CN109977846A - A kind of in-vivo detection method and system based on the camera shooting of near-infrared monocular - Google Patents
A kind of in-vivo detection method and system based on the camera shooting of near-infrared monocular Download PDFInfo
- Publication number
- CN109977846A CN109977846A CN201910221151.2A CN201910221151A CN109977846A CN 109977846 A CN109977846 A CN 109977846A CN 201910221151 A CN201910221151 A CN 201910221151A CN 109977846 A CN109977846 A CN 109977846A
- Authority
- CN
- China
- Prior art keywords
- infrared
- face
- image
- monocular
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G06V40/172—Classification, e.g. identification
-
- 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
-
- 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
Abstract
The present invention proposes a kind of biopsy method based on the camera shooting of near-infrared monocular, comprising: acquisition near-infrared image information；It detects whether comprising face in the near-infrared image, if face is not detected, judges to identify the non-true man of object；If detecting face, user is prompted to make specified facial expressions and acts；The Optical-flow Feature of the facial expressions and acts is extracted, while extracting the face-image depth characteristic of near-infrared image；The Optical-flow Feature and face-image depth characteristic are inputted into deep learning classifier；Obtain face recognition result；The present invention can effectively take precautions against video and three-dimensional mask attack, improve the accuracy rate of In vivo detection.
Description
Technical field
The present invention relates to security protection identification field more particularly to it is a kind of based on near-infrared monocular camera shooting biopsy method and
System.
Background technique
In recent years, face recognition technology commercial applications are further extensive, however face easily uses photo, video, three-dimensional surface
The modes such as tool are palmed off, therefore face In vivo detection is the important topic of recognition of face Yu Verification System safety.It is examined from living body
From the point of view of the type of the image capture device of survey, In vivo detection common at present is mainly visible images acquisition and multispectral image
Two classes are acquired, wherein multispectral Image Acquisition includes the imaging devices such as near-infrared, far infrared, thermal infrared；From In vivo detection
From the point of view of implementation method, there are interactive In vivo detection and non-interactive type biopsy method, interactive In vivo detection includes passing through use
Whether the family movements such as blink, open one's mouth to distinguish are true man.And non-interactive type In vivo detection is then not necessarily to realize by user's cooperation.
The advantage of biopsy method based on visible light be it is at low cost, for the mobile subscribers such as mobile phone do In vivo detection without
Hardware device need to be added；The disadvantage is that poor robustness, influences vulnerable to light variation etc., it can not judge that HD video is attacked.So base
It can be used only in the lower occasion of security level in the In vivo detection of visible light.
Method based on multi-spectral image processing is usually to use binocular camera, including a visible image capturing head and one
A infrared camera.There is provided multispectral information, the adaptability of the further boosting algorithm of energy for its advantage；The disadvantage is that increasing hard
Part cost and power consumption, and two cameras can not may be installed simultaneously for some portable small devices.
Interactive In vivo detection is usually to pass through visible image capturing head acquisition data, is referred to by system to user's sending movement
It enables to determine whether being true man.The disadvantage is that for commonly acting --- it blinks, open one's mouth, can not judge whether it is that true man wear
After mask, acted by the hole of mask.In addition it can not also judge the movement attack of HD video.
Non-interactive type In vivo detection mainly passes through the information of single-frame images to determine whether being true man.Its advantage is reaction speed
Degree is fast, and disadvantage is that accuracy rate is not high, is easy under attack.
Summary of the invention
In view of the above problem of the existing technology, the present invention proposes a kind of In vivo detection based on the camera shooting of near-infrared monocular
Method and system mainly solve the problems, such as prior art poor robustness and at high cost.
To achieve the goals above and other purposes, the technical solution adopted by the present invention are as follows.
A kind of biopsy method based on the camera shooting of near-infrared monocular, comprising:
Acquire near-infrared image information；
It detects whether comprising face in the near-infrared image, if face is not detected, judges to identify the non-true man of object；
If detecting face, user is prompted to make specified facial expressions and acts；
The Optical-flow Feature of the facial expressions and acts is extracted, while extracting the face-image depth characteristic of near-infrared image；
The Optical-flow Feature and face-image depth characteristic are inputted into deep learning classifier；
Obtain In vivo detection result.
