CN114091517A - Financial bill verification method based on big data and intelligent finance - Google Patents
Financial bill verification method based on big data and intelligent finance Download PDFInfo
- Publication number
- CN114091517A CN114091517A CN202111176641.9A CN202111176641A CN114091517A CN 114091517 A CN114091517 A CN 114091517A CN 202111176641 A CN202111176641 A CN 202111176641A CN 114091517 A CN114091517 A CN 114091517A
- Authority
- CN
- China
- Prior art keywords
- signature
- bill
- characteristic
- standard
- image
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000012795 verification Methods 0.000 title claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims abstract description 88
- 238000012545 processing Methods 0.000 claims abstract description 18
- 238000012550 audit Methods 0.000 claims description 50
- 230000009466 transformation Effects 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 16
- 238000003709 image segmentation Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000004148 unit process Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Abstract
The invention relates to a financial bill verification method based on big data and intelligent finance, which comprises the following steps: acquiring a pixel value of each pixel point in the non-standard bill, and updating the pixel value to obtain a second pixel value of each pixel point; taking the pixel points with the second pixel values larger than the second pixel threshold value as characteristic points; acquiring the characteristic values of all the characteristic points in the non-standard bill, and generating a bill identification matrix according to the characteristic values of all the characteristic points in the non-standard bill; processing the non-standard bill according to the bill identification matrix to obtain a plurality of first money images and first signature images; and obtaining a plurality of money images and signature images according to all the first money images and the first signature images. And the non-standard bill is identified according to all the amount images and the signature images so as to judge whether the non-standard bill is modified or not, and the non-standard bill which is not modified is sent to a corresponding auditor terminal.
Description
Technical Field
The invention relates to the field of big data and intelligent finance, in particular to a financial bill verification method based on big data and intelligent finance.
Background
The intelligent finance can classify and deeply analyze the sensed data based on modern information technologies such as cloud computing and data mining, and specifically find the development trend or existing problems of financial services, so that effective methods for solving the problems are found, and planning is made for future development. Intelligent finance can use information more effectively and fully, connect customers more generally and deeply, and communicate with customers.
In the prior art, when auditing bills, a to-be-audited party submits paper financial bills to corresponding auditors according to audit notices so that the auditors can audit the paper invoices, the mode consumes time and labor, and the paper financial bills are easily damaged and omitted in the transmission process. Meanwhile, in the existing mode of auditing the electronic financial bills through the network, when the party to be audited uploads the electronic financial bills, the contents of the electronic bills can be tampered by using the retouching software, and auditors cannot accurately identify the modified contents in the electronic bills, so that audit holes are caused.
Disclosure of Invention
In view of the above, the present invention provides a financial bill verification method based on big data and intelligent finance, including:
receiving an electronic financial bill sent by a to-be-audited terminal and judging the bill type of the electronic financial bill; the bill types comprise standard bills and non-standard bills;
when the electronic financial bill is a non-standard bill, acquiring a pixel value of each pixel point in the non-standard bill, and performing weighted summation on all components of each pixel point in the non-standard bill to perform first updating on the pixel value of each pixel point in the non-standard bill so as to obtain a first pixel value of each pixel point in the non-standard bill;
according to the first pixel threshold value, second updating is carried out on the first pixel value of each pixel point in the non-standard bill so as to obtain a second pixel value of each pixel point in the non-standard bill; acquiring second pixel values of all pixel points in the non-standard bill, and taking the pixel points with the second pixel values larger than a second pixel threshold value as feature points; acquiring the characteristic values of all the characteristic points in the non-standard bill, and generating a bill identification matrix according to the characteristic values of all the characteristic points in the non-standard bill; processing the non-standard bill according to the bill identification matrix to obtain a plurality of first money images and first signature images;
obtaining a first sum matrix and a second sum matrix according to the first sum image, and performing intersection operation on the first sum matrix and the second sum matrix to obtain a feature extraction matrix of each first sum image; extracting the sum features of each first sum image according to the feature extraction matrix of each first sum image to obtain a sum image corresponding to each first sum image;
obtaining a first signature matrix and a second signature matrix according to the first signature images, and performing intersection operation on the first signature matrix and the second signature matrix to obtain a feature extraction matrix of each first signature image; extracting signature characteristics of each first signature image according to the characteristic extraction matrix of each first signature image to obtain a signature image corresponding to each first signature image;
and the non-standard bill is identified according to all the amount images and the signature images so as to judge whether the non-standard bill is modified or not, and the non-standard bill which is not modified is sent to a corresponding auditor terminal.
