CN111462388A - Bill inspection method and device, terminal equipment and storage medium - Google Patents
Bill inspection method and device, terminal equipment and storage medium Download PDFInfo
- Publication number
- CN111462388A CN111462388A CN202010197712.2A CN202010197712A CN111462388A CN 111462388 A CN111462388 A CN 111462388A CN 202010197712 A CN202010197712 A CN 202010197712A CN 111462388 A CN111462388 A CN 111462388A
- Authority
- CN
- China
- Prior art keywords
- bill
- image
- inspected
- amount
- category
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/004—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip
- G07D7/0047—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip using checkcodes, e.g. coded numbers derived from serial number and denomination
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/005—Testing security markings invisible to the naked eye, e.g. verifying thickened lines or unobtrusive markings or alterations
- G07D7/0054—Testing security markings invisible to the naked eye, e.g. verifying thickened lines or unobtrusive markings or alterations involving markings the properties of which are altered from original properties
- G07D7/0056—Testing security markings invisible to the naked eye, e.g. verifying thickened lines or unobtrusive markings or alterations involving markings the properties of which are altered from original properties involving markings of altered colours
Abstract
The invention discloses a bill inspection method, a bill inspection device, terminal equipment and a storage medium, wherein the method comprises the steps of firstly, inputting a bill image to be inspected into a trained bill identification model to identify the type, the amount and the serial number of the bill, then comparing the bill image with preset data, judging the bill of the bill image to be inspected to be a repeated bill if bill information with the same type, the same serial number and the same amount exists, and judging the bill of the bill image to be inspected and the bill corresponding to the third bill information to have a counterfeit bill if bill information with the same type, the same serial number and the different amount exists; when the bill recognition model is used for training, various bills comprise handwritten bills and value-added tax invoices as training samples, so that the function that the bill recognition model can recognize common handwritten bills and value-added tax invoices is realized, the inspection efficiency of the handwritten bills is greatly improved, and the consumption of manpower is reduced.
Description
Technical Field
The present invention relates to the field of image recognition technologies, and in particular, to a data verification method and apparatus, a terminal device, and a storage medium.
Background
In modern society, the cooperation between enterprises is frequent, which brings about the diversification of the bill types of the financial treatment of the enterprises, including not only electronic value-added tax invoices, but also various common hand-written receipts with official seals of the enterprises.
The identification processing technology for the invoice at present is limited to be applied to fixed types of bills, such as: the electronic value-added tax invoice has high quality requirement on the bill image, the accuracy can be ensured only by requiring standard PDF scanning or high-quality electronic value-added tax image, the electronic value-added tax invoice is acquired by a special scanner or other professional image acquisition equipment, and after the content of a single bill is identified and processed, corresponding information is transmitted to a special electronic value-added tax inspection service system to verify the authenticity of the electronic value-added tax invoice. In fact, in modern enterprise activities, a greater number of receipts are involved, usually hand-written, than are electronic value-added tax invoices. Because the common handwritten receipt does not have an official special verification system, an enterprise can only check whether the common handwritten receipt is repeated or not and the authenticity of the common handwritten receipt by investing a large amount of manpower, the checking efficiency is low, and a large amount of manpower is consumed.
Disclosure of Invention
The embodiment of the invention provides a bill inspection method and device, terminal equipment and a storage medium, which can automatically inspect various bills including common handwritten receipts and improve the inspection efficiency of the conventional handwritten bills.
An embodiment of the present invention provides a bill inspection method, including: acquiring a ticket image to be inspected, inputting the ticket image to be inspected into a preset ticket identification model, so that the ticket identification model identifies the type of the ticket image to be inspected and extracts the amount and the serial number of the ticket image to be inspected;
matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
if second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database, judging the bill of the bill image to be inspected to be a repeated bill; if third bill information which is the same as the type and the bill number of the bill image to be inspected and has different money amount exists in the preset database, judging that a counterfeit bill exists in the bill of the bill image to be inspected and the bill corresponding to the third bill information;
the generation method of the bill identification model comprises the following steps:
acquiring a plurality of bill samples and extracting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing the amount of money and the filling area position characteristics of Arabic figures used for representing the serial number of the bill; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing money amount, the filling area position characteristics of Arabic figures used for representing bill numbers, Arabic figure data samples and traditional figure data samples into a deep learning network for training to generate the bill identification model.