Optionally, the near-infrared image is acquired using near-infrared monocular cam.The near-infrared monocular cam packet
The near-infrared monocular cam for including 850nm, only acquires near-infrared image, does not need visible image capturing head, reduces equipment cost
And power consumption.
Optionally, the infrared supplementary lighting sources for the interference of wiping out background light are introduced when acquiring near-infrared image.It is described infrared
The effective distance of supplementary lighting sources is shorter, can largely filtering environmental light interference, improve the accuracy rate of detection.
Optionally, the extraction face-image depth characteristic specifically includes:
The convolutional layer that will test the near-infrared image information input deep learning neural network comprising face, obtains near-infrared
Characteristics of image；Every layer of convolutional layer is acted on by the first activation primitive, activates the face characteristic in the near-infrared image feature,
Finally obtain face-image depth characteristic.
Optionally, the specific implementation procedure of the deep learning classifier includes:
The face-image depth characteristic and Optical-flow Feature are combined to the full connection for inputting the deep learning classifier
Layer, then every layer of full articulamentum is acted on by the second activation primitive and activates corresponding countenance motion characteristic, and according to described
Countenance motion characteristic judges identify whether object is real human face.
Optionally, first activation primitive includes:
Wherein, x is near-infrared image feature, and y is the output of activation primitive.
Optionally, second activation primitive includes:
Wherein, x is the assemblage characteristic of face image depth characteristic and Optical-flow Feature, and y is the output of activation primitive.Described
The ability to express of one activation primitive and the second activation primitive is strong, and good convergence can effectively improve deep neural network algorithm
Accuracy rate.
A kind of In vivo detection system based on the camera shooting of near-infrared monocular, comprising:
Near-infrared image acquisition module, for acquiring near-infrared image；
Detection module, for detecting in the near-infrared image with the presence or absence of face；
Action prompt module, for prompting user to make specified facial expressions and acts；
Characteristic extracting module, for extracting the Optical-flow Feature of countenance movement；
Deep learning identification module, the Optic flow information that facial image and countenance act for identification；
The output end connecting detection module of the near-infrared image acquisition module and the input terminal of characteristic extracting module；It is described
The output end of detection module is connect with the input terminal of action prompt module and deep learning identification module, the characteristic extracting module
Output end connect deep learning identification module.
Optionally, the near-infrared image acquisition module includes near-infrared monocular cam and infrared supplementary lighting sources.
Optionally, the deep learning identification module includes near-infrared facial image feature extraction unit and for identification
The depth recognition unit of Optical-flow Feature and near-infrared facial image feature；The near-infrared facial image feature extraction unit it is defeated
The input terminal of outlet connection depth recognition unit.
As described above, a kind of in-vivo detection method and system based on the camera shooting of near-infrared monocular of the present invention, has with following
Beneficial effect.
Infrared supplementary lighting sources are introduced, the interference of environment light can be effectively filtered out；By detecting near-infrared image face information,
It can effectively avoid the attack of video or photo；It is imaged using monocular and reduces system cost and power consumption；According to the light stream of facial expressions and acts
Feature can take precautions against the attack of three-dimensional mask.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the biopsy method based on the camera shooting of near-infrared monocular of the present invention.
Fig. 2 is a kind of structural block diagram of the In vivo detection system based on the camera shooting of near-infrared monocular of the present invention.
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification
Other advantages and efficacy of the present invention can be easily understood for disclosed content.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from
Various modifications or alterations are carried out under spirit of the invention.It should be noted that in the absence of conflict, following embodiment and implementation
Feature in example can be combined with each other.
It should be noted that illustrating the basic structure that only the invention is illustrated in a schematic way provided in following embodiment
Think, only shown in schema then with related component in the present invention rather than component count, shape and size when according to actual implementation
Draw, when actual implementation kenel, quantity and the ratio of each component can arbitrarily change for one kind, and its assembly layout kenel
It is likely more complexity.
Referring to Fig. 1, the present invention provides a kind of biopsy method based on the camera shooting of near-infrared monocular in one embodiment,
Include:
Acquire near-infrared image information；
It detects whether comprising face in the near-infrared image, if face is not detected, judges to identify the non-true man of object；
If detecting face, user is prompted to make specified facial expressions and acts；
The Optical-flow Feature of the facial expressions and acts is extracted, while extracting the face-image depth characteristic of near-infrared image；
The Optical-flow Feature and face-image depth characteristic are inputted into deep learning classifier；
Obtain In vivo detection result.
The acquisition near-infrared image information mainly passes through near-infrared monocular cam and cooperates infrared supplementary lighting sources；One
In embodiment, the near-infrared monocular cam includes using 850nm near-infrared monocular cam, can also use other shortwaves
Near-infrared camera is to reach identical effect.The infrared supplementary lighting sources infrared supplementary lighting sources shorter using effective distance, can have
The interference for imitating filtering environmental light, enhances the effect of the near-infrared image of acquisition, reduces the False Rate of detection process.
The near-infrared image is detected, is judged in the near-infrared image with the presence or absence of face.The purpose of detection is to prevent
The means such as model video attack, because video attack needs to show on one screen, which can be mobile phone, computer is shown
The equipment such as device, notebook or TV.And these displays can fully absorb near infrared light, so that near infrared light shooting screen
The picture that backstage obtains all is a piece of black, can not detect face.Therefore, if can not be detected in the near-infrared image of acquisition
To facial image, then there is not face before preceding camera in explanation, or to use the video of screen to attack.If can detect
Human face image information, then explanation is not the video or image presented by screen before preceding camera.
When including face in the near-infrared image information for detecting the acquisition, then corresponding near-infrared image is sent to
Deep learning neural network model obtains the convolutional layer of near-infrared image input deep learning neural network model close
Infrared Image Features；Acting on every layer of convolutional layer by the first activation primitive activates the face in the near-infrared image feature special
Sign, finally obtains face image depth characteristic.First activation primitive includes:
Wherein, x is near-infrared image feature, and y is the output of activation primitive.
When in the near-infrared image for detecting the acquisition including face, prompt information can be also issued the user with, is prompted
User makes specified facial expressions and acts.The prompt information includes text importing, voice prompting etc.；In one embodiment, described
Facial expressions and acts mainly include smiling and frowning, and why select the two facial expressions and acts, and do not have to relatively common blink and open
Mouth, be due to for blinking and opening one's mouth, can be worn by true man the mask of eyes and mouth borehole, do in hole movement come
Fraud system.And the two expressions of frowning and smile, involved more facial muscles, these facial muscles actuate be can not
It is simulated by three-dimensional mask.
After the near-infrared monocular cam collects user's facial expressions and acts, the sparse optical flow of two interval interframe is calculated
Feature.The sparse optical flow refers to using pixel in image sequence in the variation in time-domain and the correlation between consecutive frame
To find previous frame with corresponding relationship existing between present frame, to calculate one of the motion information of object between consecutive frame
Kind method.
By the deep learning of obtained face image depth characteristic and sparse optical flow feature input deep neural network model
The face-image depth characteristic and Optical-flow Feature are combined the full connection for inputting the deep learning classifier by classifier
Layer, then every layer of full articulamentum is acted on by the second activation primitive and activates corresponding facial expressions and acts feature.According to facial expressions and acts spy
Sign judges whether the near-infrared image of acquisition is true man's image.In one embodiment, the deep neural network model is to pass through
The deep learning model that a large amount of near-infrared facial image training obtain.
Second activation primitive includes:
Wherein, x is the assemblage characteristic of face image depth characteristic and Optical-flow Feature, and y is the output of activation primitive.
The ability to express of first activation primitive and the second activation primitive is strong, and good convergence can effectively improve depth
Spend the accuracy rate of neural network algorithm.
Referring to Fig. 2, in another embodiment, a kind of In vivo detection system based on the camera shooting of near-infrared monocular of the present invention,
Include:
Near-infrared image acquisition module, for acquiring near-infrared image；
Detection module, for detecting in the near-infrared image with the presence or absence of face；
Action prompt module, for prompting user to make specified facial expressions and acts；
Characteristic extracting module, for extracting the Optical-flow Feature of countenance movement；
Deep learning identification module, the Optic flow information that facial image and countenance act for identification；
The output end connecting detection module of the near-infrared image acquisition module and the input terminal of characteristic extracting module；It is described
The output end of detection module is connect with the input terminal of action prompt module and deep learning identification module, the characteristic extracting module
Output end connect deep learning identification module.
The near-infrared image acquisition module includes near-infrared monocular cam and infrared supplementary lighting sources.The near-infrared list
Mesh camera includes using 850nm near-infrared monocular cam, other short wavelength-NIR cameras can also be used to reach phase
Same effect.The infrared supplementary lighting sources infrared supplementary lighting sources shorter using effective distance, can effectively filter out the interference of environment light, increase
The effect of the near-infrared image acquired by force reduces the False Rate of detection process.
The detection module receives the near-infrared image of near-infrared image acquisition module acquisition, and detects the near-infrared figure
Whether include face as in, if not including face, determines not have before current monocular cam before face or camera for display
Shield the face video presented or image.If testing result is that corresponding near-infrared image is sent to depth comprising face
Practise identification module.The deep learning identification module is obtained by the training of a large amount of near-infrared facial images.The depth
Practising identification module includes near-infrared facial image feature extraction unit and Optical-flow Feature and infrared face characteristics of image for identification
Depth recognition unit.
The near-infrared image is inputted into the near-infrared facial image feature extraction unit, obtains near-infrared image spy
Sign；The face characteristic in the near-infrared image feature is activated by the first activation primitive, obtains face image depth characteristic.Institute
Stating the first activation primitive includes:
Wherein, x is near-infrared image feature, and y is the output of activation primitive.
Also will test result is sent to action prompt module to detection module simultaneously, if detecting the near-infrared figure of the acquisition
It include face as in, the action prompt module issues the user with prompt information according to detection module testing result, prompts user
Make specified facial expressions and acts.In one embodiment, the prompt information includes text importing, voice prompting.The specified table
Feelings movement includes smiling and frowning.
The near-infrared image acquisition module acquisition human face expression movement, and the corresponding near-infrared image of facial expressions and acts is sent out
Characteristic extracting module is given, sparse optical flow feature is obtained.
By the sparse optical flow feature and face image depth characteristic in conjunction with the full connection for inputting the depth recognition unit
Layer, then every layer of full articulamentum is acted on by the second activation primitive, activate corresponding facial expressions and acts feature.According to finally obtained
Facial expressions and acts feature judges whether the near-infrared image of acquisition is true man's image.
Second activation primitive includes:
Wherein, x is the assemblage characteristic of face image depth characteristic and Optical-flow Feature, and y is the output of activation primitive.
In conclusion a kind of in-vivo detection method and system based on the camera shooting of near-infrared monocular of the present invention, is used only close red
The near-infrared image of outer monocular cam acquisition carries out In vivo detection, does not need visible images, reduces system cost and function
Consumption；Infrared supplementary lighting sources effectively filter out bias light interference, improve the quality of acquisition image；It is strong using ability to express, convergence is good
Activation primitive, can effectively improve the accuracy rate of identification；Near-infrared image combination sparse optical flow feature can effectively take precautions against
The attack of video and three-dimensional mask.So the present invention effectively overcomes various shortcoming in the prior art and has high industrial benefit
With value.
The above-described embodiments merely illustrate the principles and effects of the present invention, and is not intended to limit the present invention.It is any ripe
The personage for knowing this technology all without departing from the spirit and scope of the present invention, carries out modifications and changes to above-described embodiment.Cause
This, institute is complete without departing from the spirit and technical ideas disclosed in the present invention by those of ordinary skill in the art such as
At all equivalent modifications or change, should be covered by the claims of the present invention.
Claims (10)
1. a kind of biopsy method based on the camera shooting of near-infrared monocular characterized by comprising
Acquire near-infrared image information；
It detects whether comprising face in the near-infrared image, if face is not detected, judges to identify the non-true man of object；If inspection
Face is measured, then user is prompted to make specified facial expressions and acts；
The Optical-flow Feature of the facial expressions and acts is extracted, while extracting the face-image depth characteristic of near-infrared image；
The Optical-flow Feature and face-image depth characteristic are inputted into deep learning classifier；
Obtain In vivo detection result.
2. the biopsy method according to claim 1 based on the camera shooting of near-infrared monocular, which is characterized in that using close red
Outer monocular cam acquires the near-infrared image.
3. the biopsy method according to claim 2 based on the camera shooting of near-infrared monocular, which is characterized in that acquisition is close red
The infrared supplementary lighting sources for the interference of wiping out background light are introduced when outer image.
4. the biopsy method according to claim 1 based on the camera shooting of near-infrared monocular, which is characterized in that the extraction
Face-image depth characteristic specifically includes:
The convolutional layer that will test the near-infrared image information input deep learning neural network comprising face, obtains near-infrared image
Feature；Every layer of convolutional layer is acted on by the first activation primitive, activates the face characteristic in the near-infrared image feature, finally
Obtain face-image depth characteristic.
5. the biopsy method according to claim 1 based on the camera shooting of near-infrared monocular, which is characterized in that the depth
The specific implementation procedure of Study strategies and methods includes:
The face-image depth characteristic and Optical-flow Feature are combined to the full articulamentum for inputting the deep learning classifier, then
Every layer of full articulamentum is acted on by the second activation primitive and activates corresponding countenance motion characteristic, and according to finally obtained
Countenance motion characteristic judges identify whether object is real human face.
6. the biopsy method according to claim 4 based on the camera shooting of near-infrared monocular, which is characterized in that described first
Activation primitive includes:
Wherein, x is near-infrared image feature, and y is the output of activation primitive.
7. the biopsy method according to claim 5 based on the camera shooting of near-infrared monocular, which is characterized in that described second
Activation primitive includes:
Wherein, x is the assemblage characteristic of face image depth characteristic and Optical-flow Feature, and y is the output of activation primitive.
8. a kind of In vivo detection system based on the camera shooting of near-infrared monocular characterized by comprising
Near-infrared image acquisition module, for acquiring near-infrared image；
Detection module, for detecting in the near-infrared image with the presence or absence of face；
Action prompt module, for prompting user to make specified facial expressions and acts；
Characteristic extracting module, for extracting the Optical-flow Feature of countenance movement；
Deep learning identification module, the Optic flow information that facial image and countenance act for identification；
The output end connecting detection module of the near-infrared image acquisition module and the input terminal of characteristic extracting module；The detection
The output end of module is connect with the input terminal of action prompt module and deep learning identification module, the characteristic extracting module it is defeated
Outlet connects deep learning identification module.
9. the In vivo detection system according to claim 8 based on the camera shooting of near-infrared monocular, which is characterized in that described close red
Outer image capture module includes near-infrared monocular cam and infrared supplementary lighting sources.
10. the In vivo detection system according to claim 8 based on the camera shooting of near-infrared monocular, which is characterized in that the depth
Degree study identification module includes near-infrared facial image feature extraction unit and Optical-flow Feature and near-infrared face for identification
The depth recognition unit of characteristics of image；The output end of the near-infrared facial image feature extraction unit connects depth recognition unit
Input terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910221151.2A CN109977846B (en) | 2019-03-22 | 2019-03-22 | Living body detection method and system based on near-infrared monocular photography |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910221151.2A CN109977846B (en) | 2019-03-22 | 2019-03-22 | Living body detection method and system based on near-infrared monocular photography |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109977846A true CN109977846A (en) | 2019-07-05 |
| CN109977846B CN109977846B (en) | 2023-02-10 |
Family
ID=67080001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910221151.2A Active CN109977846B (en) | 2019-03-22 | 2019-03-22 | Living body detection method and system based on near-infrared monocular photography |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977846B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991307A (en) * | 2019-11-27 | 2020-04-10 | 北京锐安科技有限公司 | Face recognition method, device, equipment and storage medium |
| CN112597932A (en) * | 2020-12-28 | 2021-04-02 | 上海汽车集团股份有限公司 | Living body detection method and device and computer readable storage medium |
| CN113066237A (en) * | 2021-03-26 | 2021-07-02 | 中国工商银行股份有限公司 | Face living body detection and identification method for automatic teller machine and automatic teller machine |
| CN113691696A (en) * | 2021-07-23 | 2021-11-23 | 杭州魔点科技有限公司 | Face recognition method based on camera module and camera module |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2546996A1 (en) * | 2005-05-17 | 2006-11-17 | Spectratech Inc. | Optical coherence tomograph |
| CN105787458A (en) * | 2016-03-11 | 2016-07-20 | 重庆邮电大学 | Infrared behavior identification method based on adaptive fusion of artificial design feature and depth learning feature |
| CN106203369A (en) * | 2016-07-18 | 2016-12-07 | 三峡大学 | Active stochastic and dynamic for anti-counterfeiting recognition of face instructs generation system |
| CN107111598A (en) * | 2014-12-19 | 2017-08-29 | 深圳市大疆创新科技有限公司 | Use the light stream imaging system and method for ultrasonic depth sense |
| CN107358181A (en) * | 2017-06-28 | 2017-11-17 | 重庆中科云丛科技有限公司 | The infrared visible image capturing head device and method of monocular judged for face live body |
| CN107368798A (en) * | 2017-07-07 | 2017-11-21 | 四川大学 | A kind of crowd's Emotion identification method based on deep learning |
| CN206672320U (en) * | 2017-04-18 | 2017-11-24 | 中国科学院重庆绿色智能技术研究院 | A kind of reverse crowd regulation warning system in region |
| CN108062546A (en) * | 2018-02-11 | 2018-05-22 | 厦门华厦学院 | A kind of computer face Emotion identification system |
| CN108664922A (en) * | 2018-05-10 | 2018-10-16 | 东华大学 | A kind of infrared video Human bodys' response method based on personal safety |
| CN108875509A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | Biopsy method, device and system and storage medium |
| CN109145817A (en) * | 2018-08-21 | 2019-01-04 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of face In vivo detection recognition methods |
| CN109308436A (en) * | 2017-07-28 | 2019-02-05 | 西南科技大学 | A kind of living body faces recognition methods based on active infrared video |
-
2019
- 2019-03-22 CN CN201910221151.2A patent/CN109977846B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2546996A1 (en) * | 2005-05-17 | 2006-11-17 | Spectratech Inc. | Optical coherence tomograph |
| CN107111598A (en) * | 2014-12-19 | 2017-08-29 | 深圳市大疆创新科技有限公司 | Use the light stream imaging system and method for ultrasonic depth sense |
| CN105787458A (en) * | 2016-03-11 | 2016-07-20 | 重庆邮电大学 | Infrared behavior identification method based on adaptive fusion of artificial design feature and depth learning feature |
| CN106203369A (en) * | 2016-07-18 | 2016-12-07 | 三峡大学 | Active stochastic and dynamic for anti-counterfeiting recognition of face instructs generation system |
| CN206672320U (en) * | 2017-04-18 | 2017-11-24 | 中国科学院重庆绿色智能技术研究院 | A kind of reverse crowd regulation warning system in region |
| CN107358181A (en) * | 2017-06-28 | 2017-11-17 | 重庆中科云丛科技有限公司 | The infrared visible image capturing head device and method of monocular judged for face live body |
| CN107368798A (en) * | 2017-07-07 | 2017-11-21 | 四川大学 | A kind of crowd's Emotion identification method based on deep learning |
| CN109308436A (en) * | 2017-07-28 | 2019-02-05 | 西南科技大学 | A kind of living body faces recognition methods based on active infrared video |
| CN108875509A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | Biopsy method, device and system and storage medium |
| CN108062546A (en) * | 2018-02-11 | 2018-05-22 | 厦门华厦学院 | A kind of computer face Emotion identification system |
| CN108664922A (en) * | 2018-05-10 | 2018-10-16 | 东华大学 | A kind of infrared video Human bodys' response method based on personal safety |
| CN109145817A (en) * | 2018-08-21 | 2019-01-04 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of face In vivo detection recognition methods |
Non-Patent Citations (3)
| Title |
|---|
| FERNÁNDEZ-CABALLERO: "Optical flow or image subtraction in human detection from infrared camera on mobile robot", 《ROBOTICS AND AUTONOMOUS SYSTEMS》 * |
| 陶明慧: "基于退化模型的图像渐晕处理技术", 《红外与激光工程》 * |
| 黄建恺: "人脸识别的活体检测技术研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991307A (en) * | 2019-11-27 | 2020-04-10 | 北京锐安科技有限公司 | Face recognition method, device, equipment and storage medium |
| CN110991307B (en) * | 2019-11-27 | 2023-09-26 | 北京锐安科技有限公司 | Face recognition method, device, equipment and storage medium |
| CN112597932A (en) * | 2020-12-28 | 2021-04-02 | 上海汽车集团股份有限公司 | Living body detection method and device and computer readable storage medium |
| CN113066237A (en) * | 2021-03-26 | 2021-07-02 | 中国工商银行股份有限公司 | Face living body detection and identification method for automatic teller machine and automatic teller machine |
| CN113691696A (en) * | 2021-07-23 | 2021-11-23 | 杭州魔点科技有限公司 | Face recognition method based on camera module and camera module |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109977846B (en) | 2023-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104915649B (en) | A kind of biopsy method applied to recognition of face | |
| CN109977846A (en) | A kind of in-vivo detection method and system based on the camera shooting of near-infrared monocular | |
| Shao et al. | Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing | |
| CN108596041B (en) | A kind of human face in-vivo detection method based on video | |
| Fathy et al. | Face-based active authentication on mobile devices | |
| WO2018040307A1 (en) | Vivo detection method and device based on infrared visible binocular image | |
| CN108470169A (en) | Face identification system and method | |
| CN105659200B (en) | For showing the method, apparatus and system of graphic user interface | |
| CN102708383B (en) | System and method for detecting living face with multi-mode contrast function | |
| CN105260726B (en) | Interactive video biopsy method and its system based on human face posture control | |
| CN109598242B (en) | Living body detection method | |
| CN109271950A (en) | A kind of human face in-vivo detection method based on mobile phone forward sight camera | |
| CN110866454B (en) | Face living body detection method and system and computer readable storage medium | |
| Le et al. | Eye blink detection for smart glasses | |
| CN108537131A (en) | A kind of recognition of face biopsy method based on human face characteristic point and optical flow field | |
| CN208351494U (en) | Face identification system | |
| CN109508706A (en) | A kind of silent biopsy method based on micro- Expression Recognition and noninductive recognition of face | |
| Farrukh et al. | FaceRevelio: a face liveness detection system for smartphones with a single front camera | |
| CN111767788A (en) | Non-interactive monocular in vivo detection method | |
| CN108363944A (en) | Recognition of face terminal is double to take the photograph method for anti-counterfeit, apparatus and system | |
| Geng | Research on athlete’s action recognition based on acceleration sensor and deep learning | |
| US9996743B2 (en) | Methods, systems, and media for detecting gaze locking | |
| CN108647650B (en) | Human face in-vivo detection method and system based on corneal reflection and optical coding | |
| CN110222647A (en) | A kind of human face in-vivo detection method based on convolutional neural networks | |
| Ali et al. | Spoofing attempt detection using gaze colocation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |