and defined as the accepting bank's correction value. The calculation method of the accepting bank's correction value is selected by the staff based on actual conditions and will not be elaborated here.

[0054] Step 203: Calculate the difference between the accepting bank's revised value and the previous fitting price, and define it as the fitting price difference.

[0055] The transaction quotation is revised according to the credit of the accepting bank, thereby increasing the transaction quotation of the bill of exchange of the accepting bank with higher credit, and reducing the transaction quotation of the bill of exchange of the accepting bank with lower credit, so as to make the transaction quotation conform to the market conditions, thereby obtaining the market transaction trend parameters according to the revised value of the accepting bank, and improving the accuracy of the market transaction trend parameters.

[0056] Reference Figure 3 , the deviation correction method includes the following steps: Step 300: Retrieve fitting price historical data from a preset fitting price database.

[0057] The historical data of the fitted price refers to the most recent part of the transaction quotation data in the fitted price database. Generally, the six latest transaction quotation data in the fitted price database are retrieved. The number of historical data of the fitted price is selected by the staff according to the actual situation and will not be elaborated here.

[0058] Step 301: Determine the average increase value based on the historical data of the fitted price.

[0059] The average increase value refers to the average value of the rise and fall in the historical data of the fitted price. For example, when the historical data of the fitted price includes When , the average increase value is .

[0060] Step 302: Calculate the difference between the absolute value of the fitted price difference and the average increase value, and define it as the increase difference.

[0061] The increase difference refers to the difference between the increase or decrease value of the current bill of exchange transaction quotation compared with the previous transaction quotation and the average increase value. The increase difference can be used to reflect whether the increase or decrease range of the current bill of exchange transaction quotation is too large.

[0062] Step 303: When the absolute value of the increase difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0.

[0063] If the absolute value of the increase difference is greater than the secondary market volatility coefficient, it means that the current transaction quotation of the bill of exchange has risen or fallen too much. At this time, the transaction quotation is too high or too low, and the transaction quotation is inconsistent with the market conditions, and further corrections are needed to the transaction quotation.

[0064] Step 304: When the fitted price difference is greater than 0, the sum of the previous fitted price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value.

[0065] If the fitted price difference is greater than 0, it means that the transaction quotation shows an upward trend. At this time, the sum of the previous fitted price and the secondary market volatility coefficient is used as the deviation correction value.

[0066] Step 305: When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value.

[0067] If the fitted price difference is not greater than 0, it means that the transaction quotation shows a downward trend. At this time, the difference between the previous fitted price and the secondary market volatility coefficient is used as the deviation correction value.

[0068] Step 306: When the absolute value of the increase difference is not greater than the preset secondary market volatility coefficient, the accepting bank's correction value is defined as the deviation correction value.

[0069] If the absolute value of the increase difference is not greater than the secondary market volatility coefficient, it means that the increase or decrease of the current transaction quotation of the bill of exchange is in line with the market conditions, and the correction value of the accepting bank can be used as the deviation correction value.

[0070] Step 307: Calculate the difference between the deviation correction value and the previous fitting price, and define it as the fitting price difference.

[0071] When the transaction quotation fluctuates too much, the transaction quotation is corrected through the secondary market volatility coefficient so that the transaction quotation conforms to the market's ups and downs, thereby improving the accuracy of the market transaction trend parameters.

[0072] Reference Figure 4 , the fitting correction method includes the following steps: Step 400: Determine the buying data and selling data according to the fitting price historical data.

[0073] Purchase data refers to the transaction quotes put forward by investors for purchasing bills of exchange, and selling data refers to the transaction quotes put forward by investors for selling bills of exchange. Purchase data and selling data are distinguished according to the transaction type of the bill of exchange, and the transaction type of the bill of exchange can be obtained by querying the matching price database.

[0074] Step 401: Determine the lowest purchase price based on the purchase data.

[0075] The lowest purchase price is the lowest transaction price in the purchase data.

