CN109376704A - A kind of human face in-vivo detection method - Google Patents
A kind of human face in-vivo detection method Download PDFInfo
- Publication number
- CN109376704A CN109376704A CN201811459708.8A CN201811459708A CN109376704A CN 109376704 A CN109376704 A CN 109376704A CN 201811459708 A CN201811459708 A CN 201811459708A CN 109376704 A CN109376704 A CN 109376704A
- Authority
- CN
- China
- Prior art keywords
- face
- vivo detection
- posture
- human face
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the present application provides a kind of human face in-vivo detection method, comprising: step 1: obtaining testing image data source, carries out Face datection to described image data source, obtain testing image data；Step 2: carrying out the testing image data to continue face tracking detection, if front and back two continuous frames face is consistent, continue step 3, if front and back two continuous frames face is inconsistent or face is lost, In vivo detection terminates, and obtains face non-living body testing result；Step 3: human face posture filtering is carried out to the facial image to be measured or face picture detected by face tracking；Step 4: carrying out living body faces detection using convolutional neural networks, solve the face In vivo detection currently based on the prior art, carries out In vivo detection only by simple judgement, when the instruction time is longer, user friendly is poor；When the instruction time is shorter, user does not have the time to cooperate, and leads to the technical problem that detection effect is poor.
Description
Technical field
The invention relates to image processing techniques and computer vision field more particularly to a kind of face In vivo detections
Method.
Background technique
With the rapid development of AI technology, deep learning has obtained explosive development in artificial intelligence field, it is solution
A certainly hot topic of computer abstract cognitive problem.In each research field of academia, have largely about deep learning
Research Literature；In industry, also there are the various intellectual products developed using deep learning.
Currently, loophole existing for face verification technology not only influences whether the safety of face identification system, but also return
User brings information leakage or property loss etc..The face identification system of one safety should must accurately and quickly determine
Face relevant information, while must also be from by the attack of illegal one's share of expenses for a joint undertaking.Comprehensive each side's factor, face In vivo detection is a kind of drop
The effective means of low face identification system attack and indispensable main ingredient, reduce people to living creature characteristic recognition system
Safety problem worry.
Currently, living body faces detection method can be divided into the detection of the living body faces based on texture, based on instruction formula face living body
It detects and the living body faces based on deep learning detects three categories.
Concrete scheme is as follows:
1, the face In vivo detection based on texture
Texture information analysis method is attacked mainly for photo, and forging face (photo) can be generated perhaps due to secondary imaging
Deformation in multi-texturing, these small variations can be caught in by texture information analysis method.Its main thought is study
The structure and multidate information of real human face microtexture
2, formula face In vivo detection based on instruction
The method of formula is mainly dynamic in view of real human face based on instruction, has some instinctive activities, for example, blink,
Mouth such as moves, shakes the head at the movement, and photo be it is static, institute attacks mainly for photo in this way.Currently, big portion in market
In vivo detection product is divided to be based on mainly being made of following situations:
(1) general network camera develops a real-time living body faces detection method, and this method is acted based on blink
Face true and false situation is differentiated, by constructing in non-directed graph frame the different phase of blink movement, and in this, as eyes
The judgment criteria opened or closed.
(2) detection method of photo face camouflage is realized by comparing the Optic flow information of different zones in video human face,
Foundation real human face is three-dimensional and video human face or photograph print are two dimensions, constructs face geometrical invariants, passes through head
Voltuntary movement counts the sizes of geometrical invariants, due to video human face or photo face be it is two-dimensional, no matter the head of people such as
What swings, which can only generate small variation；And real human face is due to being three-dimensional structure, as long as head one is put
Dynamic, which will vary widely.
(3) the face true and false is distinguished using blink movement and mouth action, can be randomly generated by system one section short
Sound or information alert can allow subject to do eyes and open or close, and mouth opens or the movement closed, if visitor person according to
The sequence of prompt is completed to may prove real human face.
