CN109635625A - Smart identity checking method, equipment, storage medium and device - Google Patents

Smart identity checking method, equipment, storage medium and device Download PDF

Info

Publication number
CN109635625A
CN109635625A CN201811206298.6A CN201811206298A CN109635625A CN 109635625 A CN109635625 A CN 109635625A CN 201811206298 A CN201811206298 A CN 201811206298A CN 109635625 A CN109635625 A CN 109635625A
Authority
CN
China
Prior art keywords
score
user
target
identity
similarity
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.)
Granted
Application number
CN201811206298.6A
Other languages
Chinese (zh)
Other versions
CN109635625B (en
Inventor
李伟
宁宁
刘伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Priority to CN201811206298.6A priority Critical patent/CN109635625B/en
Publication of CN109635625A publication Critical patent/CN109635625A/en
Application granted granted Critical
Publication of CN109635625B publication Critical patent/CN109635625B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168Feature extraction; Face representation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172Classification, e.g. identification
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification techniques
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification techniques
    • G10L17/06Decision making techniques; Pattern matching strategies
    • G10L17/08Use of distortion metrics or a particular distance between probe pattern and reference templates
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L17/00Speaker identification or verification techniques
    • G10L17/06Decision making techniques; Pattern matching strategies
    • G10L17/12Score normalisation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Acoustics & Sound (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Collating Specific Patterns (AREA)

Abstract

The invention discloses a kind of smart identity checking method, equipment, storage medium and devices, this method comprises: the facial image of acquisition user, and current face's characteristic information is extracted from facial image, target face characteristic information and current face's characteristic information are subjected to similarity comparison, and carry out similarity marking, obtain the first similarity score, first similarity score and target are compared by score threshold, when the first similarity score, which is higher than target, passes through score threshold, determine that the user identity veritification passes through.In the present invention, by obtaining the first similarity score between current face's characteristic information and target face characteristic information, to whether be more than that target judges whether identity veritification passes through by score threshold according to first similarity score, due to being directed to the different air control demands of different business scene, it is arranged and different passes through score threshold, the flexibility and efficiency of identity veritification are significantly improved, has ensured the transaction security of user.