[0076] Step 402: Determine the highest selling price according to the selling data.

[0077] The highest selling price is the price with the highest transaction quotation in the selling data.

[0078] Step 403: When the highest selling price is not 0, the highest selling price is defined as the fitting correction value.

[0079] If the highest selling price is not 0, it means that there is selling data in the historical data of the fitting price. At this time, the highest selling price will be used as the fitting correction value to improve the interests of investors.

[0080] Step 404: When the highest selling price is 0, the average of the lowest buying price and the deviation correction value is calculated and defined as the fitting correction value.

[0081] The highest selling price of 0 means that there is no selling data in the historical data of the fitting price. At this time, the average of the lowest buying price and the deviation correction value is used as the fitting correction value to correct the transaction quotation according to the lowest buying price, thereby improving the interests of investors.

[0082] Step 405: Calculate the difference between the fitting correction value and the previous fitting price, and define it as the fitting price difference.

[0083] The transaction quotation is revised according to the transaction situation of the bill of exchange. When an investor sells the bill of exchange, the highest selling price of the bill of exchange sold is used as the fitting correction value to increase the seller's profit. When no investor sells the bill of exchange, the transaction quotation is revised according to the lowest purchase price to make the transaction quotation conform to the market.

[0084] Reference Figure 5 , the transaction information acquisition method includes the following steps: Step 500: Acquire a money order image.

[0085] The bill of exchange image refers to a picture of the bill of exchange, which can be obtained through a camera or a scanner. The method for obtaining the bill of exchange image is selected by the staff based on actual conditions and will not be elaborated here.

[0086] Step 501: Determine whether the bill is torn based on the bill image.

[0087] The tearing recognition model can be used to determine whether there is a tearing situation. The tearing recognition model refers to a neural network model obtained through sample training in advance.

[0088] Step 502: When the bill of exchange is torn, the torn area is determined based on the bill of exchange image.

[0089] The tear area refers to the area on the bill of exchange where there is a tear. The tear area can be obtained through a tear area matching model. The tear area matching model refers to a neural network model obtained in advance through sample training.

[0090] Step 503: Determine whether there is text in the torn area according to the torn area.

[0091] The text recognition model can be used to determine whether there is text in the torn area. The text recognition model refers to a neural network model obtained through sample training in advance.

[0092] Step 504: When text exists in the tear area, determine the tear contour and tear extreme points based on the tear area.

[0093] The presence of text in the torn area means that the text on the bill is easily affected by the torn cracks, causing the position of the text to be distorted, thereby causing errors in the recognition result of the bill. The tear contour is the outline of the crack, and the tear extreme point is the point where the crack extends deepest into the bill. The tear contour and tear extreme point can be obtained through the crack recognition model, which refers to a neural network model obtained through sample training in advance.

[0094] Step 505: Determine the first tear boundary and the second tear boundary according to the tear contour and the tear extreme points.

[0095] The first tear boundary and the second tear boundary are two parts of the tear contour that are separated by the tear pole. When there are n tear poles, n+1 tear boundaries can be split. This embodiment takes one tear pole as an example.

[0096] Step 506: extracting tear feature points according to the first tear boundary.

[0097] The tear feature points refer to feature points extracted from the first tear boundary. The tear feature points can be obtained through a feature point extraction model. The feature point extraction model refers to a neural network model obtained in advance through sample training.

[0098] Step 507: Matching restoration feature points from the second tear boundary according to the tear feature points.

[0099] The restored feature points refer to the feature points on the second tear boundary corresponding to the tear feature points. The restored feature points can be obtained through a feature point recognition model. The feature point recognition model refers to a neural network model obtained in advance through sample training.

[0100] Step 508: Determine the restored image according to the tearing feature points, the restoration feature points and the tearing area.

