CN115841703A - Signing intention recognition method, system, equipment and medium based on handwriting characteristics - Google Patents
Signing intention recognition method, system, equipment and medium based on handwriting characteristics Download PDFInfo
- Publication number
- CN115841703A CN115841703A CN202211410547.XA CN202211410547A CN115841703A CN 115841703 A CN115841703 A CN 115841703A CN 202211410547 A CN202211410547 A CN 202211410547A CN 115841703 A CN115841703 A CN 115841703A
- Authority
- CN
- China
- Prior art keywords
- signature
- intention
- sample
- passive
- active
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The application discloses a method for recognizing signing willingness based on handwriting characteristics, wherein signing names in an active sample storage and a passive sample storage are mutually exclusive; and acquiring a handwritten electronic signature of a user on line, comparing and checking the handwritten electronic signature with a corresponding signature in the active intention sample storage, if the signature is consistent with the active intention sample storage, executing active intention setting, if the signature is inconsistent with the active intention sample storage, further comparing and checking the signature in the passive intention sample storage, if the signature is consistent with the passive intention sample storage, executing passive intention setting, starting setting early warning, and if the number of inconsistent check exceeds the preset number, locking the signature. The two will samples are kept consistent on static characteristics such as writing methods as much as possible, but dynamic characteristics are different, and the two will samples are mutually exclusive through handwriting verification algorithm verification. Is difficult to imitate, and is a passive sample keeping method with privacy and safety. Is beneficial to protecting the personal safety and the property safety of the stressed signer, and the like.
Description
Technical Field
The invention relates to the technical field of information, in particular to a method for recognizing signing will based on handwriting characteristics.
Background
With the development of informatization and digitization, signature carriers are gradually changed from traditional paper to electronic equipment such as a handwriting board, an electronic screen and the like, and electronic signatures in an electronic data storage form appear in a signature mode besides traditional paper and pen written signatures. The electronic signature legalization is promulgated by the electronic signature law of the people's republic of China, and the legal effectiveness of reliable electronic signatures and traditional signature stamping is determined to be equal. Electronic signatures are widely used in government affairs, medical treatment, bank finance, fast-moving, real estate, logistics and other fields, but how to ensure the reliability and authenticity of electronic signatures? Generally, when a relevant business process is performed, a user signs an electronic signature on an electronic document; at the moment, the business system defaults that the user is actively signed, defaults that the operation is expressed by the real intention of the user, does not consider the environmental problem and the personal safety problem of the site of the user, and does not solve the intention expression problems of whether the user is actively signed, stressed and the like, so that electronic files under certain non-active intentions are signed, and the authenticity, the legality and the reliability of the electronic signature are reduced.
The chinese patent application with publication number CN110348466A, entitled "method and apparatus for identity recognition" discloses a method and apparatus for identity recognition, which obtains the current passive features of the current user, matches the current passive features with preset passive features in a feature library, and determines the matching degree of the current passive features; when the matching degree is not smaller than a first threshold value, identifying the identity of the current user based on the current passive features; and when the matching degree is smaller than a first threshold value, identifying the identity of the current user based on the active features of the current user. The identity recognition method with the combination of active verification and passive features is adopted, and the inter-partition processing is carried out according to the matching degree of the current passive features, so that the accuracy and the safety of identity recognition and the user experience are improved.
The passive features are features of the user that can be obtained without active interaction of the user, and are used for improving the accuracy of identity authentication.
In the application of the chinese patent application, which is entitled "online handwriting authentication method with attacker identity recognition capability", in the online handwriting authentication stage of the authentication system, if the user to be logged in fails to pass the authentication, the physiological characteristic information related to the writing hand in the test handwriting characteristic information submitted by the user to be logged in is compared with the physiological characteristic information related to the writing hand in the external physiological characteristic information database to recognize the real identity of the user to be logged in. The replay attack can be resisted; an attacker cannot manufacture test handwritten characteristic information in advance; the ability to identify an attacker. Comparing the separated handwriting information and physiological characteristic information related to the handwriting with corresponding registered template information in a handwriting characteristic information database self-established by the authentication system, and judging whether the user to be logged in passes authentication or not according to a comparison result; if the authentication is not passed and the character recognition algorithm indicates that the handwriting is consistent with the presented registered standard characters, comparing the physiological characteristic information related to the handwriting hand in the test handwriting characteristic information submitted by the user to be logged with the physiological characteristic information related to the handwriting hand in the peripheral physiological characteristic information database to recognize the real identity of the user to be logged.