According to a preferred embodiment, the auditor terminal refers to electronic equipment used by auditors; and the to-be-audited terminal is electronic equipment used by audited unit personnel. The standard bill comprises a value-added tax common invoice and a value-added tax special invoice. The audit bill information comprises the information of relevant audit bills required to be prepared by audited units; the audit time information comprises an audit starting time and an audit period. The identification disturbance factor comprises a scanning type and a scanning resolution, and the scanning type comprises black-white scanning and color scanning.
According to a preferred embodiment, deriving the first amount matrix and the second amount matrix from the first amount image comprises:
carrying out image segmentation on the non-standard bills according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first money images;
vertically projecting all the first sum images to obtain a plurality of second sum images; scanning each pixel point of each second sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first sum matrix;
horizontally projecting all the first sum images to obtain a plurality of third sum images; and scanning each pixel point of each third sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the second sum matrix.
According to a preferred embodiment, deriving the first signature matrix and the second signature matrix from the first signature image comprises:
carrying out image segmentation on the non-standard bill according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first signature images;
vertically projecting all the first signature images to obtain a plurality of second signature images; scanning each pixel point of each second signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first signature matrix;
horizontally projecting all the first signature images to obtain a plurality of third signature images; and scanning each pixel point of each third signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and placing the line number of the pixel point with the pixel value not being zero in the second signature matrix.
According to a preferred embodiment, authenticating the non-marking ticket based on all of the amount image and the signature image comprises:
normalizing all the amount images into matrixes with the same size to obtain a plurality of standard amount images; normalizing all signature images into matrixes with the same size to obtain a plurality of standard signature images;
and respectively obtaining a money amount characteristic vector and a signature characteristic vector according to the standard money amount image and the standard signature image, and respectively calculating a first money amount characteristic distance and a first signature characteristic distance according to the money amount characteristic vector and the signature characteristic vector.
According to a preferred embodiment, authenticating the non-marking ticket based on all of the amount image and the signature image comprises:
respectively carrying out characteristic distance transformation on the first money characteristic distance and the first signature characteristic distance to obtain a second money characteristic distance and a second signature characteristic distance;
removing the interference of the identification disturbance factor on the second money amount characteristic distance and the second signature characteristic distance to obtain a target money amount characteristic distance and a target signature characteristic distance;
respectively comparing the target money amount characteristic distance and the target signature characteristic distance with a money amount characteristic distance threshold value and a signature characteristic distance threshold value;
and when the target money amount characteristic distance is smaller than the money amount characteristic distance threshold value and the target signature characteristic distance is smaller than the signature characteristic distance threshold value, judging that the corresponding non-standard bill is not tampered.
According to a preferred embodiment, removing the interference of the discrimination perturbation factor on the second monetary characteristic distance to obtain the target monetary characteristic distance comprises:
establishing a total disturbance variable for identifying the disturbance factors,
wherein s is a total disturbance variable, k is an index of an influence factor of a scanning type, l is an index of an influence factor of a scanning resolution, P is the number of the influence factors of the scanning type, and Q is the number of the influence factors of the scanning resolution.
According to a preferred embodiment, the feature distance transforming the first monetary feature distance to obtain the second monetary feature distance comprises:
performing characteristic distance transformation on the standard sum image to obtain a characteristic transformation image, and acquiring the position coordinate of each characteristic point in the characteristic transformation image and the position coordinate of each characteristic point in the standard sum image;
traversing all the feature points in the feature transformation graph, taking the feature points which are being traversed as first feature points, and acquiring the feature points corresponding to the first feature points in the standard sum image and taking the feature points as second feature points;
calculating the characteristic distance between the first characteristic point and the second characteristic point according to the position coordinates of the first characteristic point and the second characteristic point; and carrying out weighted average on the feature distances of all the feature points in the feature transformation graph and the standard sum image to obtain a second sum feature distance.
