CN113971804A

CN113971804A - Signature falsification detection device and method, computing device and storage medium

Info

Publication number: CN113971804A
Application number: CN202010724315.6A
Authority: CN
Inventors: 陈爽月; 李海传; 李伟; 严昱超; 穆铁马; 罗琼; 谢长弘; 马恺琳; 陈宁华; 戚靓亮
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Group Zhejiang Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Group Zhejiang Co Ltd
Priority date: 2020-07-24
Filing date: 2020-07-24
Publication date: 2022-01-25

Abstract

The invention discloses a detection device, a method, a computing device and a storage medium for signature forgery, wherein the device comprises: including interaction layer, interface layer and service layer, wherein, service layer includes: the receiving module is suitable for receiving the to-be-detected signature picture and the character mark thereof from the interface layer scheduling; the first segmentation module is suitable for carrying out single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected; the first characteristic extraction module is suitable for extracting the Hash codes of the character pictures to be detected as character picture characteristics; and the comparison module is suitable for inquiring in the character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and determining that the signature picture to be detected is a forged signature picture if the character picture features of at least one character picture to be detected are similar to the inquired character picture features. The device can improve the precision and accuracy of detecting the forged signature picture.

Description

Signature falsification detection device and method, computing device and storage medium

Technical Field

The invention relates to the technical field of data processing, in particular to a device and a method for detecting signature forgery, a computing device and a storage medium.

Background

With the rapid development of social information technology, handwriting input becomes a natural and convenient human-computer interaction form. In the digital age, the method of using handwritten signatures on smart devices (mobile phones, tablets) has been applied to more and more protocol signing scenes, such as: the user signs his own name on the hand-written signature equipment to confirm the transaction information. The legal rights and interests of the first party and the second party are maintained, meanwhile, the business handling efficiency is improved, and convenience is brought to production and life of people. However, when the client collects the signature of the user, data compliance and security verification are lacked, the client is easily tampered with externally, and the existing production frequently verifies that the signature characters of different clients are completely consistent, namely, the signature picture is the situation of synthesizing single-character forged pictures, so that huge potential safety hazards are brought to the signature system, and the signature anti-counterfeiting technology becomes a key research direction for improving the security of the signature system.

The existing signature anti-counterfeiting related technology mainly comprises two types: an anti-counterfeiting method for handwritten character recognition and page coverage invisible signature certificates. The handwritten character recognition mainly recognizes the content of the handwritten character through two technologies of deep learning or character handwriting segmentation and is used for subsequent character verification. The deep learning handwriting recognition method comprises the following steps: under the premise of a large amount of handwritten character data and labels, the deep neural network training is used, so that the model learns the characteristic mode of each character, and the aim of accurately recognizing the handwritten character is fulfilled. The character handwriting segmentation method comprises the following steps: acquiring character handwriting; acquiring a segmentation model corresponding to the language category to which the character handwriting belongs; segmenting the character handwriting by using the segmentation model to obtain a segmentation block sequence; and identifying the segmentation block sequence to obtain a character string corresponding to the character handwriting.

The page covering hidden signature certificate anti-counterfeiting method comprises the steps of forming a signature image by the signature of a certificate, printing the signature image on an area, the size and the shape of which are the same as those of the signature image, on a certificate page by utilizing metamerism color matching ink, and printing the signature image on the area and original characters and images on the certificate page together in an overlapping mode to generate the signature image hidden on the certificate page. The original characters and images printed on the page of the certificate are shown under visible light, and the signature images printed on the page of the certificate are shown under infrared light, so that whether the certificate has a counterfeiting problem or not is effectively identified.

But the inventor discovers in the process of realizing the invention that: although the handwritten character recognition method can recognize the character content, the handwritten character of the user cannot be adaptively updated in a targeted manner, and whether a character template which has almost the same characteristics as the handwritten character exists or not can not be effectively monitored, so that whether the signature is forged or not is not judged according to the method; the anti-counterfeiting of the page-covered invisible signature certificate needs to use color matching ink to print on the certificate and other signature media, and is not adaptive to paperless signature scenes of the operator industry depending on printing technology and paper media.

Disclosure of Invention

In view of the above, the present invention has been made to provide a signature falsification detection apparatus, method, computing device and storage medium that overcome or at least partially solve the above problems.