The prior art disclosed above mainly aims at identity authentication of signers, and optimizes algorithms or verification strategies implemented by improving accuracy, replay attack, repeated copy attack, and the like. In addition to the identity authentication function, signature recognition (authentication) also has a function of willingness to authenticate, which is an important distinguishing point of signature recognition and other biological characteristics. None of the above documents mention the problem of willingness to sign, or the signature that can be authenticated by signature is by default written at voluntary. However, some signatures are generated under duress and are not expressed by the active will of the signer, and such signatures are called passive-wished signatures. For example, a criminal suspect who holds a knife threatens a signer requires that the signature passes through a verification system, otherwise, the security is threatened. In this case, the signer would like to be able to sign the signature through the system without endangering his security, and would like to be recognized as an expression of the passive intention, and the system could respond to the relevant settings according to the passive intention, such as protecting property, secret alarm measures, evidence retention, etc. The prior art does not mention willingness recognition.
Disclosure of Invention
The application adopts a new handwriting recognition scheme to realize recognition of signature intentions, particularly recognition of passive intentions, so as to achieve the purposes of protecting personal safety and property safety of the stressed signer and the like.
In view of the above, the application provides a method for recognizing signing willingness based on handwriting characteristics, which includes verifying user identity through information such as an identity card and a verification code, sending a initiative willingness sample reserving prompt, collecting multiple electronic signatures of a user, storing the electronic signatures into a sample reserving library, and constructing the initiative willingness signature sample reserving library; sending a passive intention sample reserving prompt, collecting a plurality of electronic signatures of a user, carrying out mutual exclusivity judgment on the electronic signatures of the user and the electronic signatures of the corresponding user in an active intention sample reserving library, and storing the electronic signatures meeting the mutual exclusion requirement in the passive intention sample reserving library; and acquiring a handwritten electronic signature of a user on line, comparing and checking the handwritten electronic signature with a corresponding signature in the active intention sample reserving library, if the handwritten electronic signature is consistent with the corresponding signature in the active intention sample reserving library, executing active intention setting, if the handwritten electronic signature is inconsistent with the corresponding signature in the active intention sample reserving library, further comparing and checking the handwritten electronic signature with a signature in the passive intention sample reserving library, if the handwritten electronic signature is consistent with the corresponding signature in the active intention sample reserving library, executing passive intention setting, starting setting early warning, and if the inconsistent times in the verification exceed the preset times, locking and signing.
Preferably, the electronic signature satisfies the mutual exclusion requirement that the static characteristics of the electronic signature in the active will sample storage and the static characteristics of the electronic signature in the passive will sample storage are consistent, the dynamic characteristics are obviously different, the electronic signature cannot be matched through any handwriting comparison algorithm, and any sample in the active will sample storage cannot be matched through the handwriting comparison algorithm by a sample in the passive will sample storage.
Further preferably, if the comparison and verification consistency of the online electronic signature and the signature in the active intention sample reserving library is passed, the active intention response of the system is triggered, and the login user account is verified through the signature; if the comparison and verification consistency between the online electronic signature and the signature in the passive intention sample storage is passed, the signature is judged to be a passive intention signature, a system passive intention response is triggered, the foreground prompts that the signature is passed verification, the background refuses to log in a user account or limits transaction, stores transaction records and sends alarm information.
Further preferably, the on-line electronic signature is compared and verified with the comparison signature in the sample reservation library by adopting an attention weighted dynamic time warping method, the weight set for the dynamic feature of the signature data is higher than that of the static feature of the signature data, and the weight of the starting and ending feature is higher than that of the running feature.
Preferably, a dynamic time normalization algorithm is adopted in combination with signature dynamic characteristics to align the compared two signature point characteristics in signature mutual exclusion judgment and comparison verification, and in the attention weighted dynamic time warping method, the difference value between the signature dynamic sequence characteristics predicted according to the mask pre-training model and the acquired on-line signature characteristics is used as the dynamic characteristic sequence weight
And &>
Calling a formula:
Preferably, the verification distance threshold is calculated according to a plurality of signatures in the active will signature sample reserving library, all signatures with the same content of the same signer in the active will signature sample reserving library are adopted, the maximum value or the average value of the distances among the signatures is calculated to serve as the verification distance threshold, and the mutual exclusivity judgment threshold of the active will sample reserving and the passive will sample reserving is larger than or equal to the threshold of signature comparison verification in the verification stage.