According to a preferred embodiment, before the terminal of the party to be audited sends the electronic financial bill, the terminal of the auditor sends an audit notice to the terminal of the party to be audited; the audit notice comprises audit bill information and audit time information; the audit bill information comprises the information of relevant audit bills required to be prepared by audited units; the audit time information comprises an audit starting time and an audit period.
The invention has the following beneficial effects: the invention automatically identifies the truth of the non-standard electronic financial bill obtained by scanning by verifying whether the contents of money, signature and the like in the electronic financial bill are maliciously modified by a person, screens out the electronic financial bill which is not tampered to be manually audited by an auditor, improves the accuracy and efficiency of manual auditing, and simultaneously avoids auditing holes and even huge loss of relevant departments caused by maliciously tampering the contents of the electronic financial bill by a party to be auditor.
Drawings
FIG. 1 is a flowchart of a financial instrument validation method based on big data and intelligent finance, according to an exemplary embodiment.
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.
According to the method, the to-be-audited party sends the scanned electronic financial bill to the corresponding auditor, and compared with the traditional auditing mode that the to-be-audited party needs to submit a paper financial bill to a specified place, the method is more convenient and faster, and bills are not easy to miss and damage. The audit personnel can audit at any time and any place when carrying out the bill audit, need not be in the place of being bound, has improved audit personnel's audit efficiency by a wide margin.
In addition, when the electronic bill is provided, the audited party can modify the electronic bill by using the retouching software when uploading the bill, and the auditor is difficult to identify the modified content in the electronic bill, thereby causing audit holes. The invention can automatically identify the truth of the non-standard electronic financial bill obtained by scanning, namely the electronic financial bill which is not filed on the network, so as to verify whether the content of the electronic financial bill is maliciously tampered by people, improve the auditing accuracy of auditors and avoid auditing holes.
Referring to FIG. 1, in one embodiment, a financial instrument validation method based on big data and intelligent finance may include:
and S1, the auditor terminal sends the audit notice to the to-be-audited terminal, and the to-be-audited terminal sends the corresponding electronic financial bill to the bill processing server according to the audit notice.
Optionally, the auditor terminal refers to an electronic device used by an auditor; and the to-be-audited terminal is electronic equipment used by audited unit personnel. The audit notice comprises audit bill information and audit time information. The audit bill information comprises the information of the relevant audit bill which needs to be prepared by the audited unit; the audit time information includes an audit start time and an audit period. Audit side terminal and pending side terminal are for having communication function, data transmission function and the intelligent equipment of memory function, and it includes: smart phones, smart wearable devices, smart watches, tablet computers, notebook computers, and desktop computers.
Pending side of examining terminal will correspond electron financial bill according to audit notice and send to bill processing server, with traditional mode in, pending side of examining receives corresponding audit notice and need submit paper financial bill to the government affairs hall and compare, and pending side of examining need not to consume too much time, and convenient and fast more, and electron financial bill changes the save, is difficult for omitting and damaging. The audit personnel can audit at any time and any place after receiving the electronic financial bill sent by the party to be audited, and the audit efficiency is improved without being limited to the place.
And the to-be-audited party terminal sends the corresponding electronic financial bill to the bill processing server according to the audit notice so as to verify the authenticity of the electronic financial bill. In practical application, after receiving an audit notice, a to-be-audited party directly sends a corresponding electronic financial bill to a corresponding auditor for auditing, and the auditor is difficult to identify the place in the electronic financial bill, which is maliciously modified by the to-be-audited party by using a related technology and a professional image modifying tool, so that serious loss is caused by error judgment caused by the fact.
S2, the bill processing server receives the electronic financial bill sent by the terminal of the party to be examined and judges the bill type of the electronic financial bill, when the electronic financial bill is a non-standard bill, the bill processing server obtains the pixel value of each pixel point in the non-standard bill, and performs weighted summation on all components of each pixel point in the non-standard bill so as to perform first updating on the pixel value of each pixel point in the non-standard bill and obtain the first pixel value of each pixel point in the non-standard bill.
The ticket types include standard tickets and non-standard tickets. The standard bill comprises a value-added tax common invoice and a value-added tax special invoice. The standard bill is the archive of the corresponding original electronic financial bill in the network for the bill contents of money, date, purchase unit, sales unit, payee, drawer and the like in the corresponding electronic financial bill, and when auditing, the auditor inputs the serial number of the corresponding bill into the system so as to inquire the comparison electronic file of the electronic financial bill.