According to an aspect of the present invention, there is provided a signature falsification detection apparatus, including an interaction layer, an interface layer, and a service layer, wherein the service layer includes:

the receiving module is suitable for receiving the to-be-detected signature picture and the character mark thereof from the interface layer scheduling;

the first segmentation module is suitable for carrying out single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected;

the first characteristic extraction module is suitable for extracting the Hash codes of the character pictures to be detected as character picture characteristics;

and the comparison module is suitable for inquiring in the character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and determining that the signature picture to be detected is a forged signature picture if the character picture features of at least one character picture to be detected are similar to the inquired character picture features.

Optionally, the apparatus further comprises:

the second segmentation module is suitable for segmenting the acquired historical forged signature picture to obtain at least one character picture;

the second characteristic extraction module is suitable for extracting the Hash codes of at least one character picture as character picture characteristics;

the storage module is suitable for storing the character and picture characteristics of at least one character and picture into a character characteristic library; and storing different character and picture characteristics corresponding to the same character in a characteristic table corresponding to the character.

Optionally, the first feature extraction module is further adapted to:

aiming at any character picture to be detected, calculating the difference between row pixels of the character picture to be detected by using a difference hash algorithm to obtain a difference matrix, and calculating the average value of all elements in the difference matrix; comparing each element in the difference matrix with the average value, determining a characteristic character string according to a comparison result, and converting the characteristic character string into hexadecimal to obtain a hash code of the character picture to be detected;

the second feature extraction module is further adapted to: aiming at any character picture, calculating the difference between the row pixels of the character picture by using a difference hash algorithm to obtain a difference matrix, and calculating the average value of all elements in the difference matrix; and comparing each element in the difference matrix with the average value, determining a characteristic character string according to a comparison result, and converting the characteristic character string into hexadecimal to obtain the Hash code of the character picture.

Optionally, the first segmentation module is further adapted to: carrying out binarization processing on the signature picture to be detected, carrying out connected domain analysis on the signature picture to be detected after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture to be detected;

the second segmentation module is further adapted to: and carrying out binarization processing on the historical forged signature picture, carrying out connected domain analysis on the historical forged signature picture after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture.

Optionally, the apparatus further comprises:

and the filtering module is suitable for judging whether the number of different connected domains of the to-be-detected signature picture is consistent with the number of the character marks or not, and if not, filtering the to-be-detected signature picture.

Optionally, the apparatus further comprises:

and the self-adaptive adding module is suitable for storing the character and picture characteristics of the character and picture to be detected except the character and picture to be detected with character and picture characteristics similar to the inquired character and picture characteristics into the character and picture characteristic library.

Optionally, the apparatus further comprises:

the self-adaptive deleting module is suitable for sequencing the characteristics of the character pictures according to the frequency of detecting that the characteristics of the character pictures in the character characteristic library are similar to the characteristics of the character pictures to be detected; and the number of the first and second groups,

and when the number of the character and picture characteristics in the character characteristic library is detected to exceed a preset value, deleting the character and picture characteristics which are detected to be similar to the character and picture characteristics of the character and picture to be detected and have the frequency lower than the preset value.

According to another aspect of the present invention, there is provided a method for detecting signature falsification, the method being applied to a service layer and including:

receiving a signature picture to be detected and a character mark thereof from interface layer scheduling;

carrying out single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected;

extracting the Hash codes of the character pictures to be detected as character picture characteristics;

and inquiring in a character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and if the character picture features of at least one character picture to be detected are similar to the inquired character picture features, determining that the signature picture to be detected is a forged signature picture.

Optionally, the method further comprises:

the acquired historical forged signature picture is subjected to segmentation processing to obtain at least one character picture;

extracting a Hash code of at least one character picture as character picture characteristics;

storing the character and picture characteristics of at least one character and picture into a character characteristic library; and storing different character and picture characteristics corresponding to the same character in a characteristic table corresponding to the character.

Optionally, extracting the hash code of each to-be-detected text picture as the text picture feature further includes:

extracting the hash code of at least one text picture as the text picture feature further comprises:

aiming at any character picture, calculating the difference between the row pixels of the character picture by using a difference hash algorithm to obtain a difference matrix, and calculating the average value of all elements in the difference matrix; and comparing each element in the difference matrix with the average value, determining a characteristic character string according to a comparison result, and converting the characteristic character string into hexadecimal to obtain the Hash code of the character picture.

Optionally, the processing of single character segmentation on the signature picture to be detected to obtain each text picture to be detected further includes: carrying out binarization processing on the signature picture to be detected, carrying out connected domain analysis on the signature picture to be detected after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture to be detected;

the obtained historical forged signature picture is segmented, and the obtaining of at least one character picture further comprises: and carrying out binarization processing on the historical forged signature picture, carrying out connected domain analysis on the historical forged signature picture after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture.