Further, before the ticket image to be checked is input to a preset ticket recognition model, the method further comprises the following steps: carrying out image preprocessing on the bill image to be inspected; the image preprocessing comprises direction judgment, rotation correction and/or noise point removal.
Further, the direction of the bill in the bill image to be inspected is judged by identifying the direction of the text in the bill to be inspected.
Further, removing noise of the bill image to be inspected by the following modes:
generating a plurality of temporary images containing random noise points according to the ticket image to be checked;
calculating the pixel average value of pixel points with the same position in all the temporary images;
and taking the pixel average value of the pixel points at each position as the pixel value of the pixel point at the corresponding position of the ticket image to be checked.
Further, the method comprises the step of storing the type, the amount and the bill number of the bill image to be inspected into the preset database if the bill information which is the same as the type of the bill image to be inspected does not exist in the preset database or the bill information which is the same as the type of the bill image to be inspected but has a different bill number exists in the preset database.
On the basis of the above-described method item embodiments, the invention correspondingly provides device item embodiments,
an embodiment of the present invention provides a bill inspection device, including: the bill verification system comprises a bill identification module, a bill matching module, a bill inspection module and a model construction module;
the bill identification module is used for acquiring a bill image to be inspected, inputting the bill image to be inspected into a preset bill identification model, so that the bill identification model identifies the type of the bill image to be inspected and extracts the amount and the serial number of the bill image to be inspected;
the bill matching module is used for matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
the bill inspection module is used for judging the bill of the bill image to be inspected as a repeated bill when second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database; when third bill information which is the same as the type and the bill number of the bill image to be inspected and has different money amount exists in the preset database, judging that a counterfeit bill exists in the bill of the bill image to be inspected and the bill corresponding to the third bill information;
the model building module is used for acquiring a plurality of bill samples and extracting the category characteristics of each bill sample, the filling area position characteristics of traditional numbers for representing the amount of money and the filling area position characteristics of Arabic numbers for representing the serial numbers of the bills; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing money amount, the filling area position characteristics of Arabic figures used for representing bill numbers, Arabic figure data samples and traditional figure data samples into a deep learning network for training to generate the bill identification model.
On the basis of the embodiment of the method item, the invention correspondingly provides an embodiment of the terminal equipment item;
an embodiment of the present invention provides a bill inspection terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, the bill inspection terminal device implements the bill inspection method according to any one of the above method embodiments of the present invention.
On the basis of the embodiment of the method item, the invention correspondingly provides an embodiment of a storage medium item;
an embodiment of the present invention provides a storage medium, where the storage medium includes a stored computer program, where when the computer program runs, a device in which the storage medium is located is controlled to execute the bill checking method according to any one of the method embodiments of the present invention.
The embodiment of the invention has the following beneficial effects:
the embodiment of the invention provides a bill inspection method, a bill inspection device, terminal equipment and a storage medium, wherein the method comprises the steps of firstly, inputting a bill image to be inspected into a trained bill identification model to identify the type, the amount and the serial number of the bill, then comparing the bill image with preset data, judging the bill of the bill image to be inspected to be a repeated bill if bill information with the same type, the same serial number and the same serial number but different amount exists, and judging the bill of the bill image to be inspected and a counterfeit bill existing in the bill corresponding to the third bill information if bill information with the same type, the same serial number and the different amount exists; when the bill identification model is trained, various bills including handwritten bills and value-added tax invoices are taken as training samples, category characteristics for category distinguishing are extracted from the various bills, and then the category characteristics are trained with Arabic digital data samples and traditional digital data samples, so that the function that the bill identification model can identify common handwritten bills and value-added tax invoices is realized, the inspection efficiency of the handwritten bills is greatly improved, and the consumption of manpower is reduced.
Drawings
Fig. 1 is a schematic flow chart of a bill checking method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a handwritten bill in a bill inspection method according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of another handwritten ticket data in a ticket inspection method according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of another handwritten ticket in a ticket inspection method according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of another handwritten ticket in a ticket inspection method according to an embodiment of the present invention.