3, the living body faces detection based on deep learning
Deep learning developed recently is very swift and violent, is a hot spot of modern academic research.Existing network infrastructure has convolution
Neural network (Convolutional Neural Networks, CNN), limited Boltzmann machine (Restricted
Boltzmann Machine, RBM), depth confidence network (Deep Belief Nets, DBN), Recognition with Recurrent Neural Network
(Recurrent Neural Networks, RNN) etc..It by mass data, is effectively trained, obtains preferable study
Effect.
Face In vivo detection based on the prior art above carries out In vivo detection only by simple judgement, works as instruction
When time is longer, user friendly is poor；When the instruction time is shorter, user does not have the time to cooperate, and leads to detection effect
Poor technical problem.
Summary of the invention
The embodiment of the present application provide a kind of human face in-vivo detection method solve it is living currently based on the face of the prior art
Physical examination is surveyed, and carries out In vivo detection only by simple judgement, when the instruction time is longer, user friendly is poor；Work as instruction
When time is shorter, user does not have the time to cooperate, and leads to the technical problem that detection effect is poor.
The embodiment of the present application provides a kind of human face in-vivo detection method, comprising the following steps:
Step 1: obtaining testing image data source, carries out Face datection to described image data source, obtains testing image number
According to；
Step 2: carrying out the testing image data to continue face tracking detection, if front and back two continuous frames face is consistent,
Then continue step 3, if front and back two continuous frames face is inconsistent or face is lost, In vivo detection terminates, and show that face is non-
In vivo detection result；
Step 3: human face posture mistake is carried out to the facial image to be measured or face picture detected by face tracking
Filter；
Step 4: living body faces detection is carried out using convolutional neural networks.
Optionally, initialization process is carried out to the testing image data, the initialization process includes lighting process, mould
Paste processing.
Optionally, the testing image data, including facial image and/or face picture.
Optionally, the step 1 further includes carrying out information type judgement to the testing image data.
The testing image data carry out continuing face tracking detection specifically: detect the face in testing image data
Afterwards, face tracking is carried out at random, obtains the front and back two continuous frames image in video sequence comprising facial image, it is continuous according to front and back
The consistency of two frame facial images is judged.
Optionally, the human face posture filtering specifically: the change that the facial angle detected is occurred within given time
Change situation to be determined.
Optionally, the convolutional neural networks are 3D convolutional neural networks.
Specifically, the living body faces detection includes formula In vivo detection and the living body inspection based on 3D convolutional neural networks
It surveys.
The formula In vivo detection, by prompt user cooperation, transformation human face posture is detected as required, including with
Lower step:
Step 1: in background logic processing, reading within timing 1 second the value of posture mark flag, identified according to the posture
Flag judges whether the posture changes, if the posture does not change, judges that In vivo detection fails；If the posture occurs
Variation, then obtain the posture state state in timing in 2 seconds, if the posture state state is consistent with the requirement,
Classify, if the posture state state and it is described require it is inconsistent, again prompt user cooperation as required to face appearance
State is responded；
Step 2: the user being cooperated, setting expectation posture carries out transformation posture by user described in system prompt, counts
When 5 seconds, if it times out, In vivo detection fail；
The step 1 and step 2 carry out simultaneously.
Optionally, the formula In vivo detection and the In vivo detection based on 3D convolutional neural networks carry out simultaneously.
As can be seen from the above technical solutions, the embodiment of the present application has the advantage that
The embodiment of the present application provides a kind of human face in-vivo detection method, comprising: step 1: obtaining testing image data
Source carries out Face datection to described image data source, obtains testing image data；Step 2: to the testing image data into
Row continues face tracking detection, if front and back two continuous frames face is consistent, continues step 3, if front and back two continuous frames face
Inconsistent or face is lost, then In vivo detection terminates, and obtains face non-living body testing result；Step 3: to passing through face tracking
The facial image to be measured or face picture of detection carry out human face posture filtering；Step 4: it is carried out using convolutional neural networks
Living body faces detection, solves the face In vivo detection currently based on the prior art, carries out living body only by simple judgement
Detection, when the instruction time is longer, user friendly is poor；When the instruction time is shorter, user does not have the time to cooperate, and leads
The technical problem for causing detection effect poor.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of one embodiment of human face in-vivo detection method in the embodiment of the present application；
Fig. 2 is the living body faces overhaul flow chart based on 3D convolutional neural networks；
Fig. 3 is CNN In vivo detection flow chart.