Optionally, the method further comprises: and judging whether the number of different connected domains of the signature picture to be detected is consistent with the number of the character marks, and if not, filtering the signature picture to be detected.

Optionally, the method further comprises: and storing the character and picture characteristics of other character and pictures to be detected except the character and picture characteristics to be detected which are similar to the inquired character and picture characteristics into a character and picture characteristic library.

Optionally, the method further comprises: sequencing the characteristics of the character pictures according to the frequency of detecting that the characteristics of the character pictures in the character characteristic library are similar to the characteristics of the character pictures to be detected; and deleting the character and picture characteristics which are detected to be similar to the character and picture characteristics of the character and picture to be detected and have the frequency lower than the preset value when the number of the character and picture characteristics in the character and picture characteristic library is detected to exceed the preset value.

According to yet another aspect of the present invention, there is provided a computing device comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the signature falsification detection method.

According to still another aspect of the present invention, there is provided a computer storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform operations corresponding to the above-mentioned signature falsification detection method.

According to the detection device, method, computing equipment and storage medium of the signature forgery, the device comprises: including interaction layer, interface layer and service layer, wherein, service layer includes: the receiving module is suitable for receiving the to-be-detected signature picture and the character mark thereof from the interface layer scheduling; the first segmentation module is suitable for carrying out single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected; the first characteristic extraction module is suitable for extracting the Hash codes of the character pictures to be detected as character picture characteristics; and the comparison module is suitable for inquiring in the character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and determining that the signature picture to be detected is a forged signature picture if the character picture features of at least one character picture to be detected are similar to the inquired character picture features. By the method, the problem that the existing signature counterfeiting method cannot solve the problem that an electronic signature and a forged character have no fixed template in an actual business scene is solved, the individual character picture to be detected and the individual character picture in the feature library are directly compared through feature difference without comparison between pictures, the processing speed can be increased, the precision and the accuracy of detecting and synthesizing the forged signature picture can be improved, and the method has transportability and expansibility.

The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

fig. 1 is a schematic structural diagram of a signature falsification detection apparatus provided by an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a signature falsification detection apparatus according to another embodiment of the present invention;

FIG. 3a shows an example of a spurious composite signature;

FIG. 3b shows an example of an invalid signature;

fig. 4 is a schematic flow chart illustrating a method for detecting signature falsification according to another embodiment of the present invention:

FIG. 5 is a flow chart illustrating a method for detecting forgery of a signature according to another embodiment of the present invention;

fig. 6 shows a schematic structural diagram of a computing device provided by an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

Signature synthesis forgery refers to the situation that single character forgery exists in a signature picture and the signature picture is synthesized. Fig. 3a shows an example of a false composite signature, as shown in fig. 3a, the handwriting of the "li" word in the two signature pictures, i.e., "li menna" and "lie" are identical, and then the two signatures may be synthesized by using the text template of "li".

Fig. 1 shows a schematic structural diagram of a signature falsification detection apparatus provided in an embodiment of the present invention, and as shown in fig. 1, the apparatus includes an interaction layer 11, an interface layer 12, and a service layer 13.

The interaction layer 11 provides a micro-service interface with a restful style to the outside, so that the micro-service interface can be conveniently and quickly called by any client application, and has portability and expansibility, and the interface layer 12 provides functions of receiving requests, calling service modules, checking parameters, recording logs, retaining data and the like. The service layer 23 includes the following modules: a receiving module 131, a first segmentation module 132, a first feature extraction module 133, and a comparison module 134.

The receiving module 131 is adapted to receive the to-be-detected signature picture and the text label thereof from the interface layer. The character marks of the signature picture to be detected refer to each character in the signature picture to be detected.

The interface layer provides functions of receiving requests, calling service modules, checking parameters, recording logs, storing data and the like.

The first segmentation module 132 is adapted to perform single character segmentation on the signature picture to be detected to obtain each text picture to be detected. And carrying out single character segmentation processing on the received signature picture to be detected, segmenting the signature picture to be detected into a plurality of single character pictures, and obtaining each character picture to be detected.

The first feature extraction module 133 is adapted to extract the hash code of each text picture to be detected as a text picture feature. And aiming at each character picture to be detected, determining the hash code of the character picture to be detected by adopting a preset hash algorithm.