The non-standard bill is the file of the original electronic financial bill which does not correspond to the bill contents of money, date, purchase units, sales units, payees, invoices and the like in the electronic financial bill in the network, and an auditor can not accurately identify whether the corresponding electronic financial bill is tampered by a person.
The standard bill is a bill which is stored on the network, and even if the standard bill is tampered by people, an auditor can check and compare the electronic file by inputting a corresponding bill number. When people modify non-standard bills without archiving, the modified contents are difficult to see by auditors without a standardized comparison electronic file. The method comprises the steps of judging the bill type of the electronic financial bill, identifying the electronic financial bill when the bill type is a non-standard bill so as to verify whether the electronic financial bill is maliciously tampered by a person, saving computing resources of a platform, and avoiding time and computing resources waste caused by identifying a standard bill.
In one embodiment, when the electronic financial bill is a non-standard bill, the bill processing server performs blackening and denoising on the non-standard bill so as to enhance and denoise the image content of the non-standard bill, so that adverse effects on bill identification caused by noise interference are avoided, and the accuracy of bill authenticity identification is improved.
The image is often interfered by the imaging device and external environment noise in the process of generating and transmitting, and the noise reduces the quality of the image and has a plurality of adverse effects on the processing of subsequent images. Therefore, in order to suppress noise and improve image quality for higher-level processing, it is necessary to perform blackening-whitening and denoising processing on non-standard bills.
S3, the bill processing server carries out second updating on the first pixel value of each pixel point in the non-standard bill according to the first pixel threshold value so as to obtain a second pixel value of each pixel point in the non-standard bill; acquiring second pixel values of all pixel points in the non-standard bill, and taking the pixel points with the second pixel values larger than a second pixel threshold value as feature points; and acquiring the characteristic values of all the characteristic points in the non-standard bill, and generating a bill identification matrix according to the characteristic values of all the characteristic points in the non-standard bill.
And S4, the first authentication unit of the bill authentication server processes the non-standard bill according to the bill authentication matrix to obtain a plurality of first amount images and first signature images. A second identification unit of the bill identification server obtains a first sum matrix and a second sum matrix according to the first sum image, and performs intersection operation on the first sum matrix and the second sum matrix to obtain a feature extraction matrix of each first sum image; and extracting the sum features of each first sum image according to the feature extraction matrix of each first sum image to obtain the sum image corresponding to each first sum image.
In one embodiment, the obtaining of the first amount matrix and the second amount matrix by the second authentication unit according to the first amount image includes:
the second identification unit carries out image segmentation on the non-standard bills according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first money images;
the second identification unit performs vertical projection on all the first sum images to obtain a plurality of second sum images; scanning each pixel point of each second sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first sum matrix;
the second identification unit performs horizontal projection on all the first sum images to obtain a plurality of third sum images; and scanning each pixel point of each third sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the second sum matrix.
S5, a third authentication unit of the bill authentication server obtains a first signature matrix and a second signature matrix according to the first signature image, and performs intersection operation on the first signature matrix and the second signature matrix to obtain a feature extraction matrix of each first signature image; and extracting signature characteristics of each first signature image according to the characteristic extraction matrix of each first signature image to obtain a signature image corresponding to each first signature image.
In one embodiment, the third authentication unit obtaining the first signature matrix and the second signature matrix from the first signature image includes:
the third identification unit carries out image segmentation on the non-standard bill according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first signature images;
the third authentication unit performs vertical projection on all the first signature images to obtain a plurality of second signature images; scanning each pixel point of each second signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first signature matrix;
the third identification unit horizontally projects all the first signature images to obtain a plurality of third signature images; and scanning each pixel point of each third signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and placing the line number of the pixel point with the pixel value not being zero in the second signature matrix.
And S6, the fourth authentication unit of the bill authentication server authenticates the non-standard bill according to all the amount image and the signature image to judge whether the non-standard bill is modified, and sends the non-standard bill which is not modified to the corresponding auditor terminal.