Specific embodiment
The embodiment of the present application provide a kind of human face in-vivo detection method solve it is living currently based on the face of the prior art
Physical examination is surveyed, and carries out In vivo detection only by simple judgement, when the instruction time is longer, user friendly is poor；Work as instruction
When time is shorter, user does not have the time to cooperate, and leads to the technical problem that detection effect is poor.
To enable the goal of the invention, feature, advantage of the embodiment of the present application more obvious and understandable, below in conjunction with
Attached drawing in the embodiment of the present application, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that below
Described embodiment is only the embodiment of the present application a part of the embodiment, and not all embodiment.Implemented based on the application
Embodiment in example, all other implementation obtained by those of ordinary skill in the art without making creative efforts
Example belongs to the range of the embodiment of the present application protection.
Referring to Fig. 1, a kind of one embodiment of human face in-vivo detection method provided by the embodiments of the present application includes following
Step:
Step 1: obtaining testing image data source, carries out Face datection to described image data source, obtains testing image number
According to；
Step 2: carrying out the testing image data to continue face tracking detection, if front and back two continuous frames face is consistent,
Then continue step 3, if front and back two continuous frames face is inconsistent or face is lost, In vivo detection terminates, and show that face is non-
In vivo detection result；
Step 3: human face posture mistake is carried out to the facial image to be measured or face picture detected by face tracking
Filter；
Step 4: living body faces detection is carried out using convolutional neural networks.
Preferably, initialization process is carried out to the testing image data, the initialization process includes lighting process, mould
Paste processing.
Preferably, the testing image data, including facial image and/or face picture.
Preferably, the step 1 further includes carrying out information type judgement to the testing image data.
In some specific embodiments, the testing image data are carried out to continue face tracking detection specifically: inspection
After measuring the face in testing image data, face tracking is carried out at random, obtains the front and back in video sequence comprising facial image
Two continuous frames image is judged according to the consistency of front and back two continuous frames facial image.
Preferably, the human face posture filtering specifically: the change that the facial angle detected is occurred within given time
Change situation to be determined.
Preferably, the convolutional neural networks are 3D convolutional neural networks.
Preferably, the living body faces detection includes formula In vivo detection and the living body inspection based on 3D convolutional neural networks
It surveys.
Preferably, the formula In vivo detection, by prompt user cooperation, transformation human face posture is detected as required,
The following steps are included:
Step 1: in background logic processing, reading within timing 1 second the value of posture mark flag, identified according to the posture
Flag judges whether the posture changes, if the posture does not change, judges that In vivo detection fails；If the posture occurs
Variation, then obtain the posture state state in timing in 2 seconds, if the posture state state is consistent with the requirement,
Classify, if the posture state state and it is described require it is inconsistent, again prompt user cooperation as required to face appearance
State is responded；
Step 2: the user being cooperated, setting expectation posture carries out transformation posture by user described in system prompt, counts
When 5 seconds, if it times out, In vivo detection fail；
The step 1 and step 2 carry out simultaneously.
Preferably, the formula In vivo detection and the In vivo detection based on 3D convolutional neural networks carry out simultaneously.
As can be seen that determining to read information first the invention relates to based on the face In vivo detection after formula
Type (picture or video) is positioned then using existing Face datection and positioning feature point is based on by human face posture situation
To face state, and preliminary In vivo detection is provided, after the success of preliminary In vivo detection, deep learning detection window will be opened,
And then effectively adjudicate In vivo detection.