The comparison module 134 is adapted to query the text feature library according to the text labels, compare the text image features of each text image to be detected with the queried text image features, and determine that the signature image to be detected is a counterfeit signature image if the text image features of at least one text image to be detected are similar to the queried text image features.

The character feature library stores character and picture features (hash codes) corresponding to the individual characters, that is, a character template, for example, the character and picture features corresponding to a character and picture including different fonts and different writing styles of "li" characters. For any character picture to be detected, inquiring a plurality of character picture characteristics of corresponding characters in a character characteristic library according to the character mark of the character picture, then comparing the character picture characteristics of the character picture to be detected with the characteristics of the inquired character pictures, calculating whether the character picture characteristics of the character picture to be detected are similar to the characteristics of the inquired character pictures, and if the character picture characteristics of the character picture to be detected are similar to the character picture characteristics in the character characteristic library, considering that the character picture to be detected is successfully compared.

The logic for judging whether the signature picture to be detected is a forged signature picture is as follows: and if the character picture characteristics of at least one character picture to be detected are similar to the character picture characteristics in the character feature library, namely the at least one character picture to be detected is successfully compared, determining that the signature picture to be detected is a synthesized fake signature picture.

For example, if the characters of the signature picture to be detected are marked as "li", all character and picture characteristics corresponding to "li" are inquired in the character library, then the character and picture characteristics of the character and picture to be detected are compared with the character and picture characteristics corresponding to "li", if the characters and picture characteristics are similar, the characters and picture "li" in the signature picture to be detected have almost the same character template in the feature library, and if the characters and picture "li" in the signature picture to be detected are synthesized, the signature picture to be detected is determined to be a forged signature picture.

According to the signature forgery detection device provided by the embodiment, the signature picture to be detected is divided into a plurality of single character pictures to be detected by taking characters as granularity by receiving the signature picture to be detected and the character marks of the signature picture to be detected from the interface layer, then the character picture characteristics of each single character picture to be detected are extracted, the character picture characteristics in the character characteristic library are inquired according to the character marks, finally the character picture characteristics of each single character picture to be detected are compared with the inquired character picture, and if the character picture characteristics of at least one single character picture to be detected are similar to the inquired character picture characteristics, the signature picture to be detected is determined to be a forged signature picture. By the method, the problem that the existing signature counterfeiting method cannot solve the problem that an electronic signature and a forged character have no fixed template in an actual business scene is solved, the individual character picture to be detected and the individual character picture in the feature library are directly compared through feature difference without comparison between pictures, the processing speed can be increased, the precision and the accuracy of detecting and synthesizing the forged signature picture can be improved, and the method has transportability and expansibility.

Fig. 2 is a schematic structural diagram illustrating a schematic structural diagram of a signature falsification detection apparatus according to another embodiment of the present invention, and as shown in fig. 2, the apparatus includes an interaction layer 21, an interface layer 22, and a service layer 23.

The interaction layer 21 provides a micro-service interface with a restful style to the outside, so that the micro-service interface can be conveniently and quickly called by any client application, and has portability and expansibility, and the interface layer 22 provides functions of receiving a request, calling a service module, checking parameters, recording logs, retaining data and the like. The service layer 23 includes the following modules: a second segmentation module 231, a second feature extraction module 232, a storage module 233, a receiving module 234, a first segmentation module 235, a filtering module 236, a first feature extraction module 237, a comparison module 238, an adaptive addition module 239, and an adaptive deletion module 2310.

The service layer 23 is the core of the device of this embodiment, and the main functions of the service layer include: generating a forged character basic library, detecting the forging of a signature, deleting and adding the character basic library. Which will be separately described below.

Regarding the generation of the forged character basic library (i.e. the generation of the character feature library), the second segmentation module 231, the second feature extraction 232 module and the storage module 233 are mainly used for the generation of the forged character basic library, and specifically:

the second segmentation module 231 is adapted to perform single character segmentation processing on the acquired historical forged signature picture to obtain at least one text picture.

In order to detect whether the signature belongs to synthesis forgery or not, the prototype of the forged single character, namely the template character, needs to be determined.

Specifically, the second segmentation module 231 performs binarization processing on the acquired historical forged signature picture, performs connected domain analysis on the binarized historical forged signature picture, and performs segmentation processing on different connected domains to obtain each character picture.

