CN106650646A - Action recognition based living body face recognition method and system - Google Patents

Action recognition based living body face recognition method and system Download PDF

Info

Publication number
CN106650646A
CN106650646A CN201611129097.1A CN201611129097A CN106650646A CN 106650646 A CN106650646 A CN 106650646A CN 201611129097 A CN201611129097 A CN 201611129097A CN 106650646 A CN106650646 A CN 106650646A
Authority
CN
China
Prior art keywords
action
living body
human face
face image
user
Prior art date
Application number
CN201611129097.1A
Other languages
Chinese (zh)
Inventor
薛炳如
卜习栓
王金凤
Original Assignee
南京合荣欣业金融软件有限公司
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 南京合荣欣业金融软件有限公司 filed Critical 南京合荣欣业金融软件有限公司
Priority to CN201611129097.1A priority Critical patent/CN106650646A/en
Publication of CN106650646A publication Critical patent/CN106650646A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00221Acquiring or recognising human faces, facial parts, facial sketches, facial expressions
    • G06K9/00228Detection; Localisation; Normalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00885Biometric patterns not provided for under G06K9/00006, G06K9/00154, G06K9/00335, G06K9/00362, G06K9/00597; Biometric specific functions not specific to the kind of biometric
    • G06K9/00899Spoof detection
    • G06K9/00906Detection of body part being alive

Abstract

The invention discloses an action recognition based living body face recognition method and system. The method includes steps of collecting a static face image of a user; outputting an action command generated randomly; collecting video information of the user responding to the action command; extracting an image in which the action of the user is matched with the action command in the video information as a dynamic face image; comparing the static face image with the dynamic face image, determining living body face recognition success when the matching degree is greater than a recognition threshold value or else determining living body face recognition failure.

Description

A kind of living body faces recognition methods and system based on action recognition

Technical field

The present invention relates to technical field of face recognition, and in particular to a kind of living body faces based on action recognition recognize and are System.

Background technology

With the development of computer technology, the use of recognition of face is more and more extensive, and currently, face identification system is increasingly Being applied to security protection, financial field more needs the scene of authentication, such as open a bank account, gate control system, remote transaction operation test Card etc..In the application of these high safety ranks, in addition to ensuring that the human face similarity degree of authenticatee meets in database storing Bottom storehouse, in addition it is also necessary to In vivo detection is carried out to authenticatee.Existing face identification system when being identified, it sometimes appear that The video system of out-tricking that user is shot using photo or in advance completes checking, strick precaution user is also needed in addition and uses 3D faceforms Or the mode such as mask is cheated.

Ripe live body proof scheme is not also generally acknowledged in technical products in the market, or existing technology is relied on Special hardware device(Such as infrared camera, depth camera)Or simple still photo can only be taken precautions against and attacked;In addition one By the dynamic position of detection face in a little disclosed technologies, if variant, it is considered as living body faces, but this kind of method is more Passively, it is assumed that people does not have action, then detecting may failure;Or using setting sequence action, live body is carried out with user's completeness Detection, this kind of detection needs to be authenticated face does substantial amounts of expression interoperation, and man-machine interaction mode is unfriendly, easily makes user Sense is weary of in generation, and algorithm is complicated, reduces the speed of recognition of face.

The content of the invention

Goal of the invention:Present invention aim at being directed to the deficiencies in the prior art, there is provided a kind of live body based on action recognition Face identification method and system, in the case where hardware device is changed without, living body faces is solved by recognizing the action of people and are known Other problem, and recognition accuracy is high.

Technical scheme:Living body faces recognition methods based on action recognition of the present invention, comprises the steps:

S11, the Static Human Face image of collection user;

S12, the random action command for generating of output;

The video information of S13, collection user response action command;

The image that user action is matched with action command in S14, extraction video information is used as dynamic human face image;

S15, comparison Static Human Face image and dynamic human face image, when matching degree is compared recognition threshold is more than, and determines living body faces It is proved to be successful, otherwise authentication failed.

Further improve above-mentioned technical proposal, the static person that step S11 passes through the instant shooting, collecting user of camera Face image.

Further, the action command include read aloud passage, blink, open one's mouth, raise one's hand in one or more move Make.

Further, the random various motion instruction for generating is exported in step S12;Gather user in step S13 to ring The video information for answering various motion to instruct;Step S14 is extracted user action in video information and is matched with every kind of action command Image, select resolution highest image as dynamic human face image.

Further, step S13 carries out human face action detection during collection video information, if being not detected by Human face action, then collection failure, re-executes step S12.

Further, step S14 extract video information during, if do not extract user action referring to action The image of order matching, then extract failure, re-executes step S12.