Fig. 6 is a schematic diagram of a handwritten bill in a bill inspection method according to an embodiment of the present invention.
Fig. 7 is a schematic diagram of a value-added tax invoice in a bill checking method according to an embodiment of the present invention.
Fig. 8 is a schematic diagram of another value-added tax invoice in a bill verification method according to an embodiment of the present invention.
Fig. 9 is a schematic diagram of another value-added tax invoice in a bill verification method according to an embodiment of the present invention.
Fig. 10 is a schematic diagram of another handwritten ticket in a ticket inspection method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a bill checking method, including:
s101, acquiring a ticket image to be inspected, inputting the ticket image to be inspected into a preset ticket identification model, so that the ticket identification model identifies the type of the ticket image to be inspected and extracts the amount and the number of the ticket image to be inspected;
s102, matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
s103, judging the bill of the bill image to be inspected as a repeated bill when second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database; the preset database is provided with third bill information which has the same type and serial number with the bill image to be checked but different money amount, and the bill of the bill image to be checked and the bill corresponding to the third bill information are judged to have a counterfeit bill;
for step S101, firstly, a process of constructing a preset bill identification model is described:
in a preferred embodiment, the bill identifying module is constructed by:
a, acquiring a plurality of bill samples and extracting the class characteristics of each bill sample, the filling area position characteristics of traditional numbers for representing the amount of money and the filling area position characteristics of Arabic numbers for representing the serial numbers of the bills; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional numbers for representing money amount, the filling area position characteristics of Arabic numbers for representing bill numbers, the Arabic number data samples and the traditional number data samples into a deep learning network for training to generate the bill identification model.
Firstly, various types of invoices are obtained as bill samples, and the invoices can be divided into two categories, namely handwritten bills and value-added tax invoices; the handwritten bill is subdivided into several different types of handwritten bills, and the preferred specific type can be seen in fig. 2-6, and fig. 2-6 illustrate the different types of handwritten bills in fig. 5, but may also include more different types of handwritten bills; the value-added tax invoice can be subdivided into a value-added tax ordinary invoice (as shown in fig. 7), a value-added tax electronic ordinary invoice (as shown in fig. 8) and a value-added tax special invoice (as shown in fig. 9), and the feature items identified as categories are selected from the invoice according to the difference of the format contents of the various acquired invoices, for example, for the handwritten invoice shown in fig. 2, the two text contents of "payment unit" and "payment reason" can be used as the text features of the category features, the two text contents of "payment unit" and "payment reason" in fig. 2 are subjected to frame selection marking, the category label is given, the text features and the position features of the two text contents of "payment unit" and "payment reason" are extracted to be used as the category features of the handwritten invoice shown in fig. 2, and by analogy, for the handwritten invoice shown in fig. 3, the two text contents of "receipt today" and "payment reason" can be selected as the category features for identifying the handwritten invoice of the category feature of the category of the handwritten invoice of The character features of the method are also used for performing frame selection and marking on the two fields of 'received' and 'handed over' before the characters are input into the deep learning network, and giving category labels, the position features of the two fields can be determined through frame selection, and other types of invoices are not listed in a list.
Furthermore, after the category features are marked, the positions of the areas filled with traditional numbers for representing the amount items in the invoice and the positions of the areas filled with arabic numbers for representing the note numbers in the invoice are marked, and as shown in fig. 2 as an example, for the handwritten note shown in fig. 2, the filling areas of arabic numbers "4876013" following "N0" are marked by frame selection so as to extract the filling area position features of arabic numbers for representing the note numbers later, and the area of a cross bar following the "renminbi (capitalization)" field in fig. 2 is marked by frame selection so as to extract the filling area position features of traditional numbers for representing the amount later. After all the marks are finished, inputting all the marked bills, Arabic data samples and traditional Chinese digital data samples as a training set into a deep learning network for training and learning to generate the bill identification model, wherein the deep learning network can be a convolutional neural network, preferably, the marked bills, the Arabic data samples and the traditional Chinese digital data samples can be subjected to data enhancement through a GAN technology to generate the training set, the bill identification mode can identify the types of the bills according to the category characteristics, then the bill number filling areas of all the bills, the bill number filling areas and the bill number filling areas can be extracted to represent the sum of money, then the bill numbers and the traditional Chinese digital data in the two areas are respectively identified, and the Arabic numbers in the bills are extracted to represent the bill numbers, and traditional numbers used to represent amounts.
