CN107844748B - Auth method, device, storage medium and computer equipment - Google Patents

Auth method, device, storage medium and computer equipment Download PDF

Info

Publication number
CN107844748B
CN107844748B CN201710965802.XA CN201710965802A CN107844748B CN 107844748 B CN107844748 B CN 107844748B CN 201710965802 A CN201710965802 A CN 201710965802A CN 107844748 B CN107844748 B CN 107844748B
Authority
CN
China
Prior art keywords
image
certificate
data
face
human face
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.)
Active
Application number
CN201710965802.XA
Other languages
Chinese (zh)
Other versions
CN107844748A (en
Inventor
郑佳
赵骏
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 CN201710965802.XA priority Critical patent/CN107844748B/en
Priority to PCT/CN2017/112485 priority patent/WO2019075840A1/en
Publication of CN107844748A publication Critical patent/CN107844748A/en
Application granted granted Critical
Publication of CN107844748B publication Critical patent/CN107844748B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/30—Scenes; Scene-specific elements in albums, collections or shared content, e.g. social network photos or video
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/174—Facial expression recognition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/08—Network architectures or network communication protocols for network security for authentication of entities

Abstract

The present invention relates to a kind of auth method, device, storage medium and computer equipments.This method comprises: receiving authentication instruction；Multiframe target image is obtained by photographic device according to authentication instruction；Extract the human face region image and certificate area image in every frame target image；The human face region image in multiframe target image is parsed, face delta data is generated；Judge whether the face delta data matches with default delta data；When the face delta data is matched with default delta data, then corresponding certificate area image is verified according to multiframe human face region image, determined whether to pass through authentication according to verification result.Above-mentioned auth method, device, storage medium and computer equipment avoids malicious user from falsely using other people certificates and carries out the risk of authentication, to improve the safety of online identity verifying.