Present invention also offers the living body faces identifying system based on action recognition for realizing said method, including taking the photograph As head, command output module, extraction module, living body authentication module;Command output module is used for output action instruction, camera In the video information of the Static Human Face image and user's corresponding actions instruction of collection user, extraction module is used to extract video information In dynamic human face image, living body authentication module be used for compare Static Human Face image and dynamic human face image, be defined as it is same Living body faces certification success, otherwise authentication failed are determined during people.

Beneficial effect:Compared with prior art, it is an advantage of the current invention that:The present invention is being changed without the situation of hardware device Under, carry out living body faces checking by contrasting Static Human Face image and dynamic human face image, it is easy to accomplish, checking accuracy rate is high, Using active output action instruction acquisition living body faces dynamic human face image, detection success rate is improved, prevent user from passing through pre- The photo or video for first shooting is cheated;Relative to the checking for passing through detection user's execution command sequence in prior art Method, it is easier to realize, user operation difficulty is lower, while ensure that the discrimination and anti-fraud ability of checking.

Description of the drawings

Fig. 1 is the system flow chart of the present invention.

Specific embodiment

Technical solution of the present invention is described in detail below by accompanying drawing, but protection scope of the present invention is not limited to The embodiment.

Embodiment 1:Living body faces recognition methods based on action recognition as shown in Figure 1, comprises the steps:

S11, Static Human Face IMAQ:Using the Static Human Face image of the instant shooting, collecting user of camera;

S12, output action instruction " blinking ";

S13, the video information of collection user response action command " blinking ", carry out face and move during collection video information Detect, if being not detected by human face action, collection failure re-executes step S12;

S14, the image that matches with action command " blinking " of user action is extracted in video information as dynamic human face image, if The image that user action is matched with action command " blinking " is not extracted, then extracts failure, re-execute step S12;

S15, comparison Static Human Face image and dynamic human face image, when matching degree is compared recognition threshold is more than, and determines living body faces It is proved to be successful, otherwise authentication failed.

Said method is realized by following system:Including camera, command output module, extraction module, living body authentication mould Block;, for output action instruction, camera is used to gather the Static Human Face image of user and user accordingly moves for command output module The video information that instructs of work, extraction module is used to extracting the dynamic human face image in video information, living body authentication module be used for than To Static Human Face image and dynamic human face image, living body faces certification success is determined when same people is defined as, otherwise checking is lost Lose.

As described above, although the present invention has been represented and described with reference to specific preferred embodiment, it must not be explained It is to the restriction of itself of the invention.Under the premise of the spirit and scope of the present invention defined without departing from claims, can be right In the form and details various changes can be made for it.

Claims (7)

1. a kind of living body faces recognition methods based on action recognition, it is characterised in that comprise the steps:
S11, the Static Human Face image of collection user;
S12, the random action command for generating of output;
The video information of S13, collection user response action command;
The image that user action is matched with action command in S14, extraction video information is used as dynamic human face image;
S15, comparison Static Human Face image and dynamic human face image, when matching degree is compared recognition threshold is more than, and determines living body faces It is proved to be successful, otherwise authentication failed.
2. the living body faces recognition methods based on action recognition according to claim 1, it is characterised in that:The step The Static Human Face image that S11 passes through the instant shooting, collecting user of camera.
3. the living body faces recognition methods based on action recognition according to claim 1, it is characterised in that:The action refers to Order include read aloud passage, blink, open one's mouth, raise one's hand in one or more action.
4. the living body faces recognition methods based on action recognition according to claim 3, it is characterised in that:The step The random various motion instruction for generating is exported in S12;The video information of user response various motion instruction is gathered in step S13; Step S14 extracts the image that user action is matched with every kind of action command in video information, selects resolution highest figure As dynamic human face image.
5. the living body faces recognition methods based on action recognition according to claim 1, it is characterised in that:The step S13 carries out human face action detection during collection video information, if being not detected by human face action, collection failure, again Execution step S12.
6. the living body faces recognition methods based on action recognition according to claim 1, it is characterised in that:The step S14, if not extracting the image that user action is matched with action command, extracts failure during video information is extracted, Re-execute step S12.
7. a kind of living body faces identifying system based on action recognition, it is characterised in that:Including camera, command output module, Extraction module, living body authentication module;For output action instruction, camera is used to gather the static person of user command output module Face image and the video information of user's corresponding actions instruction, extraction module is used to extract the dynamic human face image in video information, Living body authentication module is used to compare Static Human Face image and dynamic human face image, determines that living body faces are recognized when same people is defined as Demonstrate,prove successfully, otherwise authentication failed.
CN201611129097.1A 2016-12-09 2016-12-09 Action recognition based living body face recognition method and system CN106650646A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611129097.1A CN106650646A (en) 2016-12-09 2016-12-09 Action recognition based living body face recognition method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611129097.1A CN106650646A (en) 2016-12-09 2016-12-09 Action recognition based living body face recognition method and system

Publications (1)

Publication Number Publication Date
CN106650646A true CN106650646A (en) 2017-05-10

Family

ID=58825436

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611129097.1A CN106650646A (en) 2016-12-09 2016-12-09 Action recognition based living body face recognition method and system

Country Status (1)

Country Link
CN (1) CN106650646A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330914A (en) * 2017-06-02 2017-11-07 广州视源电子科技股份有限公司 Face position method for testing motion and device and vivo identification method and system
CN107491759A (en) * 2017-08-21 2017-12-19 厦门中控智慧信息技术有限公司 A kind of mixed mode register method and device
CN107895155A (en) * 2017-11-29 2018-04-10 五八有限公司 A kind of face identification method and device

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103634120A (en) * 2013-12-18 2014-03-12 上海市数字证书认证中心有限公司 Method and system for real-name authentication based on face recognition
CN103886301A (en) * 2014-03-28 2014-06-25 中国科学院自动化研究所 Human face living detection method
US20160071111A1 (en) * 2012-01-13 2016-03-10 Amazon Technologies, Inc. Image analysis for user authentication
CN105518713A (en) * 2015-02-15 2016-04-20 北京旷视科技有限公司 Living human face verification method and system, computer program product
CN105989263A (en) * 2015-01-30 2016-10-05 阿里巴巴集团控股有限公司 Method for authenticating identities, method for opening accounts, devices and systems

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160071111A1 (en) * 2012-01-13 2016-03-10 Amazon Technologies, Inc. Image analysis for user authentication
CN103634120A (en) * 2013-12-18 2014-03-12 上海市数字证书认证中心有限公司 Method and system for real-name authentication based on face recognition
CN103886301A (en) * 2014-03-28 2014-06-25 中国科学院自动化研究所 Human face living detection method
CN105989263A (en) * 2015-01-30 2016-10-05 阿里巴巴集团控股有限公司 Method for authenticating identities, method for opening accounts, devices and systems
CN105518713A (en) * 2015-02-15 2016-04-20 北京旷视科技有限公司 Living human face verification method and system, computer program product

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330914A (en) * 2017-06-02 2017-11-07 广州视源电子科技股份有限公司 Face position method for testing motion and device and vivo identification method and system
CN107491759A (en) * 2017-08-21 2017-12-19 厦门中控智慧信息技术有限公司 A kind of mixed mode register method and device
CN107895155A (en) * 2017-11-29 2018-04-10 五八有限公司 A kind of face identification method and device

Similar Documents

Publication Publication Date Title
Garcia-Salicetti et al. BIOMET: A multimodal person authentication database including face, voice, fingerprint, hand and signature modalities
Jain et al. 50 years of biometric research: Accomplishments, challenges, and opportunities
KR100601963B1 (en) Authentication apparatus and method using eye gaze
CN100361135C (en) Method for acquiring human-face image, human-face discrimination and discriminating system
KR101971697B1 (en) Method and apparatus for authenticating user using hybrid biometrics information in a user device
CN100514353C (en) Living body detecting method and system based on human face physiologic moving
Barbello Continuous user authentication on mobile devices: Recent progress and remaining challenges
KR101291899B1 (en) Procedure for identifying a person by eyelash analysis and acquisition device using thereof
Juefei-Xu et al. Gait-id on the move: Pace independent human identification using cell phone accelerometer dynamics
US20140165187A1 (en) Method, Apparatus, and Computer-Readable Recording Medium for Authenticating a User
KR101242304B1 (en) Controlled access to functionality of a wireless device
Yang et al. Secure iris verification
CN104361276B (en) A kind of multi-modal biological characteristic identity identifying method and system
Parmar et al. Face recognition methods & applications
KR20100073191A (en) Method and apparatus for face liveness using range data
CN102956050A (en) Intelligent entrance guard control system
Niinuma et al. Continuous user authentication using temporal information
CN101404060A (en) Human face recognition method based on visible light and near-infrared Gabor information amalgamation
JPH05266173A (en) Face classification system
CN104166835A (en) Method and device for identifying living user
Bhatia Biometrics and face recognition techniques
Gafurov et al. Robustness of biometric gait authentication against impersonation attack
Chintalapati et al. Automated attendance management system based on face recognition algorithms
Tian et al. KinWrite: Handwriting-Based Authentication Using Kinect.
Baig et al. Fingerprint-Iris fusion based identification system using a single hamming distance matcher

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170510