The fourth authentication unit authenticates the non-standard bill according to all the amount images and the signature images, and comprises the following steps:
the fourth identification unit normalizes all the money images into matrixes with the same size to obtain a plurality of standard money images; normalizing all signature images into matrixes with the same size to obtain a plurality of standard signature images;
the fourth identification unit obtains the money amount characteristic vector and the signature characteristic vector according to the standard money amount image and the standard signature image respectively, and calculates a first money amount characteristic distance and a first signature characteristic distance according to the money amount characteristic vector and the signature characteristic vector respectively.
The fourth authentication unit performs image normalization processing on the money image and the signature image, normalizes the money image and the signature image into images with the same size, and improves authentication speed and authentication accuracy.
The fourth authentication unit authenticates the non-standard bill according to all the amount images and the signature images, and comprises the following steps:
the fourth identification unit respectively carries out characteristic distance transformation on the first money characteristic distance and the first signature characteristic distance to obtain a second money characteristic distance and a second signature characteristic distance;
the fourth identification unit removes the interference of the identification disturbance factor on the second money amount characteristic distance and the second signature characteristic distance to obtain a target money amount characteristic distance and a target signature characteristic distance;
the fourth identification unit respectively compares the target money amount characteristic distance and the target signature characteristic distance with a money amount characteristic distance threshold and a signature characteristic distance threshold;
and the fourth authentication unit judges that the corresponding non-standard bill is not tampered when the target money amount characteristic distance is smaller than the money amount characteristic distance threshold and the target signature characteristic distance is smaller than the signature characteristic distance threshold.
The fourth identification unit calculating the first money characteristic distance according to the money characteristic vector comprises:
wherein R is1Is a first amount of characteristic distance, giIs the characteristic value of the ith characteristic in the standard sum image, diThe characteristic value of the ith characteristic in the standard sum image is represented by i, and i is a characteristic index.
The fourth identification unit removing the interference of the identification disturbance factor on the second sum characteristic distance to obtain the target sum characteristic distance comprises:
the fourth discrimination unit establishes a total disturbance variable for discriminating the disturbance factor,
wherein s is a total disturbance variable, k is an index of an influence factor of a scanning type, l is an index of an influence factor of a scanning resolution, P is the number of the influence factors of the scanning type, and Q is the number of the influence factors of the scanning resolution. The identification perturbation factors include scan types and scan resolutions, and the scan types include black-and-white scanning and color scanning.
The fourth identification unit performs characteristic distance transformation on the first money characteristic distance to obtain a second money characteristic distance, and comprises:
the fourth identification unit performs characteristic distance conversion on the standard sum image to obtain a characteristic conversion map, and obtains the position coordinates of each characteristic point in the characteristic conversion map and the position coordinates of each characteristic point in the standard sum image;
the fourth identification unit traverses all the feature points in the feature transformation graph, takes the feature points which are being traversed as first feature points, acquires the feature points corresponding to the first feature points in the standard sum image and takes the feature points as second feature points;
the fourth identification unit calculates the characteristic distance between the first characteristic point and the second characteristic point according to the position coordinates of the first characteristic point and the second characteristic point; and carrying out weighted average on the feature distances of all the feature points in the feature transformation graph and the standard sum image to obtain a second sum feature distance.
The fourth identification unit performs characteristic distance transformation on the first money characteristic distance to obtain a second money characteristic distance, and comprises:
wherein R is2Is the second sum feature distance, i is the feature point index, ui(x, y) is the position coordinate of the ith feature point of the feature transformation map, vi(x, y) the position coordinates of the ith feature point of the standard sum image, and n is the number of the feature points.
In the invention, the interference of the identification disturbance factor on the second sum characteristic distance is removed to obtain the target sum characteristic distance, namely the influence of software and hardware interference factors on the scanning image imaging is removed, so that the accuracy of the electronic financial bill identification is improved. In practical applications, different scanning types and scanning resolutions affect the imaging of the scanned image, for example, the higher the scanning resolution of the scanner, the higher the definition of the electronic image generated by the scanning. On the contrary, the lower the definition of the electronic image, the greater the error between the electronic image and the actual paper image when extracting the features of the electronic image. The method removes the influence of the scanning type and the scanning resolution ratio on the accuracy of the target sum characteristic distance calculation result by removing the relevant identification disturbance factors.