Description

Auth method, device, storage medium and computer equipment
Technical field
The present invention relates to technical field of network security, more particularly to a kind of auth method, device, storage medium and Computer equipment.
Background technique
With popularizing for network, many network systems are required than system of real name social platform, internet financial platform etc. User's real name verification in registration.It usurps other people identity informations in order to avoid malicious user to register, network system is often It is required that the photo that user uploads the certificates such as identity card is audited.Network system can pass through OCR (Optical Character Recognition, optical character identification) personal information such as name identification card number on technology identification certificate photograph.
However, the auth method of traditional identity-based license piece can only by OCR technique identify certificate photograph into Row obtains the personal information of user, and is not avoided that malicious user steals other people identity cards, by other people identity card pictures into The safety of the case where row registration, online identity verifying are lower.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of auth method, device, storage medium and meter Calculate machine equipment.
A kind of auth method, comprising: receive authentication instruction；Pass through camera shooting dress according to authentication instruction It sets and obtains multiframe target image；Extract the human face region image and certificate area image in every frame target image；Parse multiframe mesh Human face region image in logo image generates face delta data；Judge the face delta data whether with default variation number According to matching；When the face delta data is matched with default delta data, then according to multiframe human face region image to corresponding card Part area image is verified, and is determined whether to pass through authentication according to verification result.
It is described in one of the embodiments, that multiframe target figure is obtained by photographic device according to authentication instruction Picture, comprising: instructed according to the authentication, multiframe is obtained by photographic device according to prefixed time interval in preset duration Image；Detect whether the both greater than default resolution ratio of the resolution ratio of every frame image；If so, using every frame image as target image； If it is not, then return continue to execute it is described in preset duration according to prefixed time interval by photographic device acquisition multiple image, Until the both greater than default resolution ratio of the resolution ratio of every frame image.
The face delta data includes expression shape change degree in one of the embodiments,；The parsing multiframe target figure Human face region image as in generates face delta data, comprising: extract the table of the human face region image in every frame target image Feelings characteristic；According to the expressive features data of every two frames adjacent target image, expression shape change degree is calculated to obtain；Described in the judgement Whether face delta data matches with default delta data, comprising: judges whether each expression shape change degree is in default variation It spends in section；When each expression shape change degree is in default change degree section, then face delta data and default change are determined Change Data Matching.
Multiframe target is obtained by photographic device according to authentication instruction described in one of the embodiments, Before image, further includes: obtain and show random expression prompt information；In the human face region extracted in every frame target image After the expressive features data of image, further includes: in the expressive features data of Detection and Extraction, if exist and the random expression The expressive features data that prompt information matches；When there are the expressive features data to match with the expression prompt information, The expressive features data according to every two frames adjacent target image are executed, expression shape change degree is calculated to obtain.
The face delta data includes face motion track in one of the embodiments,；Described according to the body Part verifying instruction is obtained by photographic device before multiframe target image, further includes: is obtained and is shown random face track prompt Information；Human face region image in the parsing multiframe target image generates face delta data, comprising: extracts every frame target figure The face location data of human face region image as in；Face moving rail is generated according to the face location data of every frame target image Mark；It is described to judge whether the face delta data matches with default delta data, comprising: when the face motion track and institute When stating the matching of random face track prompt information, then determine that face delta data is matched with default delta data.
Multiframe target is obtained by photographic device according to authentication instruction described in one of the embodiments, Before image, further includes: obtain and show random certificate track prompt information；It is described according to multiframe human face region image to phase Before answering certificate area image to be verified, further includes: extract the certificate positional number of certificate area image in every frame target image According to；Certificate motion track is generated according to the certificate position data in every frame target image；Whether judge the certificate motion track It is matched with certificate track prompt information；When the certificate motion track is matched with the random certificate track prompt information, And the face delta data is with default delta data when matching, execute it is described according to multiframe human face region image to corresponding certificate Area image is verified.
It is described in one of the embodiments, that corresponding certificate area image is tested according to multiframe human face region image Card, comprising: extract in every frame target image face area in the certificate photo characteristic of certificate area image and respective objects image The face characteristic data of area image；Judge whether there is what the certificate photo characteristic matched with the face characteristic data Target image；It is wrapped in the target image that the certificate photo characteristic matches with the face characteristic data if so, extracting The identity information data contained；Obtain verifying human face data corresponding with the identity information data in certificate database；Judgement Whether the verifying human face data matches with the certificate photo characteristic；If so, determining to pass through authentication.
A kind of authentication means, described device include: authentication command reception module, are referred to for receiving authentication It enables；Target image obtains module, for obtaining multiframe target image by photographic device according to authentication instruction；Region Image zooming-out module, for extracting human face region image and certificate area image in every frame target image；Face delta data Matching module generates face delta data for parsing the human face region image in multiframe target image；Authentication module, For judging whether the face delta data matches with default delta data；When the face delta data and default variation number When according to matching, then corresponding certificate area image is verified according to multiframe human face region image, is according to verification result judgement It is no to pass through authentication.
A kind of computer readable storage medium, is stored thereon with computer program, realization when which is executed by processor The step of above-mentioned each auth method as described in the examples.
A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage Computer program, the processor realize the step of above-mentioned each auth method as described in the examples when executing described program Suddenly.
Above-mentioned auth method, device, storage medium and computer equipment, based on the received authentication instruction obtain Multiframe target image, and simultaneously include human face region image and certificate area image in target image, by parsing target image In human face region image and generate face delta data, sentenced according to face delta data and the matching result of default delta data Whether disconnected human face region image meets the requirements, when face delta data is matched with default delta data, i.e. multiframe target image In human face region image when meeting the requirements, certificate area image is verified according to human face region image, further according to certificate The verification result of area image determines whether to pass through authentication.The dynamic of face in target image is obtained according to face delta data State, it is ensured that the authenticity of face, further according to human face region image corresponding to the face with authenticity to certificate area image It is verified and malicious user is avoided to falsely use the risk that other people certificates carry out authentication to verify to user identity, thus Improve the safety of online identity verifying.
Detailed description of the invention
Fig. 1 is the applied environment figure of auth method in one embodiment；
Fig. 2 is the flow chart of auth method in one embodiment；
Fig. 3 is the schematic diagram of target image in one embodiment；
Fig. 4 is the flow chart of auth method in another embodiment；
Fig. 5 is the flow chart of auth method in another embodiment；
Fig. 6 is the structural block diagram of authentication means in one embodiment；
Fig. 7 is the structural block diagram of authentication means in another embodiment；
Fig. 8 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the present invention, not For limiting the present invention.
It is appreciated that term " first " used in the present invention, " second " etc. can be used to describe various elements herein, But these elements are not limited by these terms.These terms are only used to distinguish the first element from the other element.Citing For, without departing from the scope of the invention, first threshold can be known as to second threshold, and similarly, it can be by Two threshold values are known as first threshold.First threshold and second threshold both threshold value, but it is not same threshold value.
Auth method provided by the embodiment of the present invention can be applied in application environment as shown in Figure 1.Referring to figure 1, which includes: terminal 102 and server 104.Terminal 102 can be mobile phone, tablet computer, personal digital assistant or Person's smart machine etc..Server 104 can be independent physical server, be also possible to the service that multiple physical servers are constituted Device cluster.Terminal 102 can be used for executing auth method provided by the embodiment of the present invention.It can be stored in server 104 Data relevant to authentication, including but not limited to identity information data and verifying human face data etc..Terminal 102 can be with clothes Business device 104 is connected to the network, and is carried out data transmission.Such as server 104 can receive the identity information number of the transmission of terminal 102 According to, and corresponding verifying human face data is searched according to identity information data, terminal 102 can also obtain testing for the transmission of server 104 Demonstrate,prove human face data etc..
In one embodiment, as shown in Fig. 2, providing a kind of auth method, this method can be used for as shown in Figure 1 Application environment in terminal 102, this method comprises:
Step S202 receives authentication instruction.
Authentication instruction can be in the authentication instruction for detecting the operation for carrying out authentication and triggering.Wherein, The operation include but is not limited to pre-set clicking operation, slide and rock operation etc. one or more of them.Into one Step ground can provide corresponding authentication interface for the operation for carrying out authentication, include corresponding on the interface for connecing The control for receiving authentication instruction, when detecting the clicking operation for acting on the control, triggering generates authentication instruction.
Step S204 obtains multiframe target image by photographic device according to authentication instruction.
Photographic device can be preset in terminal, terminal can also be with external photographic device.Wherein, photographic device can be mobile phone Camera, camera and video camera etc..Target image refers to the image comprising target information or comprising target reference object, such as Say, target reference object can be two objects of certificate and face, then can using the image simultaneously comprising certificate and face as Target image.Wherein, image can be the image that photographic device is directly shot by scanning visible area, be also possible to lead to Cross the image intercepted in the video flowing of shooting.After receiving authentication instruction, it can be clapped by photographic device It takes the photograph, and the image obtained in real time is shown by display screen, it can be using the image comprising target reference object as target image.