According to another aspect of the present application, the present invention further provides a system for recognizing a signing will based on handwriting features, including: the system comprises a signature acquisition unit, a signature comparison unit, a verification unit and an execution unit, wherein the signature acquisition unit is used for acquiring electronic signatures signed by a user for multiple times under active intention sample reserving prompts and passive intention prompts after the user passes identity verification and storing the electronic signatures in a sample reserving library to construct an active intention signature sample reserving library and a passive intention sample reserving library; acquiring a handwritten electronic signature of a user on line; a signature comparison unit: the electronic signature processing system is used for carrying out mutual exclusion judgment on the electronic signature acquired under the passive intention sample reserving prompt and the electronic signature of a corresponding user in the active intention sample reserving library, and storing the electronic signature meeting the mutual exclusion requirement into the passive intention sample reserving library; a checking unit: comparing and checking the user handwritten electronic signatures acquired on line with corresponding electronic signatures in an active intention sample reserving library and a passive intention sample reserving library respectively; an execution unit: and if the verified online electronic signature is consistent with the signature verification in the active intention sample storage, executing active intention setting, and if the verified online electronic signature is consistent with the signature verification in the passive intention sample storage, executing passive intention setting through user signature authentication, starting setting early warning, and if the verification inconsistency exceeds the preset times, locking the signature acquisition unit.
Preferably, the signature comparison unit and the verification unit adopt a dynamic time normalization algorithm to compare two signature point characteristics in combination with the signature dynamic characteristics in signature mutual exclusion judgment and comparison verification of the electronic signatureAligning, adopting an attention weighted dynamic time warping method, and taking the difference value between the signature dynamic sequence characteristic predicted by the mask pre-training model and the obtained user handwritten signature characteristic as the dynamic characteristic sequence weight
And &>
Calling a formula:
According to another aspect of the application, an electronic device is proposed, comprising: a processor; and a memory storing a program, wherein the program includes instructions that when executed by the processor cause the processor to perform a method of identifying a willingness to sign based on the handwriting features as described above.
According to another aspect of the present application, a non-transitory computer-readable storage medium is provided, having stored thereon computer instructions for causing a computer to execute the method for recognizing a willingness to sign based on handwriting features as described above.
The application relates to at least two will sample reserves, the two will sample reserves are mutually exclusive through handwriting verification algorithm verification, and the two types of sample reserves are provided that the signatures are consistent on static characteristics such as writing methods as much as possible, but have different dynamic characteristics. The passive intention sample keeping method is not easy to find characteristic differences and is difficult to imitate, and is a private and safe passive sample keeping method.
In the existing method based on time sequence alignment, point features are not weighted, so that the difference between the active intention signature and the passive intention signature is averaged. According to the method, the difference between the two signatures is calculated by adopting an attention weighting method, the weight of some unconventional pen moving characteristics is higher, the personalized characteristics different from the dynamic characteristics of a common person in the signatures are better identified by an algorithm, the accuracy of the algorithm is higher, and the method is more beneficial to the personalized characteristics of unconventional writing methods of passive signatures. In the attention weight calculation process, the degree of personalized features is reflected according to the difference value of the popular dynamic features and the actual dynamic features of the signature point features. The method for predicting the popular dynamic features can adopt a signature feature recognition method based on a mask pre-training model, and due to the fact that a large amount of data are trained, the personalized features are output smoothly, and the popular dynamic features corresponding to a signature writing method are predicted.
The existing verification system only comprises an active intention verification process and a response process, and the application also comprises a passive intention verification process and a passive intention response process, wherein the passive intention response process is configurable, such as withdrawal limit, intention deposit certificate, default alarm and the like, and is different from the active intention response process, so that the rights and interests of signers are protected.
Drawings
Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments, which is to be read in connection with the accompanying drawings, in which:
FIG. 1 is a schematic diagram illustrating a user sample-keeping library construction according to an exemplary embodiment of the present application;
fig. 2 is a schematic diagram illustrating a user sample retention intention verification process according to an exemplary embodiment of the present application;
FIG. 3 is a block diagram illustrating an exemplary electronic device that can be used to implement embodiments of the present application;
fig. 4 is a diagram illustrating DTW alignment results according to an exemplary embodiment of the present application.
Detailed Description
Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustration purposes only and are not intended to limit the scope of the present application.
It should be understood that the various steps recited in the method embodiments of the present application may be performed in a different order and/or in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present application is not limited in this respect.
The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description. It should be noted that the terms "first", "second", and the like in the present application are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
It is noted that references to "a", "an", and "the" modifications in this application are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that reference to "one or more" unless the context clearly dictates otherwise.
The names of messages or information exchanged between devices in the embodiments of the present application are for illustrative purposes only, and are not intended to limit the scope of the messages or information.