[0101] The restored image refers to the image when the torn area is not torn. The restored image can be obtained by dividing the torn area into two parts according to the torn boundary, and then overlapping the torn feature points and the corresponding restored feature points respectively located in the two parts. The restored image can be obtained by a restored image matching model. The restored image matching model refers to a neural network model obtained in advance through sample training.

[0102] Step 509: Replace the torn area by restoring the image.

[0103] The torn part of the bill of exchange is filled with the restored image, thereby obtaining the image of the bill of exchange when it is not torn.

[0104] Reference Figure 6 , the transaction information acquisition method further includes the following steps: Step 510: When there are texts in the torn area, determine whether the texts are split based on the torn area.

[0105] When text is split, cracks pass through the text. A text tearing recognition model can be used to determine whether the text is split. The text tearing recognition model refers to a neural network model obtained through sample training in advance.

[0106] Step 511: When the text is split, determine the first torn text feature point according to the torn area and the first torn boundary.

[0107] The first torn text feature point refers to the position point of the split text on the first torn boundary. The first torn text feature point can be obtained through a text tearing matching model. The text tearing matching model refers to a neural network model obtained in advance through sample training.

[0108] Step 512: Determine a second torn text feature point according to the torn area and the second torn boundary.

[0109] The second torn text feature point refers to the position point of the split text on the second torn boundary, and the second torn text feature point can be obtained through a text torn matching model.

[0110] Step 513: determining whether the torn text feature points correspond one to one according to the first torn text feature points, the second torn text feature points and the restored image.

[0111] The first torn text feature point and the second torn text feature point can be used to find the corresponding position points in the restored image, so as to determine whether the position points on the restored image are symmetrically distributed on both sides of the crack. The feature point correspondence model can be used to determine whether the torn text feature points correspond one to one. The feature point correspondence model refers to a neural network model obtained through sample training in advance.

[0112] Step 514: When the torn text feature points do not correspond, control a preset display device to issue a text missing warning.

[0113] The display device refers to a device used to display information to the staff. The market transaction trend parameters can be displayed through the display device. The display device is selected by the staff according to the actual situation and will not be described here. The non-correspondence of the torn text feature points indicates that the torn part of the text is missing. At this time, the display device is used to remind the staff that the text is missing so that the staff can repair the bill in time. The text missing warning is a prompt issued by the display device.

[0114] Reference Figure 7 , the transaction information acquisition method further includes the following steps: Step 515: When the torn text feature points do not correspond, the missing text feature points are determined according to the first torn text feature points, the second torn text feature points and the restored image.

[0115] Missing text feature points refer to characters with torn parts missing. The missing text feature points can be obtained by finding the position points that exist independently on the restored image, and then matching the corresponding first torn text feature points or second torn text feature points according to the position points. The missing text feature points can be obtained by a missing text recognition model. The missing text recognition model refers to a neural network model obtained in advance through sample training.

[0116] Step 516: Extract the missing characters based on the missing text feature points and the bill image.

[0117] The missing characters are the characters extracted from the bill of exchange image according to the missing text feature points. The missing characters can be obtained through a character extraction model. The character extraction model refers to a neural network model obtained in advance through sample training.

[0118] Step 517: Determine whether fuzzy recognition can be performed based on the missing characters.

[0119] Whether the missing characters can be fuzzily recognized can be determined through a fuzzy recognition judgment model, and the fuzzy recognition judgment model refers to a neural network model obtained through sample training in advance.

[0120] Step 518: When fuzzy recognition is possible, matching characters are determined based on the missing characters.

[0121] The matching characters refer to the characters obtained by fuzzy recognition of the missing characters. The matching characters can be obtained through a fuzzy recognition model. The fuzzy recognition model refers to a neural network model obtained in advance through sample training.

[0122] Step 519: replacing the missing characters in the restored image according to the matching characters and the missing text feature points to form a new restored image.

[0123] The characters at the positions of the missing text feature points on the restored image are replaced with matching characters to reduce the number of missing characters. The new restored image can be obtained through a character replacement model, which refers to a neural network model obtained in advance through sample training.