In one embodiment, terminal can also show corresponding prompt information, allow users to according to the prompt information Determine the position that target reference object, target reference object are in image and the movement that target reference object needs to carry out Deng.It is right for certificate and face two such as shown in figure 3, provide the schematic diagram of target image in one embodiment As that can show certificate region dotted line frame 302 and human face region dotted line frame 304 respectively, be wrapped when detecting in the dotted line frame in image It, can be using the image as target image when containing corresponding target reference object.
In one embodiment, it is instructed according to authentication and multiframe target image is obtained by photographic device, comprising: according to Authentication instruction, obtains multiple image by photographic device according to prefixed time interval in preset duration；It detects whether every The both greater than default resolution ratio of the resolution ratio of frame image.
Image can be persistently obtained according to prefixed time interval by photographic device until reaching pre-set image quantity.Than Such as, every one image of acquisition in 0.1 second or 0.2 second, until getting 20 images.It can also be shot by photographic device default The video of duration, the video flowing obtained to shooting carry out image interception according to the preset time interval.It is possible to further be directed to Image process is obtained, on a display screen by the image real-time exhibition of photographic device shooting, and is provided in respective interface and starts to obtain It takes the control of image and stops obtaining the control of image, when detecting the clicking operation for acting on control, execution is opened accordingly Begin to obtain image or stops obtaining the operation of image.After getting multiple image, every frame image in multiple image is detected Resolution ratio, when the resolution ratio of every frame image both greater than default resolution ratio, then using every frame image as target image；When detecting Less than default resolution ratio image when, then return and continue to execute and passes through photographic device according to prefixed time interval in preset duration Multiple image is obtained, until the both greater than default resolution ratio of the resolution ratio of every frame image.
Step S206 extracts human face region image and certificate area image in every frame target image.
It include two target reference objects of certificate and face in target image.Human face region image, which refers in target image, to be wrapped The area image of the part containing face, correspondingly, certificate area image refers to the area image in target image comprising certificate part. Wherein, certificate includes but is not limited to identity card, driver's license, passport etc..
Target reference object can be passed through into preset figure when detecting in target image comprising target reference object As partitioning algorithm comes out the image region segmentation that certificate part and face part occupy.Wherein, image segmentation algorithm can be Dividing method based on threshold value, the dividing method based on region, the dividing method based on edge and point based on specific theory Segmentation method etc..
In one embodiment, the human face region in target image and certificate region can also be preset.For example, will figure Certificate region dotted line frame 302 in 3 is used as certificate region, regard human face region dotted line frame 304 as human face region.It is pre- detecting If have corresponding target reference object in region, directly by the image zooming-out of predeterminable area come out be used as human face region image and Certificate area image.
Step S208 parses the human face region image in multiframe target image, generates face delta data.
Face delta data refers to the data of human face region image change process in reflection multiframe target image.Such as Expression shape change degree or face motion track.It all include human face region image in every frame target image, when extracting every frame target figure After human face region image as in, face delta data is generated according to the difference of every adjacent two frames human face region image.
Step S210, judges whether face delta data matches with default delta data.
According to the face delta data of generation and default delta data, compare whether the two matches, if so, entering step S212；Otherwise, return step S204.
It in one embodiment, can also be again when detecting that face delta data and default delta data mismatch By photographic device obtain multiframe target image, until get with corresponding to the matched face delta data of default delta data Target image.Further, recycle time can also be preset.It is instructed for same authentication, when detecting recycle time When reaching default recycle time, then do not return again to.Such as default recycle time can be 4 times, when detect the 5th face become When changing data and default delta data mismatch, then stop obtaining image, and show unmatched prompt information.
Step S212 verifies corresponding certificate area image according to multiframe human face region image, according to verification result Determine whether to pass through authentication.
The certificate area image in respective objects image can be carried out according to the human face region image in every frame icon image Verifying.Such as may include certificate photo in certificate area image, when human face region image is matched with certificate photo, it can determine that and test It demonstrate,proves successfully.Since human face target reference object itself can be dynamic, and certificate target reference object sheet is as static, therefore Certificate area image in an extractable wherein target image is as the image for verifying, by a frame or multiframe human face region Image is matched with the certificate area image that this is extracted.When certificate area image is proved to be successful, it can determine that and pass through identity Verifying；Otherwise, it is determined that not passing through authentication.
In above-described embodiment, authentication instruction based on the received obtains multiframe target image, and in target image simultaneously Including human face region image and certificate area image, face variation is generated by the human face region image in parsing target image Data judge whether human face region image meets the requirements according to the matching result of face delta data and default delta data, when When face delta data is matched with default delta data, i.e., when the human face region image in multiframe target image meets the requirements, root Certificate area image is verified according to human face region image, determines whether to pass through further according to the verification result of certificate area image Authentication.The dynamic of face in target image is obtained according to face delta data, it is ensured that the authenticity of face, further according to having Human face region image corresponding to the face of authenticity is verified certificate area image to verify to user identity, keeps away Exempt from the risk that malicious user falsely uses other people certificate progress authentications, to improve the safety of online identity verifying.
In one embodiment, face delta data includes expression shape change degree.Parse the face area in multiframe target image Area image generates face delta data, comprising: extract the expressive features data of the human face region image in every frame target image； According to the expressive features data of every two frames adjacent target image, expression shape change degree is calculated to obtain.
Expressive features data refer to the data for the face characteristic that can reflect that human face region image is included.Wherein, face is special Sign includes but is not limited to five features, face shape etc..The spy of face in human face region image can be extracted according to preset algorithm Point is levied, expressive features data are generated according to the related data of characteristic point.Expression shape change degree refers to face between two target images The difference degree for the human face expression that area image is shown.It can be according to the acquisition time sequence of multiframe target image, to every adjacent The expressive features data of two frame target images are calculated, and expression shape change degree is obtained.When detecting the people in two target images When expressive features data corresponding to face area image are inconsistent, expression shape change degree value is greater than 0.Further, expressive features Data are also denoted as the form of feature vector, can two feature vectors corresponding to adjacent two frames target image according to Cosine similarity is calculated in amount dot product formula, since the cosine similarity the big, illustrates that expression difference degree is smaller, therefore can Using the inverse of the cosine similarity as the expression shape change degree of human face region image in corresponding two frames target image.
In one embodiment, facial expression recognition model can be constructed in advance, the human face region image input that will acquire After the facial expression recognition model, corresponding expressive features data are directly generated.Three-dimensional face model can be also pre-established, is led to The case where crossing depth learning algorithm, learning the movement of true three dimensional face expression, extracts the feature on every frame human face region image After point, characteristic point is mapped to three-dimensional face model and forms threedimensional model corresponding with every frame human face region image.According to knot It closes the three-dimensional face model that characteristic point is rebuild and generates expressive features data.
Judge whether face delta data matches with default delta data, comprising: judge whether each expression shape change degree is equal In default change degree section.Wherein, presetting change degree section is the change degree area being made of first threshold and second threshold Between, refer to the value range of change degree corresponding to real user.First threshold refers to real user in two frame target images Minimum change degree generated, second threshold refer to real user maximum change degree generated in two frame target images.If The expression shape change degree of the human face region image of two frame target images is less than first threshold, illustrates table corresponding to human face region image A possibility that feelings characteristic is consistent, and pretends to be real user using static images there are other people；If two frame target images The expression shape change degree of human face region image is greater than second threshold, then it is coherent for illustrating the target image got not.Therefore, may be used Default change degree section is generated according to first threshold and second, judges whether each expression shape change degree is in default change degree area In.When each expression shape change degree is in default change degree section, then face delta data and default variation number are determined According to matching；Otherwise, it is determined that face delta data and default delta data mismatch.
In above-described embodiment, by extracting the expressive features data of the human face region image in every frame target image, and root Expression shape change degree is calculated according to expressive features data, so as to according to the dynamic of human face region image in multiframe target image Variation ensures the authenticity of face corresponding to human face region image, avoids the still photo progress body that malicious user usurps other people The risk of part verifying, to improve the safety of authentication.
In one embodiment, it is instructed according to authentication before obtaining multiframe target image by photographic device, is also wrapped It includes: obtaining and show random expression prompt information；In the expressive features number for extracting the human face region image in every frame target image According to later, further includes: in the expressive features data of Detection and Extraction, if there is the expression to match with random expression prompt information Characteristic；When there are the expressive features data to match with expression prompt information, according to every two frames adjacent target image Expressive features data calculate to obtain expression shape change degree.