It is specific as follows:
1, face detection module
For the input data source of In vivo detection, in face detection module, first have to carry out Face datection, then for figure
The clarity of picture carries out corresponding initialization process, such as lighting process, Fuzzy Processing etc., it is therefore an objective to guarantee that input picture is clear
Clear, picture quality is able to satisfy the requirement of subsequent In vivo detection.
2, face tracking
After detecting the face in input video, face tracking is carried out at random.Specific practice is using before in video sequence
Two field pictures afterwards judge the consistency of two frame face of front and back, if inconsistent or face is lost, In vivo detection failure.?
Mean the face (or picture) into camera in actual conditions again out of camera, face tracking can lose certainly at this time
It loses, therefore, In vivo detection will not can be carried out.
3, human face posture
Posture filtering is filtered primarily directed to big posture, and human face posture mainly includes a left side by spinning upside down (pitch)
Right overturning (yaw), the posture that the certain angle of plane internal rotation (roll) is constituted, by being detected before determining within given time
The case where angle of face out changes, such as blink situation, mouth are mobile, can judge the posture of face.
4, the In vivo detection based on 3D convolutional neural networks
As shown in Fig. 2, realizing that living body faces detect using 3D convolutional neural networks in one embodiment of the application；?
When designing 3D (3Dimension) convolutional neural networks structure, using the original image size of living body faces database as defeated
Enter, to reduce the excessive calculation amount of multiple dimensioned bring；By original image size as input, and living body faces database is done
Internal data library test finds the input frame number of suitable networks.
In some specific embodiments of the application, living body inspection is carried out for combination formula attitude detection and CNN network
It surveys, devises a kind of video analysis method for integrating human face posture filtering, In vivo detection, it is invalid efficiently to screen
Input data, filters out unnecessary situation in time, has not only improved the discrimination of In vivo detection, but also does not need to increase additional set
Standby, detection real-time is effectively enhanced.When illegal user takes out the photo or associated video of legitimate user it may first have to
By attitude detection, after passing through attitude detection, depth detection is carried out, is convenient for subsequent face verification.Based on 3D convolution mind
In vivo detection module through network, the ability for improving In vivo detection by extracting the feature in multiple color spaces.
As shown in figure 3, in CNN In vivo detection flow chart, progress Face datection first, then carry out lasting face with
Track after posture filtering, starts In vivo detection, system can prompt to prepare 5 seconds, and user is prompted to convert human face posture.Then, together
Shi Jinhang formula In vivo detection and In vivo detection based on 3D convolutional neural networks.There are two steps for formula In vivo detection first
Suddenly it carries out simultaneously, first, in background logic processing, reads within timing 1 second the value of posture mark flag, flag is identified according to posture
Judge whether posture changes, if posture does not change, judges that In vivo detection fails；If posture changes, at 2 seconds
It obtains posture state state in timing to classify if state is consistent with requirement, if it is inconsistent, wanting again
User is asked to respond posture；Second, the user in front end cooperates, and setting expectation posture, system prompt user converts posture
(such as blink, open one's mouth), timing 5 seconds, if it times out, providing the conclusion of In vivo detection failure.Based on 3D convolutional Neural net
The In vivo detection and formula In vivo detection of network carry out simultaneously, which can be directly entered point according to the result of image characteristics extraction
Class.
It can be seen that the embodiment of the present application, on the basis of based on the biopsy method of simple posture formula, be added
The detection module of deep learning effectively filters out video registration bring attack, obtains the accuracy rate of face In vivo detection
To large increase.
The above, above embodiments are only to illustrate the technical solution of the application, rather than its limitations；Although referring to before
Embodiment is stated the application is described in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of human face in-vivo detection method, which comprises the following steps:
Step 1: obtaining testing image data source, carries out Face datection to described image data source, obtains testing image data；
Step 2: carrying out the testing image data to continue face tracking detection, if front and back two continuous frames face is consistent, after
Continuous to carry out step 3, if front and back two continuous frames face is inconsistent or face is lost, In vivo detection terminates, and obtains face non-living body
Testing result；
Step 3: human face posture filtering is carried out to the facial image to be measured or face picture detected by face tracking；
Step 4: living body faces detection is carried out using convolutional neural networks.