Description

Smart identity checking method, equipment, storage medium and device
Technical field
The present invention relates to identity to veritify technical field more particularly to a kind of smart identity checking method, equipment, storage medium And device.
Background technique
Identity veritifies scheme, is to pass through identity card optical character identification (Optical based on computer vision technique Character Recognition, OCR) technology or user be manually entered acquisition subscriber identity information, will believe user identity according to this Breath is verified by the identification check platform that public security is specified, and obtains user in the photo of keeping on file of public security, utilizes comparison technology Confirm user identity.
Currently, identity is veritified it is widely used in user's login, insures and saves from damage etc. in various businesses scene, however, existing Identity veritify scheme be all made of unified threshold value no matter which kind of business scenario user is in and be compared to confirm user's body Part, and each business scenario has different air control demands, carries out identity veritification using unified threshold value, accuracy is lower.
Above content is only used to facilitate the understanding of the technical scheme, and is not represented and is recognized that above content is existing skill Art.
Summary of the invention
The main purpose of the present invention is to provide a kind of smart identity checking method, equipment, storage medium and devices, it is intended to It solves to carry out identity veritification, the lower technical problem of accuracy using unified threshold value in the prior art.
To achieve the above object, the present invention provides a kind of smart identity checking method, the smart identity checking method packet Include following steps:
It obtains the corresponding target of current business scene and passes through score threshold;
The subscriber identity information that instruction obtains the user is veritified according to the identity of the user received, and according to the user Identity information obtains target face characteristic information;
The facial image of the user is acquired, and extracts current face's characteristic information from the facial image;
The target face characteristic information and current face's characteristic information are subjected to similarity comparison, and carried out similar Degree marking, obtains the first similarity score;
First similarity score is compared with the target by score threshold, when first similarity is commented Point it is higher than the target when passing through score threshold, determines that the user identity veritification passes through.
Preferably, the identity of the user that the basis receives veritify instruction obtain the user subscriber identity information it Before, further includes:
Obtain the veritification precision under business scenario and targets threshold section;
Chosen in the targets threshold section meet it is described veritify precision by score threshold, and establish business scenario With the default mapping relations for passing through score threshold;
Current business scene is obtained, and searches mesh corresponding with the current business scene in the default mapping relations Mark passes through score threshold.
Preferably, the veritification precision includes target percent of pass and target False Rate;
Described choose in the targets threshold section meets the precision of veritifying and passes through score threshold, comprising:
The sample data under business scenario is obtained, the sample data includes sample facial image, the positive negativity of sample and sample This similarity score;
Successively chosen in order in the targets threshold section it is undetermined by score threshold, according to it is described it is undetermined by point Number threshold values, the positive negativity of the sample and Sample Similarity scoring calculate the current percent of pass of the sample facial image and work as Preceding False Rate;
It is not less than the target not higher than the target percent of pass and/or the current False Rate in the current percent of pass When False Rate, selection is next undetermined by score threshold, and returns described according to described undetermined by score threshold, the sample This positive negativity and Sample Similarity scoring calculate the current percent of pass of the sample facial image and the step of current False Rate Suddenly;
It is higher than the target percent of pass in the current percent of pass, and the current False Rate is lower than the target False Rate When, undetermined score threshold is passed through as the precision of veritifying is met by score threshold for described.
Preferably, it is described successively chosen in order in the targets threshold section it is undetermined by score threshold, according to institute It states undetermined score by score threshold, the positive negativity of the sample and the Sample Similarity and calculates working as the sample facial image Preceding percent of pass and current False Rate, comprising:
It is successively chosen in order in the targets threshold section and undetermined passes through score threshold;
According to Sample Similarity scoring and the veritification undetermined by score threshold judgement sample as a result, according to institute It states and veritifies the current percent of pass that result calculates the sample facial image;
According to the positive negativity of the sample and the current False Rate veritified result and calculate the sample facial image.
Preferably, it is described first similarity score and the target are compared by score threshold after, institute State method further include:
When first similarity score passes through score threshold not higher than the target, judge that first similarity is commented Divide and whether be lower than the first preset standard score, the first preset standard score passes through score threshold lower than the target;
When first similarity score is lower than the first preset standard score, it is obstructed to determine that the user identity is veritified It crosses;
When first similarity score is not less than the first preset standard score, secondary identity core is carried out to the user It tests.
Preferably, it is described when first similarity score be not less than the first preset standard score when, to the user into The secondary identity of row is veritified, comprising:
When first similarity score is not less than the first preset standard score, the audio of the user is acquired, from institute It states and extracts current vocal print feature information in audio;
Target vocal print feature information is obtained according to the subscriber identity information;
The target vocal print feature information and the current vocal print feature information are subjected to similarity comparison, and carried out similar Degree marking, obtains the second similarity score;
Second similarity score and the second preset standard score are compared, when second similarity is commented When dividing higher than the second preset standard score, determine that the user identity veritification passes through.
Preferably, the facial image of the acquisition user, and current face spy is extracted from the facial image Before reference breath, comprising:
Detect whether the user is living body;
When the user is living body, the facial image of the acquisition user is executed, and from the facial image The step of extracting current face's characteristic information;
When the user is non-living body, then determine that the user identity veritification does not pass through.
In addition, to achieve the above object, the present invention also proposes that a kind of smart identity apparatus for checking, the smart identity are veritified Equipment includes that the smart identity that can run on the memory and on the processor of memory, processor and being stored in is veritified Program, the smart identity veritify the step of program is arranged for carrying out smart identity checking method as described above.
In addition, to achieve the above object, the present invention also proposes a kind of storage medium, intelligence is stored on the storage medium Identity veritifies program, and the smart identity is veritified when program is executed by processor and realizes smart identity veritification side as described above The step of method.