The signature of the personal handwriting has characteristics, and once the writing habit is formed, the appearance form of the handwriting is fixed. The signature biological characteristics of the user including strokes, stroke order, stroke pressure, stroke speed and the like are kept relatively stable under the environment of active will. The stroke movement characteristic is correspondingly expressed in five aspects of starting stroke characteristic, stroke running characteristic, stroke closing characteristic, stroke force characteristic and stroke basic form characteristic. Even if one pen starting action, pen running action and pen drawing action are different, the basic morphological characteristics of strokes are influenced. The pen movement is the finest and most complex writing movement, so the pen movement characteristic is strong in specificity, large in stability and high in identification value.
The invention is described in detail below with reference to the drawings and specific examples.
Fig. 1 is a schematic diagram of a user sample reserving library construction process, and a user signature sample reserving database is constructed, including an active intention sample reserving library and a passive intention sample reserving library. The active will stays the appearance suggestion, stays a kind uniformity and judges, satisfies and deposits the active sample storehouse of staying, and passive will stays the appearance suggestion, and passive will stays the appearance, stays a kind uniformity and judges, stays a kind storehouse with the active in sign name and carries out the mutual exclusion and judges, satisfies the mutual exclusion requirement and deposits the passive sample storehouse of staying.
The system for recognizing the signing will based on the handwriting characteristics comprises: the system comprises a signature acquisition unit, a signature comparison unit, a verification unit and an execution unit, wherein the signature acquisition unit acquires electronic signatures signed by a user for multiple times under active intention sample reservation prompt and passive intention prompt and stores the electronic signatures in a sample reservation library, the active intention signature sample reservation library and the passive intention sample reservation library are constructed, the electronic signatures of the user are acquired, and the handwritten electronic signatures of the user are acquired online; the signature comparison unit is used for carrying out mutual exclusion judgment on the electronic signature acquired under the passive intention sample reserving prompt and the electronic signature of the corresponding user in the active intention sample reserving library, and storing the electronic signature meeting the mutual exclusion requirement into the passive intention sample reserving library; the verification unit compares and verifies the user handwritten electronic signature acquired on line with corresponding electronic signatures in the active intention sample storage and the passive intention sample storage respectively; and if the verified online electronic signature is consistent with the signature verification in the active intention sample storage, executing active intention setting, if the verified online electronic signature is consistent with the signature verification in the active intention sample storage, executing passive intention setting by the execution unit through user signature authentication, and if the verified inconsistent times exceed the preset times, starting the setting early warning by the execution unit, and locking the signature acquisition unit.
And acquiring and verifying user identity information in a trusted environment (such as handling business on site by using an identity card), and acquiring electronic signature data of the user to reserve a sample after verifying the real identity of the user. The system sends an active intention sample reserving prompt, collects a plurality of electronic signatures of a user in the same signing device and stores the electronic signatures in an active sample reserving library, and sends a passive intention sample reserving prompt and collects a plurality of electronic signatures of the user in the same signing device; and (4) carrying out mutual exclusion judgment on the electronic signature acquired under the passive intention sample reserving prompt and the corresponding electronic signature in the active sample reserving library, and storing the electronic signature meeting the mutual exclusion requirement into the passive sample reserving library.
When the initiative will sample keeping, the handwriting sample keeping specification of the system can be followed, the signature of the initiative sample keeping can be continuously and stably kept for a plurality of times (such as 3 times), the signature can be judged by a signature authentication algorithm of 1.
When the passive intention is reserved, the requirements of the active intention handwriting reservation specification can be followed, the signature which is mutually exclusive with the active intention signature is met, the two types of reserved signatures are consistent on static characteristics such as writing methods as much as possible, and meanwhile, the dynamic characteristics are different. Such passive intentions are less easily discovered and difficult to mimic. And storing the passive will sample into a passive sample storage.
In the creation of the passive intention signature database, the signature samples cannot be matched by any sample in the active intention sample-keeping signature database through a handwriting comparison algorithm. Passive intention sample retention and active intention sample retention can be similar in form, but have obvious difference in dynamic characteristics such as pressure, stroke order and the like, so that passive intention signature sample retention can be correctly distinguished by an algorithm and cannot be easily found by other people.
The active sample reserving library and the passive sample reserving library are mutually exclusive, the same signature sample of a user belongs to the active sample reserving library or the passive sample reserving library, namely any sample in the two sample reserving libraries can not be matched with a sample in the other sample reserving library through a handwriting comparison algorithm. Judging whether the signatures of the active will sample reserving library and the passive signature sample reserving database are mutually exclusive can be realized according to a handwriting comparison algorithm, namely, any combination of the active will sample reserving library and the passive will sample reserving library cannot be successfully matched, and if the matching is successful, the signatures are not mutually exclusive and the passive will signatures need to be reserved again. And (4) taking the signature which is not successfully matched with the signature in the active will sample-keeping signature library as the passive will signature sample-keeping.