The money characteristic distance threshold and the signature characteristic distance threshold are both thresholds preset by a manager according to actual conditions.
And simultaneously, identifying the authenticity of the electronic financial bill according to the target money amount characteristic distance and the target signature characteristic distance, namely carrying out multi-dimensional identification on the electronic financial bill. When the condition of multi-dimensional identification is met, the corresponding electronic financial bill is an untampered bill, namely when the target money amount characteristic distance and the target signature characteristic distance are both smaller than the corresponding characteristic distance threshold values, the corresponding untampered bill is judged to be untampered. The invention improves the accuracy of electronic financial bill identification by utilizing multi-dimensional identification.
The invention automatically identifies the authenticity of the non-standard electronic financial bill obtained by scanning by verifying whether the contents such as money amount, signature and the like in the electronic financial bill are maliciously modified by a person, so as to verify whether the contents of the electronic financial bill are maliciously tampered by the person. And the electronic financial bill which is not tampered is screened out for an auditing party to carry out manual auditing, so that the accuracy and efficiency of manual auditing are improved, and meanwhile, auditing holes and even huge losses of related departments caused by malicious tampering of the contents of the electronic financial bill by a party to be audited are avoided.
In one embodiment, a big data and intelligent finance-based financial bill validation system for performing the method of the present invention includes an auditor terminal, a candidate terminal, a bill processing server, and a bill authentication server. The bill authentication server includes: the authentication device comprises a first authentication unit, a second authentication unit, a third authentication unit and a fourth authentication unit, wherein communication connection is formed among the units.
And sending the electronic financial bill sent by the terminal of the party to be examined to a bill processing server.
The bill processing server judges the bill type of the electronic financial bill, acquires the pixel value of each pixel point in the non-standard bill when the electronic financial bill is the non-standard bill, and performs weighted summation on all components of each pixel point in the non-standard bill so as to perform first updating on the pixel value of each pixel point in the non-standard bill, thereby obtaining the first pixel value of each pixel point in the non-standard bill.
The bill processing server carries out second updating on the first pixel value of each pixel point in the non-standard bill according to the first pixel threshold value so as to obtain a second pixel value of each pixel point in the non-standard bill; acquiring second pixel values of all pixel points in the non-standard bill, and taking the pixel points with the second pixel values larger than a second pixel threshold value as feature points; acquiring the characteristic values of all the characteristic points in the non-standard bill, and generating a bill identification matrix according to the characteristic values of all the characteristic points in the non-standard bill;
the first authentication unit processes the non-standard bill according to the bill authentication matrix to obtain a plurality of first money images and first signature images.
The second identification unit obtains a first amount matrix and a second amount matrix according to the first amount image, and performs intersection operation on the first amount matrix and the second amount matrix to obtain a feature extraction matrix of each first amount image; and extracting the sum features of each first sum image according to the feature extraction matrix of each first sum image to obtain the sum image corresponding to each first sum image.
The third authentication unit obtains a first signature matrix and a second signature matrix according to the first signature image, and performs intersection operation on the first signature matrix and the second signature matrix to obtain a feature extraction matrix of each first signature image; and extracting signature characteristics of each first signature image according to the characteristic extraction matrix of each first signature image to obtain a signature image corresponding to each first signature image.
The fourth authentication unit authenticates the non-standard bill according to all the amount images and the signature images to judge whether the non-standard bill is modified or not, and sends the non-standard bill which is not modified to the corresponding auditor terminal.
The above embodiments are merely illustrative of the technical ideas and features of the present invention, and are intended to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be covered within the protection scope of the present invention.