[0124] Step 520: When fuzzy recognition cannot be performed, control a preset display device to issue a text missing warning.

[0125] When there are missing characters, try fuzzy recognition first to reduce the number of missing characters and improve the convenience and accuracy of bill recognition.

[0126] Reference Figure 8, the transaction information acquisition method further includes the following steps: Step 521: When fuzzy recognition cannot be performed, the missing fields are extracted from the restored image according to the missing text feature points.

[0127] A missing field refers to a field in the restored image that contains missing characters, where the extracted fields are bounded by spaces and punctuation marks. The missing field can be obtained through a field extraction model, and the field extraction model refers to a neural network model obtained in advance through sample training.

[0128] Step 522: Determine whether there is a synonymous field in the bill image based on the missing field.

[0129] A synonymous field refers to a field with the same or similar meaning as a missing field. The fields in the bill of exchange image can be extracted through a synonymous field matching model, and then the extracted fields are compared with the missing fields one by one to determine whether there are synonymous fields. The synonymous field matching model refers to a neural network model obtained through sample training in advance.

[0130] Step 523: When there are synonymous fields in the bill image, the synonymous fields are determined according to the missing fields and the bill image.

[0131] Synonymous fields can be extracted from the bill image through a synonymous field extraction model, and the synonymous field extraction model refers to a neural network model obtained in advance through sample training.

[0132] Step 524: Determine synonymous characters according to the synonymous fields and the missing fields.

[0133] Synonymous characters refer to characters that correspond to missing characters after semantically matching characters in the missing field and the synonymous field one by one. Synonymous characters can be obtained through a synonym matching model, which refers to a neural network model obtained in advance through sample training.

[0134] Step 525: Replace the missing characters in the restored image according to the synonymous characters and the missing text feature points to form a new restored image.

[0135] The characters at the positions of the missing text feature points on the restored image are replaced with synonymous characters to reduce the number of missing characters. The new restored image can be obtained through the character replacement model.

[0136] Based on the same inventive concept, an embodiment of the present invention provides a bill information integration system, including: The acquisition module is used to obtain transaction information and bill images.

[0137] A memory for storing a program for any of the above-mentioned methods for integrating bill of exchange information.

[0138] The program in the processor and the memory can be loaded and executed by the processor and implement any of the above-mentioned bill information integration methods.

[0139] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0140] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned bill information integration method.

[0141] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0142] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for integrating bill information, characterized in that: include: Get transaction information of bill of exchange; Determine the transaction quotation based on the transaction information, and retrieve the previous fitting price from the preset fitting price database; Calculate the difference between the transaction price and the previous fit price and define it as the fit price difference; When the absolute value of the fitted price difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0; When the fitting price difference is greater than 0, the sum of the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment; When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset fluctuation unit value is calculated and defined as the fitting price at this moment; When the absolute value of the fitting price difference is not greater than the preset secondary market volatility coefficient, the transaction quote is defined as the fitting price at this moment; Based on the existence of the fitting price at this moment, the fitting price at this moment is entered into a preset fitting price database, and the data in the preset fitting price database is retrieved to form market trading trend parameters.

2. A method for integrating bill of exchange information according to claim 1, characterized in that: The invention also includes an accepting bank correction method, wherein the accepting bank correction method includes: Determine the accepting bank based on transaction information; Determine the accepting bank parameters based on the accepting bank; Determine the accepting bank correction value based on the transaction quotation and accepting bank parameters; Calculate the difference between the accepting bank's revised value and the previous fitting price and define it as the fitting price difference.

3. A method for integrating bill of exchange information according to claim 2, characterized in that: It also includes a deviation correction method, the deviation correction method comprising: Retrieve historical fitting price data from the preset fitting price database; Determine the average increase value based on the historical data of the fitted price; Calculate the difference between the absolute value of the fitted price difference and the average increase value and define it as the increase difference; When the absolute value of the increase difference is greater than the preset secondary market volatility coefficient, determine whether the fitted price difference is greater than 0; When the fitted price difference is greater than 0, the sum of the previous fitted price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value; When the fitting price difference is not greater than 0, the difference between the previous fitting price and the preset secondary market volatility coefficient is calculated and defined as the deviation correction value; When the absolute value of the increase difference is not greater than the preset secondary market volatility coefficient, the accepting bank's correction value is defined as the deviation correction value; Calculate the difference between the deviation correction value and the previous fitting price and define it as the fitting price difference.

4. A method for integrating bill of exchange information according to claim 3, characterized in that: Also included is a fitting correction method, the fitting correction method comprising: Determine the buying and selling data based on the historical data of the fitting price; Determine the lowest purchase price based on the purchase data; Determine the highest selling price based on the selling data; When the highest selling price is not 0, the highest selling price is defined as the fitting correction value; When the highest selling price is 0, the average of the lowest buying price and the deviation correction value is calculated and defined as the fitting correction value; Calculate the difference between the fitted correction value and the previous fitted price and define it as the fitted price difference.

5. A method for integrating bill information according to claim 1, characterized in that: The invention also includes a transaction information acquisition method, wherein the transaction information acquisition method includes: Get the money order image; Determine whether the bill of exchange is torn based on the bill of exchange image; When there is a tear in the bill, the tear area is determined based on the bill image; Judging whether there is text in the torn area according to the torn area; When text exists in the tearing area, a tearing contour and a tearing extreme point are determined based on the tearing area; Determine a first tear boundary and a second tear boundary according to the tear contour and the tear extreme point; Extracting tear feature points according to the first tear boundary; Matching restoration feature points from the second tear boundary according to the tear feature points; Determine a restored image according to the tearing feature points, the restoration feature points and the tearing area; Replaces the torn area by restoring the image.

6. A method for integrating bill information according to claim 5, characterized in that: The transaction information acquisition method further includes: When there are words in the torn area, judging whether the words are split based on the torn area; When there is a situation where the text is split, determining the first torn text feature point according to the torn area and the first torn boundary; Determine a second torn text feature point according to the torn area and the second torn boundary; Determining whether the torn text feature points correspond to each other one by one according to the first torn text feature points, the second torn text feature points and the restored image; When the torn text feature points do not correspond, a preset display device is controlled to issue a text missing warning.

7. A method for integrating bill information according to claim 6, characterized in that: The transaction information acquisition method further includes: When the torn text feature points do not correspond, determining the missing text feature points according to the first torn text feature points, the second torn text feature points and the restored image; Extract missing characters based on missing text feature points and bill of exchange image; Determine whether fuzzy recognition can be performed based on the missing characters; When fuzzy recognition is possible, matching characters are determined based on missing characters; replacing the missing characters in the restored image according to the matching characters and the missing text feature points to form a new restored image; When fuzzy recognition cannot be performed, a preset display device is controlled to issue a text missing warning.

8. A method for integrating bill information according to claim 7, characterized in that: The transaction information acquisition method further includes: When fuzzy recognition is not possible, the missing fields are extracted from the restored image based on the missing text feature points; Determine whether there are synonymous fields in the bill of exchange image based on the missing fields; When there are synonymous fields in the bill of exchange image, the synonymous fields are determined based on the missing fields and the bill of exchange image; Determine synonym characters based on synonym fields and missing fields; The missing characters in the restored image are replaced according to the synonymous characters and the missing text feature points to form a new restored image.

9. A bill of exchange information integration system, characterized in that: include: An acquisition module, used to acquire transaction information and bill images; A memory for storing a program of a method for integrating bill information according to any one of claims 1 to 8; The program in the memory can be loaded and executed by the processor and implement a method for integrating bill information as described in any one of claims 1 to 8.

10. An intelligent terminal, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and executes a method for integrating bill of exchange information as claimed in any one of claims 1 to 8.

Images