For example, binarization processing is carried out on an acquired historical forged signature picture, connected domain analysis is carried out on the signature in the picture by using an 8-neighborhood pixel method, one connected domain is composed of foreground pixel points with the same pixel value and adjacent positions and represents a single character, and the signature picture is divided into single character pictures by means of dividing different connected domains.

Optionally, to facilitate the representation of the picture hash feature, the second segmentation module 231 is further adapted to: and zooming each character picture into a preset size, and carrying out picture graying processing on each zoomed character picture. For example, the divided individual character pictures are collectively scaled to 12 × 11 size, and the scaled pictures are grayed.

The second extraction feature module 232 is adapted to extract a hash code of at least one text picture as a text picture feature.

Specifically, for any character picture, the difference between the row pixels of the character picture is calculated by using a difference hash algorithm to obtain a difference matrix, and the average value of all elements in the difference matrix is calculated; and comparing each element in the difference matrix with the average value, determining a characteristic character string according to a comparison result, and converting the characteristic character string into hexadecimal to obtain the Hash code of the character picture.

Following the above example, using a DHA algorithm (difference hash algorithm), a difference matrix G with a size of 11 × 11 is obtained by calculating a difference value between pixels in a row, calculating an average value a of all elements in the G matrix, representing a character string after difference calculation by dhash, traversing pixels of the matrix G from left to right in rows, if G (i, j) > a, then dhash + '1', if G (i, j) < a, then dhash + '0', obtaining a character string composed of 121 digits after difference calculation, converting the character string into hexadecimal, and then obtaining a hash code corresponding to a single-character picture.

The storage module 233 is adapted to store the text and image characteristics of at least one text and image into a text and image characteristic library; and storing different character and picture characteristics corresponding to the same character in a characteristic table corresponding to the character. And each single character constructs a data table, and different character and picture characteristics corresponding to the same single character are stored in the data table corresponding to the single character, so that a character characteristic library is generated.

The receiving module 234 receives the to-be-detected signature picture and the text label thereof from the interface layer scheduling.

The first segmentation module 235 is adapted to perform single character segmentation on the signature picture to be detected to obtain each text picture to be detected.

The first segmentation module and the second segmentation module are similar in principle, and only the processed object is the to-be-detected signature picture, specifically, the to-be-detected signature picture is subjected to binarization processing, connected domain analysis is performed on the to-be-detected signature picture after binarization processing, and different connected domains are subjected to segmentation processing to obtain each to-be-detected character picture.

And the filtering module 236 is adapted to determine whether the number of different connected domains of the signature picture to be detected is consistent with the number of the text marks, and if not, filter the signature picture to be detected.

The filtering module plays a role in filtering invalid signatures, and combines the distance characteristics between single characters appearing in a signature synthesis and counterfeiting scene, when a picture is divided, the number of different connected domains in the same picture is counted, if the number of the different connected domains is inconsistent with the number of character marks, the situation that continuous strokes or deletion possibly exist is judged, synthesis and counterfeiting cannot occur under the situation, the signature to be detected is judged to be invalid, the filtering of the invalid signatures is better realized, and the identification accuracy is improved. Fig. 3b shows an example of an invalid signature in which the individual words are hyphenated and typically not synthetically forged.

The first feature extraction module 237 is adapted to extract hash codes of the text pictures to be detected as text picture features.

The principle of the first feature extraction module is similar to that of the second feature extraction module, except that the processed object is each character picture to be detected, which is obtained by dividing the signature picture to be detected, specifically, for any character picture to be detected, the difference between the row pixels of the character picture to be detected is calculated by using a difference hash algorithm, so as to obtain a difference matrix, and the average value of all elements in the difference matrix is calculated; and comparing each element in the difference matrix with the average value, determining a characteristic character string according to a comparison result, and converting the characteristic character string into hexadecimal to obtain the Hash code of the character picture to be detected.

The comparison module 238 is adapted to query the text feature library according to the text labels, compare the text image features of each text image to be detected with the queried text image features, and determine that the signature image to be detected is a counterfeit signature image if the text image features of at least one text image to be detected are similar to the queried text image features.

The comparison module 238 mainly realizes the query and comparison of the signature picture to be detected and the character feature library. Specifically, for any character picture to be detected, inquiring a plurality of character picture features in a character feature library according to the character mark of the character picture to be detected, calculating the similarity between the character picture features of the character picture to be detected and the characteristics of each inquired character picture, if the similarity is greater than a similarity threshold value, determining that the character picture to be detected is similar to the characteristics of the inquired character picture, successfully comparing the character picture to be detected, otherwise, indicating that the character picture to be detected is unsuccessfully compared.