Claims (9)
1. A financial bill verification method based on big data and intelligent finance is characterized in that an electronic financial bill sent by a to-be-audited terminal is received, and the bill type of the electronic financial bill is judged; the bill types comprise standard bills and non-standard bills;
when the electronic financial bill is a non-standard bill, acquiring a pixel value of each pixel point in the non-standard bill, and performing weighted summation on all components of each pixel point in the non-standard bill to perform first updating on the pixel value of each pixel point in the non-standard bill so as to obtain a first pixel value of each pixel point in the non-standard bill;
according to the first pixel threshold value, second updating is carried out on the first pixel value of each pixel point in the non-standard bill so as to obtain a second pixel value of each pixel point in the non-standard bill; acquiring second pixel values of all pixel points in the non-standard bill, and taking the pixel points with the second pixel values larger than a second pixel threshold value as feature points; acquiring the characteristic values of all the characteristic points in the non-standard bill, and generating a bill identification matrix according to the characteristic values of all the characteristic points in the non-standard bill; processing the non-standard bill according to the bill identification matrix to obtain a plurality of first money images and first signature images;
obtaining a first sum matrix and a second sum matrix according to the first sum image, and performing intersection operation on the first sum matrix and the second sum matrix to obtain a feature extraction matrix of each first sum image; extracting the sum features of each first sum image according to the feature extraction matrix of each first sum image to obtain a sum image corresponding to each first sum image;
obtaining a first signature matrix and a second signature matrix according to the first signature images, and performing intersection operation on the first signature matrix and the second signature matrix to obtain a feature extraction matrix of each first signature image; extracting signature characteristics of each first signature image according to the characteristic extraction matrix of each first signature image to obtain a signature image corresponding to each first signature image;
and the non-standard bill is identified according to all the amount images and the signature images so as to judge whether the non-standard bill is modified or not, and the non-standard bill which is not modified is sent to a corresponding auditor terminal.
2. The method of claim 1, wherein deriving the first matrix of amounts and the second matrix of amounts from the first moth-eye image comprises:
carrying out image segmentation on the non-standard bills according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first money images;
vertically projecting all the first sum images to obtain a plurality of second sum images; scanning each pixel point of each second sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first sum matrix;
horizontally projecting all the first sum images to obtain a plurality of third sum images; and scanning each pixel point of each third sum image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the second sum matrix.
3. The method of any of claims 1 to 2, wherein deriving the first signature matrix and the second signature matrix from the first signature image comprises:
carrying out image segmentation on the non-standard bill according to the arrangement sequence of the characteristic points in the bill identification matrix and the characteristic values of the characteristic points to obtain a plurality of first signature images;
vertically projecting all the first signature images to obtain a plurality of second signature images; scanning each pixel point of each second signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and putting the line number of the pixel point with the pixel value not being zero into the first signature matrix;
horizontally projecting all the first signature images to obtain a plurality of third signature images; and scanning each pixel point of each third signature image from top to bottom and from left to right, recording the line number of the pixel point with the pixel value not being zero, and placing the line number of the pixel point with the pixel value not being zero in the second signature matrix.
4. The method of claim 3, wherein authenticating the non-marking ticket based on all of the amount image and the signature image comprises:
normalizing all the amount images into matrixes with the same size to obtain a plurality of standard amount images; normalizing all signature images into matrixes with the same size to obtain a plurality of standard signature images;
and respectively obtaining a money amount characteristic vector and a signature characteristic vector according to the standard money amount image and the standard signature image, and respectively calculating a first money amount characteristic distance and a first signature characteristic distance according to the money amount characteristic vector and the signature characteristic vector.
5. The method of any one of claims 1 to 4, wherein authenticating the non-marking ticket based on all of the amount image and the signature image comprises:
respectively carrying out characteristic distance transformation on the first money characteristic distance and the first signature characteristic distance to obtain a second money characteristic distance and a second signature characteristic distance;
removing the interference of the identification disturbance factor on the second money amount characteristic distance and the second signature characteristic distance to obtain a target money amount characteristic distance and a target signature characteristic distance;
respectively comparing the target money amount characteristic distance and the target signature characteristic distance with a money amount characteristic distance threshold value and a signature characteristic distance threshold value;
and when the target money amount characteristic distance is smaller than the money amount characteristic distance threshold value and the target signature characteristic distance is smaller than the signature characteristic distance threshold value, judging that the corresponding non-standard bill is not tampered.
6. The method of claim 5, wherein the identified perturbation factors comprise a scan type and a scan resolution, the scan type comprising a black and white scan and a color scan.
7. The method of claim 6, wherein removing the perturbation to identify the perturbation factor from the second monetary characteristic distance to obtain the target monetary characteristic distance comprises:
establishing a total disturbance variable for identifying the disturbance factors,
wherein s is a total disturbance variable, k is an index of an influence factor of a scanning type, l is an index of an influence factor of a scanning resolution, P is the number of the influence factors of the scanning type, and Q is the number of the influence factors of the scanning resolution.
8. The method of claim 7, wherein performing a feature distance transformation on the first monetary feature distance to obtain the second monetary feature distance comprises:
performing characteristic distance transformation on the standard sum image to obtain a characteristic transformation image, and acquiring the position coordinate of each characteristic point in the characteristic transformation image and the position coordinate of each characteristic point in the standard sum image;
traversing all the feature points in the feature transformation graph, taking the feature points which are being traversed as first feature points, and acquiring the feature points corresponding to the first feature points in the standard sum image and taking the feature points as second feature points;
calculating the characteristic distance between the first characteristic point and the second characteristic point according to the position coordinates of the first characteristic point and the second characteristic point; and carrying out weighted average on the feature distances of all the feature points in the feature transformation graph and the standard sum image to obtain a second sum feature distance.
9. The method according to claim 8, characterized in that before the electronic financial bill is sent by the terminal of the party to be audited, the terminal of the auditor sends an audit notice to the terminal of the party to be audited; the audit notice comprises audit bill information and audit time information; the audit bill information comprises the information of relevant audit bills required to be prepared by audited units; the audit time information comprises an audit starting time and an audit period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111176641.9A CN114091517A (en) | 2021-10-09 | 2021-10-09 | Financial bill verification method based on big data and intelligent finance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111176641.9A CN114091517A (en) | 2021-10-09 | 2021-10-09 | Financial bill verification method based on big data and intelligent finance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091517A true CN114091517A (en) | 2022-02-25 |
Family
ID=80296663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111176641.9A Pending CN114091517A (en) | 2021-10-09 | 2021-10-09 | Financial bill verification method based on big data and intelligent finance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091517A (en) |
-
2021
- 2021-10-09 CN CN202111176641.9A patent/CN114091517A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11030752B1 (en) | System, computing device, and method for document detection | |
| US10607073B2 (en) | Systems and methods for classifying payment documents during mobile image processing | |
| US10885644B2 (en) | Detecting specified image identifiers on objects | |
| CN110046644B (en) | Certificate anti-counterfeiting method and device, computing equipment and storage medium | |
| US9230382B2 (en) | Document image capturing and processing | |
| US8290237B1 (en) | Systems and methods to use a digital camera to remotely deposit a negotiable instrument | |
| US8320657B1 (en) | Systems and methods to use a digital camera to remotely deposit a negotiable instrument | |
| KR102406432B1 (en) | Identity authentication methods and devices, electronic devices and storage media | |
| US20170287252A1 (en) | Counterfeit Document Detection System and Method | |
| WO2022156178A1 (en) | Image target comparison method and apparatus, computer device and readable storage medium | |
| WO2021259096A1 (en) | Identity authentication method, apparatus, electronic device, and storage medium | |
| CN113569863B (en) | Document checking method, system, electronic equipment and storage medium | |
| CN111259891A (en) | Method, device, equipment and medium for identifying identity card in natural scene | |
| US11393272B2 (en) | Systems and methods for updating an image registry for use in fraud detection related to financial documents | |
| CN112215225B (en) | KYC certificate verification method based on computer vision technology | |
| CN112560683A (en) | Method and device for identifying copied image, computer equipment and storage medium | |
| US20210090086A1 (en) | Systems and methods for fraud detection for images of financial documents | |
| CN114091517A (en) | Financial bill verification method based on big data and intelligent finance | |
| CN113077355B (en) | Insurance claim settlement method and device, electronic equipment and storage medium | |
| CN111046899B (en) | Identification card authenticity identification method, device, equipment and storage medium | |
| CN110929708A (en) | Method, equipment and storage medium for identifying national flag in Thai identity card | |
| US11900755B1 (en) | System, computing device, and method for document detection and deposit processing | |
| CN116109914B (en) | Method and device for identifying authenticity of bank running water image, electronic equipment and medium | |
| Patil et al. | A Machine Learning Based Approach for Image Quality Assessment of Forged Document Images | |
| CN113723212A (en) | Electronic contract verification system based on big data and intelligent finance |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20220225 |
|
| WD01 | Invention patent application deemed withdrawn after publication |