Setting up different willingness system responses. And judging whether the electronic signature data is in an active or passive intention state, triggering different responses according to an intention system to which the signature belongs, and setting system responses of identity and the active intention of the user under the active intention state.
The system collects the electronic signature of the user, compares the electronic signature with the electronic signature reserved sample in the database for judgment, if the electronic signature is judged to be the initiative intention signature, confirms that the signed electronic signature is the real expression of the intention of the user, triggers the initiative intention response of the system, and triggers the next operation of the system through signature authentication. Such as bank transfer, withdrawal, contract signing, online transaction and the like, and the next step of starting transfer, transaction completion and the like is carried out.
Passive will sign, i.e. some signs are produced under duress and are not expressed by the signer's active will. For example, a criminal suspect who holds a knife to threaten a signer requires that the criminal suspect must sign to pass through a verification system, otherwise, the criminal suspect threatens safety. In this case, the signer wants to obtain the system prompt of passing the verification through the signature without endangering the security of the signer, and also wants to prevent the property of the signer from being lost, and the signer can be recognized as the passive intention expression. The foreground prompts that the verification is passed, the background does not enter into substantial transaction processing operation, relevant evidence is kept, and password alarm measures are taken to send alarm information and the like.
The system can respond to relevant settings according to passive wishes, such as transfer rejection, secret alarm, evidence retention, even false user information displayed by the system, and the like.
If on-line transaction scene, the initiative will can set 1 ten thousand yuan per transaction, and the transaction times are not limited every day. The transaction number of single transaction is set to be 1000 yuan under passive will, the transaction number is not more than 3 times per day, and meanwhile, the system stores the transaction information and sends early warning information to a system background manager for future warning and evidence-taking.
The method greatly protects the signed user under the state of coercion, and subsequently provides reliable evidence for loss recovery, judicial evidence proofing and the like.
FIG. 2 is a schematic diagram showing a user sample leaving intention verification process, in which a system prompts a user to write, the user writes according to the prompted intention, based on a signature handwriting verification algorithm, the user performs active intention verification, the user signature is compared with signatures in an active intention library, if the user signature passes consistency detection, active intention setting and response are performed, if the user signature does not pass consistency detection, passive intention verification is performed, the user signature is compared with signatures in a passive intention library, and if the user signature passes consistency detection, passive intention setting and response are performed; if the consistency check is not passed, the signing is prompted to be carried out again, the consistency detection is carried out again, the setting is early warned, and the signing is locked.
In the system verification stage, firstly, the electronic signature of a signer is compared with the electronic signature of the active intention sample storage through a handwriting comparison algorithm, whether the electronic signature belongs to the real intention expression of the principal is verified, if so, the system executes active intention setting, and if not, the system enters passive intention verification.
And when the electronic signature of the signer does not pass the active intention verification, performing passive intention verification, comparing the electronic signature with the electronic signature of the passive intention sample storage through a handwriting comparison algorithm, verifying whether the electronic signature belongs to the signature signed by the party under the coerced state, and if so, executing passive intention setting. If not, judging that the sample is not consistent with the samples of the active sample reserving library and the passive sample reserving library, rewriting, locking and signing and other processes.
The user may choose to re-write when the user is not adapting or the writing style changes due to the writing device. When the signatures are not passed through for multiple times, the situation that other people have the signatures is considered to be possible, and relevant situations such as prompting the form of the leave-sample signatures or locking the signature verification function can be executed according to settings.
Signature handwriting comparison or signature handwriting verification methods are used in the user signature verification process and in the sample leaving process for signature stability verification and mutual exclusivity verification.
According to the characteristics of active intention sample retention signature and passive intention sample retention signature data, the method pays more attention to distinguished dynamic characteristics, adopts an attention weighting method to calculate and compare the difference of two comparison signatures, sets higher weight for dynamic characteristics including stroke movement characteristics (such as stroke pressure, stroke speed and stroke order) and the like, and sets lower weight for static characteristics including stroke shape, character spacing, structure and the like. The method is beneficial to identifying the dynamic characteristics of the signature, which are different from the dynamic characteristics of the ordinary person, by the algorithm, the accuracy rate of the algorithm is higher, and the method is more beneficial to the personalized characteristics of the unconventional writing method of the passive signature. The algorithm accuracy of sample keeping and matching of the active will and the passive will can be effectively improved. The method can be realized by the following steps:
generally, personal handwriting signatures are characterized in that once a writing habit is formed, the appearance of the handwriting is fixed. The signature biological characteristics of the user including strokes, stroke order, stroke pressure, stroke speed and the like are kept relatively stable under the environment of active will. The pen movement is the finest and most complex writing movement, so the pen movement characteristic is strong in specificity, large in stability and high in identification value. The moving stroke characteristics of all the point positions are not different greatly from person to person, and if any stroke is written in any one frame, the three aspects of starting stroke, running stroke and drawing stroke are included, and the characteristic difference between the starting stroke and the drawing stroke is usually larger than that of the drawing stroke. Generally, a dtw dynamic time warping algorithm is adopted to consider all point positions as the same weight more, and then difference is calculated, and no attention is paid to more distinctive features.