Specifically, a hamming distance dist between the character picture features of the character picture to be detected and the inquired character picture features is calculated, and then the similarity between the character picture features and the inquired character picture features is calculated according to a formula similarity sim ═ 1-dist/(n × n), wherein n represents the transverse number of the differential matrix. Wherein, the similarity threshold is obtained according to the existing negative data and the mark.

The logic of the comparison module 238 determining whether the signature picture to be detected is a fake signature picture is as follows: and if the character picture characteristics of at least one character picture to be detected are similar to the character picture characteristics in the character feature library, namely the at least one character picture to be detected is successfully compared, determining that the signature picture to be detected is a synthesized fake signature picture.

Optionally, the comparison module 238 is further adapted to: firstly, all the character pictures to be detected corresponding to the signature pictures to be detected are compared, whether the character pictures to be detected are successfully compared or not is judged, if all the character pictures to be detected are successfully compared, the single characters in the signature pictures to be detected are forged and synthesized, and the signature pictures to be detected are directly judged to be signature fake. If the partial comparison is successful, the single characters in the signature picture to be detected are forged and synthesized, and the signature picture to be detected is still judged to be signature fake.

Optionally, the apparatus further comprises: the result returning module is suitable for returning the result of whether the signature picture to be detected is a forged signature picture to the interface layer; and returning the position information of the character picture to be detected, which has character picture characteristics similar to the inquired character picture characteristics, corresponding to the signature picture to be detected to the interface layer.

In the device of the embodiment, each character is correspondingly stored with the picture characteristics of the character in different writing methods, the picture characteristics of the character are quickly searched through single character marks, the characters to be detected and the characters in the character characteristic library are compared by using the difference of the hash characteristics, the comparison between the pictures is not needed, the picture difference is quickly and effectively judged, and therefore the forged signature picture can be positioned.

The adaptive adding module 239 is adapted to store the text and picture characteristics of the text and picture to be detected, except the text and picture characteristics similar to the inquired text and picture characteristics, in the text and picture characteristic library.

The adaptive adding module 239 mainly expands the character templates in the character feature library, determines the content to be added to the character feature library according to the result of the comparison module, and specifically stores the character and picture features of the character and picture to be detected, except the successfully compared character and picture to be detected, in the character feature library. If the characters corresponding to the character picture to be detected exist in the character feature library, directly adding the character picture features to a feature table corresponding to the characters; and if the characters corresponding to the character picture to be detected do not exist in the character feature library, creating a character feature table according to the single character marks, and adding the character picture features into the created feature table.

The reason why the incremental data are continuously recorded on the character feature library of the stock data in the device of the embodiment is as follows: (1) the signature synthesis has randomness, the fake words do not have fixed templates, and the sources of the fake words are unknown; (2) different channels use the same character to synthesize a plurality of signatures to implement counterfeiting behaviors, so that the character feature library is continuously enlarged, and the counterfeiting hit rate can be improved.

The self-adaptive deleting module 2310 is suitable for sequencing the character and picture characteristics according to the frequency of detecting that the character and picture characteristics of the character and picture in the character characteristic library are similar to the character and picture characteristics of the character and picture to be detected; and deleting the character and picture characteristics which are detected to be similar to the character and picture characteristics of the character and picture to be detected and have the frequency lower than the preset value when the number of the character and picture characteristics in the character and picture characteristic library is detected to exceed the preset value.

Unlimited addition of text picture features to the text feature library results in an overflow of storage space and an increase in comparison time, based on this. The device of the embodiment also provides a self-adaptive deleting module for properly deleting the character and picture characteristics in the character characteristic library.

For example, for one character picture characteristic of the "li" in the character characteristic library, when three signature pictures to be detected are detected, the character picture characteristic of the "li" in the three pictures to be detected is similar to the character picture characteristic of the "li" in the character characteristic library, that is, the character picture characteristic of the "li" is used for forging the signature for three times. And (4) according to the frequency of the character and picture characteristics detected to synthesize the forged signature, arranging the character and picture characteristics in sequence from big to small according to the frequency.

Considering that the text and picture features added to the text feature library are not counterfeit frequently, the text and picture features which are used for synthesizing the forged signatures frequently can be deleted. Specifically, when the number of the character and picture features stored in the character feature library exceeds the limit, the character and picture features with the frequency lower than the preset value are deleted. By the method, the character and picture features which are used for forging the signature and are low in frequency are deleted, so that useless characters are deleted, and new space is released for adding new potential character and picture features.