2. human face in-vivo detection method according to claim 1, which is characterized in that carried out just to the testing image data
Beginningization processing, the initialization process includes lighting process, Fuzzy Processing.
3. human face in-vivo detection method according to claim 1 or 2, which is characterized in that the testing image data, including
Facial image and/or face picture.
4. human face in-vivo detection method according to claim 1, which is characterized in that the step 1 further include to it is described to
Altimetric image data carry out information type judgement.
5. human face in-vivo detection method according to claim 1, which is characterized in that held to the testing image data
Continuous face tracking detection specifically: after detecting the face in testing image data, carry out face tracking at random, obtain video sequence
Include the front and back two continuous frames image of facial image in column, is judged according to the consistency of front and back two continuous frames facial image.
6. human face in-vivo detection method according to claim 1, which is characterized in that the human face posture filtering specifically:
The situation of change that the facial angle detected occurs within given time determines.
7. human face in-vivo detection method according to claim 1, which is characterized in that the convolutional neural networks are 3D convolution
Neural network.
8. human face in-vivo detection method according to claim 1, which is characterized in that the living body faces detection includes cooperation
Formula In vivo detection and In vivo detection based on 3D convolutional neural networks.
9. human face in-vivo detection method according to claim 1, which is characterized in that the formula In vivo detection passes through
Transformation human face posture detects as required for prompt user cooperation, comprising the following steps:
Step 1: in background logic processing, reading within timing 1 second the value of posture mark flag, flag is identified according to the posture and is sentenced
Whether the posture of breaking changes, if the posture does not change, judges that In vivo detection fails；If the posture changes,
The posture state state is then obtained in timing in 2 seconds, if the posture state state is consistent with the requirement, is divided
Class, if the posture state state and it is described require it is inconsistent, again prompt user cooperation as required to human face posture carry out
Response；
Step 2: the user being cooperated, setting expectation posture carries out transformation posture, timing 5 by user described in system prompt
Second, if it times out, In vivo detection fails；
The step 1 and step 2 carry out simultaneously.
10. human face in-vivo detection method according to claim 8, which is characterized in that the formula In vivo detection and institute
It states the In vivo detection based on 3D convolutional neural networks while carrying out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459708.8A CN109376704A (en) | 2018-11-30 | 2018-11-30 | A kind of human face in-vivo detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459708.8A CN109376704A (en) | 2018-11-30 | 2018-11-30 | A kind of human face in-vivo detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109376704A true CN109376704A (en) | 2019-02-22 |
Family
ID=65374988
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811459708.8A Pending CN109376704A (en) | 2018-11-30 | 2018-11-30 | A kind of human face in-vivo detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376704A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110069983A (en) * | 2019-03-08 | 2019-07-30 | 深圳神目信息技术有限公司 | Vivo identification method, device, terminal and readable medium based on display medium |
| CN111325157A (en) * | 2020-02-24 | 2020-06-23 | 高新兴科技集团股份有限公司 | Face snapshot method, computer storage medium and electronic device |
| CN112163493A (en) * | 2020-09-21 | 2021-01-01 | 中国科学院信息工程研究所 | Video false face detection method and electronic device |
| CN114639129A (en) * | 2020-11-30 | 2022-06-17 | 北京君正集成电路股份有限公司 | Paper medium living body detection method for access control system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
| CN108182409A (en) * | 2017-12-29 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Biopsy method, device, equipment and storage medium |
| CN108268864A (en) * | 2018-02-24 | 2018-07-10 | 达闼科技（北京）有限公司 | Face identification method, system, electronic equipment and computer program product |
| CN108549854A (en) * | 2018-03-28 | 2018-09-18 | 中科博宏（北京）科技有限公司 | A kind of human face in-vivo detection method |
| US20180276455A1 (en) * | 2017-03-27 | 2018-09-27 | Samsung Electronics Co., Ltd. | Apparatus and method for image processing |
-
2018
- 2018-11-30 CN CN201811459708.8A patent/CN109376704A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106557723A (en) * | 2015-09-25 | 2017-04-05 | 北京市商汤科技开发有限公司 | A kind of system for face identity authentication with interactive In vivo detection and its method |
| US20180276455A1 (en) * | 2017-03-27 | 2018-09-27 | Samsung Electronics Co., Ltd. | Apparatus and method for image processing |
| CN108182409A (en) * | 2017-12-29 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Biopsy method, device, equipment and storage medium |
| CN108268864A (en) * | 2018-02-24 | 2018-07-10 | 达闼科技（北京）有限公司 | Face identification method, system, electronic equipment and computer program product |
| CN108549854A (en) * | 2018-03-28 | 2018-09-18 | 中科博宏（北京）科技有限公司 | A kind of human face in-vivo detection method |
Non-Patent Citations (1)
| Title |
|---|
| 李山路: "基于3D卷积神经网络的活体人脸检测研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110069983A (en) * | 2019-03-08 | 2019-07-30 | 深圳神目信息技术有限公司 | Vivo identification method, device, terminal and readable medium based on display medium |
| CN111325157A (en) * | 2020-02-24 | 2020-06-23 | 高新兴科技集团股份有限公司 | Face snapshot method, computer storage medium and electronic device |
| CN112163493A (en) * | 2020-09-21 | 2021-01-01 | 中国科学院信息工程研究所 | Video false face detection method and electronic device |
| CN114639129A (en) * | 2020-11-30 | 2022-06-17 | 北京君正集成电路股份有限公司 | Paper medium living body detection method for access control system |
| CN114639129B (en) * | 2020-11-30 | 2024-05-03 | 北京君正集成电路股份有限公司 | Paper medium living body detection method for access control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376704A (en) | A kind of human face in-vivo detection method | |
| CN103440479B (en) | A kind of method and system for detecting living body human face | |
| CN103106397B (en) | Human face in-vivo detection method based on bright pupil effect | |
| CN108764058B (en) | Double-camera face in-vivo detection method based on thermal imaging effect | |
| CN108140123A (en) | Face living body detection method, electronic device and computer program product | |
| US20190095701A1 (en) | Living-body detection method, device and storage medium | |
| CN108898112A (en) | A kind of near-infrared human face in-vivo detection method and system | |
| CN107341481A (en) | It is identified using structure light image | |
| CN107169483A (en) | Tasks carrying based on recognition of face | |
| US20090135188A1 (en) | Method and system of live detection based on physiological motion on human face | |
| CN106372601A (en) | In vivo detection method based on infrared visible binocular image and device | |
| CN105956572A (en) | In vivo face detection method based on convolutional neural network | |
| CN109684965A (en) | A kind of face identification system based near infrared imaging and deep learning | |
| CN108021892B (en) | Human face living body detection method based on extremely short video | |
| TWI603293B (en) | Method and apparatus for detecting blink | |
| CN110866454B (en) | Face living body detection method and system and computer readable storage medium | |
| CN109508706B (en) | Silence living body detection method based on micro-expression recognition and non-sensory face recognition | |
| CN108875461A (en) | A kind of human face in-vivo detection method and device | |
| CN103123690B (en) | Information acquisition device, information acquisition method, identification system and identification method | |
| KR101640014B1 (en) | Iris recognition apparatus for detecting false face image | |
| CN109886137A (en) | Infant sleeping posture detection method, device and computer readable storage medium | |
| CN110069983A (en) | Vivo identification method, device, terminal and readable medium based on display medium | |
| CN110929705A (en) | Living body detection method and device, identity authentication method and system and storage medium | |
| CN108446690A (en) | A kind of human face in-vivo detection method based on various visual angles behavioral characteristics | |
| CN107862298A (en) | It is a kind of based on the biopsy method blinked under infrared eye |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190222 |
|
| RJ01 | Rejection of invention patent application after publication |