The human face expression that random expression prompt information refers to that terminal obtains at random from default expression library prompts to believe accordingly Breath.User can make corresponding expression in face of photographic device according to the random expression prompt information of display.Wherein, expression library is preset There are many related data of human face expression, human face expression includes but is not limited to smile, blink, opening one's mouth, frowning, stucking out one's tongue for middle storage One or more combination in head, the expressions such as nod, shake the head.Random expression is determined by random algorithm, and is got corresponding Random expression prompt information.Wherein, can be using the corresponding text information of the random expression as random expression prompt information, it can be with Describe determining random expression as random expression prompt information by lines.
After determining random expression, default expression characteristic corresponding with the prompt of random expression is obtained.Extract every frame After the expressive features data of human face region image in target image, the expressive features data of more every frame human face region image Whether matched with default expression characteristic.Such as the mouth arc for the expression of smile, in detection expressive features data When degree reaches preset radian threshold value, then it can determine that the expressive features data and default expression characteristic of the frame human face region image According to matching, it can be performed and calculate expression shape change degree.
In one embodiment, face delta data includes face motion track, is taken the photograph being passed through according to authentication instruction Before device acquisition multiframe target image, further includes: obtain and show random face track prompt information；Parse multiframe target Human face region image in image generates face delta data, comprising: extracts the human face region image in every frame target image Face location data；Face motion track is generated according to the face location data of every frame target image；Judge face delta data Whether matched with default delta data, comprising: when face motion track is matched with random face track prompt information, then determine Face delta data is matched with default delta data.
Face motion track refers to the multiframe human face region image moving rail that present position is constituted in respective objects image Mark.It, can be random by this after determining face track corresponding to the prompt information of random face track according to preset random algorithm The text information of face track can also describe determining random face rail as random face track prompt information by lines Mark is as random face track prompt information.Such as text information can be " to moving to left ", " moving right ", " to moving up " etc., go back Face moving direction and distance etc. can be shown by arrow.Face location data are to refer to indicate human face region image corresponding The data of present position in target image.Such as in human face region image human face characteristic point position data.Wherein, face Characteristic point can be the assemblage characteristic point of the one or more of them such as canthus, the corners of the mouth, nose.According to the people in every frame target image The face location data of face area image can generate face motion track according to preset curve fitting algorithm.Judge that face moves Dynamic rail mark whether with random face path matching corresponding to the prompt information of random face track, if so, can determine that face become Change data to match with default delta data, corresponding certificate area image can be verified according to multiframe human face region image, Determined whether to pass through authentication according to verification result；Otherwise, it can determine that face delta data and default delta data mismatch.
In one embodiment, before obtaining multiframe target image by photographic device according to authentication instruction, also Include: acquisition and shows random certificate track prompt information；According to multiframe human face region image to corresponding certificate area image Before being verified, further includes: extract the certificate position data of certificate area image in every frame target image；According to every frame target Certificate position data in image generates certificate motion track；Judge certificate motion track whether with certificate track prompt information Match；When certificate motion track is matched with random certificate track prompt information, and face delta data is matched with default delta data When, execution verifies corresponding certificate area image according to multiframe human face region image.
Certificate motion track refers to the multiframe certificate area image moving rail that present position is constituted in respective objects image Mark.It, can be random by this after determining certificate track corresponding to the prompt information of random certificate track according to preset random algorithm The text information of certificate track can also describe determining random certificate rail as random certificate track prompt information by lines Mark is as random certificate track prompt information.Such as text information can be " to moving to left ", " moving right ", " to moving up " etc., go back Certificate moving direction and distance etc. can be shown by arrow.Certificate position data is to refer to indicate certificate area image corresponding The data of present position in target image.Such as the characteristic point of one or more characters in certificate can be extracted in corresponding mesh Position data in logo image is as certificate position data.According to the certificate position of the certificate area image in every frame target image Data can generate certificate motion track according to preset curve fitting algorithm.Judge certificate motion track whether with random certificate Random certificate path matching corresponding to the prompt information of track, if so, and face delta data when being matched with default delta data, Execution verifies corresponding certificate area image according to multiframe human face region image.
In one embodiment, as shown in figure 4, above-mentioned steps S212, comprising:
Step S402 extracts the certificate photo characteristic and respective objects image of certificate area image in every frame target image The face characteristic data of middle human face region image.
Certificate corresponding to certificate area image in target image includes certificate photo.Certificate photo characteristic refers to basis The characteristic of certificate photo in the certificate area image that face recognition technology is extracted, correspondingly, face characteristic data refer to The characteristic of face in the human face region image extracted according to face recognition technology.Wherein, face recognition technology includes But it is not limited to the recognizer based on human face characteristic point, the recognizer based on whole picture facial image, the identification calculation based on template Method, the algorithm identified using neural network etc..
In one embodiment, the face recognition algorithms identified using neural network can be used, training is for identification The neural network of human face region image.It, can be for the human face region image of photographic device acquisition after initializing neural network As training sample set, acquisition identical image can also be used the certificate area image of the user as training verifying collection, passed through Training sample set and training verifying collection are trained neural network, extract corresponding certificate characteristic and face characteristic number According to until neural network is greater than default discrimination threshold value for the discrimination of certificate characteristic and face characteristic data.It will obtain The target image got inputs trained neural network and is identified, for extracting the certificate photo characteristic in the target image According to face characteristic data, identify whether the human face region image in the target image matches with certificate area image.
Step S404 judges whether there is the target image that certificate photo characteristic and face characteristic data match.
Certificate photo characteristic and face characteristic data can be extracted from every frame target image, when detecting the presence of certificate When the target image to match according to characteristic and face characteristic data, step S406 is executed；Otherwise, S407 is entered step, no It deals with.
Step S406 extracts the identity for including in the target image that certificate photo characteristic and face characteristic data match Information data.
Identity information data refers to the related data of identity information included in certificate area image.Such as work as certificate When certificate corresponding to area image is identity card, then identity information includes but is not limited to name, identification card number, date of birth Day, home address etc..
Identity information data included in certificate area image can be identified by OCR technique.Wherein, OCR technique is logical The text of various bills, newpapers and periodicals, books, manuscript and other printed matters is converted image letter by the optics input modes such as overscanning Breath recycles character recognition technology to convert image information to the computer input technology that can be used.Since discrimination is limited, It can recognize that multiple results.Can be to a variety of results identified according to information legitimacy, similarity, frequency of occurrence etc., it will Multiple identity information datas identified are ranked up, and user can be showed to confirm optimal result.It wherein, can basis Information legitimacy excludes illegal result.For example, whether identification card number, birthdate that detection identifies meet Whether the surname in name that preset value range and detection identify is legal surname, if not, then it is assumed that identity letter It is illegal to cease data, can exclude.Further, it can also be carried out according to similarity, frequency of occurrence of each recognition result etc. Recognition result more than preset threshold is showed user as optional information by weighted calculation.If optional information be it is multiple, can will Multiple optional informations are shown allows user to select in drop-down list.The optional information that user is selected is as in the target image The identity information data for including.
Step S408 obtains verifying human face data corresponding with identity information data in certificate database.
Certificate database refers to that the related data of the preset certificate for being stored with all users, including but not limited to identity are believed Cease data and verifying human face data.Wherein, verifying human face data refers to for verifying certificate photo authenticity in certificate area image Human face data.Verifying human face data can be the characteristic of certificate photo in the certificate extracted in advance.
For example, certificate area image can be ID Card Image, have identified identification card number in ID Card Image it Afterwards, certificate photo corresponding with the identification card number can be obtained in public security system database according to the identification card number.
Step S410 judges to verify whether human face data matches with certificate photo characteristic.
When verifying human face data is matched with certificate photo characteristic, S412 is entered step；When verifying human face data and card When part is mismatched according to characteristic, S414 is entered step.
Step S412, judgement pass through authentication.
Step S414, judgement do not pass through authentication.
In above-described embodiment, according to face characteristic data verification certificate photo characteristic and then according to identity information number According to the verifying human face data obtained in certificate database, by avoiding according to verifying face data verification certificate photo characteristic Malicious user falsely uses other people certificates and patch covers the risk of oneself photo at certificate photo, improves the safety of online identity verifying Property.
In one embodiment, another auth method is provided, as shown in connection with fig. 5, this method comprises:
Step S502 receives authentication instruction.
Terminal can provide the control for triggering authentication instruction, when detecting the trigger action to the control, touching Send out authentication instruction.
Step S504 is obtained and is shown random expression prompt information.
Expression data library is preset in terminal, after getting random expression prompt information according to preset random algorithm, The random expression prompt information is shown by display screen.