According to the detection device for signature falsification provided by the embodiment, the device provides an interface for any client to call through the interaction layer, and has portability and expansibility. By processing the signature picture and converting the signature picture into a characteristic sequence to be stored in a database, when the signature picture to be detected is detected, on one hand, invalid signatures are filtered according to a connected domain in the picture, so that the detection accuracy can be improved, on the other hand, the single character picture to be detected and the single character picture in the characteristic library are directly compared through the characteristics, the comparison between the pictures is not needed, the detection time can be shortened, the forged signature picture can be quickly and effectively positioned, and the detection accuracy can be ensured; in addition, the content in the character feature library can be adaptively added and deleted, and the hit rate of detecting the synthetic forged signature can be improved.

Fig. 4 is a schematic flow chart of a method for detecting signature falsification according to another embodiment of the present invention, as shown in fig. 4, the method includes the following steps:

and step S410, receiving the to-be-detected signature picture and the character mark thereof from the interface layer scheduling.

And step S420, carrying out single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected.

Step S430, extracting the Hash codes of the character pictures to be detected as character picture characteristics.

Step S440, inquiring in the character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and if the character picture features of at least one character picture to be detected are similar to the inquired character picture features, determining that the signature picture to be detected is a forged signature picture.

By the method, the problem that the existing signature counterfeiting method cannot solve the problem that an electronic signature and a forged character have no fixed template in an actual business scene is solved, the individual character picture to be detected and the individual character picture in the feature library are directly compared through feature difference without comparison between pictures, the processing speed can be increased, the precision and the accuracy of detecting and synthesizing the forged signature picture can be improved, and the method has transportability and expansibility.

Fig. 5 is a schematic flow chart of a signature falsification detection method according to another embodiment of the present invention, which is applied to a service layer, and as shown in fig. 5, the method includes the following steps:

step S501, receiving the signature picture to be detected and the character mark thereof from the interface layer scheduling.

And step S502, carrying out binarization processing on the signature picture to be detected, carrying out connected domain analysis on the signature picture to be detected after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture to be detected.

Step S503, judging whether the number of different connected domains of the to-be-detected signature picture is consistent with the number of the character marks, if so, executing step S505; if not, go to step S504.

Step S504, the signature is confirmed to be invalid, and the execution of step S513 is skipped. And if the number of different connected domains of the picture of the signature to be detected is inconsistent with the number of the character marks, confirming that the signature to be detected is invalid, and outputting the result that the signature to be detected is invalid to the interface layer.

And step S505, extracting the hash codes of the character pictures to be detected as character picture characteristics.

Step S506, inquiring in the character feature library according to the character marks, and comparing the character picture features of each character picture to be detected with the inquired character picture features.

Step S507, judging whether all the character pictures to be detected are successfully compared; if yes, go to step S508; if not, step S509 is executed.

Step S508, confirming that the signature picture to be detected is a forged signature picture, and skipping to execute step S513; and if all the character pictures to be detected are successfully compared, directly confirming that the signature picture to be detected is a fake signature picture, and outputting a result that the signature picture to be detected is the fake signature picture to the interface layer.

Step S509, determining whether the comparison of the partial text pictures to be detected is successful, if yes, executing step S510; if not, executing step S511;

and step S510, confirming that the signature picture to be detected is a forged signature picture, marking the position of a forged single character, and skipping to execute the step S512 and the step S513. If some character pictures to be detected are successfully compared, the signature picture to be detected is confirmed to be a forged signature picture, the position information of a forged single character in the signature picture to be detected is determined, the result is output by the interface layer, and the character picture characteristics corresponding to the normal characters are added to the character characteristic library.

And step S511, determining that the signature picture to be detected is a normal signature picture, and skipping to execute the step S512 and the step S513. And if the comparison of the text pictures to be detected is not successful, determining that the signature picture to be detected is a normal signature picture, adding the text picture characteristics corresponding to the normal text into a text characteristic library, and outputting the result that the signature picture to be detected is the normal signature picture to the interface layer.

Therefore, by processing the signature picture and converting the signature picture into a feature sequence to be stored in the database, when the signature picture to be detected is detected, on one hand, invalid signatures are filtered according to a connected domain in the picture, so that the detection accuracy can be improved, on the other hand, the single character picture to be detected and the single character picture in the feature library are directly compared through the features, the comparison between the pictures is not needed, the detection time can be shortened, the forged signature picture can be quickly and effectively positioned, and the detection accuracy can be ensured; in addition, the content in the character feature library can be adaptively added and deleted, and the hit rate of detecting the synthetic forged signature can be improved.