In addition, to achieve the above object, the present invention also proposes that a kind of smart identity veritifies device, and the smart identity is veritified Device includes:
Threshold value obtains module, passes through score threshold for obtaining the corresponding target of current business scene;
Fisrt feature obtains module, veritifies user's body that instruction obtains the user for the identity according to the user received Part information, and target face characteristic information is obtained according to the subscriber identity information;
Second feature obtains module, extracts for acquiring the facial image of the user, and from the facial image Current face's characteristic information;
Scoring modules, for the target face characteristic information and current face's characteristic information to be carried out similarity ratio It is right, and similarity marking is carried out, obtain the first similarity score;
It veritifies module and works as institute for comparing first similarity score by score threshold with the target When stating the first similarity score and passing through score threshold higher than the target, determine that user identity veritification passes through.
In the present invention, commented by obtaining the first similarity between current face's characteristic information and target face characteristic information Point, and obtain the corresponding target of current business scene by score threshold, thus according to first similarity score whether be more than Target judges whether user identity veritification passes through by score threshold, due to being directed to the different air control demands of different business scene, It is arranged different by score threshold, significantly improves the flexibility and efficiency of identity veritification, improve user experience, ensure The transaction security of user.
Detailed description of the invention
Fig. 1 is the smart identity apparatus for checking structural schematic diagram for the hardware running environment that the embodiment of the present invention is related to;
Fig. 2 is the flow diagram of smart identity checking method first embodiment of the present invention;
Fig. 3 is the flow diagram of smart identity checking method second embodiment of the present invention;
Fig. 4 is the flow diagram of smart identity checking method 3rd embodiment of the present invention;
Fig. 5 is the structural block diagram that smart identity of the present invention veritifies device first embodiment.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Referring to Fig.1, Fig. 1 is the smart identity apparatus for checking structure for the hardware running environment that the embodiment of the present invention is related to Schematic diagram.
As shown in Figure 1, the smart identity apparatus for checking may include: processor 1001, such as central processing unit (Central Processing Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005.Wherein, communication bus 1002 is for realizing the connection communication between these components.User interface 1003 may include display Shield (Display), optional user interface 1003 can also include standard wireline interface and wireless interface, for user interface 1003 wireline interface can be USB interface in the present invention.Network interface 1004 optionally may include standard wireline interface, Wireless interface (such as Wireless Fidelity (WIreless-FIdelity, WI-FI) interface).Memory 1005 can be the random of high speed Memory (Random Access Memory, RAM) memory is accessed, stable memory (Non-volatile is also possible to Memory, NVM), such as magnetic disk storage.Memory 1005 optionally can also be the storage independently of aforementioned processor 1001 Device.
It will be understood by those skilled in the art that structure shown in Fig. 1 does not constitute the limit to smart identity apparatus for checking It is fixed, it may include perhaps combining certain components or different component layouts than illustrating more or fewer components.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium Believe that module, Subscriber Interface Module SIM and smart identity veritify program.
In smart identity apparatus for checking shown in Fig. 1, network interface 1004 is mainly used for connecting background server, with institute It states background server and carries out data communication;User interface 1003 is mainly used for connecting user equipment;The smart identity veritification is set It is standby to call the smart identity stored in memory 1005 to veritify program by processor 1001, and offer of the embodiment of the present invention is provided Smart identity checking method.
Based on above-mentioned hardware configuration, the embodiment of smart identity checking method of the present invention is proposed.
It is the flow diagram of smart identity checking method first embodiment of the present invention referring to Fig. 2, Fig. 2, proposes the present invention Smart identity checking method first embodiment.
In the first embodiment, the smart identity checking method the following steps are included:
Step S10: it obtains the corresponding target of current business scene and passes through score threshold.
It should be noted that the executing subject of the present embodiment is smart identity apparatus for checking, wherein the smart identity core Testing equipment can be the electronic equipments such as PC, smart phone, tablet computer.It, will in the case where existing various identity veritify scene Active user's feature is compared with target user's feature, obtains similarity score, while being arranged one by score threshold, works as phase When passing through score threshold higher than described in like degree scoring, determine that the identity veritification of active user passes through, when similarity score is not higher than It is described when passing through score threshold, determine that current user identities veritification does not pass through.However, there is each business scenario different air controls to need It asks, such as the business scenario air control demand that user logs in is lower, settable lower by score threshold, the business that user saves from damage Scene air control demand is higher, settable higher to pass through score threshold.
Therefore, the present embodiment predefines various businesses scene, and various businesses scene is provided with and meets it Air control demand by score threshold, establish and save business scenario and mapping relations by score threshold.Pass through in user Smart identity apparatus for checking enter identity veritify interface when, will from the identity veritify interface extraction current business scene, and Target corresponding with the current business scene is searched in the mapping relations passes through score threshold.
Step S20: veritifying according to the identity of the user received and instruct the subscriber identity information for obtaining the user, and according to The subscriber identity information obtains target face characteristic information.
It should be noted that the identity veritifies the instruction that the initiation identity that instruction is user's input is veritified, wherein including Subscriber identity information, the subscriber identity information include in the information such as user name, user account or user identity card number extremely One item missing.Face characteristic information be according to the shape description of human face and they the distance between characteristic facilitated The characteristic of face classification, characteristic component generally include Euclidean distance, curvature and angle between characteristic point etc..Face is by eye Eyeball, nose, mouth, chin etc. are locally constituted, and to these local and structural relation between them geometric descriptions, can be used as identification people The important feature of face.It is veritified in instruction from the identity and extracts the subscriber identity information, can believed according to the user identity Breath obtains target face characteristic information, which is face characteristic corresponding with subscriber identity information letter Breath namely real human face characteristic information.
It is understood that subscriber identity information and the first corresponding relationship of face characteristic information can be stored in intelligent body Part apparatus for checking local is stored in background server.When first corresponding relationship is stored in smart identity apparatus for checking local When, search speed can be accelerated, improve user experience;When first corresponding relationship is stored in background server, local can be saved Memory space, reduce processing locality pressure.
