CN109255282A

CN109255282A - A kind of biometric discrimination method, device and system

Info

Publication number: CN109255282A
Application number: CN201710573675.9A
Authority: CN
Inventors: 黄欢; 赵刚
Original assignee: Shanghai Jinghong Electronic Technology Co Ltd
Current assignee: Shenzhen Jinghong Technology Co., Ltd
Priority date: 2017-07-14
Filing date: 2017-07-14
Publication date: 2019-01-22
Anticipated expiration: 2037-07-14
Also published as: CN109255282B

Abstract

The present invention provides a kind of biometric discrimination method, system and device, which comprises S1, acquisition carry the first picture signal of user's head biological information and carry the second picture signal of user's eye biological information；S2, according to the depth information of the first picture signal and the second picture signal composite calulation user, and judge whether user is in bio-identification regional scope；S3, bio-identification is carried out to the user in bio-identification regional scope.User, which is carried out, by dual biological characteristic identifies certification；The image that at least one biological characteristic can not only be carried out is shot, the image that can also carry out at least one biological characteristic is acquired and handles, and realize identity registration and the certification of biology, complete the identification of biology, convenience is provided for the acquisition and use of information security, and reach current highest identification security level, to promote the bigger progress of community service Yu people's life and work.

Description

A kind of biometric discrimination method, device and system

Technical field

The present invention relates to human-computer interaction technique fields, more particularly, to a kind of biometric discrimination method, system and device.

Background technique

With the rapid development of economy, community network is more complicated, the mobility of personnel is even more frequent, therefore for personnel Identification safety authentication demand it is also more urgent, while authentication condition is also more harsh.In numerous personal identification methods, By human body have by oneself in the way of the superiority that biological nature is identified is other peripheral hardwares incomparable, safety and reliable Property has been widely recognized.In view of uniqueness and constancy feature of the iris in living individual, in biological identification technology At one of the technology by high praise.More due to its low-down error recognition rate, so that this technology is more applicable for having In the field that high safety guarantee requires.Characteristic based on human body biological characteristics, propose based on face, palmmprint, fingerprint, iris, The biometrics identification technology of vein, hand-type, person's handwriting etc..

In the example embodiment of the relevant concept of the invention of one or more biological identification devices, face recognition, retina In identification, eyeprint identification, iris recognition etc. have also been employed that, it is desirable that the biology that user captures user in appropriately distance is special The image of sign carries out bio-identification, handles image procossing, and image is compared with pre-stored registered images.Matched It is set in the camera for capturing user biological image, the acquisition of biometric image, nothing is usually carried out only with single camera Method captures more biological characteristics according to distance, and the information collection of image is caused to be not easy to be used to assess the effective of biological characteristic Property, while the image information that uses of single camera or being colored or be black and white, to influence the body of user It tests.

With the rise of smart phone, binocular camera or more mesh cameras it is progressive enter people's lives, but they are logical It is often used only for shooting living scene or low-light (level) enhancing or first takes pictures to focus afterwards or optical zoom etc., such function It can not be the Image Acquisition applied to biological characteristic, therefore be unable to satisfy the requirement of bio-identification.

Just because of this, biological recognition system, which is shown using the result of distance measurement sensor measurement user distance, positions use The opereating specification at family, and while defining only in opereating specification, can just be used for the Image Acquisition of bio-identification, an otherwise just only meeting For general scene capture.Due to bio-identification no matter daytime or night, normal use is required, especially in dark situation Under, therefore the effect that will affect Image Acquisition if the supplement of not infrared light is arranged infrared on biological recognition system The supplement of light is very important, and will increase the effect of lower acquisition.

Summary of the invention

The present invention provides a kind of a kind of bio-identification side for overcoming the above problem or at least being partially solved the above problem Method, system and device solve the acquisition for carrying out biometric image only with single camera in the prior art, can not basis Distance captures more biological characteristics, causes the information collection of image to be not easy to be used to assess asking for the validity of biological characteristic Topic.

According to an aspect of the present invention, a kind of biometric discrimination method is provided, comprising:

S1, acquisition carry the first picture signal of user's head biological information and carry user's eye biological characteristic letter Second picture signal of breath；

S2, according to the depth information of the first picture signal and the second picture signal composite calulation user, and judge that user is It is no to be in bio-identification regional scope；

S3, bio-identification is carried out to the user in bio-identification regional scope.

Preferably, the head that the first image signal is shooting user captures the image letter of corresponding visible rays Number；Second picture signal is that the picture signal of corresponding infrared light is captured by shooting the eye being used for.

Preferably, the head biological information includes face mask, vein distributed intelligence；The eye biology Characteristic information includes iris, retina, eyeprint information.

Preferably, in step sl, the first image signal is RGB image signal, second picture signal For B/W image signal.

Preferably, the step S2 is specifically included:

The luminance component image and RGB component image of the first picture signal are extracted, the gray scale point of the second picture signal is extracted Spirogram picture；

The brightness classification image of first picture signal is synthesized with the gray component image of the second picture signal, is contained The customer position information of head position and eye locations, and obtain the depth information of user；

Judge whether user is in bio-identification regional scope according to user's depth information.

Preferably, the depth information of the user is by binocular vision technology, according to extract the first picture signal and The feature point feature of second picture signal simultaneously carries out Feature Points Matching, and the certain point of appearance in two images simultaneously is filtered out Come, and restores the three-dimensional coordinate of match point, and then obtain depth information.

Preferably, the step S3 is specifically included:

S31, to the user in the bio-identification region, user is calculated according to the RGB component image of the first picture signal Biological characteristic, obtain the face mask and/or vein distributed intelligence of user；

S32, obtained face mask and/or vein distributed intelligence are matched with log-on data, if matching is unsuccessful Then authentification failure terminates certification；S33 is entered step if successful match；

S33, the biological characteristic that user is calculated according to the second picture signal, obtain the eye iris and/or retina of user And/or eyeprint information, and matched with log-on data, the authentification failure if matching is unsuccessful terminates certification；If successful match Then certification passes through, and terminates certification.

Preferably, the step S3 further includes, when detecting that user is not in bio-identification region or authenticates mistake When losing, the first picture signal and the second picture signal are handled, synthesized, and by the image after processing synthesis and is exported aobvious Show.

Preferably, in the step S2 and step S3, after matching, authentication output consequential signal, the certification Consequential signal includes picture signal, voice signal.

A kind of biological recognition system, including image collection module and system processor；

Described image obtains module and is used to obtain the first picture signal for carrying user's head biological information and carries Second picture signal of user's eye biological information；

The system processor is used to be believed according to the first picture signal and the depth of the second picture signal composite calulation user Breath, and identification is carried out based on depth information, user's head biological information, user's eye biological information.

Preferably, described image obtains the visible rays optical signalling of module capture user's head and exports the first figure As signal, captures the infrared light optical signalling of user's eye and export the second picture signal, the first image signal is RGB Picture signal, second picture signal are B/W image signal.

Preferably, the system processor include color image processor, black white image processor, synthesis unit, Recognition unit and judging unit；

The color image processor is used to extract the luminance component image and RGB component image of the first picture signal；

The black white image processor is used to extract the gray component image of the second picture signal；

The synthesis unit is used for the gray component of brightness the classification image and the second picture signal of the first picture signal Image synthesis, obtains the customer position information containing head position and eye locations, and obtain the depth information of user；

The judging unit is used to judge whether user is in bio-identification regional scope according to user's depth information；

The recognition unit is used to carry out bio-identification to the user in bio-identification regional scope.

Preferably, further include prompt unit, after the prompt unit is for matching, authentication output result letter Number, the authentication result signal includes picture signal, voice signal.

Preferably, the recognition unit includes the first recognition unit and the second recognition unit；

First recognition unit is used to calculate the biological characteristic of user according to the RGB component image of the first picture signal, The face mask and/or vein distributed intelligence of user is obtained, by obtained face mask and/or vein distributed intelligence and registration number According to progress match cognization；

Second recognition unit is used to calculate the biological characteristic of user according to the second picture signal, obtains the eye of user Iris and/or retina and/or eyeprint information, and match cognization is carried out with log-on data.

Preferably, the synthesis unit includes that depth obtains sub-block and image synthesis sub-block；

The depth obtains sub-block and is used to be believed by binocular vision technology according to the first picture signal of extraction and the second image Number feature point feature and carry out Feature Points Matching, screened occurring certain point in two images simultaneously, and restore The three-dimensional coordinate of match point, and then obtain depth information；

Described image synthesis sub-block will be located for being handled the first picture signal and the second picture signal, being synthesized It manages the image after synthesizing and exports display.

Preferably, further including display unit, the display unit is for showing user images information and authentication result Information.

A kind of biological identification device, including binocular camera and system processor chip；

The binocular camera includes the first camera and second camera, and first camera carries use for obtaining First picture signal of account portion biological information；The second camera carries user's eye biological characteristic letter for obtaining Second picture signal of breath；

The system processing chip is used for the depth according to the first picture signal and the second picture signal composite calulation user Information, and identification is carried out based on depth information, user's head biological information, user's eye biological information.

Preferably, the time for exposure of first camera and the second camera is identical or different.

Preferably, first camera is visible rays filming apparatus, the second camera is infrared light Line filming apparatus.

Preferably, first camera includes the first camera lens, the first optical filter and the first image sensor chip； The second camera includes the second camera lens, the second optical filter and the second image sensor chip；First camera.

Preferably, first optical filter is the optical filter of infrared light wave cut-off, for filtering shooting visible light Infrared light outside line, second optical filter are the optical filter that infrared light filters that section passes through, for filtering shooting infrared light Visible rays in addition.

Preferably, the first image sensor is RGB image sensor, for shooting colored image；It is described Second image sensor is B/W image sensor, for shooting the image of black and white.

Preferably, first camera lens is wide-angle lens, for grabbing user's head image, second camera lens is Narrow angle mirror head, for grabbing the eyes image of user.

Preferably, further including wiring board assembly, the binocular camera and system processor chip installation In on the wiring board assembly.

Preferably, first camera further includes the first base with through-hole, the second camera further includes Second base with through-hole, first optical filter, the first camera lens are by being installed on wiring board assembly in the first base On, second optical filter, the second camera lens in the second base by being installed on wiring board assembly.

Preferably, the binocular camera further includes active focusing mechanism, the active focusing mechanism is for adjusting The relative position of first camera lens and the first image sensor chip, the second camera lens and the second image sensor chip relative position And first camera and second camera optical axis position.

Preferably, the first image sensor chip further includes first control unit, the first control unit For controlling the first pixel of the first image sensor chip, reader and the time for exposure of the first pixel；Second image Sensor chip further includes the second control unit, and second control unit is for controlling the second of the second image sensor chip The time for exposure of pixel, reader and the second pixel.

Preferably, first camera is fixed friendship office camera module or automatic focusing camera head mould group or light Learn zoom camera mould group.

Preferably, the system processor chip includes image processor, judging unit and recognition unit；

Described image processor is used for the brightness classification image of the first picture signal and the gray scale point of the second picture signal The synthesis of spirogram picture, obtains the customer position information containing head position and eye locations, and obtain the depth information of user；

Preferably, described image processor includes color image processor, black white image processor and synthesis unit；

The synthesis unit is used for the gray component of brightness the classification image and the second picture signal of the first picture signal Image synthesis, obtains the customer position information containing head position and eye locations, and obtain the depth information of user.

The application proposes a kind of biometric discrimination method, system and device, by carrying user's head biological characteristic letter respectively First picture signal of breath and the second picture signal for carrying user's eye biological information, and the first image signal is The head of shooting user captures the picture signal of corresponding visible rays；Second picture signal is by shooting the eye being used for Portion captures the picture signal of corresponding infrared light, and carries out user by dual biological characteristic and identify certification；It not only can be into The image of at least one biological characteristic of row is shot, and the image that can also carry out at least one biological characteristic is acquired and locates Reason, and realize identity registration and the certification of biology, the identification of biology is completed, is provided just for the acquisition and use of information security Benefit, and reach current highest identification security level, to promote the bigger of community service and people's life and work Progress.

Detailed description of the invention

Fig. 1 is the biometric discrimination method flow diagram according to the embodiment of the present invention 1；

Fig. 2 is the biometric discrimination method idiographic flow schematic diagram according to the embodiment of the present invention 1；

Fig. 3 is the self-timer schematic diagram according to the biological recognition system method of the embodiment of the present invention 1；

Fig. 4 is the head living things feature recognition schematic diagram according to the biological recognition system method of the embodiment of the present invention 1；

Fig. 5 is the eye living things feature recognition schematic diagram according to the biometric discrimination method of the embodiment of the present invention 1；

Fig. 6 is the biometric discrimination method application schematic diagram according to the embodiment of the present invention 2；

Fig. 7 is the biological recognition system structural block diagram according to the embodiment of the present invention 3；

Fig. 8 is the system identification flow diagram according to the embodiment of the present invention 3；

Fig. 9 is the biological recognition system application schematic diagram according to the embodiment of the present invention 4；

Figure 10 is the biological identification device application schematic diagram according to the embodiment of the present invention 5；

Figure 11 is the biological identification device structure chart according to the embodiment of the present invention 5；

Figure 12 is the frame diagram according to the binocular camera chip of the embodiment of the present invention 5.

Specific embodiment

With reference to the accompanying drawings and examples, specific embodiments of the present invention will be described in further detail.Implement below Example is not intended to limit the scope of the invention for illustrating the present invention.

Embodiment 1

As depicted in figs. 1 and 2, a kind of biometric discrimination method is shown in figure, comprising:

In the present embodiment, the first image signal is that the head of shooting user captures the image of corresponding visible rays Signal；Second picture signal is that the picture signal of corresponding infrared light is captured by shooting the eye being used for.

Preferably, the step S2 is specifically included:

Binocular vision technology belongs to a branch of computer vision technique, thought be by two width of processing or several Two dimensional image, obtains the three-dimensional information of object, then restores the three-dimensional live figure of corresponding scene, to reach reconstruction real world The purpose of model passes through view using imaging position difference of the same target on imaging plane separated by a distance, i.e. parallax Difference obtains the depth information of object.The research achievement of binocular vision technology can be applied to space flight, aviation, military affairs, education, amusement Etc. every field.The moon in space industry surveys landform vehicle, realizes the accurate of the three-dimensional data of the equipment of complex condition Measurement and its equipment positioning when acquiring the first picture signal and the second picture signal, can pass through binocular in the present embodiment Vision system is acquired respectively, and two camera lenses of binocular camera obtain target object and its surrounding scenes from different location simultaneously Two width digital pictures, or with one camera in different moments from two width digital pictures of different position acquisition surrounding scenes, base The three-dimensional spatial information of target object is recovered, is gone forward side by side between known two camera on the basis of positional relationship in principle of parallax One step rebuilds target object, enables to captured object reduction in world coordinate system, and then pass through binocular vision technology reality The acquisition of existing depth information.

Preferably, the step S3 is specifically included:

It in the present embodiment, is biometric discrimination method application provided in this embodiment specifically, as shown in Fig. 3 to Fig. 6 Into mobile phone, when user 2 picks up mobile phone (image processing system), user 2 takes the photograph in face of the binocular of mobile phone (image processing system) When as head 21, the scene including user 2 will be shot together image by binocular camera 21, and show user over the display 2 ' pictures；If detecting user 2 to be in bio-identification region, 2 " picture of user's head or use are shown over the display 2 " ' face face of family eye.Image conjunction will be carried out with luminance component image 1a and gray component image 2 ' in 1 signal of image At, and based on this composite result calculate between binocular camera 21 and user apart from information.Based on depth information and quilt The bio-identification regional scope parameter of storage determines whether user is in bio-identification regional scope after being matched.Based on coming from The second component image 1b of 1 signal of image of RBG firmware come calculate biological information and record the face mask of user 2 with/ Or vein distribution.Based on depth information and biological information for the user 2 in bio-identification region face mask with/ Or vein distribution carries out matching log-on data, if the face mask of the user 2 in bio-identification region and/or vein distribution With one image grabbing command signal of output after passing through.According to crawl command signal shooting based on 2 signal of image from BW firmware come Calculate the biological information progress of the eye iris and/or retina and/or eyeprint of the user 2 in bio-identification region With log-on data, if 2 signal of image is equal to log-on data, user authentication, which succeeds and exports one, " authenticates successfully " signal； If 2 signal of image is not equal to log-on data, user authentication failure simultaneously exports one " authentification failure " signal.

When detecting not in living things feature recognition region and/or receive living things feature recognition " authentification failure " after, respectively 1 signal of image and 2 signal of image are handled, the first processed image and the second image data are processed into First and second image datas are finally carried out composograph by processing and showing again over the display by new composograph Show.

Embodiment 2

As shown in fig. 6, in the present embodiment, image processing system 1 can be any one with biological recognition system 20 Equipment, including but not limited to mobile device, fixed equipment etc., such as door lock, gate inhibition, attendance, arms locker.Work as image processing system 1 ' when being a kind of gate inhibition, this is the another embodiment of image processing system 1, when user 2 is in bio-identification regional scope Interior, the biological recognition system 20 of image processing system 1 ' can capture the eyes image 2 " of user 2, to realize bio-identification.

Embodiment 3

As shown in Figure 7 and Figure 8, a kind of biological recognition system, including image collection module and system processing are shown in figure Device；

Fig. 8 is the flow chart that bio-identification is carried out according to the system of the present embodiment, and image may include Color Image Processing Device (image processor 1 in figure), black white image processor (image processor 2 in figure), synthesis unit, judging unit, first are known Other unit and the second recognition unit.Image processor can handle 1 signal of image from binocular camera, and 1 signal of image is One rgb signal, therefore extract the luminance component image 1a and the 2nd RBG component image of 1 signal of image of RGB firmware 1b.Image processor can handle 2 signal of image from binocular camera, and 2 signal of image is a BW signal, therefore extract The gray component image 2 ' of 2 signal of image of BW firmware.Synthesis unit will have luminance component image 1a in 1 signal of image With gray component image 2 ' carry out image synthesis, and based on this composite result calculate between binocular camera and user away from From information.Meanwhile synthesis unit can be calculated in binocular camera and by the eye of the head position and user that synthesize user The distance between user of position information.Second component image of first recognition unit based on 1 signal of image from BW firmware 1b calculates biological characteristic, this biological information, which passes through binocular camera and shoots and can be shown unit, to be shown The spatial information of spatial information and/or user face the vein distribution of user's face mask picture.Judging unit judgement is based on deep Degree information determines whether user is in bio-identification regional scope, only when judgement user is in bio-identification regional scope symbol Biological recognition system can just be operated by closing in the value pre-seted；Judging unit is regarded by being centered in the optics of binocular camera simultaneously The image of the shape of face profile on the head of user and/or the image of vein distribution and two eyes can be detected in field areas. A "No" signal can be exported when judging unit detects user not in bio-identification region, prompt unit receives "No" An image and the failure of auditory tone cues bio-identification can be exported after signal or are lost in one prompting message bio-identification of display It loses.When the image that judging unit detects the shape of face profile on the head of user and/or vein is distributed, judging unit can be exported One shooting command signal is given to the first recognition unit, and the detecting of the first recognition unit is from binocular camera according to shooting order letter The shape of face profile on the head of number capture user and/or the image 1b signal of vein distribution, and combine judging unit by detecting The picture signal of face contour and/or the vein distribution on the head of user determines whether the first image data matches registration number According to if output " it is that " signal, judging unit can continue to output the iris and/or view of the eye of a shooting order capture user The location drawing picture of nethike embrane and/or eyeprint, the detecting of the second recognition unit are used from binocular camera according to shooting command information capture The iris and/or retina of the eye at family and/or 2 signal of image of eyeprint pass through the rainbow of the eye of user in conjunction with judging unit The picture signal of film and/or retina and/or eyeprint determines whether the second image data matches log-on data, if output " be " signal, successful match.If output " is no, " signal exports an acoustic information prompt bio-identification failure or display one The failure of prompting message bio-identification.

Embodiment 4

As shown in figure 9, for the system concrete application schematic diagram in embodiment 2, the shape of image processing system 1 can be one A intelligent mobile terminal, including but not limited to smart phone, tablet computer, digital camera, internet of things equipment or internet device Deng.The image processing system 1 is answered including a display screen 10, a biological recognition system 20, a light emitting device 30, a memory 40 and one With processor 50.The application processor 50 of the image processing system 1, which can control, shows image data by display screen 10；Its It can control the image information that biological recognition system 20 captures the biological characteristic of user's head and eye, the biological recognition system 20 The depth information and biological information of user are captured by binocular structure；It can control light emitting device 30 and provides towards eye The infrared light in portion；Memory 40 can store the program for controlling image processing system 1.Application processor 50 can be executed and be deposited Store up the program in memory 40.When biological recognition system 20 detects biological characteristic when within bio-identification region, using place Reason device 50 can control the number that biological recognition system 20 exports the data and/or visible rays image that are modified to infrared light image According to；When biological recognition system 20 detects biological characteristic when other than bio-identification region, application processor 50 can control life Object identifying system 20 only exports the data of visible rays image.

Embodiment 5

Present embodiment illustrates a kind of biological identification devices, including binocular camera and system processor chip；

The depth obtains sub-block and is used to be believed by binocular vision technology according to the first picture signal of extraction and the second image Number feature point feature and carry out Feature Points Matching, screened occurring certain point in two images simultaneously, and restore The three-dimensional coordinate of match point, and then obtain depth information；Binocular vision technology belongs to a branch of computer vision technique, Thought is to obtain the three-dimensional information of object by two width of processing or two-dimensional images, then restore the three of corresponding scene Realistic picture is tieed up, to achieve the purpose that rebuild Real-world model, using same target on imaging plane separated by a distance Imaging position it is different, i.e. parallax obtains the depth information of object by parallax.The research achievement of binocular vision technology can answer For every field such as space flight, aviation, military affairs, education, amusements.The moon in space industry surveys landform vehicle, realizes complexity Under the conditions of equipment three-dimensional data precise measurement and its equipment positioning, in the present embodiment, acquire the first picture signal It when with the second picture signal, can be acquired respectively by binocular vision system, two camera lenses of binocular camera are from different positions Set while obtaining two width digital pictures of target object and its surrounding scenes, or with one camera in different moments from different location Two width digital pictures of surrounding scenes are obtained, principle of parallax is based on, between known two camera on the basis of positional relationship, are restored The three-dimensional spatial information of target object out, and target object is further rebuild, enable to captured object reduction to sit in the world In mark system, and then pass through the acquisition of binocular vision technology realization depth information.

When carrying out visual identity (i.e. face recognition, retina identification, iris recognition), multiple are shot by camera first Biometric image selects most suitable picture as selected biometric image, assesses the quality of image (if met The image of biometric identification criteria), then it is compared with the biometric image registered before, sees whether biological characteristic is consistent, if one Sample then enters unlocking screen, if it is inconsistent, prompt again attempts to identify.

Data show that the misclassification rate of fingerprint recognition is about 0.8%, and recognition of face is about 2%.Iris recognition is as " ultimate There is BOSS " advantages, the misclassification rates such as individual uniqueness, consistency, antifalsification can apply down to hundred a ten thousandths, therefore gradually In the scene for having high request to security level, such as the fields such as medical treatment, social security, credit, finance.Such scheme is used by identification The biological characteristic at family is to determine whether being legitimate user and being unlocked.

Iris recognition mode is used in the present embodiment, and the eye structure of people is by sclera, iris, pupil crystalline lens, view The part such as film forms.Iris is the annular formations between black pupil and white sclera, and it includes have many mutually friendships The minutia of wrong spot, filament, coronal, striped, crypts etc..And iris is after prenatal development stage is formed, entire It will be to maintain in life course constant.These features determine the uniqueness of iris feature, while also determining identification Uniqueness.It therefore, can be using the iris feature of eyes as everyone identification object.Compared to other lifes such as face Object feature, the Physiologic Studies of iris show it not only and have textural characteristics abundant, but also have uniqueness, stability, non- The features such as the property invaded, natural antifalsification, high bioactivity, are that the safest bio-identification for having been able to come into operation at present is recognized Card technology.Therefore, the identity recognizing technology based on iris is current research hotspot, will be in access control and attendance, firearms national treasury, identity The demanding fields of safety certifications such as card passport, social security system, Financial Information play a significant role.But it is fine due to iris Property and vision hide biological nature, at present iris recognition technology extensive industrialization, commercialization still have some difficulties, such as The acquisition equipment of high quality iris image, reduce user's fitness, evaluation iris image quality, iris source whether be living body with And iris self-adaptive processing, validity feature matching etc..

As shown in Figure 10 to Figure 12, the shape of image processing system 1 can be an intelligent mobile terminal, including but unlimited In smart phone, tablet computer, digital camera, internet of things equipment or internet device etc..The image processing system 1 includes one aobvious Display screen 10, a biological recognition system 20, a light emitting device 30, a memory 40 and an application processor 50.The image processing system 1 Application processor 50 can control image data is shown by display screen 10；It can control the capture of biological recognition system 20 The image information of the biological characteristic of user's head and eye, the biological recognition system 20 capture the depth of user by binocular structure Spend information and biological information；It can control infrared light of the offer of light emitting device 30 towards eye；Memory 40 can be deposited Store up the program for controlling image processing system 1.Application processor 50 can execute the program being stored in memory 40.Work as biology Identifying system 20 detects biological characteristic when within bio-identification region, and application processor 50 can control biological recognition system 20 outputs are modified to the data of infrared light image and/or the data of visible rays image；When biological recognition system 20 detects When other than bio-identification region, application processor 50 can control biological recognition system 20 and only exports visible rays biological characteristic The data of image.

The biological recognition system 20 includes a binocular camera 21 and a biological recognition system processor chips 22.At these Managing device and other processing can be by biological identification device 20 as on the similar memory 40 of the machine executable instructions that can be performed It instructs and equally goes to execute.Binocular camera 21 and biological recognition system processor chips 22 are as dedicated for the hard of allocation processing Part goes to execute, and processing task reasonably can be distributed or be shared.Binocular camera 21 is used to capture head and the eye figure of user As information, and based on shooting picture output 1 signal of image and 2 signal of image.Binocular camera 21 include the first camera 21a and Second camera 21b, the first camera 21a and second camera 21b are installed on wiring board assembly 110, the first camera shooting Head 21a includes the first image sensor chip 120, the first optical filter 130, the first camera lens 140 and first base 150；Second Camera 21b includes the first image sensor chip 160, the second optical filter 170, the second camera lens 180 and second base 190. Wiring board assembly 110 is for installing various devices, including the first camera 21a, second camera 21b and bio-identification System processor chip 22.First camera lens 140 and the second camera lens 180 may include the mirror for possessing different field angles according to focal length Piece；First camera lens 140 possesses the wide-angle lens that one big field angle goes capture user's head, and is the mirror of a short focus Head；Second camera lens 160 possesses one and neglects the narrow angle mirror head that capture user's eye is removed in rink corner, and can be a fixed focal length Camera lens or an auto-focusing camera lens.First optical filter 130 can permit the wavelength by visible rays, and The wavelength of barrier infrared light passes through, and is one piece of cutoff filter；Second optical filter 170, which can permit, passes through infrared light Filter it is long, and the filter length for obstructing visible rays passes through, and is one piece infrared to pass through optical filter.First image sensor chip 120 It is a color cmos chip, for collecting the optical signalling that infrared cutoff is crossed, and converts output 1 signal of image；Second image Sensor chip 160 is a black and white CMOS chip, for collecting the infrared optical signalling by crossing, and converts output image 2 Signal.First base 150 is the cover with through-hole, and through hole is used to place the first camera lens 140 and the first optical filter 130, First base 150 carries out shading and protection to the first image sensor chip 120 simultaneously；Second base 190 is also a band logical The cover in hole, through-hole piece is for placing the second camera lens 180 and the second optical filter 170, while second base 190 is to the second image sense It surveys device chip 160 and carries out shading and protection.

With reference to Figure 12, there are two independent image sensors, including the first image sensor chip 120 for binocular camera 21 With the second image sensor chip 160.The first control unit 122 of first image sensor chip 120 can be answered according to coming from The first pixel 121 and reader 123 are controlled with the control signal 1 of processor 50.The second of second image sensor chip 160 Control unit 162 can control the second pixel 161 and reader 163 according to the control signal 2 from application processor 50.The One control unit 122 can control the time for exposure of the first pixel 121, and the second control unit 162 can control second as 161 Time for exposure, their time for exposure are different.

Consider that the environmental condition lower time for exposure of light emitting device 30 can be different control, the second image sense according to variation The sensitivity for surveying device 160 has selection wavelength etc..When bio-identification is under a dark situation, first control unit 122 can To control the pixel exposure for the picture that the pixel exposure time for the picture for corresponding to visible rays goes collection to correspond to infrared light Time；The pixel exposure time that second control unit 162 can control the picture corresponding to infrared light goes collection to correspond to can Depending on the pixel exposure time of the picture of light；The difference of the two time for exposure can be depending on changes in environmental conditions.In order to The efficiency of the pixel size of respective image is promoted, arbitrary pixel can include or merge and may carry out difference to these pixels The comparison of size, bigger Pixel Dimensions be include the second pixel 161 export corresponding infrared image and the first pixel 121 output The resulting pixel size of corresponding visible rays image, it is corresponding that brighter resulting pixel size is that the second pixel 161 exports Infrared image exports the resulting pixel size that corresponding visible rays image carries out luminance compensation to the first pixel 121, apparent Resulting pixel size be the first pixel 121 output visible rays image just exports the second pixel 161 it is corresponding infrared The resulting pixel size of image progress details compensation.

It is required to reach image synthesis, the first camera 21a and second camera 21b can be controlled by active focusing Optimize position precision, between the first camera 21a and second camera 21b position precision assembly to be focused by active come it is real Existing, secondly the first image sensor chip 120 and the first camera lens 140 can be controlled by active focusing to optimize the first camera The optical axis position precision of 21a, the second image sensor core sign 160 can be by active focusing controls come excellent with the second camera lens 180 Change second camera 21b position precision, last first camera 21a and second camera 21b can be controlled by active focusing To optimize 21 position precision of binocular camera." actively focusing control " herein is commonly referred to as the assembly of camera module monomer The assembly technology of technique and binocular camera entirety, camera module monomer are fitted by the first camera lens of adjustment 140 relatively Height, rotation angle, tilt angle between the first image sensor chip 120 etc. influence the factor of optical characteristics at one It completes in technique, by adjusting height of second camera lens 180 relative to second image sensor chip 160 between, rotates angle, inclines The factor that rake angle etc. influences optical characteristics is completed in a technique；Finally by adjusting the first camera 21a and second Optical axis between camera 21b, which is controlled, just completes the assembly of binocular camera 21 in geometrical relationship.The first of binocular camera 21 Reader 123 in the signal for receiving the first pixel 121, under the control of first control unit 122 first reader 123 generate and 1 signal of image is exported, 1 signal of image is the image to the visible rays for being applied to capture user's head；Second reader 163 exists The signal for receiving the second pixel 161, second reader 163 generates under the control of the second control unit 162 and output image 2 is believed Number, 2 signal of image is the image to the infrared light for being applied to capture user's eye.

Finally, the present processes are only preferable embodiment, it is not intended to limit the scope of the present invention.It is all Within the spirit and principles in the present invention, any modification, equivalent replacement, improvement and so on should be included in protection of the invention Within the scope of.

Claims

1. a kind of biometric discrimination method characterized by comprising

S1, acquisition carry the first picture signal of user's head biological information and carry user's eye biological information Second picture signal；

S2, according to the depth information of the first picture signal and the second picture signal composite calulation user, and judge whether user locates In in bio-identification regional scope；

2. biometric discrimination method according to claim 1, which is characterized in that the first image signal is shooting user Head captures the picture signal of corresponding visible rays；Second picture signal is to be corresponded to by shooting the eye being used for capture Infrared light picture signal.

3. biometric discrimination method according to claim 1, which is characterized in that the head biological information includes face Profile, vein distributed intelligence；The eye biological information includes iris, retina, eyeprint information.

4. biometric discrimination method according to claim 1, which is characterized in that in step sl, the first image signal For RGB image signal, second picture signal is B/W image signal.

5. biometric discrimination method according to claim 4, which is characterized in that the step S2 is specifically included:

The luminance component image and RGB component image of the first picture signal are extracted, the gray component figure of the second picture signal is extracted Picture；

The brightness classification image of first picture signal is synthesized with the gray component image of the second picture signal, is obtained containing head The customer position information of position and eye locations, and obtain the depth information of user；

6. biometric discrimination method according to claim 5, which is characterized in that the depth information of the user passes through binocular vision Feel technology according to the feature point feature for extracting the first picture signal and the second picture signal and carries out Feature Points Matching, will simultaneously The certain point occurred in two images screens, and restores the three-dimensional coordinate of match point, and then obtain depth information.

7. biometric discrimination method according to claim 5, which is characterized in that the step S3 is specifically included:

S31, to the user in the bio-identification region, the life of user is calculated according to the RGB component image of the first picture signal Object feature obtains the face mask and/or vein distributed intelligence of user；

S32, obtained face mask and/or vein distributed intelligence are matched with log-on data, is recognized if matching is unsuccessful Card failure, terminates certification；S33 is entered step if successful match；

S33, according to the second picture signal calculate user biological characteristic, obtain user eye iris and/or retina and/or Eyeprint information, and matched with log-on data, the authentification failure if matching is unsuccessful terminates certification；Recognize if successful match Card passes through, and terminates certification.

8. biometric discrimination method according to claim 7, which is characterized in that the step S3 further includes, when detecting use Family is not in bio-identification region or when authentification failure, is handled the first picture signal and the second picture signal, is synthesized, And by the image after processing synthesis and export display.

9. biometric discrimination method according to claim 6, which is characterized in that in the step S2 and step S3, matching knot Shu Hou, authentication output consequential signal, the authentication result signal include picture signal, voice signal.

10. a kind of biological recognition system, which is characterized in that including image collection module and system processor；

Described image obtains module and is used to obtain the first picture signal for carrying user's head biological information and carries user Second picture signal of eye biological information；

The system processor is used for the depth information according to the first picture signal and the second picture signal composite calulation user, and Identification is carried out based on depth information, user's head biological information, user's eye biological information.

11. biological recognition system according to claim 9, which is characterized in that described image obtains module capture and uses account The visible rays optical signalling in portion simultaneously exports the first picture signal, captures the infrared light optical signalling of user's eye and exports the Two picture signals, the first image signal are RGB image signal, and second picture signal is B/W image signal.

12. biometric discrimination method according to claim 10, which is characterized in that the head biological information includes face Contouring, vein distributed intelligence；The eye biological information includes iris, retina, eyeprint information.

13. biometric discrimination method according to claim 11, which is characterized in that the system processor includes color image Processor, black white image processor, synthesis unit, recognition unit and judging unit；

The synthesis unit is used for the gray component image of brightness the classification image and the second picture signal of the first picture signal Synthesis, obtains the customer position information containing head position and eye locations, and obtain the depth information of user；

14. biological recognition system according to claim 13, which is characterized in that further include prompt unit, the prompt is single After member is for matching, authentication output consequential signal, the authentication result signal includes picture signal, voice signal.

15. biological recognition system according to claim 13, which is characterized in that the recognition unit includes that the first identification is single Member and the second recognition unit；

First recognition unit is used to calculate the biological characteristic of user according to the RGB component image of the first picture signal, obtains The face mask and/or vein distributed intelligence of user, by obtained face mask and/or vein distributed intelligence and log-on data into Row match cognization；

Second recognition unit is used to calculate the biological characteristic of user according to the second picture signal, obtains the eye iris of user And/or retina and/or eyeprint information, and match cognization is carried out with log-on data.

16. biological recognition system according to claim 13, which is characterized in that the synthesis unit includes that depth obtains son Block and image synthesize sub-block；

The depth obtains sub-block and is used for through binocular vision technology, according to extracting the first picture signal and the second picture signal Feature point feature simultaneously carries out Feature Points Matching, the certain point of appearance in two images simultaneously is screened, and restore matching The three-dimensional coordinate of point, and then obtain depth information；

Described image synthesis sub-block is closed for being handled the first picture signal and the second picture signal, being synthesized, and by processing Image after simultaneously exports display.

17. biological recognition system according to claim 13, which is characterized in that further include display unit, the display is single Member is for showing user images information and authentication result information.

18. a kind of biological identification device, which is characterized in that including binocular camera and system processor chip；

The binocular camera includes the first camera and second camera, and first camera uses account for obtaining to carry First picture signal of portion's biological information；The second camera carries user's eye biological information for obtaining Second picture signal；

The system processing chip is used for the depth information according to the first picture signal and the second picture signal composite calulation user, And identification is carried out based on depth information, user's head biological information, user's eye biological information.

19. biological identification device according to claim 18, which is characterized in that first camera and described second is taken the photograph As the time for exposure of head is identical or different.

20. biological identification device according to claim 18, which is characterized in that first camera is visible rays Filming apparatus, the second camera are infrared light filming apparatus.

21. biological identification device according to claim 19, which is characterized in that first camera includes the first mirror Head, the first optical filter and the first image sensor chip；The second camera includes the second camera lens, the second optical filter and second Image sensor chip；First camera.

22. biological identification device according to claim 21, which is characterized in that first optical filter is infrared light wave The optical filter of cut-off, for filtering the infrared light outside shooting visible rays, second optical filter is that infrared light filter section is logical The optical filter crossed, for filtering the visible rays other than shooting infrared light.

23. biological identification device according to claim 21, which is characterized in that the first image sensor is RGB figure As sensor, for shooting colored image；Second image sensor is B/W image sensor, for shooting the figure of black and white Picture.

24. biological identification device according to claim 21, which is characterized in that first camera lens is wide-angle lens, is used In crawl user's head image, second camera lens is narrow angle mirror head, for grabbing the eyes image of user.

25. biological identification device according to claim 21, which is characterized in that it further include wiring board assembly, it is described double Mesh camera and the system processor chip are installed on the wiring board assembly.

26. biological identification device according to claim 25, which is characterized in that first camera further includes band through-hole First base, the second camera further includes the second base with through-hole, and first optical filter, the first camera lens pass through institute It states and is installed in first base on wiring board assembly, second optical filter, the second camera lens in the second base by pacifying Loaded on wiring board assembly.

27. biological identification device according to claim 21, which is characterized in that the binocular camera further includes actively right Coke installation, the active focusing mechanism are used to adjust relative position, the second mirror of the first camera lens and the first image sensor chip Head and the relative position of the second image sensor chip and the optical axis position of the first camera and second camera.

28. biological identification device according to claim 21, which is characterized in that the first image sensor chip also wraps Include first control unit, the first control unit be used to control the first pixel of the first image sensor chip, reader and The time for exposure of first pixel；The second image sensor chip further includes the second control unit, second control unit For controlling the second pixel of the second image sensor chip, reader and the time for exposure of the second pixel.

29. biological identification device according to claim 18, which is characterized in that the head biological information includes face Contouring, vein distributed intelligence；The eye biological information includes iris, retina, eyeprint information.

30. biological identification device according to claim 18, which is characterized in that first camera is that fixed friendship office is taken the photograph As head mould group or automatic focusing camera head mould group or optical zoom pick-up head mould group.

31. biological identification device according to claim 18, which is characterized in that the system processor chip includes image Processor, judging unit and recognition unit；

Described image processor is used for the gray component figure of brightness the classification image and the second picture signal of the first picture signal As synthesis, the customer position information containing head position and eye locations is obtained, and obtain the depth information of user；

32. biological identification device according to claim 31, which is characterized in that described image processor includes color image Processor, black white image processor and synthesis unit；

The synthesis unit is used for the gray component image of brightness the classification image and the second picture signal of the first picture signal Synthesis, obtains the customer position information containing head position and eye locations, and obtain the depth information of user.

33. biological identification device according to claim 32, which is characterized in that the synthesis unit includes that depth obtains son Block and image synthesize sub-block；

34. biological identification device according to claim 18, which is characterized in that further include display unit, the display is single Member is for showing user images information and authentication result information.

35. biological identification device according to claim 18, which is characterized in that further include prompt unit, the prompt is single After member is for matching, authentication output consequential signal, the authentication result signal includes picture signal, voice signal.

36. biological recognition system according to claim 31, which is characterized in that the recognition unit includes that the first identification is single Member and the second recognition unit；