CN106874871A - A kind of recognition methods of living body faces dual camera and identifying device - Google Patents
A kind of recognition methods of living body faces dual camera and identifying device Download PDFInfo
- Publication number
- CN106874871A CN106874871A CN201710081460.5A CN201710081460A CN106874871A CN 106874871 A CN106874871 A CN 106874871A CN 201710081460 A CN201710081460 A CN 201710081460A CN 106874871 A CN106874871 A CN 106874871A
- Authority
- CN
- China
- Prior art keywords
- image
- white
- black
- camera
- living body
- 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
- 230000009977 dual effect Effects 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 31
- 210000000056 organ Anatomy 0.000 claims abstract description 60
- 238000000605 extraction Methods 0.000 claims abstract description 51
- 230000001815 facial effect Effects 0.000 claims abstract description 20
- 241001465754 Metazoa Species 0.000 claims abstract description 18
- 239000000284 extract Substances 0.000 claims description 26
- 230000003287 optical effect Effects 0.000 claims description 8
- 239000004744 fabric Substances 0.000 claims 1
- 210000000887 face Anatomy 0.000 description 59
- 238000005516 engineering process Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 201000008217 Aggressive systemic mastocytosis Diseases 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 210000004709 eyebrow Anatomy 0.000 description 2
- 210000003128 head Anatomy 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- 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)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of living body faces dual camera recognition methods, including：One black white image and an infrared image are obtained by the black and white camera of the dual camera for configuring；Colour imagery shot obtains a coloured image；Face part binding characteristic extraction algorithm in black white image and near-infrared image is extracted into two-dimensional state organ characteristic's point；Organ characteristic's point of three-dimensional state is formed, the feature of living body faces is recognized by face domestic animal algorithm, judge whether facial image is living body faces.The invention also discloses living body faces dual camera identifying device.The present invention gathers coloured image and near-infrared image by dual camera, by background calculus of differences, background picture i.e. face part is gone in acquisition, and organ characteristic point is extracted by feature extraction algorithm, the feature of living body faces is recognized by face characteristic algorithm, judges whether facial image is living body faces.Recognition methods reliability of the invention is high, convenient and practical, cost of implementation is low.
Description
Technical field
The present invention relates to living body faces identification technology field, and in particular to a kind of recognition methods of living body faces dual camera and
Identifying device.
Background technology
As the technology of security protection is constantly updated, face recognition technology is applied also more and more extensive in life.Especially in political affairs
Mansion department, frontier juncture and financial industry, there is irreplaceable intelligent safety monitoring to security protection.Face recognition technology
Reach its maturity, commercial applications extensive but face further is easily replicated with modes such as photo, videos, therefore to legal
The personation of user's face, is recognition of face, and especially living body faces identification Verification System constitutes important threat.In these years,
Living body faces detection technique has made some progress, but in the security reliability and cost-effectivenes of the existing method of practical application
Balance very high can not be obtained.
Existing living body faces identification technology, mainly detects whether to meet face characteristic by a common camera,
The entity head portrait such as plastic cement being still easily counterfeited is out-tricked.Also have plenty of by professional infra-red radiation imaging lens, pass through
The trickle biological characteristic of scanning living human face, in addition it is trickle to the vascular distribution that can be seen inside live body face.But it is this
Equipment is very expensive, and this has been resulted in can only be adapted to some specific occasions, and can not be widely used.
The content of the invention
It is an object of the invention to provide a kind of living body faces dual camera recognition methods high, convenient and practical of reliability and
Identifying device, to solve the problems, such as to be proposed in above-mentioned background technology.
To achieve the above object, the present invention provides following technical scheme：
A kind of living body faces dual camera recognition methods, it is characterised in that the recognition methods includes：
By the dual camera for configuring, the dual camera is a black and white camera and a colour imagery shot, black and white camera
The black white image that one is generation under natural light or white light conditions is obtained, and obtains another to be produced under the conditions of near infrared light
Near-infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
The black white image that black and white camera in the dual camera is obtained is combined with the face part in near-infrared image
Feature extraction algorithm extracts two-dimensional state organ characteristic's point；
The face part binding characteristic extraction algorithm of the coloured image that colour imagery shot in the dual camera is obtained is carried
Take organ characteristic's point and extract organ characteristic point with the near-infrared image that black and white camera is obtained, form the organ characteristic of three-dimensional state
Point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
The recognition methods also includes：
When being judged to living body faces, a complete face characteristic figure is obtained from the two-dimensional state organ characteristic point for having extracted
Piece and data value, are matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
I.e. output display or control after the completion of matching.
Black white image and near-infrared image are obtained by the black and white camera of the dual camera for configuring, including：
Calculus of differences is carried out to black white image and near-infrared image, calculus of differences includes：
Using black white image as the still image of current environment, using near-infrared image as active light source image, by static state
Image and active light source image carry out calculus of differences, obtain difference image, are drawn according to active light source feature and complete go background
Active light source image.