In the concrete realization, user name, user account or user are inputted according to prompt when user veritifies in interface in identity The subscriber identity informations such as identification card number, after clicking identity veritification instruction, the subscriber identity information will be veritified with the identity and be instructed It is sent to smart identity apparatus for checking local or background server simultaneously, smart identity apparatus for checking locally will be according to the use Family identity information searches corresponding target face characteristic information, or receives the target face characteristic letter of background server feedback Breath.
Step S30: acquiring the facial image of the user, and current face's feature letter is extracted from the facial image Breath.
It should be noted that will carry out recognition of face after obtaining target face characteristic information to user, and acquire the use The facial image at family, and feature extraction is carried out to the facial image, obtain current face's characteristic information.
In the concrete realization, from the facial image extract current face's characteristic information before, due to the people of acquisition Face image is limited by various conditions and random disturbances, tends not to directly use, it is necessary in the early stage of image procossing The image preprocessings such as gray correction, noise filtering are carried out to it.For facial image, preprocessing process mainly includes people Light compensation, greyscale transformation, histogram equalization, normalization, geometric correction, filtering and sharpening of face image etc..
Step S40: carrying out similarity comparison for the target face characteristic information and current face's characteristic information, and Similarity marking is carried out, the first similarity score is obtained.
It should be noted that since the target face characteristic information is the corresponding face characteristic letter of subscriber identity information Breath, if current face's characteristic information and the target face characteristic information similarity are higher, it is believed that the user and The corresponding user of the subscriber identity information is same people, that is, can determine that the user identity veritification passes through.Thus, by the mesh It marks face characteristic information and current face's characteristic information and carries out similarity comparison, and carry out similarity marking, can get the One similarity score.
Step S50: first similarity score is compared with the target by score threshold, when described first When similarity score passes through score threshold higher than the target, determine that the user identity veritification passes through.
It is understood that could be veritified to which kind of degree by identity for similar, it will be according to the current business field The corresponding target of scape is determined by score threshold.First similarity score target corresponding with current business scene is passed through Score threshold compares, if first similarity score passes through score threshold higher than the target, determines the user Identity veritification passes through.
In the first embodiment, by obtaining the first phase between current face's characteristic information and target face characteristic information It scores like degree, and obtains the corresponding target of current business scene by score threshold, to be according to first similarity score It is no to judge whether user identity veritification passes through by score threshold more than target, due to being directed to the different air controls of different business scene Demand, is arranged different by score threshold, significantly improves the flexibility and efficiency of identity veritification, improves user experience, The transaction security of user is ensured.
It is the flow diagram of smart identity checking method second embodiment of the present invention referring to Fig. 3, Fig. 3, is based on above-mentioned Fig. 2 Shown in first embodiment, propose the second embodiment of smart identity checking method of the present invention.
In a second embodiment, the step S10, comprising:
Step S101: the veritification precision under business scenario and targets threshold section are obtained.
Step S102: chosen in the targets threshold section meet it is described veritify precision by score threshold, and build Vertical business scenario and the default mapping relations for passing through score threshold.
It is understood that the precision of veritifying refers to the percent of pass veritified and False Rate, each business scenario is to veritification Required precision is different, such as under the business scenario of user's login, it is desirable that the percent of pass that user veritifies is higher, the business that user saves from damage Under scene, it is desirable that the False Rate that user veritifies is lower, therefore, by the veritification precision different to different business scene setting.It is described Targets threshold section refers to wanting comprising target by one of score threshold substantially section, such as the business scenario that user logs in The targets threshold section asked is about 40-60 points, and the targets threshold section for the business scenario requirement that user saves from damage is about 65-90 Between, subsequent will determine from the targets threshold section final passes through score threshold.
Step S103: obtaining current business scene, and searches and the current business field in the default mapping relations The corresponding target of scape passes through score threshold.
It, can be according to working as it should be noted that after establishing business scenario with by the default mapping relations of score threshold Preceding business scenario searches corresponding target in the default mapping relations and passes through score threshold.
Further, the veritification precision includes target percent of pass and target False Rate;
The step S102, comprising:
The sample data under business scenario is obtained, the sample data includes sample facial image, the positive negativity of sample and sample This similarity score.
Successively chosen in order in the targets threshold section it is undetermined by score threshold, according to it is described it is undetermined by point Number threshold values, the positive negativity of the sample and Sample Similarity scoring calculate the current percent of pass of the sample facial image and work as Preceding False Rate.
It is not less than the target not higher than the target percent of pass and/or the current False Rate in the current percent of pass When False Rate, selection is next undetermined by score threshold, and returns described according to described undetermined by score threshold, the sample This positive negativity and Sample Similarity scoring calculate the current percent of pass of the sample facial image and the step of current False Rate Suddenly.
It is higher than the target percent of pass in the current percent of pass, and the current False Rate is lower than the target False Rate When, undetermined score threshold is passed through as the precision of veritifying is met by score threshold for described.
It should be noted that being higher than the target percent of pass in the current percent of pass, and the current False Rate is lower than When the target False Rate, illustrate it is described it is undetermined the veritification precision is met by score threshold, undetermined pass through score for described Threshold value passes through score threshold as the precision of veritifying is met under business scenario.
In the concrete realization, it is illustrated by taking the business scenario that user saves from damage as an example, the business scenario saved from damage due to user To veritifying, required precision is higher, and it is that target percent of pass is not less than 70% that initial setting, which veritifies precision, and target False Rate is not higher than 10% and targets threshold section be 65-90 point, in the section, in order by each score value successively as threshold value undetermined point Value, under different threshold scores undetermined, calculates separately current percent of pass and current False Rate;Current percent of pass is missed with current The threshold scores undetermined for sentencing rate satisfaction veritification precision pass through score threshold as target.It is allocated as example, working as 60 as threshold value undetermined point When value, percent of pass 80%, False Rate 20%;When 72 are allocated as threshold scores undetermined, percent of pass 75%, False Rate is 8%;When 85 are allocated as threshold scores undetermined, percent of pass 65%, False Rate 8%.It therefore meets veritifying the undetermined of precision It is 72 points by score threshold, so 72 are allocated as to pass through score threshold for target.
Further, it is described successively chosen in order in the targets threshold section it is undetermined by score threshold, according to Undetermined scored by score threshold, the positive negativity of the sample and the Sample Similarity calculates the sample facial image Current percent of pass and current False Rate, comprising:
It is successively chosen in order in the targets threshold section and undetermined passes through score threshold;
According to Sample Similarity scoring and the veritification undetermined by score threshold judgement sample as a result, according to institute It states and veritifies the current percent of pass that result calculates the sample facial image;
According to the positive negativity of the sample and the current False Rate veritified result and calculate the sample facial image.
In the concrete realization, it is similar that multiple positive samples under each business scenario, multiple negative samples and corresponding sample are obtained Degree scoring, for each business scenario, selection one is undetermined in the targets threshold section of the business scenario passes through score threshold Value, when Sample Similarity scoring lower than it is described it is undetermined pass through score threshold when, it is believed that Sample Similarity scores corresponding sample Veritify result be do not pass through, otherwise it is assumed that veritify result be pass through, according to it is described veritification result be by sample number divided by total Sample number calculates current percent of pass;It according to the positive negativity of sample and veritifies result simultaneously and calculates current False Rate, when sample is positive sample This, and veritify result be by when, be denoted as correct judgment, when sample be positive sample, and veritify result be it is obstructed out-of-date, be denoted as and sentence Dislocation miss, when sample be negative sample, and veritify result be by when, be denoted as misjudgment, when sample be negative sample, and veritify tie Fruit be it is obstructed out-of-date, correct judgment is denoted as, to calculate current False Rate according to judicious sample number and total number of samples.
In a second embodiment, veritification precision and the targets threshold section under business scenario are obtained;In the target threshold It chooses in value section and meets the precision of veritifying and pass through score threshold.Based on the sample data of many experiments room, acquisition most can Meet the veritification precision of each business scenario passes through score threshold.
It is the flow diagram of smart identity checking method 3rd embodiment of the present invention referring to Fig. 4, Fig. 4, is based on above-mentioned Fig. 2 Shown in first embodiment, propose the 3rd embodiment of smart identity checking method of the present invention.
In the third embodiment, after the step S50, the method also includes:
Step S501: when first similarity score passes through score threshold not higher than the target, judge described the Whether one similarity score is lower than the first preset standard score, and the first preset standard score passes through score lower than the target Threshold value.
It is understood that when first similarity score is higher than the target and passes through score threshold, described in judgement User identity veritification passes through;When first similarity score passes through score threshold not higher than the target, if directly determining User identity veritification does not pass through, and is likely to result in part real user and is mistaken for puppet and emits user, therefore, will be arranged secondary Identity is veritified to improve identity veritification ground accuracy and specifically determine whether to carry out secondary body by the first preset standard score Part is veritified.
Step S502: when first similarity score is lower than the first preset standard score, determine the user identity Veritification does not pass through.
It should be noted that the first preset standard score passes through score threshold lower than the target, if described first Similarity score still is below the first preset standard score, illustrates that a possibility that user is real user is very low, at this point, Determine that the user identity veritification does not pass through.
Step S503: when first similarity score is not less than the first preset standard score, the user is carried out Secondary identity is veritified.
It is understood that when first similarity score is not less than the first preset standard score, user's tool Having certain may be real user, will carry out secondary identity veritification to the user.
Further, the step S503, specifically includes:
When first similarity score is not less than the first preset standard score, obtained according to the subscriber identity information Target vocal print feature information;
The audio for acquiring the user extracts current vocal print feature information from the audio;
The target vocal print feature information and the current vocal print feature information are subjected to similarity comparison, and carried out similar Degree marking, obtains the second similarity score;
Second similarity score and the second preset standard score are compared, when second similarity is commented When dividing higher than the second preset standard score, determine that the user identity veritification passes through.
It should be noted that being known when first similarity score is not less than the first preset standard score by vocal print It is other that secondary identity veritification is carried out to the user.
In the concrete realization, target vocal print feature information, the target vocal print feature are obtained according to the subscriber identity information Information is vocal print feature information corresponding with the subscriber identity information, namely true vocal print feature information.Subscriber identity information with Second corresponding relationship of vocal print feature information can be stored in smart identity apparatus for checking local or be stored in background server; When second corresponding relationship is stored in smart identity apparatus for checking local, search speed can be accelerated, improve user experience;When this When second corresponding relationship is stored in background server, local memory space can be saved, reduces processing locality pressure.Smart identity Apparatus for checking will locally search corresponding target vocal print feature information according to the subscriber identity information, or receive background service The target vocal print feature information of device feedback.
It is understood that will carry out Application on Voiceprint Recognition after obtaining target vocal print feature information to user, and acquire the use The audio at family, and feature extraction is carried out to the audio, obtain current vocal print feature information.Due to the target vocal print feature information For the corresponding vocal print feature information of subscriber identity information, if the current vocal print feature information and the target vocal print feature information Similarity is higher, then it is believed that user user corresponding with the subscriber identity information is same people, that is, can determine that described User identity veritification passes through.Thus, the target vocal print feature information and the current vocal print feature information are subjected to similarity It compares, and carries out similarity marking, obtain the second similarity score.The second preset standard score is what Application on Voiceprint Recognition passed through By score threshold, second similarity score and the second preset standard score are compared, if second similarity When scoring is higher than the second preset standard score, determine that the user identity veritification passes through.
In the third embodiment, before the step S30, the method also includes:
Detect whether the user is living body;
When the user is living body, the facial image of the acquisition user is executed, and from the facial image The step of extracting current face's characteristic information;
When the user is non-living body, then determine that the user identity veritification does not pass through.
It should be noted that in order to avoid the user not instead of true man, photo, video, static state or 3D model Attack, will test whether the user is living body, when the active user is non-living body, terminates identity and veritifies, work as described When preceding user is living body, continue identity veritification.
In the concrete realization, whether the detection active user is living body, comprising:
Prompt the user to carry out specified limb action, such as the right hand takes left shoulder, detect the user movement whether Correctly, when the movement of the user is consistent with specified limb action, assert that the user is living body;Alternatively, by applying journey Sequence (such as wechat) detects whether the user is living body.
In the third embodiment, whether it is living body by detecting the user, can effectively avoids photo, video, static state Or the attack of 3D model;And after recognition of face, secondary identity veritification is carried out by Application on Voiceprint Recognition to improve identity veritification Ground accuracy avoids part real user from being mistaken for puppet and emits user, improves user experience.
In addition, the embodiment of the present invention also proposes a kind of storage medium, smart identity veritification is stored on the storage medium Program, the smart identity veritify the step that smart identity checking method as described above is realized when program is executed by processor Suddenly.
In addition, the embodiment of the present invention also proposes that a kind of smart identity veritifies device, and the smart identity is veritified referring to Fig. 5 Device includes:
Threshold value obtains module 10, passes through score threshold for obtaining the corresponding target of current business scene;
Fisrt feature obtains module 20, veritifies the user that instruction obtains the user for the identity according to the user received Identity information, and target face characteristic information is obtained according to the subscriber identity information;
Second feature obtains module 30, extracts for acquiring the facial image of the user, and from the facial image Current face's characteristic information out;
Scoring modules 40, for the target face characteristic information and current face's characteristic information to be carried out similarity It compares, and carries out similarity marking, obtain the first similarity score;
Module 50 is veritified, for first similarity score to be compared with the target by score threshold, when When first similarity score passes through score threshold higher than the target, determine that the user identity veritification passes through.
In one embodiment, the smart identity veritifies device further include: threshold determination module, for obtaining business scenario Under veritification precision and targets threshold section;
Chosen in the targets threshold section meet it is described veritify precision by score threshold, and establish business scenario With the default mapping relations for passing through score threshold;
Current business scene is obtained, and searches mesh corresponding with the current business scene in the default mapping relations Mark passes through score threshold.
In one embodiment, the threshold determination module is also used to obtain the sample data under business scenario, the sample Data include sample facial image, the positive negativity of sample and Sample Similarity scoring;
Successively chosen in order in the targets threshold section it is undetermined by score threshold, according to it is described it is undetermined by point Number threshold values, the positive negativity of the sample and Sample Similarity scoring calculate the current percent of pass of the sample facial image and work as Preceding False Rate;
It is not less than the target not higher than the target percent of pass and/or the current False Rate in the current percent of pass When False Rate, selection is next undetermined by score threshold, and returns described according to described undetermined by score threshold, the sample This positive negativity and Sample Similarity scoring calculate the current percent of pass of the sample facial image and the step of current False Rate Suddenly;
It is higher than the target percent of pass in the current percent of pass, and the current False Rate is lower than the target False Rate When, undetermined score threshold is passed through as the precision of veritifying is met by score threshold for described.
In one embodiment, the threshold determination module is also used to successively select in order in the targets threshold section It takes and undetermined passes through score threshold;
According to Sample Similarity scoring and the veritification undetermined by score threshold judgement sample as a result, according to institute It states and veritifies the current percent of pass that result calculates the sample facial image;
According to the positive negativity of the sample and the current False Rate veritified result and calculate the sample facial image.
In one embodiment, the smart identity veritifies device further include: secondary veritification module, for working as first phase When passing through score threshold not higher than the target like degree scoring, judge whether first similarity score is lower than the first pre- bidding Quasi- score, the first preset standard score pass through score threshold lower than the target;
When first similarity score is lower than the first preset standard score, it is obstructed to determine that the user identity is veritified It crosses;
When first similarity score is not less than the first preset standard score, secondary identity core is carried out to the user It tests.
In one embodiment, the secondary veritification module is also used to pre- not less than first when first similarity score If when criterion score, acquiring the audio of the user, current vocal print feature information is extracted from the audio;
Target vocal print feature information is obtained according to the subscriber identity information;
The target vocal print feature information and the current vocal print feature information are subjected to similarity comparison, and carried out similar Degree marking, obtains the second similarity score;
Second similarity score and the second preset standard score are compared, when second similarity is commented When dividing higher than the second preset standard score, determine that the user identity veritification passes through.
In one embodiment, the smart identity veritifies device further include: In vivo detection module, for detecting the user It whether is living body;
When the user is living body, the facial image of the acquisition user is executed, and from the facial image The step of extracting current face's characteristic information;
When the user is non-living body, then determine that the user identity veritification does not pass through.
Smart identity of the present invention veritifies the other embodiments of device or specific implementation can refer to above-mentioned each method Embodiment, details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.If listing equipment for drying Unit claim in, several in these devices, which can be, to be embodied by the same item of hardware.Word first, Second and the use of third etc. do not indicate any sequence, can be title by these word explanations.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art The part contributed out can be embodied in the form of software products, which is stored in a storage medium (such as read-only memory mirror image (Read Only Memory image, ROM)/random access memory (Random Access Memory, RAM), magnetic disk, CD) in, including some instructions use so that a terminal intelligent identity apparatus for checking (can be Mobile phone, computer, server, air conditioner or network intelligence identity apparatus for checking etc.) it executes described in each embodiment of the present invention Method.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills Art field, is included within the scope of the present invention.

Claims (10)

1. a kind of smart identity checking method, which is characterized in that the smart identity checking method the following steps are included:
It obtains the corresponding target of current business scene and passes through score threshold;
The subscriber identity information that instruction obtains the user is veritified according to the identity of the user received, and according to the user identity Acquisition of information target face characteristic information;
The facial image of the user is acquired, and extracts current face's characteristic information from the facial image;
The target face characteristic information and current face's characteristic information are subjected to similarity comparison, and carries out similarity and beats Point, obtain the first similarity score;
First similarity score is compared with the target by score threshold, when first similarity score is high When the target passes through score threshold, determine that the user identity veritification passes through.
2. smart identity checking method as described in claim 1, which is characterized in that the acquisition current business scene is corresponding Target passes through score threshold, comprising:
Obtain the veritification precision under business scenario and targets threshold section;
Chosen in the targets threshold section meet it is described veritify precision by score threshold, and establish business scenario and logical Cross the default mapping relations of score threshold;
Current business scene is obtained, and searches target corresponding with the current business scene in the default mapping relations and leads to Cross score threshold.
3. smart identity checking method as claimed in claim 2, which is characterized in that the veritification precision includes target percent of pass And target False Rate;
Described choose in the targets threshold section meets the precision of veritifying and passes through score threshold, comprising:
The sample data under business scenario is obtained, the sample data includes sample facial image, the positive negativity of sample and sample phase It scores like degree;
It is successively chosen in order in the targets threshold section undetermined by score threshold, undetermined passes through score threshold according to described It is worth, the positive negativity of the sample and Sample Similarity scoring calculate the current percent of pass of the sample facial image and currently miss Sentence rate;
It is judged by accident not higher than the target percent of pass and/or the current False Rate not less than the target in the current percent of pass When rate, choose it is next undetermined by score threshold, and return it is described according to it is described it is undetermined by score threshold, the sample just Negativity and Sample Similarity scoring calculate the step of current percent of pass of the sample facial image is with current False Rate;
The current percent of pass be higher than the target percent of pass, and the current False Rate be lower than the target False Rate when, Undetermined score threshold is passed through as the precision of veritifying is met by score threshold for described.
4. smart identity checking method as claimed in claim 3, which is characterized in that described to be pressed in the targets threshold section Sequence is successively chosen undetermined by score threshold, undetermined passes through score threshold, the positive negativity of the sample and the sample according to described This similarity score calculate the sample facial image current percent of pass and current False Rate, comprising:
It is successively chosen in order in the targets threshold section and undetermined passes through score threshold;
According to Sample Similarity scoring and the veritification undetermined by score threshold judgement sample as a result, according to the core Test the current percent of pass that result calculates the sample facial image;
According to the positive negativity of the sample and the current False Rate veritified result and calculate the sample facial image.
5. such as smart identity checking method of any of claims 1-4, which is characterized in that described by first phase After being compared with the target by score threshold like degree scoring, the method also includes:
When first similarity score passes through score threshold not higher than the target, judge that first similarity score is No to be lower than the first preset standard score, the first preset standard score passes through score threshold lower than the target;
When first similarity score is lower than the first preset standard score, determine that the user identity veritification does not pass through;
When first similarity score is not less than the first preset standard score, secondary identity veritification is carried out to the user.
6. smart identity checking method as claimed in claim 5, which is characterized in that described to work as first similarity score not When lower than the first preset standard score, secondary identity veritification is carried out to the user, comprising:
When first similarity score is not less than the first preset standard score, the audio of the user is acquired, from the sound Current vocal print feature information is extracted in frequency;
Target vocal print feature information is obtained according to the subscriber identity information;
The target vocal print feature information and the current vocal print feature information are subjected to similarity comparison, and carries out similarity and beats Point, obtain the second similarity score;
Second similarity score and the second preset standard score are compared, when second similarity score is high When the second preset standard score, determine that the user identity veritification passes through.
7. such as smart identity checking method of any of claims 1-4, which is characterized in that the acquisition user Facial image, and before extracting current face's characteristic information in the facial image, comprising:
Detect whether the user is living body;
When the user is living body, the facial image of the acquisition user is executed, and is extracted from the facial image The step of current face's characteristic information out;
When the user is non-living body, then determine that the user identity veritification does not pass through.
8. a kind of smart identity apparatus for checking, which is characterized in that the smart identity apparatus for checking includes: memory, processor And be stored in the smart identity that can be run on the memory and on the processor and veritify program, the smart identity is veritified The step of smart identity checking method as described in any one of claims 1 to 7 is realized when program is executed by the processor.
9. a kind of storage medium, which is characterized in that be stored with smart identity on the storage medium and veritify program, the intelligence body Part veritifies the step that the smart identity checking method as described in any one of claims 1 to 7 is realized when program is executed by processor Suddenly.
10. a kind of smart identity veritifies device, which is characterized in that the smart identity veritifies device and includes:
Threshold value obtains module, passes through score threshold for obtaining the corresponding target of current business scene;
Fisrt feature obtains module, veritifies the user identity letter that instruction obtains the user for the identity according to the user received Breath, and target face characteristic information is obtained according to the subscriber identity information;
Second feature obtains module, extracts currently for acquiring the facial image of the user, and from the facial image Face characteristic information;
Scoring modules, for the target face characteristic information and current face's characteristic information to be carried out similarity comparison, And similarity marking is carried out, obtain the first similarity score;
Module is veritified, for comparing first similarity score by score threshold with the target, when described the When one similarity score passes through score threshold higher than the target, determine that the user identity veritification passes through.
CN201811206298.6A 2018-10-16 2018-10-16 Intelligent identity verification method, equipment, storage medium and device Active CN109635625B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811206298.6A CN109635625B (en) 2018-10-16 2018-10-16 Intelligent identity verification method, equipment, storage medium and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811206298.6A CN109635625B (en) 2018-10-16 2018-10-16 Intelligent identity verification method, equipment, storage medium and device

Publications (2)

Publication Number Publication Date
CN109635625A true CN109635625A (en) 2019-04-16
CN109635625B CN109635625B (en) 2023-08-18

Family

ID=66066349

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811206298.6A Active CN109635625B (en) 2018-10-16 2018-10-16 Intelligent identity verification method, equipment, storage medium and device

Country Status (1)

Country Link
CN (1) CN109635625B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110298243A (en) * 2019-05-22 2019-10-01 深圳壹账通智能科技有限公司 Testimony verification method, apparatus, computer equipment and computer readable storage medium
CN111583451A (en) * 2020-04-09 2020-08-25 惠州拓邦电气技术有限公司 Identity verification method and device of electronic lock, computer equipment and storage medium
CN111914769A (en) * 2020-08-06 2020-11-10 腾讯科技(深圳)有限公司 User validity judging method, device, computer readable storage medium and equipment
CN113705455A (en) * 2021-08-30 2021-11-26 平安银行股份有限公司 Identity verification method and device, electronic equipment and readable storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101751551A (en) * 2008-12-05 2010-06-23 比亚迪股份有限公司 Method, device, system and device for identifying face based on image
CN105512535A (en) * 2016-01-08 2016-04-20 广东德生科技股份有限公司 User authentication method and user authentication device
WO2016131063A1 (en) * 2015-02-15 2016-08-18 Alibaba Group Holding Limited System and method for user identity verification, and client and server by use thereof
CN106961418A (en) * 2017-02-08 2017-07-18 北京捷通华声科技股份有限公司 Identity identifying method and identity authorization system
CN107066983A (en) * 2017-04-20 2017-08-18 腾讯科技(上海)有限公司 A kind of auth method and device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101751551A (en) * 2008-12-05 2010-06-23 比亚迪股份有限公司 Method, device, system and device for identifying face based on image
WO2016131063A1 (en) * 2015-02-15 2016-08-18 Alibaba Group Holding Limited System and method for user identity verification, and client and server by use thereof
CN105512535A (en) * 2016-01-08 2016-04-20 广东德生科技股份有限公司 User authentication method and user authentication device
CN106961418A (en) * 2017-02-08 2017-07-18 北京捷通华声科技股份有限公司 Identity identifying method and identity authorization system
CN107066983A (en) * 2017-04-20 2017-08-18 腾讯科技(上海)有限公司 A kind of auth method and device

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110298243A (en) * 2019-05-22 2019-10-01 深圳壹账通智能科技有限公司 Testimony verification method, apparatus, computer equipment and computer readable storage medium
CN111583451A (en) * 2020-04-09 2020-08-25 惠州拓邦电气技术有限公司 Identity verification method and device of electronic lock, computer equipment and storage medium
CN111914769A (en) * 2020-08-06 2020-11-10 腾讯科技(深圳)有限公司 User validity judging method, device, computer readable storage medium and equipment
CN111914769B (en) * 2020-08-06 2024-01-26 腾讯科技(深圳)有限公司 User validity determination method, device, computer readable storage medium and equipment
CN113705455A (en) * 2021-08-30 2021-11-26 平安银行股份有限公司 Identity verification method and device, electronic equipment and readable storage medium
CN113705455B (en) * 2021-08-30 2024-03-19 平安银行股份有限公司 Identity verification method and device, electronic equipment and readable storage medium

Also Published As

Publication number Publication date
CN109635625B (en) 2023-08-18

Similar Documents

Publication Publication Date Title
CN109635625A (en) Smart identity checking method, equipment, storage medium and device
CN109657554B (en) Image identification method and device based on micro expression and related equipment
CN101661557B (en) Face recognition system and face recognition method based on intelligent card
CN102945366B (en) A kind of method and device of recognition of face
CN106295482B (en) A kind of update method and device of face database
US20170286790A1 (en) Biometric Enrollment Systems And Methods
CN106295591A (en) Gender identification method based on facial image and device
CN101216884A (en) A method and system for face authentication
CN108734067A (en) A kind of authentication method, system and camera that the testimony of a witness compares
CN104036254A (en) Face recognition method
CN109829071A (en) Face image searching method, server, computer equipment and storage medium
CN109389098A (en) A kind of verification method and system based on lip reading identification
CN112836661A (en) Face recognition method and device, electronic equipment and storage medium
CN112241667A (en) Image detection method, device, equipment and storage medium
CN106709488A (en) Business card identification method and device
CN113076886A (en) Face individual identification device and method for cat
CN109508685A (en) Power communication dispatching method based on face recognition technology
CN113630721A (en) Method and device for generating recommended tour route and computer readable storage medium
CN103984415B (en) A kind of information processing method and electronic equipment
CN111062345B (en) Training method and device for vein recognition model and vein image recognition device
CN112766065A (en) Mobile terminal examinee identity authentication method, device, terminal and storage medium
CN106548112A (en) A kind of personal identification method and system
CN112560728B (en) Target object identification method and device
CN111753656B (en) Feature extraction method, device, equipment and computer readable storage medium
CN116503918A (en) Palm vein image classification method, device, equipment and medium based on ViT network

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
GR01 Patent grant
GR01 Patent grant