The active intention signature sample retention and the passive intention signature sample retention related to the exemplary embodiment of the present application may be consistent on many feature points, and a point difference of point location features needs to be concerned more.
Before signature comparison verification and sample selection of a reserved sample library, preprocessing acquired related electronic signatures, unifying sampling rate, removing interference strokes and other conventional signature preprocessing, and supposing to obtain a preprocessed active intention signature reserved sample library signature A and a preprocessed passive intention signature reserved sample library signature B; extracting normalized dynamic signature characteristics such as signature stroke speed, acceleration and the like; aligning the characteristics of two signature points by combining the dynamic signature characteristics by adopting a dynamic time normalization algorithm
And &>
Respectively representing the dynamic characteristic sequences of the signature A and the signature B, wherein n is the total length of the aligned dynamic characteristic sequences, and a formula is called:
Wherein d represents the quantization difference between the signature B and the signature A, the smaller d represents the more similar the signatures are,
and &>
Represents the weight of the signature A and the signature B respectively in the dynamic characteristic sequence of the point i, the bigger the dynamic weight of a certain point position is, the more the difference of the point position is amplified, and the greater the weight of the point position is>
And &>
The weighting for the point is represented by a larger value.
The difference of the two signatures is calculated using an attention weighted Dynamic Time Warping (DTW) method. If the difference is larger than the threshold value, the two signatures are judged to be inconsistent, and if the difference is smaller than the threshold value, the two signatures are consistent.
And calculating a verification distance threshold according to a plurality of signatures in the active will signature sample reserving library, wherein all signatures with the same content of the same signer in the active will signature sample reserving library can be adopted, and the maximum value or the average value of the distances between the signatures is calculated according to the formula to serve as the distance threshold. The judgment threshold of the stable consistency of the signatures of the active will sample reserving library is greater than or equal to the judgment threshold of the mutual exclusion of the active will sample reserving library and the passive will sample reserving library, and is greater than or equal to the verification threshold of the signatures in the verification stage.
The method can pay more attention to the influence of the characteristic obvious change area on the signature handwriting, and the method of attention weighted dynamic characteristic difference is adopted, so that the algorithm is greatly improved for correctly distinguishing the active intention sample retention and the passive intention sample retention.
Obtaining signature characteristics, adopting a signature characteristic identification method based on a mask pre-training model, and taking the difference value between the signature dynamic sequence characteristics predicted by the mask pre-training model after signature characteristic pre-training and the signature real characteristic size as the weight of the dynamic characteristic sequence
And &>
The basis for the calculation may be taken as a two-norm distance.
The embodiment of the application adopts a pre-training method, and due to the fact that a large amount of data are trained, the output of the pre-training method smoothes out the personalized features, and the popular dynamic features corresponding to the signature writing method are predicted. And the magnitude of the difference may reflect the degree of individualization. The method is more favorable for identifying the dynamic characteristics of the signature which are individualized and different from those of the ordinary person, has higher accuracy of the algorithm, and is more favorable for the individualized characteristics of the unconventional writing method of the passive signature.
An exemplary embodiment of the present application also provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor. The memory stores a computer program executable by the at least one processor, the computer program, when executed by the at least one processor, is for causing the electronic device to perform a method according to an embodiment of the application.
The exemplary embodiments of this application also provide a non-transitory computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to perform a method according to an embodiment of this application.
The exemplary embodiments of this application also provide a computer program product comprising a computer program, wherein the computer program, when being executed by a processor of a computer, is adapted to cause the computer to carry out the method according to embodiments of this application.
Referring to fig. 3, a block diagram of a structure of an electronic device 300, which may be a server or a client of the present application, which is an example of a hardware device that may be applied to aspects of the present application, will now be described. Electronic device is intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.
As shown in fig. 3, the electronic device 300 includes a computing unit 301 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 302 or a computer program loaded from a storage unit 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the device 300 can also be stored. The calculation unit 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.
A number of components in the electronic device 300 are connected to the I/O interface 305, including: an input unit 306, an output unit 307, a storage unit 308, and a communication unit 309. The input unit 306 may be any type of device capable of inputting information to the electronic device 300, and the input unit 306 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device. Output unit 307 may be any type of device capable of presenting information and may include, but is not limited to, a display, speakers, a video/audio output terminal, a vibrator, and/or a printer. The storage unit 308 may include, but is not limited to, a magnetic disk, an optical disk. The communication unit 309 allows the electronic device 300 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers and/or chipsets, such as bluetooth devices, wiFi devices, wiMax devices, cellular communication devices, and/or the like.
The computing unit 301 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 301 performs the respective methods and processes described above. For example, in some embodiments, the reconstruction and decomposition of the original trajectory of a signed stroke to redraw its trajectory of muscle motion, the decomposition of its logarithmic speed curve, and the like, may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 300 via the ROM 302 and/or the communication unit 309. In some embodiments, the computing unit 301 may be configured to perform the signature script dynamic acquisition implementation method in any other suitable manner (e.g., by means of firmware).
Program code for implementing the methods of the present application may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
Fig. 4 shows the DTW alignment result, with thicker dots representing higher weights. It can be seen that "Jiu" on the left has more thick points, the overall personalized features are stronger, the weight of the points at some details is obvious, and the dynamic features of such points are given more weight to participate in the calculation of the difference degree.
Although only two will, active and passive, are listed, the same is true for scenarios with more than two will. Meanwhile, the method is also suitable for a sample reservation library, is not limited to signatures, and is also suitable for symbols of dynamic sequences.
Claims (10)
1. A method for recognizing signing willingness based on handwriting characteristics is characterized by verifying user identity, sending initiative willingness sample reserving prompt, collecting multiple electronic signatures of a user, storing the electronic signatures into a sample reserving library, and constructing an initiative willingness signature sample reserving library; sending a passive intention sample reserving prompt, collecting a plurality of electronic signatures of a user, carrying out mutual exclusivity judgment on the electronic signatures of the user corresponding to the active intention sample reserving library, and storing the electronic signatures meeting the mutual exclusion requirement into the passive intention sample reserving library; and acquiring a handwritten electronic signature of a user on line, comparing and checking the handwritten electronic signature with a corresponding signature in the active intention sample storage, if the signature is consistent with the active intention sample storage, executing active intention setting, if the signature is inconsistent with the active intention sample storage, further comparing and checking the signature in the passive intention sample storage, if the signature is consistent with the passive intention sample storage, executing passive intention setting, starting setting early warning, and if the number of inconsistent check exceeds the preset number, locking the signature.
2. The method as claimed in claim 1, wherein the electronic signature satisfies the mutual exclusion requirement that the electronic signature in the active will sample library is consistent with the electronic signature in the passive will sample library in static characteristics, dynamic characteristics are obviously different, any handwriting comparison algorithm cannot match any sample in the active will sample library, and any sample in the passive will sample library cannot be matched by the handwriting comparison algorithm.
3. The method of claim 1, wherein if the on-line electronic signature passes comparison and verification consistency with the signature in the active intention sample reserving library, the system active intention response is triggered, and the login user account is verified through the signature; if the comparison and verification consistency between the online electronic signature and the signature in the passive intention sample storage is passed, the signature is judged to be a passive intention signature, a system passive intention response is triggered, the foreground prompts that the signature is passed verification, the background refuses to log in a user account or limits transaction, stores transaction records and sends alarm information.
4. The method according to any one of claims 1 to 3, wherein the on-line electronic signature is compared and verified by adopting an attention-weighted dynamic time warping method and compared signatures in a sample reservation library, the dynamic features of the signature data are set with higher weights than the static features of the signature data, and the initial and final feature weights are higher than the stroke feature weights.
5. The method as claimed in claim 4, wherein a dynamic time warping algorithm is used in combination with the signature dynamic features to align the compared two signature point features in the mutual exclusion determination and comparison verification of the signature, and in the attention weighted dynamic time warping method, the difference between the signature dynamic sequence feature predicted by the mask pre-training model and the acquired on-line signature feature is used as the dynamic feature sequence weight
And
calling a formula:
6. The method according to one of claims 1 to 3 or 5, characterized in that a verification distance threshold is calculated according to a plurality of signatures in the active intention signature sample reservation library, all signatures with the same content of the same signer in the active intention signature sample reservation library are adopted, the maximum value or the average value of the distances between the signatures is calculated to be used as the verification distance threshold, and the mutual exclusivity judgment threshold of the active intention sample reservation and the passive intention sample reservation is larger than or equal to the threshold of signature comparison and verification in the verification stage.