After the bill identification model is built, an image of a bill to be inspected, namely the image of the bill to be inspected is obtained through a camera device and then is input into the bill identification model, the type of the bill image to be inspected, the amount of money represented by traditional numbers and the serial number of the bill represented by Arabic numbers are identified.
In a preferred embodiment, before the ticket image to be checked is input to a preset ticket recognition model, the method further comprises the following steps: carrying out image preprocessing on the bill image to be inspected; the image preprocessing comprises direction judgment, rotation correction and/or noise point removal.
The direction is first determined, in a preferred embodiment, by identifying the direction of the text in the document to be inspected, the direction of the document in the image of the document to be inspected is determined. Specifically, an identification model for detecting the direction of the characters can be constructed on the basis of the convolutional neural network, then the image of the ticket to be detected is input into the model, the direction of the characters in the ticket is identified, and then the direction of the ticket in the image of the ticket to be detected is judged. If the direction of the bill in the image of the bill to be inspected is inverted, the rotation correction can be performed by utilizing the existing OPENCV technology. Next, noise point removing processing is carried out on the image, specifically, in a preferred embodiment, a plurality of temporary images containing random noise points are generated according to the ticket image to be checked; calculating the pixel average value of pixel points with the same position in all the temporary images; and taking the pixel average value of the pixel points at each position as the pixel value of the pixel point at the corresponding position of the ticket image to be checked. Of course, the noise of the ticket image to be inspected can be removed by the existing noise removing technology.
In step S102, the category information, the amount of money, and the ticket number extracted in step S101 may be searched and matched in a preset database, where a part of the ticket information is stored in advance, where the ticket information refers to the category, the amount of money, and the ticket number of the ticket.
For step S103, if the bill information with the same category, amount and bill number can be found in the database, i.e. the second bill information, it indicates that the bill in the bill image to be inspected is a duplicate bill.
If the bill information with the same category and the same bill number but different bill amounts can be found in the database, that is, the third bill information indicates that one bill is a counterfeit bill among the bills in the bill image to be inspected and the bills corresponding to the third bill information, a prompt message can be generated for prompting the user, and the prompt message can be character information displayed on a display interface.
In a preferred embodiment, if the preset database does not have bill information with the same type as the bill image to be inspected or bill information with the same type as the bill image to be inspected but different bill numbers, the type, the amount and the bill number of the bill image to be inspected are stored in the preset database. In the embodiment, the bill information in the database is mainly expanded to improve the accuracy of subsequent inspection.
On the basis of the above method item embodiments, the present invention correspondingly provides apparatus item embodiments;
as shown in fig. 10, an embodiment of the present invention provides a bill inspection device, including: the bill verification system comprises a bill identification module, a bill matching module, a bill inspection module and a model construction module;
the bill identification module is used for acquiring a bill image to be inspected, inputting the bill image to be inspected into a preset bill identification model, so that the bill identification model identifies the type of the bill image to be inspected and extracts the amount and the serial number of the bill image to be inspected;
the bill matching module is used for matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
the bill inspection module is used for judging the bill of the bill image to be inspected as a repeated bill when second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database; when third bill information which is the same as the type and the bill number of the bill image to be inspected and has different money amount exists in the preset database, judging that a counterfeit bill exists in the bill of the bill image to be inspected and the bill corresponding to the third bill information;
the model building module is used for acquiring a plurality of bill samples and extracting the category characteristics of each bill sample, the filling area position characteristics of traditional numbers for representing the amount of money and the filling area position characteristics of Arabic numbers for representing the serial numbers of the bills; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing money amount, the filling area position characteristics of Arabic figures used for representing bill numbers, Arabic figure data samples and traditional figure data samples into a deep learning network for training to generate the bill identification model.
It should be noted that the above-mentioned embodiments of the apparatus correspond to the embodiments of the method of the present invention, and the method of checking a bill according to any one of the above-mentioned embodiments of the method of the present invention can be implemented. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
On the basis of the embodiment of the method item, the invention correspondingly provides an embodiment of the terminal equipment item;
another embodiment of the present invention provides a bill inspection terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor executes the computer program to implement the bill inspection method according to any one of the above method embodiments of the present invention.
Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory and executed by the processor to implement the invention.
The bill inspection terminal equipment can be computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud server. The bill inspection terminal equipment can include, but is not limited to, a processor, and a memory.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, which is the control center for the bill inspection terminal device, connecting the various parts of the overall bill inspection terminal device with various interfaces and lines.
The memory may be used to store the computer programs and/or modules, and the processor may implement the various functions of the bill validator/terminal device by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
On the basis of the above method item embodiment, the present invention provides a storage medium item embodiment correspondingly, and another trial example of the present invention provides a storage medium, where the storage medium includes a stored computer program, where when the computer program runs, a device in which the storage medium is located is controlled to execute the bill checking method according to any one of the above method item embodiments of the present invention.
The storage medium is a computer-readable storage medium, all or part of the processes in the method according to the embodiments of the present invention may be implemented by instructing related hardware through a computer program, where the computer program may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc.
By implementing the embodiment of the invention, the handwritten bills and the value added tax bills can be simultaneously inspected, the inspection efficiency of the handwritten bills is greatly improved, and the consumption of manpower and material resources is reduced.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (8)
1. A method of bill inspection, comprising:
acquiring a ticket image to be inspected, inputting the ticket image to be inspected into a preset ticket identification model, so that the ticket identification model identifies the type of the ticket image to be inspected and extracts the amount and the serial number of the ticket image to be inspected;
matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
if second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database, judging the bill of the bill image to be inspected to be a repeated bill; if third bill information which is the same as the type and the bill number of the bill image to be inspected and has different money amount exists in the preset database, judging that a counterfeit bill exists in the bill of the bill image to be inspected and the bill corresponding to the third bill information;
the generation method of the bill identification model comprises the following steps:
acquiring a plurality of bill samples and extracting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing the amount of money and the filling area position characteristics of Arabic figures used for representing the serial number of the bill; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing money amount, the filling area position characteristics of Arabic figures used for representing bill numbers, Arabic figure data samples and traditional figure data samples into a deep learning network for training to generate the bill identification model.
2. The document inspection method according to claim 1, further comprising, before inputting the document image to be inspected to a preset document recognition model: carrying out image preprocessing on the bill image to be inspected; the image preprocessing comprises direction judgment, rotation correction and/or noise point removal.
3. The document inspection method according to claim 2, wherein the direction of the document in the image of the document to be inspected is judged by recognizing the direction of the text in the document to be inspected.
4. A document inspection method according to claim 3, wherein noise is removed from the image of the document to be inspected by:
generating a plurality of temporary images containing random noise points according to the ticket image to be checked;
calculating the pixel average value of pixel points with the same position in all the temporary images;
and taking the pixel average value of the pixel points at each position as the pixel value of the pixel point at the corresponding position of the ticket image to be checked.
5. The bill inspection method according to claim 1, further comprising storing the category, amount and bill number of the bill image to be inspected in said preset database if there is no bill information in the same category as the bill image to be inspected or bill information in the same category as the bill image to be inspected but different bill number.
6. A bill validator comprising: the bill verification system comprises a bill identification module, a bill matching module, a bill inspection module and a model construction module;
the bill identification module is used for acquiring a bill image to be inspected, inputting the bill image to be inspected into a preset bill identification model, so that the bill identification model identifies the type of the bill image to be inspected and extracts the amount and the serial number of the bill image to be inspected;
the bill matching module is used for matching in a pre-stored database according to the type, the bill number and the amount of the bill image to be checked;
the bill inspection module is used for judging the bill of the bill image to be inspected as a repeated bill when second bill information which is the same as the type, the bill number and the amount of the bill image to be inspected exists in the preset database; when third bill information which is the same as the type and the bill number of the bill image to be inspected and has different money amount exists in the preset database, judging that a counterfeit bill exists in the bill of the bill image to be inspected and the bill corresponding to the third bill information;
the model building module is used for acquiring a plurality of bill samples and extracting the category characteristics of each bill sample, the filling area position characteristics of traditional numbers for representing the amount of money and the filling area position characteristics of Arabic numbers for representing the serial numbers of the bills; the category characteristics comprise text characteristics selected as bill category identification and position characteristics corresponding to each text characteristic, and the bill sample comprises a plurality of handwritten bill images of different categories and a plurality of value-added tax invoice images of different categories;
inputting the category characteristics of each bill sample, the filling area position characteristics of traditional figures used for representing money amount, the filling area position characteristics of Arabic figures used for representing bill numbers, Arabic figure data samples and traditional figure data samples into a deep learning network for training to generate the bill identification model.
7. A ticket checking terminal device, characterized by comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the ticket checking method according to any one of claims 1 to 5 when executing the computer program.
8. A storage medium, characterized in that the storage medium comprises a stored computer program, wherein the apparatus on which the storage medium is located is controlled to execute the bill inspection method according to any one of claims 1 to 5 when the computer program is run.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010197712.2A CN111462388A (en) | 2020-03-19 | 2020-03-19 | Bill inspection method and device, terminal equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010197712.2A CN111462388A (en) | 2020-03-19 | 2020-03-19 | Bill inspection method and device, terminal equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111462388A true CN111462388A (en) | 2020-07-28 |
Family
ID=71683575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010197712.2A Pending CN111462388A (en) | 2020-03-19 | 2020-03-19 | Bill inspection method and device, terminal equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111462388A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931784A (en) * | 2020-09-17 | 2020-11-13 | 深圳壹账通智能科技有限公司 | Bill recognition method, system, computer device and computer-readable storage medium |
| CN112861841A (en) * | 2021-03-22 | 2021-05-28 | 北京百度网讯科技有限公司 | Bill confidence value model training method and device, electronic equipment and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751194A (en) * | 2015-04-27 | 2015-07-01 | 陈包容 | Processing method and processing device for financial expense reimbursement |
| CN107067044A (en) * | 2017-05-31 | 2017-08-18 | 北京空间飞行器总体设计部 | A kind of finance reimbursement unanimous vote is according to intelligent checks system |
| CN108171127A (en) * | 2017-12-13 | 2018-06-15 | 广东电网有限责任公司清远供电局 | A kind of invoice automatic identifying method based on deep learning |
| CN108446621A (en) * | 2018-03-14 | 2018-08-24 | 平安科技（深圳）有限公司 | Bank slip recognition method, server and computer readable storage medium |
| CN110188743A (en) * | 2019-05-13 | 2019-08-30 | 武汉大学 | A kind of taxi invoice identifying system and method |
| CN110414927A (en) * | 2019-07-05 | 2019-11-05 | 西安网算数据科技有限公司 | A kind of bill processing automatically generates the method and device of voucher |
| CN110490198A (en) * | 2019-08-12 | 2019-11-22 | 上海眼控科技股份有限公司 | Text orientation bearing calibration, device, computer equipment and storage medium |
| CN110516664A (en) * | 2019-08-16 | 2019-11-29 | 咪咕数字传媒有限公司 | Bank slip recognition method, apparatus, electronic equipment and storage medium |
| CN110634223A (en) * | 2019-09-27 | 2019-12-31 | 中国银行股份有限公司 | Bill verification method and device |
| CN110689658A (en) * | 2019-10-08 | 2020-01-14 | 北京邮电大学 | Taxi bill identification method and system based on deep learning |
| CN110866495A (en) * | 2019-11-14 | 2020-03-06 | 杭州睿琪软件有限公司 | Bill image recognition method, bill image recognition device, bill image recognition equipment, training method and storage medium |
-
2020
- 2020-03-19 CN CN202010197712.2A patent/CN111462388A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751194A (en) * | 2015-04-27 | 2015-07-01 | 陈包容 | Processing method and processing device for financial expense reimbursement |
| CN107067044A (en) * | 2017-05-31 | 2017-08-18 | 北京空间飞行器总体设计部 | A kind of finance reimbursement unanimous vote is according to intelligent checks system |
| CN108171127A (en) * | 2017-12-13 | 2018-06-15 | 广东电网有限责任公司清远供电局 | A kind of invoice automatic identifying method based on deep learning |
| CN108446621A (en) * | 2018-03-14 | 2018-08-24 | 平安科技（深圳）有限公司 | Bank slip recognition method, server and computer readable storage medium |
| CN110188743A (en) * | 2019-05-13 | 2019-08-30 | 武汉大学 | A kind of taxi invoice identifying system and method |
| CN110414927A (en) * | 2019-07-05 | 2019-11-05 | 西安网算数据科技有限公司 | A kind of bill processing automatically generates the method and device of voucher |
| CN110490198A (en) * | 2019-08-12 | 2019-11-22 | 上海眼控科技股份有限公司 | Text orientation bearing calibration, device, computer equipment and storage medium |
| CN110516664A (en) * | 2019-08-16 | 2019-11-29 | 咪咕数字传媒有限公司 | Bank slip recognition method, apparatus, electronic equipment and storage medium |
| CN110634223A (en) * | 2019-09-27 | 2019-12-31 | 中国银行股份有限公司 | Bill verification method and device |
| CN110689658A (en) * | 2019-10-08 | 2020-01-14 | 北京邮电大学 | Taxi bill identification method and system based on deep learning |
| CN110866495A (en) * | 2019-11-14 | 2020-03-06 | 杭州睿琪软件有限公司 | Bill image recognition method, bill image recognition device, bill image recognition equipment, training method and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111931784A (en) * | 2020-09-17 | 2020-11-13 | 深圳壹账通智能科技有限公司 | Bill recognition method, system, computer device and computer-readable storage medium |
| CN112861841A (en) * | 2021-03-22 | 2021-05-28 | 北京百度网讯科技有限公司 | Bill confidence value model training method and device, electronic equipment and storage medium |
| CN112861841B (en) * | 2021-03-22 | 2023-06-13 | 北京百度网讯科技有限公司 | Training method and device for bill confidence value model, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717545B (en) | Bill identification method and system based on mobile phone photographing | |
| CN108777021B (en) | Bill identification method and system based on scanner mixed scanning | |
| CN110705952A (en) | Contract auditing method and device | |
| CN112395996A (en) | Financial bill OCR recognition and image processing method, system and readable storage medium | |
| CN109271951A (en) | A kind of method and system promoting book keeping operation review efficiency | |
| CN110516664A (en) | Bank slip recognition method, apparatus, electronic equipment and storage medium | |
| CN112395995A (en) | Method and system for automatically filling and checking bill according to mobile financial bill | |
| CN111462388A (en) | Bill inspection method and device, terminal equipment and storage medium | |
| CN112232336A (en) | Certificate identification method, device, equipment and storage medium | |
| CN113673528B (en) | Text processing method, text processing device, electronic equipment and readable storage medium | |
| CN114511866A (en) | Data auditing method, device, system, processor and machine-readable storage medium | |
| CN110197140A (en) | Material checking method and equipment based on Text region | |
| WO2022103564A1 (en) | Fraud detection via automated handwriting clustering | |
| CN108090728B (en) | Express information input method and system based on intelligent terminal | |
| CN111199240A (en) | Training method of bank card identification model, and bank card identification method and device | |
| CN112613367A (en) | Bill information text box acquisition method, system, equipment and storage medium | |
| CN112632926A (en) | Data processing method and device for bill, electronic equipment and storage medium | |
| CN112560855A (en) | Image information extraction method and device, electronic equipment and storage medium | |
| CN111414889A (en) | Financial statement identification method and device based on character identification | |
| CN111582115A (en) | Financial bill processing method, device and equipment and readable storage medium | |
| CN115083024A (en) | Signature identification method, device, medium and equipment based on region division | |
| CN113077355B (en) | Insurance claim settlement method and device, electronic equipment and storage medium | |
| CN114549177A (en) | Insurance letter examination method, device, system and computer readable storage medium | |
| CN107240185A (en) | A kind of crown word number identification method, device, equipment and storage medium | |
| CN113343968A (en) | Multi-template certificate rapid verification method, system, medium and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200728 |
|
| RJ01 | Rejection of invention patent application after publication |