Embodiments of the present invention provide a non-volatile computer storage medium, where at least one executable instruction is stored in the computer storage medium, and the computer executable instruction may execute the signature falsification detection method in any of the above method embodiments.

The executable instructions may be specifically configured to cause the processor to:

inquiring in a character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and if the character picture features of at least one character picture to be detected are similar to the inquired character picture features, determining that the signature picture to be detected is a forged signature picture

In an alternative, the executable instructions cause the processor to:

carrying out binarization processing on the signature picture to be detected, carrying out connected domain analysis on the signature picture to be detected after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture to be detected;

and carrying out binarization processing on the historical forged signature picture, carrying out connected domain analysis on the historical forged signature picture after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture.

In an alternative, the executable instructions cause the processor to: and judging whether the number of different connected domains of the signature picture to be detected is consistent with the number of the character marks, and if not, filtering the signature picture to be detected.

In an alternative, the executable instructions cause the processor to: and storing the character and picture characteristics of other character and pictures to be detected except the character and picture characteristics to be detected which are similar to the inquired character and picture characteristics into a character and picture characteristic library.

In an alternative, the executable instructions cause the processor to: sequencing the characteristics of the character pictures according to the frequency of detecting that the characteristics of the character pictures in the character characteristic library are similar to the characteristics of the character pictures to be detected; and deleting the character and picture characteristics which are detected to be similar to the character and picture characteristics of the character and picture to be detected and have the frequency lower than the preset value when the number of the character and picture characteristics in the character and picture characteristic library is detected to exceed the preset value.

Fig. 6 is a schematic structural diagram of an embodiment of a computing device according to the present invention, and a specific embodiment of the present invention does not limit a specific implementation of the computing device.

As shown in fig. 6, the computing device may include: a processor (processor)602, a communication Interface 604, a memory 606, and a communication bus 608.

Wherein: the processor 602, communication interface 604, and memory 606 communicate with one another via a communication bus 608. A communication interface 604 for communicating with network elements of other devices, such as clients or other servers. The processor 602, configured to execute the program 610, may specifically perform relevant steps in the above-described embodiment of the method for detecting signature falsification of the computing device.

In particular, program 610 may include program code comprising computer operating instructions.

The processor 602 may be a central processing unit CPU or an application Specific Integrated circuit asic or one or more Integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.

And a memory 606 for storing a program 610. Memory 606 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

The program 610 may specifically be configured to cause the processor 602 to perform the following operations:

In an alternative, the program 610 causes the processor 602 to:

optionally, the method further comprises:

The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components according to embodiments of the present invention. The present invention may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specified otherwise.

Claims

1. An apparatus for detecting forgery of a signature, the apparatus comprising an interaction layer, an interface layer, and a service layer, wherein the service layer comprises:

and the comparison module is suitable for inquiring in the character feature library according to the character marks, comparing the character picture features of each character picture to be detected with the inquired character picture features, and if the character picture features of at least one character picture to be detected are similar to the inquired character picture features, determining that the signature picture to be detected is a forged signature picture.

2. The apparatus of claim 1, wherein the apparatus further comprises:

the second characteristic extraction module is suitable for extracting the Hash codes of the at least one character picture as character picture characteristics;

the apparatus further comprises: the storage module is suitable for storing the character and picture characteristics of the at least one character and picture into a character characteristic library; and storing different character and picture characteristics corresponding to the same character in a characteristic table corresponding to the character.

3. The apparatus of claim 1 or 2, wherein the first feature extraction module is further adapted to:

4. The apparatus of claim 2, wherein the first segmentation module is further adapted to: carrying out binarization processing on the signature picture to be detected, carrying out connected domain analysis on the signature picture to be detected after binarization processing, and carrying out segmentation processing on different connected domains to obtain each character picture to be detected;

5. The apparatus of claim 4, wherein the apparatus further comprises:

6. The apparatus of claim 1, wherein the apparatus further comprises:

7. The apparatus of claim 1, wherein the apparatus further comprises:

8. A method for detecting signature forgery, the method being applied to a service layer and comprising:

performing single character segmentation processing on the signature picture to be detected to obtain each character picture to be detected;

9. A computing device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operation corresponding to the signature falsification detection method according to claim 8.

10. A computer storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the method of detecting signature falsification recited in claim 8.