7. A system for recognizing a willingness to sign based on handwriting characteristics, comprising: the system comprises a signature acquisition unit, a signature comparison unit, a verification unit and an execution unit, wherein the signature acquisition unit is used for respectively acquiring electronic signatures signed by a user for multiple times under active intention prompt and passive intention prompt after the user passes identity verification and storing the electronic signatures in a sample reserving library, constructing an active intention signature sample reserving library and a passive intention signature sample reserving library with mutually exclusive signature characteristics, and acquiring a handwritten electronic signature of the user on line; a signature comparison unit: the electronic signature processing system is used for carrying out mutual exclusion judgment on the electronic signature acquired under the passive intention sample reserving prompt and the electronic signature of a corresponding user in the active intention sample reserving library, and storing the electronic signature meeting the mutual exclusion requirement into the passive intention sample reserving library; a checking unit: comparing and checking the handwritten electronic signatures acquired on line with corresponding electronic signatures in an active intention sample reserving library and a passive intention sample reserving library respectively; an execution unit: and if the verified online electronic signature is consistent with the signature verification in the active intention sample storage, executing active intention setting, and if the verified online electronic signature is consistent with the signature verification in the passive intention sample storage, executing passive intention setting through user signature authentication, starting setting early warning, and if the verification inconsistency exceeds the preset times, locking the signature acquisition unit.
8. The system of claim 7, wherein the signature comparison unit and the checklistIn the mutual exclusion judgment and comparison verification of the electronic signature, a dynamic time normalization algorithm is adopted to align the compared two signature point characteristics in combination with the signature dynamic characteristics, an attention weighted dynamic time warping method is adopted, and the difference value between the signature dynamic sequence characteristics predicted according to a mask pre-training model and the acquired user handwritten signature characteristics is taken as the dynamic characteristic sequence weight
And
calling a formula:
9. An electronic device, comprising: a processor; and a memory storing a program, wherein the program includes instructions which, when executed by the processor, cause the processor to perform a method of recognizing a willingness to sign based on handwriting characteristics according to any one of claims 1-6.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method for recognizing a willingness to sign based on handwriting characteristics according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211410547.XA CN115841703A (en) | 2022-11-11 | 2022-11-11 | Signing intention recognition method, system, equipment and medium based on handwriting characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211410547.XA CN115841703A (en) | 2022-11-11 | 2022-11-11 | Signing intention recognition method, system, equipment and medium based on handwriting characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115841703A true CN115841703A (en) | 2023-03-24 |
Family
ID=85577088
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211410547.XA Pending CN115841703A (en) | 2022-11-11 | 2022-11-11 | Signing intention recognition method, system, equipment and medium based on handwriting characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115841703A (en) |
-
2022
- 2022-11-11 CN CN202211410547.XA patent/CN115841703A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Unobservable re-authentication for smartphones. | |
| US11100481B2 (en) | Image authentication and security system and method | |
| Feher et al. | User identity verification via mouse dynamics | |
| Impedovo et al. | Automatic signature verification in the mobile cloud scenario: survey and way ahead | |
| EP3321853A1 (en) | Facial recognition similarity threshold adjustment | |
| US9202035B1 (en) | User authentication based on biometric handwriting aspects of a handwritten code | |
| Bours et al. | A login system using mouse dynamics | |
| WO2018082011A1 (en) | Living fingerprint recognition method and device | |
| CN106603563A (en) | Information safety realization method and system based on biometric features identification | |
| Tolosana et al. | BioTouchPass: Handwritten passwords for touchscreen biometrics | |
| US10552596B2 (en) | Biometric authentication | |
| CN106778151B (en) | Handwriting-based user identity recognition method and device | |
| Huang et al. | Development of a typing behaviour recognition mechanism on android | |
| Smith-Creasey et al. | A continuous user authentication scheme for mobile devices | |
| Van Nguyen et al. | Finger-drawn pin authentication on touch devices | |
| Palma et al. | Biometric-based human recognition systems: an overview | |
| CN111563746A (en) | Method, device, electronic equipment and medium for user identity authentication | |
| JPH10261082A (en) | Registered signature data preparing method for computer signature collating system | |
| Mondal et al. | A continuous combination of security & forensics for mobile devices | |
| Dybczak et al. | Continuous authentication on mobile devices using behavioral biometrics | |
| Fenu et al. | Leveraging continuous multi-modal authentication for access control in mobile cloud environments | |
| Canfora et al. | Silent and continuous authentication in mobile environment | |
| US11669604B2 (en) | Methods and systems for authenticating a user | |
| SE1650354A1 (en) | Method and system for evaluating fingerprint templates | |
| CN115841703A (en) | Signing intention recognition method, system, equipment and medium based on handwriting characteristics |
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 |