CN109325462A - Recognition of face biopsy method and device based on iris - Google Patents
Recognition of face biopsy method and device based on iris Download PDFInfo
- Publication number
- CN109325462A CN109325462A CN201811183736.1A CN201811183736A CN109325462A CN 109325462 A CN109325462 A CN 109325462A CN 201811183736 A CN201811183736 A CN 201811183736A CN 109325462 A CN109325462 A CN 109325462A
- Authority
- CN
- China
- Prior art keywords
- face
- iris
- image
- recognition
- characteristic value
- 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
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/19—Sensors therefor
-
- 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
-
- 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
- G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Ophthalmology & Optometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention provides a kind of recognition of face biopsy method and device based on iris, and this method includes generating random motion track；Control servo central moves along random motion track, shoots iris image and facial image；The human eye of acquisition is jumped into apparent path and random motion track compares and carries out iris test, face lookup is carried out to video image using deep learning algorithm, it is split if face exists, it carries out characteristic point and the characteristic value extracted is compared with the previous face characteristic value for being stored in local face database, if comparing the image successfully illustrated in local face database there are testee, unsuccessfully there is no the images of testee for explanation for comparison, the accuracy of present invention guarantee recognition of face In vivo detection, it does not need testee and does excessive movement, promote the experience of testee, and the algorithm speed used is fast, accuracy is high, the resource occupied simultaneously is small, it can be run on desktop end or ARM, enrich exercisable practical scene.
Description
Technical field
The present invention relates to identity identification technical field more particularly to a kind of recognition of face biopsy methods based on iris
And device.
Background technique
With the improvement of people's life quality, the Internet of Things networking of life house, office high building modernization, vehicles day is increasingly
Up to diversification etc., these undoubtedly all at every moment affect the various aspects of people, so that people are small to social security is arrived greatly
Safety to house address is more worried；The people of the public places such as government department and airport, high-speed rail station, subway station, customs
Flow control, security administration, potential crime analysis etc. demand are growing；Large-scale public place such as stadium, football pitch or
The monitoring to the stream of people and identity such as person financial center CBD, requires to use human face detection and recognition technology.
With biological identification technology, the development of mode identification technology, artificial intelligence technology, the technology of field of face identification exists
Tend to be mature from the point of view of currently available technology angle, the people inside face algorithm such as Opencv is detected in traditional pattern-recognition
Face Haar feature classifiers and dlib Face datection；Such as LBP recognition of face scheduling algorithm can be very for face feature extraction method
Good carry out human face detection and recognition, but the requirement such as system that above method is unable to satisfy current various complex environments is small-sized
Change, operation high efficiency, recognition accuracy height and anti-fraud etc. composite factor.
Existing recognition of face In vivo detection mainly uses following three kinds of schemes: it is complete that corresponding actions are made in 1. users cooperation
At In vivo detection, user needs to cooperate a series of actions of shaking the head, blink, open one's mouth etc., causes user experience low；
2. being analyzed and determined using the video image that algorithm acquires single visible image capturing head, due to the video figure of camera acquisition
Whether piece is two-dimensional space, so being living body faces regardless of being all difficult to distinguish using what algorithm in face of camera, cause
Recognition accuracy is low；3. 3D modeling is carried out to measured using depth of field camera, by 3D model to determine whether being living body faces,
But the depth of field camera cost that this method uses is high, and 3D modeling algorithm needs to expend a large amount of computer computation abilities, fortune
It is extremely low to calculate speed.
Summary of the invention
The purpose of the present invention is to provide a kind of recognition of face biopsy method and device based on iris, it is existing to solve
The problem that user experience is low and recognition accuracy is low when having in technology to measured's progress recognition of face.
The invention is realized in this way first aspect present invention provides a kind of recognition of face In vivo detection side based on iris
Method, the recognition of face biopsy method based on iris include:
Generate random motion track；
Control servo central moves along the random motion track, and shoots iris image and facial image；
It is tested the motion profile of pupil center's point of target according to the iris image acquisition, and judges in the pupil
Whether the motion profile of heart point and the registration of the random motion track are less than preset value；
When the registration is less than preset value, determine that the tested target is non-living body face；
When the registration is not less than preset value, then judge the facial image with the presence or absence of face；
When face is not present in the facial image, determine that the tested target is non-living body face；
When the facial image is there are when face, the facial image is split, from the facial image divided
Feature point extraction characteristic value, and extracted characteristic value and the face characteristic value in face database are compared；
When comparing successfully, determine that there are the facial images in the face database；
When comparing failure, determine that there is no the facial images in the face database.
Second aspect of the present invention provides a kind of recognition of face living body detection device based on iris, the people based on iris
Face identification living body detection device includes: infrared camera, visible image capturing head, servo-actuated target generator, motion profile acquisition mould
Block, living body iris distinguished number module, Face datection algoritic module and human face data library module；
The servo-actuated target generator controls servo central along the random motion track for generating random motion track
Movement；
The infrared camera is for shooting iris image；
The visible image capturing head is for shooting facial image；
The motion profile obtains pupil center point of the module for being tested target according to the iris image acquisition
Motion profile；
The living body iris distinguished number module be used for judge pupil center's point motion profile and the random fortune
Whether the registration of dynamic rail mark is less than preset value, when the registration is less than preset value, determines that the tested target is non-
Living body faces；
The Face datection algoritic module is used to determine that the registration is not small when the living body iris distinguished number module
When preset value, then the facial image is judged with the presence or absence of face, when face is not present in the facial image, described in judgement
Tested target is non-living body face；
The face recognition algorithms module is used for when there are people for the Face datection algoritic module judgement facial image
When face, the facial image is split, from the feature point extraction characteristic value in the facial image divided, and will be extracted
Characteristic value and face database in face characteristic value compare；
The human face data library module is used for when the face recognition algorithms module compares successfully, determines the face number
According to there are the facial image, when the face recognition algorithms module compares failure, determining in the face database in library
There is no the facial images.
The present invention provides a kind of recognition of face biopsy method and device based on iris, and this method includes generating at random
Motion profile；Control servo central moves along random motion track, shoots iris image and facial image；The human eye of acquisition is jumped
Apparent path and random motion track, which compare, carries out iris test, carries out face lookup to video image using deep learning algorithm,
It is split if face exists, carries out characteristic point and by the characteristic value extracted and is previously stored in local face database
Face characteristic value is compared, if comparing the image successfully illustrated in local face database there are testee, if right
Than unsuccessfully illustrating the image there is no testee, technical solution of the present invention may be implemented to guarantee the standard of recognition of face In vivo detection
True property, while not needing testee and doing excessive movement, the algorithm speed for improving the experience of user testee, and using
Degree is fast, and accuracy is high, while the resource occupied is small, can run on desktop end or ARM, enrich exercisable practical field
Scape.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of process for recognition of face biopsy method based on iris that an embodiment of the present invention provides
Figure；
Fig. 2 is that a kind of execution for recognition of face biopsy method based on iris that an embodiment of the present invention provides is set
Standby structural schematic diagram；
Fig. 3 is random in a kind of recognition of face biopsy method based on iris that an embodiment of the present invention provides
Motion track schematic diagram；
Fig. 4 is the step in a kind of recognition of face biopsy method based on iris that an embodiment of the present invention provides
The flow chart of S30；
Fig. 5 is a kind of another stream for recognition of face biopsy method based on iris that an embodiment of the present invention provides
Cheng Tu；
Fig. 6 is that a kind of structure for recognition of face living body detection device based on iris that an embodiment of the present invention provides is shown
It is intended to.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
In order to illustrate technical solution of the present invention, the following is a description of specific embodiments.
The embodiment of the present invention provides a kind of recognition of face biopsy method based on iris, as shown in Figure 1, described be based on
The recognition of face biopsy method of iris includes:
Step S10. generates random motion track.
Step S20. control servo central moves along the random motion track, and shoots iris image and facial image.
In step S10 and step S20, as an implementation, as shown in Figures 2 and 3, the equipment for executing this method
Including video image display area 1, servo-actuated target generator 2, infrared camera 3, infrared light compensating lamp 4 and visible image capturing head
5, random motion track is generated by the way that servo-actuated target generator 2 is arranged, being equipped with one in servo-actuated target generator 2 can move at random
Dynamic servo central 12 is based on this biological characteristic since the movement of human eyeball has the characteristics that jump view, can be using servo-actuated
Target generator generates a random movement track 13 by servo central, while backstage records and stores the random motion track
Track 13 is servo-actuated target generator 2 in servo central 13 and generates pipper, moved according to random movement track 13,
Prompt tone, which is issued, before servo central 12 is mobile prompts testee, it is desirable that the eyeball of testee follows servo central 12 to move,
Iris image and facial image are shot by camera simultaneously, it is preferred that iris image can be shot by infrared camera, led to
Cross visible image capturing head shooting facial image.
Step S30. is tested the motion profile of pupil center's point of target according to the iris image acquisition, and judges institute
Whether the registration of the motion profile and the random motion track of stating pupil center's point is less than preset value, when the registration is small
When preset value, step S40 is executed, when the registration is not less than preset value, executes step S50.
In step s 30, specifically, as an implementation, infrared camera meeting while servo central is mobile
The iris video image that synchronous acquisition measured's eyeball jumps view will be regarded by handling collected iris video image
Center coordinate of eye pupil moving track calculation in frequency image, finally by servo central's motion track and center coordinate of eye pupil motion profile
Registration comparison is carried out, whether the similarity by comparison is more than preset value to determine whether being living body iris, for example, can incite somebody to action
Center coordinate of eye pupil motion profile be scaled to servo central's motion track same size, respectively to center coordinate of eye pupil move rail
Coordinate system is arranged in mark servo central motion track, obtains the coordinate of the point of predetermined number on center coordinate of eye pupil motion profile, and
Judge whether the coordinate is overlapped with the coordinate put on servo central's motion track, when the number ratio of the coordinate points of coincidence is greater than in advance
If value, such as when 80%, determine the iris image from living body iris.
As an implementation, as shown in figure 4, it is described tested according to the iris image acquisition in step S30
The motion profile of pupil center's point of target, comprising:
Step S301. handles each frame iris image, and obtains and be tested target in each frame iris image
The position of pupil center's point.
It is described that each frame iris image is handled in step S301, comprising:
Iris in each frame iris image is positioned, is normalized and image enhancement.
Specifically, the Iris Location include iris inward flange and outer peripheral positioning, inward flange, that is, pupil boundary,
The boundary of outer edge, that is, iris and sclera, iris is not overlapped with the center of pupil under normal circumstances, so the two edges are all
It needs to position, being normalized after positioning inward flange and outer edge using edge detection operator can fit with image enhancement
The radius of inside and outside circle and the center of circle.
The position of multiple pupil center's points is connected the motion profile for generating pupil center's point by step S302..
In step s 302, the center of circle of acquired every frame image is connected to the movement rail to form pupil center's point
Mark.
The motion profile for accurately obtaining pupil center's point may be implemented in present embodiment.
It should be noted that the technical program can also use the infrared camera of different focal length, single infrared camera
Infrared camera difference with two different focal lengths is that two collected iris images of infrared camera include more information,
Wherein, the depth of field of iris image is richer, and stereovision is also stronger.Two collected iris images of infrared photography are more conducive to
It is further to be handled, wherein equally take positioning, normalization and the image enhancement to iris, while can also to iris into
The feature extraction of row iris pattern carries out iris recognition and comparison to measured target, while face recognition algorithms being cooperated to carry out quilt
The dual identification of target is surveyed, the accuracy rate of a whole set of recognition of face biopsy method based on iris is improved.
As an implementation, IR wavelength used in infrared camera is between 700-900 nanometers, energy
It is minimum, it can only break through in the gap of atom and molecule, and it cannot be penetrated into the inside of atom, molecule, it is lower than international safety standard,
To eyes almost without injury, even if, again smaller than the radiation to human body in 1 minute is played, illumination is strong using 100,000 radiation
Degree tends to natural light intensity, and it is more than to be closed after 9 seconds automatically that infrared camera, which can also light the time in infrared LED indicator light,
It closes, to protect the eyes of user.
Need to illustrate when, it is seen that light video camera head module infrared camera acquire iris video image while to people
Face video image synchronizes acquisition, it is ensured that the iris video image and facial video image of acquisition come from the same person.
As an implementation, the color image that will acquire is converted into grayscale image, carries out denoising to image；It obtains
The subregion that gray value is minimum is taken, binary image is divided the image into according to lowest gray value, using all less than threshold value
Point (x k, y k), k=1 ... N, N are the numbers for the point that gray scale is less than lowest gray value in histogram, slightly fixed in binary image
Rectangular area centered on it is set as region of interest ROI according to slightly fixed pupil center by position pupil center location；
An initial threshold is set using maximum variance between clusters, then the ROI image after segmentation is assessed, if worked as from utilizing
Search for the hot spot less than the condition that meets in image after preceding Threshold segmentation, then adjust automatically threshold value, segmented image again, again into
Row search, directly searches out whole hot spots of the condition of satisfaction；Using centroid method, the centre coordinate of five hot spots is determined；Based on several
What relationship match hot spot and light source；Using corneal reflection spot area obtained in hot spot extraction process, by the point of spot area
The gray scale of collection is substituted for the optimal self-adaptive threshold value in hot spot extraction process, then using classical Threshold sementation to image into
Image is split according to the optimal self-adaptive threshold value in hot spot extraction process, obtains the general area of pupil by row binaryzation；
Extract the edge point set of pupil；Using ellipse fitting, the centre coordinate of pupil is obtained: using least square ellipse fitting process first,
Ellipse fitting is carried out to candidate point, removes the point of the hypertelorism in candidate marginal from elliptical center；Circulation is fitted, directly
To a stable ellipse center location is obtained, this stable ellipse center location is exactly the centre coordinate of pupil.
Step S40. determines that the tested target is non-living body face when the registration is less than preset value.
Step S50. then judges that the facial image whether there is face when the registration is not less than preset value, when
When face is not present in the facial image, step S40 is executed, when the facial image is there are when face, executes step S70.
Step S60. is split the facial image, from the people divided when the facial image is there are when face
Feature point extraction characteristic value in face image, and the face characteristic value in extracted characteristic value and face database is carried out pair
Than.
In step S60, the different database of face database stores the road of facial image characteristic value, facial image respectively
Diameter and facial image, and the path of facial image characteristic value, facial image and facial image correspond to each other, when being extracted
Characteristic value and face database in face characteristic value compare successfully after, face can be searched out according to the path of facial image
Image, it can illustrate that there are the facial images in face database.
Step S70. determines that there are the facial images in the face database when comparing successfully.
Step S80. determines that there is no the facial images in the face database when comparing failure.
Further, in step s 50, the judgement facial image whether there is face, comprising:
Judge the facial image with the presence or absence of face by MTCNN Face datection algorithm.
Specifically, being started after the trigger condition for receiving living body iris distinguished number using MTCNN Face datection algorithm
Face lookup is carried out to the video image of visible image capturing head module acquisition, if do not found in collected video image
Face then illustrates that target is not present, and terminates detection.Deep learning algorithm MTCNN Face datection algorithm is broadly divided into three steps: initially
Multi-scale transform first is carried out to image, obtains image pyramid, obtains Image Multiscale information, and then realize the inspection to face
It surveys, MTCNN Face datection algorithm has more robustness, Face datection to light, angle and human face expression variation in natural environment
Effect is more preferable than the conventional face of congenerous detection and other deep learning detection algorithms, while the low embedded core of resources occupation rate
Piece can carry, and miniaturization real-time face detection may be implemented.
Further, in step S60, the feature point extraction characteristic value from the facial image divided, comprising:
By LightenedCNN face recognition algorithms from the feature point extraction characteristic value in the facial image divided.
Specifically, after receiving the trigger condition of Face datection algorithm, using LightenedCNN face recognition algorithms pair
The every frame video image being partitioned into carries out human face characteristic point extraction, and the characteristic value extracted is stored in local face number with previous
It is compared according to the face characteristic value in library, system can write off value after comparing calculation lower than 80 points, i.e., local human face data
There is no the people in visible image capturing head acquisition video pictures in library, it is considered as success equal to and above 80 points, i.e., local face number
According in library, there are the people in visible image capturing head acquisition video pictures.
Wherein, carry out face verification with LightenedCNN and be divided into three kinds: one is the task training for using face classification
CNN extracts feature, then judges to be the same person with classifier.Second is direct optimization verifying loss.The third is
Recognition of face and validation task are carried out simultaneously, Lightened CNN face recognition algorithms are a kind of light-duty CNN, are being obtained
Simultaneously, network structure simplifies relatively good effect, and time and space are all optimized, and may operate in embedded device and shifting
In dynamic equipment.
The embodiment of the present invention is moved by control servo central along random motion track, using infrared camera and visible light
Camera system to carry out video image acquisition to iris and face respectively；It can be and then servo-actuated in target generator according to human eye
The human eye of infrared camera acquisition is jumped apparent path and carry out rainbow is compared with random motion track by the mobile feature of servo central
Film test, complete iris test after using deep learning algorithm to video image carry out face lookup, if face exist if into
The face being partitioned into is carried out the processing such as feature point extraction by row segmentation；By the local face of the characteristic value extracted and previously deposit
The face characteristic value of database is compared, and successfully illustrates there is visible image capturing at this time in local face database if compared
People in head acquisition video image, if comparison unsuccessfully illustrates that there is no visible image capturing heads to acquire video in local human face data
People in image, the embodiment of the present invention may be implemented to ensure that the accurate of recognition of face In vivo detection when applying to different scenes
Property.
Further, as another embodiment, iris image can also be identified, as shown in figure 5, step S10
Described in shooting iris image and facial image, later further include:
Step S101. carries out a point window to the iris image and handles, and carries out WAVELET PACKET DECOMPOSITION to each window and obtain son
Band image.
Step S102. carries out screening to the sub-band images and obtains the sub-band images that iris feature is concentrated.
Step S103. carries out singular value decomposition to the sub-band images that the iris feature is concentrated to obtain feature vector.
Step S104. compares described eigenvector and the iris feature vector in iris database.
Step S105. determines that there are the iris images in the iris database when comparing successfully.
Step S106. determines that there is no the iris images in the iris database when comparing failure.
Specifically, carrying out iris recognition and comparison to the iris of measured target in above-mentioned steps S101 into step S106
Iris image feature will be extracted, a point window first is carried out to iris image and is handled, and WAVELET PACKET DECOMPOSITION is carried out to each window,
Screening Treatment is done to the sub-band images of each window on this basis, singular value then is carried out to the sub-band images that iris feature is concentrated
It decomposes to extract feature, finally according to the property of singular value features, compression processing is done to the feature vector after decomposition, as final
Recognition feature vector, compare identification decision whether in iris database further according to feature vector, the present embodiment can be with
Iris image identification is carried out before facial image identification, further enhances the accuracy to measured target identification.
Further, as another embodiment, described that face in extracted characteristic value and face database is special
Value indicative compares, before further include:
Into the face database after additions and deletions facial image, modified human face data library file is loaded in memory；
It is described to compare extracted characteristic value and the face characteristic value in face database, further includes:
Extracted characteristic value and the face characteristic value in the face database in the memory are compared.
Specifically, human face data library module is local database, by specifically reading and writing interface to database
The face characteristic value of middle storage carries out additions and deletions, starts in equipment operation, and memory will load human face data library file, when face is known
When other algorithm carries out face characteristic value comparison, directly data loaded into memory can be compared, rather than human face data
Library file itself；When calling interface increases face into face database, can modify first to the data in memory, when
Equipment will be updated local database files when closing operation, while next secondary device run again will load newest local number
According to library file into memory.
Another kind embodiment of the invention provides a kind of recognition of face living body detection device based on iris, as shown in fig. 6, institute
The recognition of face living body detection device based on iris is stated to include: infrared camera 401, visible image capturing head 402, send out with moving-target
Raw device 403, motion profile obtain module 404, living body iris distinguished number module 405, Face datection algoritic module 406, face
Recognizer module 407 and human face data library module 408；
The servo-actuated target generator 406 controls servo central along the random motion for generating random motion track
Track movement；
The infrared camera 401 is for shooting iris image；
The visible image capturing head 402 is for shooting facial image；
The motion profile obtains pupil center's point that module 404 is used to be tested target according to the iris image acquisition
Motion profile；
The living body iris distinguished number module 405 is used to judge the motion profile of pupil center's point and described random
Whether the registration of motion profile is less than preset value, when the registration is less than preset value, determines that the tested target is
Non-living body face；
The Face datection algoritic module 406 is used to determine the registration not when the living body iris distinguished number module
When less than preset value, then judge that the facial image determines institute when face is not present in the facial image with the presence or absence of face
Stating tested target is non-living body face, when the Face datection algoritic module 406 determine the facial image there are when face,
The facial image is split；
The face recognition algorithms module 407 is used for the feature point extraction characteristic value from the facial image divided, and
Extracted characteristic value and the face characteristic value in face database are compared；
The human face data library module 408 is used for when the face recognition algorithms module 407 compares successfully, described in judgement
There are the facial images in face database determines the face number when the face recognition algorithms module compares failure
According in library, there is no the facial images.
The motion profile obtains module 404 and is specifically used for:
Each frame iris image is handled, and obtains the pupil center's point for being tested target in each frame iris image
Position；
The position of multiple pupil center's points is connected to the motion profile for generating pupil center's point.
The motion profile obtains module 404 and is also used to position the iris in each frame iris image, normalize
And image enhancement.
Technical solution of the present invention provides a kind of recognition of face biopsy method and device based on iris, can not only solve
Certainly current single camera carries out the video that recording cannot be identified when recognition of face or the picture of shooting carries out asking for deceptive practices
Topic, and in recognition speed faster than existing multi-angle multi-cam 3D modeling vivo identification scheme；What this system was used
Not only speed is fast for algorithm, accuracy is high, while its resource occupied is also very small, and can run in desktop end can also be in ARM
Upper operation, thus exercisable practical scene very abundant；While human eye is close to defined infrared camera pickup area
Visible image capturing head also can synchronous acquisition facial video image, when living body iris distinguished number determine eyeball exist jump apparent time people
Face detection algorithm just can to visible image capturing head acquire facial video image analyze, ensure that infrared camera and
The video image of visible image capturing head acquisition is from the same person, therefore carry out scheme of the present invention can apply to different fields
The accuracy of recognition of face In vivo detection is also ensured while scape.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of recognition of face biopsy method based on iris, which is characterized in that the recognition of face based on iris is living
Body detecting method includes:
Generate random motion track；
Control servo central moves along the random motion track, and shoots iris image and facial image；
It is tested the motion profile of pupil center's point of target according to the iris image acquisition, and judges pupil center's point
Motion profile and the registration of the random motion track whether be less than preset value；
When the registration is less than preset value, determine that the tested target is non-living body face；
When the registration is not less than preset value, then judge the facial image with the presence or absence of face；
When face is not present in the facial image, determine that the tested target is non-living body face；
When the facial image is there are when face, the facial image is split, the spy from the facial image divided
Sign point extracts characteristic value, and extracted characteristic value and the face characteristic value in face database are compared；
When comparing successfully, determine that there are the facial images in the face database；
When comparing failure, determine that there is no the facial images in the face database.
2. recognition of face biopsy method as described in claim 1, which is characterized in that described to be obtained according to the iris image
Take the motion profile of pupil center's point of tested target, comprising:
Each frame iris image is handled, and obtains the position for being tested pupil center's point of target in each frame iris image
It sets；
The position of multiple pupil center's points is connected to the motion profile for generating pupil center's point.
3. recognition of face biopsy method as claimed in claim 2, which is characterized in that it is described to each frame iris image into
Row processing, comprising:
Iris in each frame iris image is positioned, is normalized and image enhancement.
4. recognition of face biopsy method as described in claim 1, which is characterized in that the shooting iris image and face
Image, later further include:
A point window is carried out to the iris image to handle, and WAVELET PACKET DECOMPOSITION is carried out to each window and obtains sub-band images；
Screening is carried out to the sub-band images and obtains the sub-band images that iris feature is concentrated；
Singular value decomposition is carried out to obtain feature vector to the sub-band images that the iris feature is concentrated；
Described eigenvector and the iris feature vector in iris database are compared；
When comparing successfully, determine that there are the iris images in the iris database；
When comparing failure, determine that there is no the iris images in the iris database.
5. recognition of face biopsy method as described in claim 1, which is characterized in that described to judge that the facial image is
It is no that there are faces, comprising:
Judge the facial image with the presence or absence of face by MTCNN Face datection algorithm.
6. recognition of face biopsy method as described in claim 1, which is characterized in that described from the facial image divided
In feature point extraction characteristic value, comprising:
By LightenedCNN face recognition algorithms from the feature point extraction characteristic value in the facial image divided.
7. recognition of face biopsy method as described in claim 1, which is characterized in that it is described by extracted characteristic value with
Face characteristic value in face database compares, before further include:
Into the face database after additions and deletions facial image, modified human face data library file is loaded in memory；
It is described to compare extracted characteristic value and the face characteristic value in face database, further includes:
Extracted characteristic value and the face characteristic value in the face database in the memory are compared.
8. a kind of recognition of face living body detection device based on iris, which is characterized in that the recognition of face based on iris is living
Body detection device includes: infrared camera, visible image capturing head, servo-actuated target generator, motion profile acquisition module, living body rainbow
Film distinguished number module, Face datection algoritic module and human face data library module；
For generating random motion track, control servo central transports the servo-actuated target generator along the random motion track
It is dynamic；
The infrared camera is for shooting iris image；
The visible image capturing head is for shooting facial image；
The motion profile obtains the movement that module is used to be tested pupil center's point of target according to the iris image acquisition
Track；
The living body iris distinguished number module is used to judge the motion profile and the random motion rail of pupil center's point
Whether the registration of mark is less than preset value, when the registration is less than preset value, determines that the tested target is non-living body
Face；
The Face datection algoritic module is used to determine the registration not less than pre- when the living body iris distinguished number module
If when value, then judging the facial image with the presence or absence of face, when face is not present in the facial image, determining described tested
Examination target is non-living body face, when the Face datection algoritic module determines the facial image there are when face, to the people
Face image is split；
The face recognition algorithms module will be extracted for the feature point extraction characteristic value from the facial image divided
Characteristic value and face database in face characteristic value compare；
The human face data library module is used for when the face recognition algorithms module compares successfully, determines the face database
In there are the facial image, when the face recognition algorithms module compares failure, determine not deposit in the face database
In the facial image.
9. recognition of face living body detection device as claimed in claim 8, which is characterized in that the motion profile obtains module tool
Body is used for:
Each frame iris image is handled, and obtains the position for being tested pupil center's point of target in each frame iris image
It sets；
The position of multiple pupil center's points is connected to the motion profile for generating pupil center's point.
10. recognition of face living body detection device as claimed in claim 9, which is characterized in that the motion profile obtains module
It is also used to position the iris in each frame iris image, normalize and image enhancement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811183736.1A CN109325462B (en) | 2018-10-11 | 2018-10-11 | Face recognition living body detection method and device based on iris |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811183736.1A CN109325462B (en) | 2018-10-11 | 2018-10-11 | Face recognition living body detection method and device based on iris |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109325462A true CN109325462A (en) | 2019-02-12 |
| CN109325462B CN109325462B (en) | 2021-03-12 |
Family
ID=65261284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811183736.1A Active CN109325462B (en) | 2018-10-11 | 2018-10-11 | Face recognition living body detection method and device based on iris |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109325462B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532933A (en) * | 2019-08-26 | 2019-12-03 | 淮北师范大学 | A kind of living body faces detection head pose returns the acquisition methods and detection method of device |
| CN111402480A (en) * | 2020-02-29 | 2020-07-10 | 深圳壹账通智能科技有限公司 | Visitor information management method, device, system, equipment and storage medium |
| CN111797677A (en) * | 2020-05-13 | 2020-10-20 | 南京中科道置智能科技有限公司 | Face recognition living body detection method based on face iris recognition and thermal imaging technology |
| CN112164169A (en) * | 2020-09-21 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Access control management method and device, access control equipment and computer readable storage medium |
| CN112287909A (en) * | 2020-12-24 | 2021-01-29 | 四川新网银行股份有限公司 | Double-random in-vivo detection method for randomly generating detection points and interactive elements |
| CN113255400A (en) * | 2020-02-10 | 2021-08-13 | 深圳市光鉴科技有限公司 | Training and recognition method, system, equipment and medium of living body face recognition model |
| CN114005019A (en) * | 2021-10-29 | 2022-02-01 | 北京有竹居网络技术有限公司 | Method for identifying copied image and related equipment thereof |
| CN116994325A (en) * | 2023-07-27 | 2023-11-03 | 山东睿芯半导体科技有限公司 | Iris recognition method, chip and terminal |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6954553B2 (en) * | 2000-10-20 | 2005-10-11 | Fujitsu Limited | Personal authentication system using fingerprint information, registration-and-authentication method for the system, determination method for the system, and computer-readable recording medium |
| CN101483652A (en) * | 2009-01-10 | 2009-07-15 | 五邑大学 | Living creature characteristic recognition system |
| JP2010136223A (en) * | 2008-12-05 | 2010-06-17 | Sony Corp | Imaging device and imaging method |
| CN101833646A (en) * | 2009-03-11 | 2010-09-15 | 中国科学院自动化研究所 | Living iris detection method |
| US20110157347A1 (en) * | 2009-12-31 | 2011-06-30 | Peter Kalocsai | Unintrusive biometric capture device, system and method for logical access control |
| CN104992085A (en) * | 2015-06-13 | 2015-10-21 | 东莞市微模式软件有限公司 | Method and device for human body in-vivo detection based on touch trace tracking |
| CN105184277A (en) * | 2015-09-29 | 2015-12-23 | 杨晴虹 | Living body human face recognition method and device |
| CN105956536A (en) * | 2016-04-26 | 2016-09-21 | 北京森科赛德科技有限公司 | Pretreatment method and device for iris recognition |
| CN106203372A (en) * | 2016-07-19 | 2016-12-07 | 奇酷互联网络科技（深圳）有限公司 | Biopsy methods based on eyes, device and terminal unit |
| CN106339931A (en) * | 2015-12-02 | 2017-01-18 | 广西职业技术学院 | Multifunctional taxation service system |
| CN107451547A (en) * | 2017-07-17 | 2017-12-08 | 广东欧珀移动通信有限公司 | Identify the method and Related product of live body |
| CN107463883A (en) * | 2017-07-18 | 2017-12-12 | 广东欧珀移动通信有限公司 | Biometric discrimination method and Related product |
| CN107506721A (en) * | 2017-08-18 | 2017-12-22 | 上海爱优威软件开发有限公司 | A kind of more method for collecting iris and system |
| WO2018164364A1 (en) * | 2017-03-10 | 2018-09-13 | 크루셜텍 (주) | Contactless multiple body part recognition method and multiple body part recognition device, using multiple biometric data |
-
2018
- 2018-10-11 CN CN201811183736.1A patent/CN109325462B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6954553B2 (en) * | 2000-10-20 | 2005-10-11 | Fujitsu Limited | Personal authentication system using fingerprint information, registration-and-authentication method for the system, determination method for the system, and computer-readable recording medium |
| JP2010136223A (en) * | 2008-12-05 | 2010-06-17 | Sony Corp | Imaging device and imaging method |
| CN101483652A (en) * | 2009-01-10 | 2009-07-15 | 五邑大学 | Living creature characteristic recognition system |
| CN101833646A (en) * | 2009-03-11 | 2010-09-15 | 中国科学院自动化研究所 | Living iris detection method |
| US20110157347A1 (en) * | 2009-12-31 | 2011-06-30 | Peter Kalocsai | Unintrusive biometric capture device, system and method for logical access control |
| CN104992085A (en) * | 2015-06-13 | 2015-10-21 | 东莞市微模式软件有限公司 | Method and device for human body in-vivo detection based on touch trace tracking |
| CN105184277A (en) * | 2015-09-29 | 2015-12-23 | 杨晴虹 | Living body human face recognition method and device |
| CN106339931A (en) * | 2015-12-02 | 2017-01-18 | 广西职业技术学院 | Multifunctional taxation service system |
| CN105956536A (en) * | 2016-04-26 | 2016-09-21 | 北京森科赛德科技有限公司 | Pretreatment method and device for iris recognition |
| CN106203372A (en) * | 2016-07-19 | 2016-12-07 | 奇酷互联网络科技（深圳）有限公司 | Biopsy methods based on eyes, device and terminal unit |
| WO2018164364A1 (en) * | 2017-03-10 | 2018-09-13 | 크루셜텍 (주) | Contactless multiple body part recognition method and multiple body part recognition device, using multiple biometric data |
| CN107451547A (en) * | 2017-07-17 | 2017-12-08 | 广东欧珀移动通信有限公司 | Identify the method and Related product of live body |
| CN107463883A (en) * | 2017-07-18 | 2017-12-12 | 广东欧珀移动通信有限公司 | Biometric discrimination method and Related product |
| CN107506721A (en) * | 2017-08-18 | 2017-12-22 | 上海爱优威软件开发有限公司 | A kind of more method for collecting iris and system |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532933A (en) * | 2019-08-26 | 2019-12-03 | 淮北师范大学 | A kind of living body faces detection head pose returns the acquisition methods and detection method of device |
| CN113255400A (en) * | 2020-02-10 | 2021-08-13 | 深圳市光鉴科技有限公司 | Training and recognition method, system, equipment and medium of living body face recognition model |
| CN111402480A (en) * | 2020-02-29 | 2020-07-10 | 深圳壹账通智能科技有限公司 | Visitor information management method, device, system, equipment and storage medium |
| CN111797677A (en) * | 2020-05-13 | 2020-10-20 | 南京中科道置智能科技有限公司 | Face recognition living body detection method based on face iris recognition and thermal imaging technology |
| CN111797677B (en) * | 2020-05-13 | 2024-04-02 | 南京中科道置智能科技有限公司 | Face recognition living body detection method based on face iris recognition and thermal imaging technology |
| CN112164169A (en) * | 2020-09-21 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Access control management method and device, access control equipment and computer readable storage medium |
| CN112287909A (en) * | 2020-12-24 | 2021-01-29 | 四川新网银行股份有限公司 | Double-random in-vivo detection method for randomly generating detection points and interactive elements |
| CN114005019A (en) * | 2021-10-29 | 2022-02-01 | 北京有竹居网络技术有限公司 | Method for identifying copied image and related equipment thereof |
| CN114005019B (en) * | 2021-10-29 | 2023-09-22 | 北京有竹居网络技术有限公司 | Method for identifying flip image and related equipment thereof |
| CN116994325A (en) * | 2023-07-27 | 2023-11-03 | 山东睿芯半导体科技有限公司 | Iris recognition method, chip and terminal |
| CN116994325B (en) * | 2023-07-27 | 2024-02-20 | 山东睿芯半导体科技有限公司 | Iris recognition method, chip and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109325462B (en) | 2021-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109325462A (en) | Recognition of face biopsy method and device based on iris | |
| CN106951867B (en) | Face identification method, device, system and equipment based on convolutional neural networks | |
| KR102174595B1 (en) | System and method for identifying faces in unconstrained media | |
| CN100397410C (en) | Method and device for distinguishing face expression based on video frequency | |
| TWI754806B (en) | System and method for locating iris using deep learning | |
| CN109819208A (en) | A kind of dense population security monitoring management method based on artificial intelligence dynamic monitoring | |
| Kähm et al. | 2d face liveness detection: An overview | |
| KR102554391B1 (en) | Iris recognition based user authentication apparatus and method thereof | |
| US9892315B2 (en) | Systems and methods for detection of behavior correlated with outside distractions in examinations | |
| CN107438854A (en) | The system and method that the image captured using mobile device performs the user authentication based on fingerprint | |
| CN105138954A (en) | Image automatic screening, query and identification system | |
| CN104951773A (en) | Real-time face recognizing and monitoring system | |
| CN108563999A (en) | A kind of piece identity's recognition methods and device towards low quality video image | |
| CN106372629A (en) | Living body detection method and device | |
| CN109858375A (en) | Living body faces detection method, terminal and computer readable storage medium | |
| Zhang et al. | A survey on face anti-spoofing algorithms | |
| CN106599785A (en) | Method and device for building human body 3D feature identity information database | |
| CN109977867A (en) | A kind of infrared biopsy method based on machine learning multiple features fusion | |
| CN110929570B (en) | Iris rapid positioning device and positioning method thereof | |
| Mady et al. | Efficient real time attendance system based on face detection case study “MEDIU staff” | |
| CN111860394A (en) | Gesture estimation and gesture detection-based action living body recognition method | |
| RU2316051C2 (en) | Method and system for automatically checking presence of a living human face in biometric safety systems | |
| CN109299641A (en) | A kind of train dispatcher's fatigue monitoring image adaptive Processing Algorithm | |
| JP4708835B2 (en) | Face detection device, face detection method, and face detection program | |
| Yuan et al. | Ear detection based on CenterNet |
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 |