The feature of living body faces is recognized by face domestic animal algorithm, including：
The near-infrared image that the coloured image that colour imagery shot in dual camera is obtained is obtained with black and white camera is carried out
Optical flow estimation model, otherness distribution estimation, judges whether image is living body faces, if so, then passing through data obtaining module
Binding characteristic extraction algorithm extracts organ characteristic's point of two-dimensional state, if it is not, then terminating.
A kind of living body faces dual camera identifying device, the identifying device includes：
Image collection module, for the dual camera by configuring, the dual camera is that a black and white camera and one are colored
Camera, black and white camera obtains the black white image that is generation under natural light or white light conditions, and another of acquisition is
The near-infrared image produced under the conditions of near infrared light；Colour imagery shot obtains the coloured silk that is generation under natural light or white light conditions
Color image；
Data obtaining module, for the black white image and near-infrared image that obtain black and white camera in the dual camera
In face part binding characteristic extraction algorithm extract two-dimensional state organ characteristic's point；
Judge module, the face part of the coloured image for colour imagery shot in the dual camera to be obtained combines spy
Levy extraction algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three-dimensional shape
Organ characteristic's point of state, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
The identifying device also includes：
Matching module, for being judged to during living body faces, from the two-dimensional state organ characteristic point for having extracted obtain one it is complete
Whole face characteristic picture and data value, are matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
Display or control module, for i.e. output display or control after the completion of matching.
The black and white camera of dual camera that described image acquisition module is additionally operable to by configuring obtains black white image and near
Infrared image, including：
Calculus of differences is carried out to black white image and near-infrared image, calculus of differences includes：
Using black white image as the still image of current environment, using near-infrared image as active light source image, by static state
Image and active light source image carry out calculus of differences, obtain difference image, are drawn according to active light source feature and complete go background
Active light source image.
The judge module is additionally operable to obtain the coloured image that colour imagery shot in dual camera is obtained with black and white camera
The near-infrared image for taking carries out Optical flow estimation model, and otherness distribution estimation judges whether image is living body faces, if so,
Organ characteristic's point of two-dimensional state is then extracted by data obtaining module binding characteristic extraction algorithm, if it is not, then terminating.
A kind of living body faces dual camera identifying device, including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
By the dual camera for configuring, the dual camera is a black and white camera and a colour imagery shot, black and white camera
The black white image that one is generation under natural light or white light conditions is obtained, and obtains another to be produced under the conditions of near infrared light
Near-infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
The black white image that black and white camera in the dual camera is obtained is combined with the face part in near-infrared image
Feature extraction algorithm extracts two-dimensional state organ characteristic's point；
The face part binding characteristic extraction algorithm of the coloured image that colour imagery shot in the dual camera is obtained is carried
Take organ characteristic's point and extract organ characteristic point with the near-infrared image that black and white camera is obtained, form the organ characteristic of three-dimensional state
Point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
The dual camera is each configured with infrared LED and white light LEDs.
Beneficial effects of the present invention：The present invention gathers coloured image and near-infrared image by dual camera, by background
Background picture i.e. face part is gone in calculus of differences, acquisition, organ characteristic point is extracted by feature extraction algorithm, by face characteristic
Algorithm recognizes the feature of living body faces, judges whether facial image is living body faces.Recognition methods reliability of the invention is high, side
Just practical, cost of implementation is low.
Below in conjunction with the accompanying drawings with specific embodiment, the present invention is described in more detail.
Brief description of the drawings
Fig. 1 is the flow chart of the living body faces dual camera recognition methods of embodiment 1；
The flow chart of the feature of the step of Fig. 2 is embodiment 1 102 face domestic animal algorithms identification living body faces；
Fig. 3 is the block diagram of the living body faces dual camera identifying device of embodiment 1；
Fig. 4 is the structural representation of the living body faces dual camera of embodiment 1；
Fig. 5 is the hardware block diagram of the living body faces dual camera identifying device of embodiment 1.
Fig. 6 is the flow chart of the living body faces dual camera recognition methods of embodiment 2；
Fig. 7 is the living body faces dual camera recognition methods FB(flow block) of embodiment 2；
Fig. 8 is the block diagram of the living body faces dual camera identifying device of embodiment 2；
In figure, 1, camera lens, 2, uniform light board, 3, infrared transmitting tube, 4, lens base, 5, mainboard and lamp plate connector, 6, USB
Connector, 7, power connector, 8, mainboard and sensor connector for substrate, 9, lamp plate and motherboard connector, 10, master control borad, 11,
Sensor plates, 12, lamp plate.
Specific embodiment
Embodiment 1, referring to the living body faces dual camera recognition methods that Fig. 1, the present embodiment are provided, the recognition methods bag
Include：
Step 101, by the dual camera for configuring, the dual camera be a black and white camera and a colour imagery shot, it is black
White camera obtains the black white image that is generation under natural light or white light conditions, and it is near-infrared striation to obtain another
The near-infrared image produced under part；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Wherein, background calculus of differences is carried out to black white image and near-infrared image, background calculus of differences includes：With artwork master
As the still image as current environment, using near-infrared image as active light source image, by still image and active light source figure
As carrying out calculus of differences, difference image is obtained, drawn according to active light source feature and complete remove background active light source image；
Background calculus of differences basic process is：In adjacent two frame of image, (the 1st frame is black for current environment static background image
White image, the 2nd frame is near-infrared image for active light source image) between use figure extracted by closing value based on pixel difference
Active light source parts of images as in.First, adjacent two field picture respective pixel value is subtracted each other and obtains difference image, if correspondence picture
When plain value changes are less than pre-determined threshold value, it is believed that be herein background pixel:If the pixel value changes of image-region
It is very big, it is believed that this is caused due to active light source, it is foreground pixel by these zone markers, using the pixel region of mark
Domain can determine active light source target position in the picture.Because the time interval of adjacent two interframe is very short, former frame is used
Image has preferable real-time as the background model of present frame, and its background is not accumulated, and renewal speed is fast, algorithm is simple,
Amount of calculation is small；
Step 102, the face part binding characteristic of the coloured image that colour imagery shot in the dual camera is obtained is carried
Take algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three-dimensional state
Organ characteristic's point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces；
Wherein, referring to Fig. 2, the feature of living body faces is recognized by face domestic animal algorithm, including：Colour in dual camera is taken the photograph
As the coloured image that head is obtained, it is changed into black white image, the near-infrared image obtained with black and white camera carries out Optical flow estimation
Model, otherness distribution estimation, judges whether image is living body faces, if so, then being carried by data obtaining module binding characteristic
Take algorithm and extract organ characteristic point, if it is not, then terminating.
In step 103, the process for extracting organ characteristic point by data obtaining module binding characteristic extraction algorithm includes straight
See two steps of Extraction of Geometrical Features and shape and texture feature extraction：
Intuitive geometry feature extraction：Extract the representative position of face face, such as eyebrow, eyes, nose, face and people
Face contour, intuitive geometry figure is extracted using its each genius loci relation；
Shape and texture feature extraction：Shape Feature Extraction is to extract facial image edge, profile or some key points
Coordinate vector, is a kind of binary feature, with very strong resisting illumination variation ability；And texture feature extraction is facial image pixel
Gray value, be the useful supplement of shape facility, ASMs/AAMs models are counted using shape and textural characteristics by PCA
Modeling.
Referring to Fig. 3, the living body faces dual camera identifying device that the present embodiment is also provided, the identifying device includes：
Image collection module 301, is a black and white camera and a colour imagery shot, black and white camera for the dual camera
The black white image that one is generation under natural light or white light conditions is obtained, and obtains another to be produced under the conditions of near infrared light
Near-infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Described image acquisition module 301 is additionally operable to carry out background calculus of differences, background calculus of differences to two facial images
Including：Using black white image as the still image of current environment, using near-infrared image as active light source image, by still image
Calculus of differences is carried out with active light source image, difference image is obtained, is drawn according to active light source feature and complete is gone background actively
Light source image.
Data obtaining module 302, for the black white image and near-infrared that obtain black and white camera in the dual camera
Face part binding characteristic extraction algorithm in image extracts two-dimensional state organ characteristic's point；
Judge module 303, the face part knot of the coloured image for colour imagery shot in the dual camera to be obtained
Close feature extraction algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three
Organ characteristic's point of dimension state, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is live body people
Face；
The judge module 303 is additionally operable to the coloured image for obtaining colour imagery shot in dual camera, is changed into black and white
Image, the near-infrared image obtained with black and white camera carries out Optical flow estimation model, and otherness distribution estimation judges that image is
No is living body faces, if so, organ characteristic's point of two-dimensional state is then extracted by data obtaining module binding characteristic extraction algorithm,
If it is not, then terminating.
The living body faces dual camera identifying device that this implementation is also provided, including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
By the dual camera for configuring, the dual camera is a black and white camera and a colour imagery shot, black and white camera
The black white image that one is generation under natural light or white light conditions is obtained, and obtains another to be produced under the conditions of near infrared light
Near-infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
The black white image that black and white camera in the dual camera is obtained is combined with the face part in near-infrared image
Feature extraction algorithm extracts two-dimensional state organ characteristic's point；
The face part binding characteristic extraction algorithm of the coloured image that colour imagery shot in the dual camera is obtained is carried
Take organ characteristic's point and extract organ characteristic point with the near-infrared image that black and white camera is obtained, form the organ characteristic of three-dimensional state
Point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
The dual camera is each configured with infrared LED and white light LEDs,
Referring to Fig. 5, Sensor1 is near infrared spectrum image, and light source is provided by infrared LED, can be 850nm or
940nm light sources, image is obtained by black and white camera；
Sensor 2 is coloured image, and light source is provided by white light LEDs, and image is obtained by colour imagery shot；
Control LED driver so as to control Push And Release or the flicker of LED by DSP, be that Sensor1 and Sensor2 are provided most
Good light source state, can add the equal tabula rasa of light source, uniform diffusing light above LED.
Dsp controls can distinguish attribute (resolution ratio, frame per second, time for exposure, gain, brightness etc. of Sensor1 and Sensor2
Deng)；
According to external environment condition light source, the electric current of DSP precise controls LED and LED times and Sensor1 and Sensor2 are lighted
Attribute it is obtained optimal image.
Referring to Fig. 4, in this embodiment, active infrared light source and day that living body faces dual camera identifying device is collected
Energy contrast in light or white LED lamp between respective wavelength, is the core for influenceing living body faces dual camera identifying device performance
Heart factor, whole living body faces dual camera identifying device is designed with lifting this ratio basis.
Within the possible range, should try one's best and increase energy output and the utilization ratio of infrared light supply, while reducing daylight as far as possible
Energy interference；
The LED infrared light supplies that selection spectral energy is concentrated, make its work in the state of big power output and shorter pulse times
Make, with improving energy efficiency ratio；
Designed by suitable even light and the IR of LED is concentrated and image acquisition region is uniformly illuminated, subtracted as far as possible
Few light ray energy loss；
Camera lens uses bandpass filter, is entered with intercepting the spectral energy outside LED output areas, is done with reducing daylight
Disturb；
Imageing sensor is as far as possible sensitive for preferential with response in LED wave-length coverages, to be kept for the as far as possible short time for exposure, enters
And working pulse time of LED is reduced, reduce system energy consumption；
In terms of living body faces dual camera identifying device composition, including white light and infrared two IMAQ passages, in vain
Light provides image preview, keeps image smooth with frame per second high；Infrared image acquisition based on meeting algorithm requirement, try one's best by selection
Few frame per second output, to reduce overall energy consumption, reduces heating.
Embodiment 2, referring to the living body faces dual camera recognition methods that Fig. 6, the present embodiment are provided, the recognition methods bag
Include：
Step 101, by the dual camera for configuring, the dual camera be a black and white camera and a colour imagery shot, it is black
White camera obtains the black white image that is generation under natural light or white light conditions, and it is near-infrared striation to obtain another
The near-infrared image produced under part；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Wherein, background calculus of differences is carried out to black white image and near-infrared image, background calculus of differences includes：With artwork master
As the still image as current environment, using near-infrared image as active light source image, by still image and active light source figure
As carrying out calculus of differences, obtain difference image, according to active light source feature draw it is complete remove background active light source image, i.e., only
The image of surplus face part；
Step 102, the face in the black white image that black and white camera in the dual camera is obtained and near-infrared image
Part binding characteristic extraction algorithm extracts two-dimensional state organ characteristic's point；
Step 103, the face part binding characteristic of the coloured image that colour imagery shot in the dual camera is obtained is carried
Take algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three-dimensional state
Organ characteristic's point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces；
Wherein, the feature of living body faces is recognized by face domestic animal algorithm, including：Colour imagery shot in dual camera is obtained
Coloured image, be changed into black white image, with black and white camera obtain near-infrared image carry out Optical flow estimation model, difference
Property distribution estimation, judge whether image is living body faces, if so, then by data obtaining module binding characteristic extraction algorithm extraction
Organ characteristic's point, if it is not, then terminating.
Step 104, when being judged to living body faces, a complete people is obtained from the two-dimensional state organ characteristic point for having extracted
Face feature image and data value, are matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
Step 105, i.e. output display or control after the completion of matching.
Wherein, in step 103, the process bag of organ characteristic point is extracted by data obtaining module binding characteristic extraction algorithm
Include intuitive geometry feature extraction and shape and two steps of texture feature extraction：
Intuitive geometry feature extraction：Extract the representative position of face face, such as eyebrow, eyes, nose, face and people
Face contour, intuitive geometry figure is extracted using its each genius loci relation；
Shape and texture feature extraction：Shape Feature Extraction is to extract facial image edge, profile or some key points
Coordinate vector, is a kind of binary feature, with very strong resisting illumination variation ability；And texture feature extraction is facial image pixel
Gray value, be the useful supplement of shape facility, ASMs/AAMs models are counted using shape and textural characteristics by PCA
Modeling.
Referring to Fig. 7, Fig. 7 is living body faces dual camera recognition methods FB(flow block), and upper computer end (can be arbitrary behaviour
Make system such as Windows, MAC OS, iOS, Android, Linux etc.) obtain near by USB2.0 dual cameras identifying device
Infrared image and coloured image, first image is pre-processed (as sharpened, again binaryzation, go background etc.), feature point extraction will
The feature extraction of face part out, goes to recognize the feature of living body faces and false proof (such as human face photo, figure and features through face domestic animal algorithm
Feature is close etc.), accurate complete a face characteristic picture and data value are obtained, go contrast in conjunction with Feature Correspondence Algorithm
Property data base, it in this map office can also be high in the clouds picture library data that property data base can be, it is also possible to newly-built current people
Face data.Be after the completion of matching output display and control (can be any action or equipment, such as unlocking, work attendance, identification,
Visitor's record etc.).
Referring to Fig. 8, the living body faces dual camera identifying device that the present embodiment is also provided, the identifying device includes：
Image collection module 301, is a black and white camera and a colour imagery shot, black and white camera for the dual camera
The black white image that one is generation under natural light or white light conditions is obtained, and obtains another to be produced under the conditions of near infrared light
Near-infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Described image acquisition module 301 is additionally operable to carry out background calculus of differences, background calculus of differences to two facial images
Including：Using black white image as the still image of current environment, using near-infrared image as active light source image, by still image
Calculus of differences is carried out with active light source image, difference image is obtained, is drawn according to active light source feature and complete is gone background actively
Light source image.
Data obtaining module 302, for the black white image and near-infrared that obtain black and white camera in the dual camera
Face part binding characteristic extraction algorithm in image extracts two-dimensional state organ characteristic's point；
Judge module 303, the face part knot of the coloured image for colour imagery shot in the dual camera to be obtained
Close feature extraction algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three
Organ characteristic's point of dimension state, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is live body people
Face；
The judge module 303 is additionally operable to the coloured image for obtaining colour imagery shot in dual camera, is changed into black and white
Image, the near-infrared image obtained with black and white camera carries out Optical flow estimation model, and otherness distribution estimation judges that image is
No is living body faces, if so, organ characteristic's point of two-dimensional state is then extracted by data obtaining module binding characteristic extraction algorithm,
If it is not, then terminating；
Matching module 304, for being judged to during living body faces, obtains an accurate complete face characteristic picture and data
Value, is matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
Display or control module 305, for i.e. output display or control after the completion of matching.
The present invention is not limited to above-mentioned implementation method, using or approximation method identical with the above embodiment of the present invention or dress
Put, and other living body faces dual camera recognition methods for obtaining and identifying device, within protection scope of the present invention.
Claims (10)
1. a kind of living body faces dual camera recognition methods, it is characterised in that the recognition methods includes：
By the dual camera for configuring, the dual camera is a black and white camera and a colour imagery shot, and black and white camera is obtained
One is the black white image produced under natural light or white light conditions, and it is produce under the conditions of near infrared light near to obtain another
Infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Face part binding characteristic in black white image and near-infrared image that black and white camera in the dual camera is obtained
Extraction algorithm extracts two-dimensional state organ characteristic's point；
The face part binding characteristic extraction algorithm extractor of the coloured image that colour imagery shot in the dual camera is obtained
Official's characteristic point extracts organ characteristic point with the near-infrared image that black and white camera is obtained, and forms organ characteristic's point of three-dimensional state,
The feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
2. recognition methods according to claim 1, it is characterised in that the recognition methods also includes：
When being judged to living body faces, from the two-dimensional state organ characteristic point for having extracted obtain a complete face characteristic picture and
Data value, is matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
I.e. output display or control after the completion of matching.
3. recognition methods according to claim 1, it is characterised in that obtained by the black and white camera of the dual camera for configuring
Black white image and near-infrared image are taken, including：
Calculus of differences is carried out to black white image and near-infrared image, calculus of differences includes：
Using black white image as the still image of current environment, using near-infrared image as active light source image, by still image
Calculus of differences is carried out with active light source image, difference image is obtained, is drawn according to active light source feature and complete is gone background actively
Light source image.
4. recognition methods according to claim 2, it is characterised in that the spy of living body faces is recognized by face domestic animal algorithm
Levy, including：
The near-infrared image that the coloured image that colour imagery shot in dual camera is obtained is obtained with black and white camera carries out light stream
Model is estimated in field, and otherness distribution estimation judges whether image is living body faces, if so, then being combined by data obtaining module
Feature extraction algorithm extracts organ characteristic's point of two-dimensional state, if it is not, then terminating.
5. a kind of living body faces dual camera identifying device, it is characterised in that the identifying device includes：
Image collection module, for the dual camera by configuring, the dual camera is a black and white camera and a colored shooting
Head, black and white camera obtains the black white image that is generation under natural light or white light conditions, and it is near red to obtain another
The near-infrared image produced under outer optical condition；Colour imagery shot obtains the cromogram that is generation under natural light or white light conditions
Picture；
Data obtaining module, for by the black white image and near-infrared image of the acquisition of black and white camera in the dual camera
Face part binding characteristic extraction algorithm extracts two-dimensional state organ characteristic's point；
Judge module, the face part binding characteristic of the coloured image for colour imagery shot in the dual camera to be obtained is carried
Take algorithm and extract the near-infrared image extraction organ characteristic point that organ characteristic point is obtained with black and white camera, form three-dimensional state
Organ characteristic's point, the feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
6. identifying device according to claim 5, it is characterised in that the identifying device also includes：
Matching module, for being judged to during living body faces, from the two-dimensional state organ characteristic point for having extracted obtain one it is complete
Face characteristic picture and data value, are matched in conjunction with Feature Correspondence Algorithm with contrast characteristic's database；
Display or control module, for i.e. output display or control after the completion of matching.
7. identifying device according to claim 5, it is characterised in that described image acquisition module is additionally operable to by configuring
The black and white camera of dual camera obtains black white image and near-infrared image, including：
Calculus of differences is carried out to black white image and near-infrared image, calculus of differences includes：
Using black white image as the still image of current environment, using near-infrared image as active light source image, by still image
Calculus of differences is carried out with active light source image, difference image is obtained, is drawn according to active light source feature and complete is gone background actively
Light source image.
8. identifying device according to claim 6, it is characterised in that the judge module is additionally operable to prize dual camera
The near-infrared image that the coloured image that color camera is obtained is obtained with black and white camera carries out Optical flow estimation model, otherness point
Cloth estimation, judges whether image is living body faces, if so, then extracting two dimension by data obtaining module binding characteristic extraction algorithm
Organ characteristic's point of state, if it is not, then terminating.
9. a kind of living body faces dual camera identifying device, it is characterised in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
By the dual camera for configuring, the dual camera is a black and white camera and a colour imagery shot, and black and white camera is obtained
One is the black white image produced under natural light or white light conditions, and it is produce under the conditions of near infrared light near to obtain another
Infrared image；Colour imagery shot obtains the coloured image that is generation under natural light or white light conditions；
Face part binding characteristic in black white image and near-infrared image that black and white camera in the dual camera is obtained
Extraction algorithm extracts two-dimensional state organ characteristic's point；
The face part binding characteristic extraction algorithm extractor of the coloured image that colour imagery shot in the dual camera is obtained
Official's characteristic point extracts organ characteristic point with the near-infrared image that black and white camera is obtained, and forms organ characteristic's point of three-dimensional state,
The feature of living body faces is recognized by face domestic animal algorithm, judges whether facial image is living body faces.
10. identifying device according to claim 9, it is characterised in that the dual camera be each configured with infrared LED with
White light LEDs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710081460.5A CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710081460.5A CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106874871A true CN106874871A (en) | 2017-06-20 |
| CN106874871B CN106874871B (en) | 2020-06-05 |
Family
ID=59167378
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710081460.5A Expired - Fee Related CN106874871B (en) | 2017-02-15 | 2017-02-15 | Living body face double-camera identification method and identification device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106874871B (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107633198A (en) * | 2017-07-25 | 2018-01-26 | 百度在线网络技术（北京）有限公司 | Biopsy method, device, equipment and storage medium |
| CN107657248A (en) * | 2017-10-26 | 2018-02-02 | 广州云从信息科技有限公司 | A kind of infrared binocular In vivo detections of Android based on recognition of face certification |
| CN107688775A (en) * | 2017-07-11 | 2018-02-13 | 浙江新再灵科技股份有限公司 | Binocular camera recognition of face right discriminating system and method based on elevator scene |
| CN108197586A (en) * | 2017-12-12 | 2018-06-22 | 北京深醒科技有限公司 | Recognition algorithms and device |
| CN108280418A (en) * | 2017-12-12 | 2018-07-13 | 北京深醒科技有限公司 | The deception recognition methods of face image and device |
| CN108304789A (en) * | 2017-12-12 | 2018-07-20 | 北京深醒科技有限公司 | Recognition algorithms and device |
| CN108344738A (en) * | 2018-01-22 | 2018-07-31 | 翰飞骏德（北京）医疗科技有限公司 | Imaging method and its device for hydroxyapatite |
| CN108388889A (en) * | 2018-03-23 | 2018-08-10 | 百度在线网络技术（北京）有限公司 | Method and apparatus for analyzing facial image |
| CN108737728A (en) * | 2018-05-03 | 2018-11-02 | Oppo广东移动通信有限公司 | A kind of image capturing method, terminal and computer storage media |
| CN108830229A (en) * | 2018-06-20 | 2018-11-16 | 哈尔滨理工大学 | The vivo identification method of Face datection is combined under a kind of frame based on caffe |
| CN109426762A (en) * | 2017-08-22 | 2019-03-05 | 上海荆虹电子科技有限公司 | A kind of biological recognition system, method and bio-identification terminal |
| CN110008878A (en) * | 2019-03-27 | 2019-07-12 | 中控智慧科技股份有限公司 | A kind of anti-false method of Face datection and the face identification device for having anti-false function |
| CN110096861A (en) * | 2019-04-12 | 2019-08-06 | 檀鹏程 | A kind of bi-directional distributed formula authentication system based on biological characteristic |
| WO2019218943A1 (en) * | 2018-05-15 | 2019-11-21 | Oppo广东移动通信有限公司 | Front dual camera-based security verification method and electronic device |
| CN110555931A (en) * | 2019-08-31 | 2019-12-10 | 华南理工大学 | Face detection and gate inhibition system device based on deep learning recognition |
| CN110895678A (en) * | 2018-09-12 | 2020-03-20 | 耐能智慧股份有限公司 | Face recognition module and method |
| CN111325139A (en) * | 2020-02-18 | 2020-06-23 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| CN111582118A (en) * | 2020-04-29 | 2020-08-25 | 福州瑞芯微电子股份有限公司 | Face recognition method and device |
| US10771689B2 (en) | 2018-04-28 | 2020-09-08 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Image processing method and device, computer-readable storage medium and electronic device |
| TWI734454B (en) * | 2020-04-28 | 2021-07-21 | 鴻海精密工業股份有限公司 | Identity recognition device and identity recognition method |
| CN113822222A (en) * | 2021-10-11 | 2021-12-21 | 中国平安人寿保险股份有限公司 | Human face anti-cheating method and device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102221937A (en) * | 2010-04-15 | 2011-10-19 | 上海天派无线科技有限公司 | Real-time video image coordinate recognition system and method |
| CN104147778A (en) * | 2014-08-20 | 2014-11-19 | 赵东方 | Desktop projection interactive game machine host |
| CN104318237A (en) * | 2014-10-28 | 2015-01-28 | 厦门大学 | Fatigue driving warning method based on face identification |
| CN105023005A (en) * | 2015-08-05 | 2015-11-04 | 王丽婷 | Face recognition apparatus and recognition method thereof |
-
2017
- 2017-02-15 CN CN201710081460.5A patent/CN106874871B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102221937A (en) * | 2010-04-15 | 2011-10-19 | 上海天派无线科技有限公司 | Real-time video image coordinate recognition system and method |
| CN104147778A (en) * | 2014-08-20 | 2014-11-19 | 赵东方 | Desktop projection interactive game machine host |
| CN104318237A (en) * | 2014-10-28 | 2015-01-28 | 厦门大学 | Fatigue driving warning method based on face identification |
| CN105023005A (en) * | 2015-08-05 | 2015-11-04 | 王丽婷 | Face recognition apparatus and recognition method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 张铎: "《生物识别技术基础》", 30 April 2009, 武汉大学出版社 * |
| 杨铁军: "《产业专利分析报告 第33册 智能识别》", 30 June 2015, 中国矿业大学出版社 * |
| 赵小川: "《MATLAB图像处理 程序实现与模块化仿真》", 31 January 2014, 北京航空航天大学出版社 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688775A (en) * | 2017-07-11 | 2018-02-13 | 浙江新再灵科技股份有限公司 | Binocular camera recognition of face right discriminating system and method based on elevator scene |
| CN107633198A (en) * | 2017-07-25 | 2018-01-26 | 百度在线网络技术（北京）有限公司 | Biopsy method, device, equipment and storage medium |
| CN109426762A (en) * | 2017-08-22 | 2019-03-05 | 上海荆虹电子科技有限公司 | A kind of biological recognition system, method and bio-identification terminal |
| CN107657248A (en) * | 2017-10-26 | 2018-02-02 | 广州云从信息科技有限公司 | A kind of infrared binocular In vivo detections of Android based on recognition of face certification |
| CN108280418A (en) * | 2017-12-12 | 2018-07-13 | 北京深醒科技有限公司 | The deception recognition methods of face image and device |
| CN108197586B (en) * | 2017-12-12 | 2020-04-21 | 北京深醒科技有限公司 | Face recognition method and device |
| CN108304789A (en) * | 2017-12-12 | 2018-07-20 | 北京深醒科技有限公司 | Recognition algorithms and device |
| CN108197586A (en) * | 2017-12-12 | 2018-06-22 | 北京深醒科技有限公司 | Recognition algorithms and device |
| CN108344738A (en) * | 2018-01-22 | 2018-07-31 | 翰飞骏德（北京）医疗科技有限公司 | Imaging method and its device for hydroxyapatite |
| CN108388889A (en) * | 2018-03-23 | 2018-08-10 | 百度在线网络技术（北京）有限公司 | Method and apparatus for analyzing facial image |
| US10771689B2 (en) | 2018-04-28 | 2020-09-08 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Image processing method and device, computer-readable storage medium and electronic device |
| CN108737728A (en) * | 2018-05-03 | 2018-11-02 | Oppo广东移动通信有限公司 | A kind of image capturing method, terminal and computer storage media |
| CN108737728B (en) * | 2018-05-03 | 2021-06-11 | Oppo广东移动通信有限公司 | Image shooting method, terminal and computer storage medium |
| WO2019218943A1 (en) * | 2018-05-15 | 2019-11-21 | Oppo广东移动通信有限公司 | Front dual camera-based security verification method and electronic device |
| CN108830229A (en) * | 2018-06-20 | 2018-11-16 | 哈尔滨理工大学 | The vivo identification method of Face datection is combined under a kind of frame based on caffe |
| CN110895678A (en) * | 2018-09-12 | 2020-03-20 | 耐能智慧股份有限公司 | Face recognition module and method |
| CN110008878A (en) * | 2019-03-27 | 2019-07-12 | 中控智慧科技股份有限公司 | A kind of anti-false method of Face datection and the face identification device for having anti-false function |
| CN110008878B (en) * | 2019-03-27 | 2021-07-30 | 熵基科技股份有限公司 | Anti-fake method for face detection and face recognition device with anti-fake function |
| CN110096861A (en) * | 2019-04-12 | 2019-08-06 | 檀鹏程 | A kind of bi-directional distributed formula authentication system based on biological characteristic |
| CN110555931A (en) * | 2019-08-31 | 2019-12-10 | 华南理工大学 | Face detection and gate inhibition system device based on deep learning recognition |
| CN111325139A (en) * | 2020-02-18 | 2020-06-23 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| CN111325139B (en) * | 2020-02-18 | 2023-08-04 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| TWI734454B (en) * | 2020-04-28 | 2021-07-21 | 鴻海精密工業股份有限公司 | Identity recognition device and identity recognition method |
| CN111582118A (en) * | 2020-04-29 | 2020-08-25 | 福州瑞芯微电子股份有限公司 | Face recognition method and device |
| CN113822222A (en) * | 2021-10-11 | 2021-12-21 | 中国平安人寿保险股份有限公司 | Human face anti-cheating method and device, computer equipment and storage medium |
| CN113822222B (en) * | 2021-10-11 | 2024-09-06 | 中国平安人寿保险股份有限公司 | Face anti-cheating method, device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106874871B (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106874871A (en) | A kind of recognition methods of living body faces dual camera and identifying device | |
| CN108229362B (en) | Binocular face recognition living body detection method based on access control system | |
| CN109902604B (en) | High-safety face comparison system and method based on Feiteng platform | |
| CN105874472B (en) | Multiband bio-identification camera system with iris color identification | |
| US11227368B2 (en) | Method and device for controlling an electronic device based on determining a portrait region using a face region detection and depth information of the face region detected | |
| KR101286454B1 (en) | Fake face identification apparatus and method using characteristic of eye image | |
| CN102622588B (en) | Dual-certification face anti-counterfeit method and device | |
| Zou et al. | Illumination invariant face recognition: A survey | |
| CN107832677A (en) | Face identification method and system based on In vivo detection | |
| CN101493884B (en) | Multi-optical spectrum image collecting device and method | |
| CN106778518A (en) | A kind of human face in-vivo detection method and device | |
| CN104794458A (en) | Fuzzy video person identifying method | |
| CN207491128U (en) | A kind of RGB+IR image capture devices | |
| CN109086754A (en) | A kind of human posture recognition method based on deep learning | |
| CN105975938A (en) | Smart community manager service system with dynamic face identification function | |
| US10893594B2 (en) | Method of identifying light sources and a corresponding system and product | |
| CN107239772A (en) | Palm print and palm vein image collecting device and clearance gate | |
| CN110458041A (en) | A kind of face identification method and system based on RGB-D camera | |
| Wang et al. | A new multispectral method for face liveness detection | |
| Ghiass et al. | Vesselness features and the inverse compositional AAM for robust face recognition using thermal IR | |
| WO2021217764A1 (en) | Human face liveness detection method based on polarization imaging | |
| CN106650370A (en) | Non-contact encryption method and system for computer | |
| CN110008878A (en) | A kind of anti-false method of Face datection and the face identification device for having anti-false function | |
| CN206849035U (en) | A kind of image processing apparatus and face identification system | |
| CN110032932A (en) | A kind of human posture recognition method based on video processing and decision tree given threshold |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200703 Address after: 230000 west side of Xianghe North Road, Feidong Economic Development Zone, Feidong County, Hefei City, Anhui Province Patentee after: ANHUI GUANGZHEN PHOTOELECTRIC TECHNOLOGY Co.,Ltd. Address before: 523000 Guangdong Province, Dongguan City Industrial Zone Qingxi Town Silver Star Patentee before: GUANGDONG LITE ARRAY Co.,Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200605 |