For example, if the random expression prompt information obtained is to smile, it can show that text prompt is believed on a display screen It ceases " smile ", can also allow users to be made according to random expression prompt information in the portrait of predeterminable area displaying smiling face Corresponding smile expression.
Step S506 obtains multiple image by photographic device according to prefixed time interval in preset duration.
It is preset with photographic device in terminal, multiple image can be obtained by the photographic device every 0.1 second, and by aobvious Display screen real-time display image.
Step S508 detects whether the both greater than default resolution ratio of the resolution ratio of every frame image.
If it is not, then return step S506, until the both greater than default resolution ratio of the resolution ratio of every frame image；If so, into Step S510.
Step S510, using every frame image as target image.
Target image refers to the image comprising two target reference objects of certificate and face, and the resolution of every frame target image The both greater than default resolution ratio of rate.
Step S512 extracts human face region image and certificate area image in every frame target image.
Human face region image refers to the area image in target image comprising human face target reference object, certificate area image Refer to the area image in target image comprising certificate target reference object.It can be by preset image partition method by target figure Human face region image and certificate area image as in are split, and are extracted.
Step S514 extracts the expressive features data of the human face region image in every frame target image.
For example, the face recognition algorithms based on human face characteristic point can be used, extract in every frame human face region image The key features such as eye shape and mouth radian.Analytical calculation is carried out to key feature, it is corresponding to generate every frame human face region image Expressive features data.
Step S516, in the expressive features data of Detection and Extraction, if exist and match with random expression prompt information Expressive features data.
If so, entering step S618；If it is not, then return step S604.For example, it is mentioned for the random expression of acquisition Show the case where information is smiles, then needs to detect the expressive features data expression corresponding with " smile " that human face region image includes Whether characteristic matches.For example, whether mouth radian reaches preset threshold.
Step S518 calculates to obtain expression shape change degree according to the expressive features data of every two frames adjacent target image.
Expression shape change degree refers to the difference degree for the human face expression that human face region image is shown between two target images. For example, it can be shown according to human face region expression in two target images
Step S520, judges whether each expression shape change degree is in default change degree section.
If so, entering step S522；If it is not, entering step S534.
Step S522 extracts the certificate photo characteristic and respective objects image of certificate area image in every frame target image The face characteristic data of middle human face region image.
The certificate photo characteristic and human face region figure of certificate area image can be extracted according to identical face recognition algorithms The face characteristic data of picture, and compare whether the two matches.
Step S524 judges whether there is the target image that certificate photo characteristic and face characteristic data match.
If so, entering step S526；If it is not, entering step S534.
Step S526 extracts the identity for including in the target image that certificate photo characteristic and face characteristic data match Information data.
For example, certificate can be identity card, as the spy for detecting that identity card included in certificate area image shines When sign data and the face characteristic data of user match, the identity letter for including in the target image can be extracted by OCR technique Cease data.Wherein, identity information data includes but is not limited to identification card number.
Step S528 obtains verifying human face data corresponding with identity information data in certificate database.
For example, for certificate be identity card the case where, can according to the identification card number extracted public security system body The verifying human face data that corresponding identity card shines is searched in part card database.
Step S530 judges to verify whether human face data matches with certificate photo characteristic.
If so, entering step S532；If it is not, entering step S534.
Step S532, judgement pass through authentication.
Step S534, judgement do not pass through authentication.
In above-described embodiment, by multiple deterministic process, to human face region image in target image and certificate area image It is verified, it is ensured that the authenticity of face corresponding to human face region image, while also avoiding malicious user and falsely using his testimony of a witness Part carries out the risk of authentication, improves the safety of authentication.
In one embodiment, as shown in fig. 6, providing a kind of authentication means 600, which includes: that identity is tested Command reception module 602 is demonstrate,proved, for receiving authentication instruction；Target image obtains module 604, for being referred to according to authentication It enables and multiframe target image is obtained by photographic device；Area image extraction module 606, for extracting the people in every frame target image Face area image and certificate area image；Face delta data matching module 608, for parsing the face in multiframe target image Area image generates face delta data；Authentication module 610, for judge face delta data whether with default variation Data Matching, when face delta data is matched with default delta data, then according to multiframe human face region image to corresponding certificate Area image is verified, and is determined whether to pass through authentication according to verification result.
In one embodiment, target image obtains module 604 and is also used to be instructed according to authentication, in preset duration Multiple image is obtained by photographic device according to prefixed time interval；Detect whether both greater than default point of resolution ratio of every frame image Resolution；If so, using every frame image as target image；If it is not, then return continue to execute in preset duration according to it is default when Between interval by photographic device obtain multiple image, until every frame image resolution ratio both greater than preset resolution ratio.
In one embodiment, face delta data includes expression shape change degree；Face delta data matching module 608 is also used In the expressive features data for extracting the human face region image in every frame target image, according to the expression of every two frames adjacent target image Characteristic calculates to obtain expression shape change degree；It is pre- that authentication module 610 is also used to judge whether each expression shape change degree is in If in change degree section, when each expression shape change degree is in default change degree section, then determine face delta data with Default delta data matching.
In one embodiment, as shown in fig. 7, providing another kind of authentication means 700, the device further include: Random prompt information display module 603, for obtaining and showing random expression prompt information；Face delta data matching module 608 are also used in the expressive features data of Detection and Extraction, if there are the expressive features to match with random expression prompt information Data；When there are the expressive features data to match with expression prompt information, execute according to every two frames adjacent target image Expressive features data calculate to obtain expression shape change degree.
In one embodiment, face delta data includes face motion track；Random prompt information display module 703 is also For obtaining and showing random face track prompt information；Face delta data matching module 708 is also used to extract every frame target The face location data of human face region image in image；It is mobile that face is generated according to the face location data of every frame target image Track；Authentication module 610 is also used to then determine people when face motion track is matched with random face track prompt information Face delta data is matched with default delta data.
In one embodiment, random prompt information display module 603 is also used to obtain and show that random certificate track mentions Show information；Face delta data matching module 608 is also used to extract the certificate position of certificate area image in every frame target image Data；Certificate motion track is generated according to the certificate position data in every frame target image；Judge certificate motion track whether with The matching of certificate track prompt information；Authentication module 610 is also used to believe when certificate motion track and the prompt of random certificate track Breath matching, and when face delta data is matched with default delta data, execution is according to multiframe human face region image to corresponding certificate Area image is verified.
In one embodiment, authentication module 610 is also used to extract certificate area image in every frame target image The face characteristic data of human face region image in certificate photo characteristic and respective objects image；Judge whether there is certificate photo spy The target image that sign data and face characteristic data match；If so, extracting certificate photo characteristic and face characteristic data The identity information data for including in the target image to match；Obtain corresponding with identity information data in certificate database test Demonstrate,prove human face data；Judge to verify whether human face data matches with certificate photo characteristic；If so, determining to pass through authentication.
Above-mentioned authentication means can be implemented as a kind of form of computer program, and computer program can be in such as Fig. 8 Shown in run in computer equipment.
Modules in above-mentioned authentication means can be realized fully or partially through software, hardware and combinations thereof. Above-mentioned each module can be embedded in the form of hardware or independently of in the memory of terminal, can also be stored in terminal in a software form Memory in, execute the corresponding operation of above modules in order to which processor calls.The processor can be central processing Unit (CPU), microprocessor, single-chip microcontroller etc..
As shown in figure 8, for the schematic diagram of internal structure of computer equipment in one embodiment.A kind of computer is provided to set It is standby, including memory, processor and the computer program that can be run on a memory and on a processor is stored, processor is being held It is performed the steps of when line program and receives authentication instruction；Multiframe mesh is obtained by photographic device according to authentication instruction Logo image；Extract the human face region image and certificate area image in every frame target image；Parse the people in multiframe target image Face area image generates face delta data；Judge whether face delta data matches with default delta data；When face changes When data are matched with default delta data, then corresponding certificate area image is verified according to multiframe human face region image, root Determine whether to pass through authentication according to verification result.
In one embodiment, when above-mentioned processor executes program, that is realized passes through camera shooting according to authentication instruction Device obtains multiframe target image, comprising: is instructed according to authentication, according to prefixed time interval by taking the photograph in preset duration As device obtains multiple image；Detect whether the both greater than default resolution ratio of the resolution ratio of every frame image；If so, by every frame image As target image；It is obtained according to prefixed time interval by photographic device in preset duration if it is not, then returning and continuing to execute Multiple image, until the both greater than default resolution ratio of the resolution ratio of every frame image.
In one embodiment, face delta data includes expression shape change degree, when above-mentioned processor executes program, is realized Parsing multiframe target image in human face region image, generate face delta data the step of, specifically includes the following steps: mentioning Take the expressive features data of the human face region image in every frame target image；According to the expressive features of every two frames adjacent target image Data calculate to obtain expression shape change degree；That is realized judges the step of whether face delta data matches with default delta data, tool Body is the following steps are included: judge whether each expression shape change degree is in default change degree section；When each expression shape change degree When being in default change degree section, then determine that face delta data is matched with default delta data.
In one embodiment, when above-mentioned processor executes program, that is realized passes through camera shooting according to authentication instruction It is further comprising the steps of: to obtain and show random expression prompt information before device obtains the step of multiframe target image；Institute is real It is further comprising the steps of after the step of expressive features data of human face region image in the existing every frame target image of extraction: In the expressive features data of Detection and Extraction, if there are the expressive features data to match with random expression prompt information；When depositing In the expressive features data to match with expression prompt information, the expressive features number according to every two frames adjacent target image is executed According to calculating to obtain expression shape change degree.
In one embodiment, face delta data includes face motion track, and when above-mentioned processor executes program, institute is real It is existing according to authentication instruction by before the step of photographic device acquisition multiframe target image, it is further comprising the steps of: to obtain It takes and shows random face track prompt information；Human face region image in the parsing multiframe target image realized generates face The step of delta data, specifically includes the following steps: extracting the face location number of the human face region image in every frame target image According to；Face motion track is generated according to the face location data of every frame target image；The judgement face delta data realized is It no the step of being matched with default delta data, specifically includes: when face motion track is matched with random face track prompt information When, then determine that face delta data is matched with default delta data.
In one embodiment, when above-mentioned processor executes program, that is realized passes through camera shooting according to authentication instruction It is further comprising the steps of: to obtain and show random certificate track prompt information before device obtains the step of multiframe target image； Realized the step of verifying according to multiframe human face region image to corresponding certificate area image before, further include following step It is rapid: to extract the certificate position data of certificate area image in every frame target image；According to the certificate position in every frame target image Data generate certificate motion track；Judge whether certificate motion track matches with certificate track prompt information；When certificate moving rail Mark is matched with random certificate track prompt information, and when face delta data is matched with default delta data, is executed according to multiframe Human face region image verifies corresponding certificate area image.
In one embodiment, when above-mentioned processor executes program, realized according to multiframe human face region image to phase The step of answering certificate area image to be verified, specifically includes the following steps: extracting certificate area image in every frame target image Certificate photo characteristic and respective objects image in human face region image face characteristic data；Judge whether there is certificate photo The target image that characteristic and face characteristic data match；If so, extracting certificate photo characteristic and face characteristic number According to the identity information data for including in the target image to match；It obtains corresponding with identity information data in certificate database Verify human face data；Judge to verify whether human face data matches with certificate photo characteristic；If so, determining to test by identity Card.
In one embodiment, above-mentioned computer equipment can be used as server, including but not limited to independent physics clothes The server cluster for device or the multiple physical servers composition of being engaged in, the computer equipment also act as terminal, including but unlimited Mobile phone, tablet computer, personal digital assistant or smart machine etc..As shown in figure 8, the computer equipment includes total by system Processor, non-volatile memory medium, built-in storage, display screen, photographic device and the network interface of line connection.
Wherein, the processor of the computer equipment supports entire computer equipment for providing calculating and control ability Operation.The non-volatile memory medium of computer equipment is stored with operating system and computer program.The computer program can quilt Performed by processor, for realizing a kind of auth method provided by above each embodiment.In computer equipment Built-in storage in non-volatile memory medium operating system and computer program the running environment of cache is provided.Display Screen can be touch screen, for example be capacitance plate or electrical screen, and the point of the control of the touch screen display can be acted on by receiving Operation is hit, corresponding instruction is generated.Photographic device can be for network interfaces such as mobile phone camera, camera and video cameras Ethernet card or wireless network card etc., for being communicated with external terminal or server.
It will be understood by those skilled in the art that the structure of computer equipment shown in Fig. 8, only and application scheme The block diagram of relevant part-structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specifically Computer equipment may include perhaps combining certain components than more or fewer components as shown in the figure or there is difference Component layout.For example, the computer equipment in the figure also may not include photographic device, but realized by external photographic device Auth method in above-mentioned each implementation.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, the journey It is performed the steps of when sequence is executed by processor and receives authentication instruction；It is instructed according to authentication and is obtained by photographic device Take multiframe target image；Extract the human face region image and certificate area image in every frame target image；Parse multiframe target figure Human face region image as in generates face delta data；Judge whether face delta data matches with default delta data；When When face delta data is matched with default delta data, then corresponding certificate area image is carried out according to multiframe human face region image Verifying, determines whether to pass through authentication according to verification result.
In one embodiment, when which is executed by processor, that is realized passes through camera shooting according to authentication instruction Device obtains multiframe target image, comprising: is instructed according to authentication, according to prefixed time interval by taking the photograph in preset duration As device obtains multiple image；Detect whether the both greater than default resolution ratio of the resolution ratio of every frame image；If so, by every frame image As target image；It is obtained according to prefixed time interval by photographic device in preset duration if it is not, then returning and continuing to execute Multiple image, until the both greater than default resolution ratio of the resolution ratio of every frame image.
In one embodiment, face delta data includes that expression shape change degree is realized when the program is executed by processor Parsing multiframe target image in human face region image, generate face delta data the step of, specifically includes the following steps: mentioning Take the expressive features data of the human face region image in every frame target image；According to the expressive features of every two frames adjacent target image Data calculate to obtain expression shape change degree；That is realized judges the step of whether face delta data matches with default delta data, tool Body is the following steps are included: judge whether each expression shape change degree is in default change degree section；When each expression shape change degree When being in default change degree section, then determine that face delta data is matched with default delta data.
In one embodiment, when which is executed by processor, that is realized passes through camera shooting according to authentication instruction It is further comprising the steps of: to obtain and show random expression prompt information before device obtains the step of multiframe target image；Institute is real It is further comprising the steps of after the step of expressive features data of human face region image in the existing every frame target image of extraction: In the expressive features data of Detection and Extraction, if there are the expressive features data to match with random expression prompt information；When depositing In the expressive features data to match with expression prompt information, the expressive features number according to every two frames adjacent target image is executed According to calculating to obtain expression shape change degree.
In one embodiment, face delta data includes face motion track, and when which is executed by processor, institute is real It is existing according to authentication instruction by before the step of photographic device acquisition multiframe target image, it is further comprising the steps of: to obtain It takes and shows random face track prompt information；Human face region image in the parsing multiframe target image realized generates face The step of delta data, specifically includes the following steps: extracting the face location number of the human face region image in every frame target image According to；Face motion track is generated according to the face location data of every frame target image；The judgement face delta data realized is It no the step of being matched with default delta data, specifically includes: when face motion track is matched with random face track prompt information When, then determine that face delta data is matched with default delta data.
In one embodiment, when which is executed by processor, that is realized passes through camera shooting according to authentication instruction It is further comprising the steps of: to obtain and show random certificate track prompt information before device obtains the step of multiframe target image； Realized the step of verifying according to multiframe human face region image to corresponding certificate area image before, further include following step It is rapid: to extract the certificate position data of certificate area image in every frame target image；According to the certificate position in every frame target image Data generate certificate motion track；Judge whether certificate motion track matches with certificate track prompt information；When certificate moving rail Mark is matched with random certificate track prompt information, and when face delta data is matched with default delta data, is executed according to multiframe Human face region image verifies corresponding certificate area image.
In one embodiment, when which is executed by processor, realized according to multiframe human face region image to phase The step of answering certificate area image to be verified, specifically includes the following steps: extracting certificate area image in every frame target image Certificate photo characteristic and respective objects image in human face region image face characteristic data；Judge whether there is certificate photo The target image that characteristic and face characteristic data match；If so, extracting certificate photo characteristic and face characteristic number According to the identity information data for including in the target image to match；It obtains corresponding with identity information data in certificate database Verify human face data；Judge to verify whether human face data matches with certificate photo characteristic；If so, determining to test by identity Card.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with Relevant hardware is instructed to complete by computer program, the program can be stored in a non-volatile computer and can be read In storage medium, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, storage medium can For magnetic disk, CD, read-only memory (Read-Only Memory, ROM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance Shield all should be considered as described in this specification.
Only several embodiments of the present invention are expressed for above embodiments, and the description thereof is more specific and detailed, but can not Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art, Under the premise of not departing from present inventive concept, various modifications and improvements can be made, and these are all within the scope of protection of the present invention. Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.

Claims (9)

1. a kind of auth method, comprising:
Receive authentication instruction；
Target image is persistently obtained according to prefixed time interval until reaching by photographic device according to authentication instruction Goal-selling amount of images, the prefixed time interval are 0.1 or 0.2 second；Be preset in every frame target image human face region and Certificate region；
Extract the corresponding certificate of the corresponding human face region image of human face region described in every frame target image and the certificate region Area image；
The human face region image in multiframe target image is parsed, face delta data is generated；
Judge whether the face delta data matches with default delta data；
Extract the expressive features data of the human face region image in every frame target image；
Human face region image in the parsing multiframe target image, generates face delta data, comprising:
According to the expressive features data of every two frames adjacent target image, expression shape change degree is calculated to obtain；The expression shape change degree refers to The difference degree for the human face expression that human face region image is shown between two target images；
It is described to judge whether the face delta data matches with default delta data, comprising:
Judge whether each expression shape change degree is in default change degree section；The default change degree section is by the first threshold The value range of change degree corresponding to the real user that value and second threshold are constituted, first threshold is real user in two frame mesh Minimum change degree in logo image, second threshold are maximum change degree of the real user in two frame target images；
When each expression shape change degree is in default change degree section, then face delta data and default delta data are determined Matching；
When the face delta data is matched with default delta data, then according to multiframe human face region image to corresponding certificate area Area image is verified, and is determined whether to pass through authentication according to verification result.
2. the method according to claim 1, wherein described pass through photographic device according to authentication instruction Target image is persistently obtained according to prefixed time interval until reaching goal-selling amount of images, comprising:
It is instructed according to the authentication, multiframe figure is obtained by photographic device according to prefixed time interval in preset duration Picture；
Detect whether the both greater than default resolution ratio of the resolution ratio of every frame image；
If so, using every frame image as target image；
Described pass through photographic device acquisition multiframe figure according to prefixed time interval in preset duration if it is not, then returning to continue to execute Picture, until the both greater than default resolution ratio of the resolution ratio of every frame image.
3. the method according to claim 1, wherein passing through camera shooting dress according to authentication instruction described It sets before persistently obtaining target image according to prefixed time interval until reaching goal-selling amount of images, further includes:
It obtains and shows random expression prompt information；
After the expressive features data for extracting the human face region image in every frame target image, further includes:
In the expressive features data of Detection and Extraction, if there is the expressive features number to match with the random expression prompt information According to；
When there are the expressive features data to match with the expression prompt information, execute described according to every two frames adjacent target The expressive features data of image calculate to obtain expression shape change degree.
4. the method according to claim 1, wherein the face delta data includes face motion track；
According to authentication instruction persistently to obtain according to prefixed time interval target image by photographic device straight described To before reaching goal-selling amount of images, further includes:
It obtains and shows random face track prompt information；
Human face region image in the parsing multiframe target image generates face delta data, comprising:
Extract the face location data of the human face region image in every frame target image；
Face motion track is generated according to the face location data of every frame target image；
It is described to judge whether the face delta data matches with default delta data, comprising:
When the face motion track is matched with the random face track prompt information, then face delta data and pre- is determined If delta data matches.
5. the method according to claim 1, wherein passing through camera shooting dress according to authentication instruction described It sets before persistently obtaining target image according to prefixed time interval until reaching goal-selling amount of images, further includes:
It obtains and shows random certificate track prompt information；
It is described corresponding certificate area image is verified according to multiframe human face region image before, further includes:
Extract the certificate position data of certificate area image in every frame target image；
Certificate motion track is generated according to the certificate position data in every frame target image；
Judge whether the certificate motion track matches with certificate track prompt information；
When the certificate motion track is matched with the random certificate track prompt information, and the face delta data and default When delta data matches, execution is described to verify corresponding certificate area image according to multiframe human face region image.
6. the method according to claim 1, wherein it is described according to multiframe human face region image to corresponding certificate area Area image is verified, comprising:
Extract in every frame target image human face region figure in the certificate photo characteristic of certificate area image and respective objects image The face characteristic data of picture；
Judge whether there is the target image that the certificate photo characteristic and the face characteristic data match；
If so, extracting the body for including in the target image that the certificate photo characteristic and the face characteristic data match Part information data；
Obtain verifying human face data corresponding with the identity information data in certificate database；
Judge whether the verifying human face data matches with the certificate photo characteristic；
If so, determining to pass through authentication.
7. a kind of authentication means, which is characterized in that described device includes:
Authentication command reception module, for receiving authentication instruction；
Target image obtains module, for being continued by photographic device according to prefixed time interval according to authentication instruction Target image is obtained until reaching goal-selling amount of images, the prefixed time interval is 0.1 or 0.2 second；Every frame target figure Human face region and certificate region are preset with as in；
Area image extraction module corresponds to human face region image and institute for extracting the human face region in every frame target image State the corresponding certificate area image in certificate region；
Face delta data matching module extracts the expressive features data of the human face region image in every frame target image；According to The expressive features data of every two frames adjacent target image, calculate to obtain expression shape change degree；The expression shape change degree refers to two targets The difference degree for the human face expression that human face region image is shown between image；
Authentication module, for judging whether each expression shape change degree is in default change degree section；The default change Change degree section is the value range of change degree corresponding to the real user that is made of first threshold and second threshold, first threshold Minimum change degree of the real user in two frame target images, second threshold be real user in two frame target images most Big change degree；When each expression shape change degree is in default change degree section, then face delta data and default change are determined Change Data Matching；When the face delta data is matched with default delta data, then according to multiframe human face region image to phase It answers certificate area image to be verified, is determined whether to pass through authentication according to verification result.
8. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor The step of any one of claim 1 to 6 the method is realized when row.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor Calculation machine program, which is characterized in that the processor realizes side described in any one of claim 1 to 6 when executing described program The step of method.
CN201710965802.XA 2017-10-17 2017-10-17 Auth method, device, storage medium and computer equipment Active CN107844748B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201710965802.XA CN107844748B (en) 2017-10-17 2017-10-17 Auth method, device, storage medium and computer equipment
PCT/CN2017/112485 WO2019075840A1 (en) 2017-10-17 2017-11-23 Identity verification method and apparatus, storage medium and computer device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710965802.XA CN107844748B (en) 2017-10-17 2017-10-17 Auth method, device, storage medium and computer equipment

Publications (2)

Publication Number Publication Date
CN107844748A CN107844748A (en) 2018-03-27
CN107844748B true CN107844748B (en) 2019-02-05

Family

ID=61661436

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710965802.XA Active CN107844748B (en) 2017-10-17 2017-10-17 Auth method, device, storage medium and computer equipment

Country Status (2)

Country Link
CN (1) CN107844748B (en)
WO (1) WO2019075840A1 (en)

Families Citing this family (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109255299A (en) * 2018-08-13 2019-01-22 北京市商汤科技开发有限公司 Identity identifying method and device, electronic equipment and storage medium
JP7165746B2 (en) * 2018-08-13 2022-11-04 ベイジン・センスタイム・テクノロジー・デベロップメント・カンパニー・リミテッド ID authentication method and device, electronic device and storage medium
CN109359502A (en) * 2018-08-13 2019-02-19 北京市商汤科技开发有限公司 False-proof detection method and device, electronic equipment, storage medium
CN110197108A (en) * 2018-08-17 2019-09-03 平安科技（深圳）有限公司 Auth method, device, computer equipment and storage medium
CN108830512B (en) * 2018-08-20 2020-01-21 华润守正招标有限公司 User registration auditing method, device and equipment of electronic bidding platform
CN109299692A (en) * 2018-09-26 2019-02-01 深圳壹账通智能科技有限公司 A kind of personal identification method, computer readable storage medium and terminal device
CN109410138B (en) * 2018-10-16 2021-10-01 北京旷视科技有限公司 Method, device and system for modifying double chin
CN109829362A (en) * 2018-12-18 2019-05-31 深圳壹账通智能科技有限公司 Safety check aided analysis method, device, computer equipment and storage medium
CN109871845B (en) 2019-01-10 2023-10-31 平安科技（深圳）有限公司 Certificate image extraction method and terminal equipment
CN111507143B (en) * 2019-01-31 2023-06-02 北京字节跳动网络技术有限公司 Expression image effect generation method and device and electronic equipment
CN110223710A (en) * 2019-04-18 2019-09-10 深圳壹账通智能科技有限公司 Multiple joint authentication method, device, computer installation and storage medium
CN110210312A (en) * 2019-04-29 2019-09-06 众安信息技术服务有限公司 A kind of method and system verifying certificate and holder
CN111866589A (en) * 2019-05-20 2020-10-30 北京嘀嘀无限科技发展有限公司 Video data verification method and device, electronic equipment and storage medium
CN110176024B (en) * 2019-05-21 2023-06-02 腾讯科技（深圳）有限公司 Method, device, equipment and storage medium for detecting target in video
CN110298246A (en) * 2019-05-22 2019-10-01 深圳壹账通智能科技有限公司 Unlocking verification method, device, computer equipment and storage medium
CN110414454A (en) * 2019-07-31 2019-11-05 南充折衍智能光电科技有限公司 A kind of testimony of a witness unification identifying system based on machine vision
CN112395906A (en) * 2019-08-12 2021-02-23 北京旷视科技有限公司 Face living body detection method and device, face living body detection equipment and medium
CN110705350B (en) * 2019-08-27 2020-08-25 阿里巴巴集团控股有限公司 Certificate identification method and device
CN110730169A (en) * 2019-09-29 2020-01-24 北京东软望海科技有限公司 Processing method, device and system for guaranteeing account security
CN112764824B (en) * 2019-10-21 2023-10-10 腾讯科技（深圳）有限公司 Method, device, equipment and storage medium for triggering identity verification in application program
TWI731461B (en) * 2019-11-01 2021-06-21 宏碁股份有限公司 Identification method of real face and identification device using the same
CN111079712B (en) * 2019-12-31 2023-04-21 中国银行股份有限公司 Permission management method and device based on face recognition
KR20210087792A (en) * 2020-01-03 2021-07-13 엘지전자 주식회사 User verification
TWI825371B (en) * 2020-10-20 2023-12-11 普匯金融科技股份有限公司 Device and method of legal person real-name authentication
KR102502631B1 (en) * 2020-11-16 2023-02-23 고큐바테크놀로지 주식회사 Technique for authenticating a user
US11068908B1 (en) * 2020-12-22 2021-07-20 Lucas GC Limited Skill-based credential verification by a credential vault system (CVS)
CN112862458A (en) * 2021-03-02 2021-05-28 岭东核电有限公司 Nuclear power test procedure supervision method and device, computer equipment and storage medium
TWI810548B (en) * 2021-04-15 2023-08-01 臺灣網路認證股份有限公司 Living body recognition system for integrating image processing and deep learning and method thereof
CN113312972A (en) * 2021-04-26 2021-08-27 国家能源集团新能源有限责任公司 Self-service hydrogenation method and device thereof
CN115083002A (en) * 2022-08-16 2022-09-20 深圳市海清视讯科技有限公司 Image processing method, device and equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103678984A (en) * 2013-12-20 2014-03-26 湖北微模式科技发展有限公司 Method for achieving user authentication by utilizing camera
CN106599772A (en) * 2016-10-31 2017-04-26 北京旷视科技有限公司 Living body authentication method, identity authentication method and device
CN106778525A (en) * 2016-11-25 2017-05-31 北京旷视科技有限公司 Identity identifying method and device
CN107181852A (en) * 2017-07-19 2017-09-19 维沃移动通信有限公司 A kind of method for sending information, method for information display and mobile terminal

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8705813B2 (en) * 2010-06-21 2014-04-22 Canon Kabushiki Kaisha Identification device, identification method, and storage medium
US9495586B1 (en) * 2013-09-18 2016-11-15 IDChecker, Inc. Identity verification using biometric data
CN104834905A (en) * 2015-04-29 2015-08-12 河南城建学院 Facial image identification simulation system and method
CN105005779A (en) * 2015-08-25 2015-10-28 湖北文理学院 Face verification anti-counterfeit recognition method and system thereof based on interactive action
CN106548121A (en) * 2015-09-23 2017-03-29 阿里巴巴集团控股有限公司 A kind of method of testing and device of vivo identification
CN106709402A (en) * 2015-11-16 2017-05-24 优化科技（苏州）有限公司 Living person identity authentication method based on voice pattern and image features
CN106488130A (en) * 2016-11-15 2017-03-08 上海斐讯数据通信技术有限公司 A kind of screening-mode changing method and its switched system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103678984A (en) * 2013-12-20 2014-03-26 湖北微模式科技发展有限公司 Method for achieving user authentication by utilizing camera
CN106599772A (en) * 2016-10-31 2017-04-26 北京旷视科技有限公司 Living body authentication method, identity authentication method and device
CN106778525A (en) * 2016-11-25 2017-05-31 北京旷视科技有限公司 Identity identifying method and device
CN107181852A (en) * 2017-07-19 2017-09-19 维沃移动通信有限公司 A kind of method for sending information, method for information display and mobile terminal

Also Published As

Publication number Publication date
CN107844748A (en) 2018-03-27
WO2019075840A1 (en) 2019-04-25

Similar Documents

Publication Publication Date Title
CN107844748B (en) Auth method, device, storage medium and computer equipment
US11669607B2 (en) ID verification with a mobile device
US11539525B2 (en) Systems and methods for secure tokenized credentials
Komulainen et al. Complementary countermeasures for detecting scenic face spoofing attacks
Pereira et al. Face liveness detection using dynamic texture
Freitas Pereira et al. Face liveness detection using dynamic texture
Agarwal et al. Swapped! digital face presentation attack detection via weighted local magnitude pattern
CN108804884B (en) Identity authentication method, identity authentication device and computer storage medium
Hernandez-Ortega et al. Introduction to face presentation attack detection
EP3905102A1 (en) Method and device for determining operation based on facial expression groups, and electronic device
Das et al. Recent advances in biometric technology for mobile devices
Smith-Creasey et al. Continuous face authentication scheme for mobile devices with tracking and liveness detection
CN106529255B (en) Method for identifying ID and device based on person's handwriting fingerprint
EP3642756A1 (en) Detecting artificial facial images using facial landmarks
Kant et al. Fake face recognition using fusion of thermal imaging and skin elasticity
Shen et al. Iritrack: Face presentation attack detection using iris tracking
US20240037995A1 (en) Detecting wrapped attacks on face recognition
Dubey et al. Echelon Based Pose Generalization of Facial Images Approaches
Tyagi et al. A survey of unconstrained face recognition algorithm and its applications
Zolotarev et al. Liveness detection methods implementation to face identification reinforcement in gaming services
Li ◾ Biometrics in Social Media Applications
Mishra et al. Integrating State-of-the-Art Face Recognition and Anti-Spoofing Techniques into Enterprise Information Systems
US11973877B2 (en) Systems and methods for secure tokenized credentials
Findling Pan shot face unlock: Towards unlocking personal mobile devices using stereo vision and biometric face information from multiple perspectives
RU2791846C2 (en) Method and device for decision-making to perform operation based on groups of face expressions and electronic device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 1245948

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant