CN109766806A - Efficient face identification method and electronic equipment - Google Patents

Efficient face identification method and electronic equipment Download PDF

Info

Publication number
CN109766806A
CN109766806A CN201811628411.XA CN201811628411A CN109766806A CN 109766806 A CN109766806 A CN 109766806A CN 201811628411 A CN201811628411 A CN 201811628411A CN 109766806 A CN109766806 A CN 109766806A
Authority
CN
China
Prior art keywords
image
face
display screen
camera
sub
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201811628411.XA
Other languages
Chinese (zh)
Inventor
肖振中
许星
闫敏
王兆民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Orbbec Co Ltd
Original Assignee
Shenzhen Orbbec Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Orbbec Co Ltd filed Critical Shenzhen Orbbec Co Ltd
Priority to CN201811628411.XA priority Critical patent/CN109766806A/en
Publication of CN109766806A publication Critical patent/CN109766806A/en
Pending legal-status Critical Current

Links

Abstract

The present invention proposes efficient face identification method and electronic equipment, wherein face identification method includes: S1: receiving the first, second IR image comprising face;S2: the face in detection the first, second IR image;S3: will test, facial image is input to the neural network model trained in advance to carry out recognition of face;S4: relevant instruction is executed according to recognition result.The present invention also proposes a kind of electronic equipment, is able to carry out above-mentioned face identification method, and the first, second camera, light emitting mould group are arranged in the back side of display screen.Two width IR images are directly carried out recognition of face, need not move through the biggish algorithmic procedure of this calculation amount of disparity computation by the efficient face identification method of the present invention, can significantly promote recognition of face efficiency, promote user experience;Electronic equipment of the invention can not only carry out accurate efficient recognition of face, screen accounting also with higher.

Description

Efficient face identification method and electronic equipment
Technical field
The present invention relates to electronic technology field, in particular to a kind of efficient face identification method and electronic equipment.
Background technique
The positive direction of screen comprehensively such as smart machine such as mobile phone, plate is developed, and screen means its of these smart machines comprehensively He optimizes device needs, for example the extra large shape of current lift camera model, postposition fingerprint module, double screen module, stream is not advised Then display screen etc. designs, and is to realize higher screen accounting to promote visual effect by the optimization of component.
Nevertheless, some rigid demands have contradiction with comprehensive screen always, such as based on the face recognition scheme taken pictures Screen accounting certainly will be will affect (to shield comprehensively and take pictures although lift camera model can be taken into account, equally bring poor user experience, device The problems such as part service life).In addition, with universal, comprehensive screen side of high-precision, high security face recognition scheme based on 3D rendering Case facing challenges are increasing.
Summary of the invention
The purpose of the present invention is to solve exist between comprehensive screen scheme in the prior art and 3D face recognition technology The problem of contradiction, proposes a kind of efficient face identification method and electronic equipment.
The efficient face identification method of the present invention, comprising the following steps: S1: the first IR image comprising face and the are received Two IR images;S2: the face in detection the first IR image and the 2nd IR image;S3: the first IR image that will test and Facial image in 2nd IR image is input to the neural network model trained in advance to carry out recognition of face;S4: root Relevant instruction is executed according to recognition result.
In a preferred embodiment, the step S3 includes following sub-step: S31: the first IR image of detection and the Whether the face in two IR images is stereoscopic face;S32: to the face in the first IR image and/or the 2nd IR image into Row identification.
In a preferred embodiment, process trained in advance described in the step S3 are as follows: to more than two true Face and more than two false faces acquire the first IR image and the 2nd IR image respectively, and mark to these images Note;Then these images are input in neural network model, to be trained to parameters in model, so that after training Neural network model will have three-dimensional judgement and two kinds of function of recognition of face to the first IR image of input and the 2nd IR image Energy.It is highly preferred that the neural network model includes the first sub-neural network model and the second sub-neural network model, in training Stage, the first IR image being collected in advance, the 2nd IR image and whether be real human face markup information be input to it is described First sub-neural network model;The first IR image being collected in advance and/or the 2nd IR image are input to the second son mind Through network model;The first sub-neural network model is for detecting whether facial image to be measured is stereoscopic face, and described second Sub-neural network model is used to carry out recognition of face to the facial image to be measured.
The present invention also proposes a kind of electronic equipment, comprising: shell;It is arranged in positive display screen, first camera, the second phase Machine, light emitting mould group, memory and processor;The first camera and second camera are used to acquire external people or object red Infrared image under outer light irradiation, the light emitting mould group is for providing infrared illumination;The first camera and second camera Binocular depth camera is formed, the back side of the display screen is arranged in the first camera, second camera, light emitting mould group;It is described Memory storage has program;The processor runs described program, is distinguished with controlling the first camera and the second camera Acquisition includes the first IR image and the 2nd IR image of face, and the control electronic equipment execute it is any of the above described Face identification method.
In a preferred embodiment, the first camera is different from the field angle of second camera, resolution ratio or focal length.? In preferred embodiment, the display screen includes the first sub- display screen, the second sub- display screen and the sub- display screen of third, and described the One camera and second camera are separately positioned on the back side of the described first sub- display screen and the sub- display screen of third.It is highly preferred that described First sub- display screen and the sub- display screen of third are transparent display screen, of third described in the resolution ratio of the first sub- display screen The resolution ratio of display screen is low, the display aperture opening ratio of the display aperture opening ratio display screen than the third of the first sub- display screen It is high.
In a preferred embodiment, it is also set between the first camera and display screen, between second camera and display screen It is equipped with optical filter, for reducing the transmission of visible light.It is highly preferred that the optical filter is optical switch, when power is off in non- Pellucidity, light can not pass through, and pellucidity is in when being powered, and light can pass through;Alternatively, the optical filter is unidirectional Perspective film, a side plane of the single-way perspective membrane towards optics module are less than reflectivity for the transmitance of visible light, towards One side plane of display screen is greater than reflectivity for the transmitance of infrared light, so that external infrared light be allowed to pass through display screen Internal light is prevented to pass through display screen into optics module;Alternatively, the optical filter is optical filter, for preventing visible light logical It crosses and to pass through with the light beam in second camera corresponding wavelength section;And it is described when the optical filter is optical filter Second camera only includes imaging sensor and imaging len.
Compared with prior art, the beneficial effects of the present invention are as follows:
Two width IR images are directly carried out recognition of face, need not move through parallaxometer by the efficient face identification method of the present invention The biggish algorithmic procedure of this calculation amount is calculated, recognition of face efficiency can be significantly promoted, promotes user experience;Electricity of the invention Sub- equipment carries out recognition of face using the binocular depth camera of first camera and second camera composition, and by the first, second phase Machine is arranged under screen, can not only carry out accurate efficient recognition of face, screen accounting also with higher.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with The attached drawing of other embodiments is obtained according to these attached drawings.
Fig. 1 is the positive structure schematic of electronic equipment in one embodiment of the invention.
Fig. 2 is the structure composition schematic diagram of electronic equipment in one embodiment of the invention.
Fig. 3 is the cross section structure schematic diagram of electronic equipment in one embodiment of the invention.
Fig. 4 is the cross section structure schematic diagram of electronic equipment in another embodiment of the present invention.
Fig. 5 is face identification method flow diagram in another embodiment of the present invention.
Specific embodiment
With reference to embodiment and compares attached drawing the present invention is described in further details.Wherein identical attached drawing Label indicates identical component, unless stated otherwise.It is emphasized that following the description is only exemplary, without It is to limit the scope of the invention and its apply.
Fig. 1 is electronic equipment front schematic view according to an embodiment of the invention.Electronic equipment 10 include shell 105, Display screen 106, the binocular vision module being made of first camera 101 and second camera 102, can also include optical transmitter module The sensors 104 such as 103 and some sensors, such as loudspeaker, environment light/proximity sensor.
Display screen 106 can be liquid crystal display (LCD), organic light-emitting diode (OLED) display screen (OLED) etc., for showing Application program image is also possible to the display screen of any other form according to actual needs.Display screen can also include touching simultaneously Setting capacitance touching control electrode in function, such as display screen is touched, using the input unit as human-computer interaction.
Sensor is for sending or collecting the information such as electronic device exterior information, such as illumination, sound.Multiple sensors can The different parts that also can be set at top in other positions or scattering device in electronic equipment are arranged.In some implementations In example, some sensors also be can be set at the back side of electronic equipment.
Binocular vision module includes first camera 101 and second camera 102, in order to enable the electronic equipment have it is round-the-clock Visual capacity, first camera 101 and second camera 102 be configured to infrared camera, such as to the close of 840nm or 950nm The light beam of infrared wavelength is sensitive, certainly, in other examples can also light beam to other infrared wavelengths it is sensitive.Optical transmitter module 103, for providing corresponding infrared beam illumination, can be infrared floodlight, infrared structure light projector etc..In infrared beam Under illumination, first camera 101 can acquire the left and right infrared image of target, two width figures with second camera 102 under different perspectives Parallax as between pixel then reflects the three-dimensional information of target, in some embodiments, then can be based in two images Parallax information can then recover the 3-D image of target.In order to promote screen accounting, first camera 101, second camera 102 with And optical transmitter module 103 is arranged on 103 back side of display screen, on display screen 103 with first camera 101, second camera 102 with And the corresponding pixel region of optical transmitter module 103 can still carry out content and show.The specific plan of establishment will be set forth later.
In one embodiment, electronic equipment can also include third camera (not shown), and third camera is for adopting Collect visible images, such as color camera for the application such as self-timer.
In one embodiment, electronic equipment can also include more infrared cameras, such as the of infrared imaging Four cameras etc. carry out binocular by two or more infrared cameras or multi-vision visual are imaged.
Fig. 2 is the structure composition schematic diagram of electronic equipment according to an embodiment of the invention.Electronic equipment 20 includes aobvious Display screen 211, environment light/proximity sensor 201, infrared camera 210 (include at least the first infrared camera and the second infrared camera), Color camera 212, processor 206, microphone 202, radio frequency and baseband processor 203, interface 204, memory 205, battery 207, MEMS sensor 208, audio devices 209 etc., between different units can by circuit connection realize data transmission with Signal communication.Here it is only illustrated with the composed structure of one embodiment, in other embodiments, electronic equipment also can wrap Containing less structure or include other more composed structures.Electronic equipment can be mobile phone, computer, game machine, plate, electricity Depending on, wearable device etc..
Processor 206 is used to carry out entire electronic equipment whole control, and processor 206 can be single processor It may include multiple processor units, for example be made of the processor unit of different function.In some embodiments, processor 206 are also possible to integrated system-on-chip (SOC), include central processing unit, on-chip memory, controller, communication interface etc..? In some embodiments, processor 206 is application processor AP, such as mobile application processor, is mainly responsible in electronic equipment and removes Communicate the realization, such as text-processing, image procossing, recognition of face etc. of outer other function.
Display screen 211 is presented to the user for showing image under the control of processor 206 to apply etc., is in addition shown Screen 211 also may include touch function, and display is also used as man-machine interactive interface at this time, for receiving the input of user.Generally Protective cover, such as glass cover-plate etc. can be also arranged on 211 surface of display screen in order to protect display screen 211 not damaged in ground.
Microphone 202 can be used to realize the interactive voice with user for receiving voice messaging.
Radio frequency and baseband processor 203 are responsible for the communication function of electronic equipment, such as reception and translated speech or text etc. Signal is to realize the information interchange between remote user.
Interface 204 is used for so that electronic equipment is attached with outside, to further realize data transmission, power transmission etc. Function, interface 204 are controlled by the communication interface in processor 206.Interface 204 may include USB interface, WIFI interface etc..
Memory 205 is used to save data, for example application data, system data, processor 206 are in the process of implementation The temporary designator and data of preservation.Memory 205 can be made of single or multiple memories, can be RAM, FLASH etc. Any form of memory that can be used for saving data.It is understood that memory may act as a part of electronic equipment, Can also exist independently of electronic equipment, such as cloud memory, the data saved can be waited by interface 204 and be set with electronics Standby communication.Application program such as face recognition application are generally held in non-volatile readable storage medium, when the execution application When, processor will call corresponding program to execute from the storage medium, and specific recognition of face will illustrate later.
Environment light/proximity sensor 201 can be integrated single-sensor and be also possible to independent ambient light sensing Device and proximity sensor.Wherein ambient light sensor is used to obtain the Lighting information of current environment locating for electronic equipment, one In a embodiment, the adjust automatically of screen intensity may be implemented based on the Lighting information, human eye is more comfortably shown with providing Brightness;Whether proximity sensor can measure has object close to electronic equipment, some functions may be implemented based on this, for example connecing The touch function for closing screen when answering a call when face sufficiently closes to prevents from accidentally touching.In some embodiments, proximity sensor It can also quickly judge the approximate distance between face and electronic equipment.
Battery 207 is for providing electric power.Loudspeaker 209 is for realizing voice output.
MEMS sensor 208 is for obtaining the current status information of electronic equipment, such as position, direction, acceleration, gravity Deng, therefore MEMS sensor 208 may include the sensors such as accelerometer, gravitometer, gyroscope.In one embodiment, MEMS sensor 208 can be used to activate some face recognition applications, such as when user picks up electronic equipment, MEMS sensor 208 this available variation, while this variation is transferred to processor 206, processor 206 adjusts memory-aided face to know Other application program is to carry out face recognition application.
Color camera 212 is for acquiring image, and, processor controls phase Machine 212 acquires image, and image transmitting is shown to display.In some embodiments, camera 101 may include light field Camera, wide angle camera, focal length camera etc..Color camera 212 can be set in any position of electronic equipment, such as preposition plane The positions such as the top or bottom end of (identical as plane where display) etc., postposition plane, for example coloured silk is set in electronic equipment front Form and aspect machine 212 is to acquire user's facial image;Color camera 212 is also disposed in postposition plane for taking pictures to scene Deng.In one embodiment, color camera 212 is arranged on preposition and postposition plane, and the two can also be with independent acquisition image It can also be controlled with device 206 processed with synchronous acquisition image.
Infrared camera 210 contains at least two infrared camera, as shown in Figure 1, multiple infrared cameras composition binocular or more mesh Vision module.Infrared camera 210 is used to acquire the Infrared Targets image under the irradiation of corresponding infrared wavelength light beam, such as The infrared image of 840nm or 950nm, provide illumination can be light emitting mould group as shown in Figure 1, can also be with electronic equipment Except light source, such as environment light.After infrared camera 210 collects several Infrared Targets images, it is based on several Infrared Targets Parallax between image can reflect or restore the depth information of target, and be realized further based on parallax or depth information Using, such as recognition of face etc..
Fig. 3 is electronic equipment schematic cross-section according to an embodiment of the invention, only schematically provides electronics in figure and sets Standby part of devices, the electronic equipment include the first infrared camera 33, the second infrared camera 34, display screen 32 and cover board 31. Electronic equipment can also include light emitting mould group (not shown), for emit and the light beam of infrared camera corresponding wavelength, with Infrared illumination is provided to target.Wherein, the first infrared camera 33 and the second infrared camera 34 are for acquiring exterior object red The lower infrared image of outer light irradiation, the first infrared camera 33 and the second infrared camera 34 composition binocular vision depth camera, into When row Image Acquisition, the first infrared camera 33 is with the second infrared camera 34 due to leading between each other in the presence of interval (also known as baseline) When same object is imaged in cause, there are parallaxes between the image of acquisition, and parallax then reflects the actual distance of object, i.e., Depth information.
In one embodiment, field angle, resolution ratio, focal length of the first infrared camera 33 and the second infrared camera 34 etc. is joined At least one of number is not identical.Such as first infrared camera 33 high resolution in the second infrared camera 34, the first infrared phase First infrared image captured by machine 33 since high resolution can be applied further, such as face characteristic extract with Recognition of face, and the second infrared image captured by the second infrared camera 34 will carry out parallax estimation or depth with the first infrared image Degree calculates.
In one embodiment, protection is played in the surface that cover board 31 is covered on the display screen 32 for imaging for glass material Effect, cover board 31 are also possible to other transparent materials and are made.In one embodiment, cover board 31 is also possible to the one of display screen 32 Part.
In some embodiments, electronic equipment further includes optical filter 35, and setting is in the first infrared camera 33 and display screen 32 Between, between the second infrared camera 34 and display screen 32, be configured for reducing visible light-transmissive so that external use Family cannot observe directly the first infrared camera 33 and the second infrared camera 34, promote visual sense of beauty.
In one embodiment, optical filter 35 is optical switch, such as liquid crystal shutter, is in nontransparent shape when power is off State, light can not pass through;Pellucidity is in when being powered, light can pass through.Optical switch is also possible to other types material Material is made, such as electrochromic material, thermochromatic material, or allows to whether change light by certain optical texture Pass through.
In one embodiment, optical filter 35 is single-way perspective membrane, and single-way perspective membrane allows external light logical as far as possible It crosses display screen to enter optics module and internal light is prevented to pass through display screen, i.e., single-way perspective membrane is towards one side plane of optics module The transmitance of visible light is less than reflectivity (for example transmitance is 5%~30%, and reflectivity is 90%~95%), and towards One side plane of display screen for visible light transmitance be greater than reflectivity (such as transmitance be 60%~95%, reflectivity 5% ~30%).For infrared camera, single-way perspective membrane should be configured to infrared light light transmittance with higher, further, single There is higher transmitance for infrared light towards one side plane of display screen to perspective film, transmitance generally required to be greater than reflectivity, To ensure image quality.
In one embodiment, optical filter 35 is optical filter, for preventing visible light by making and infrared camera pair The light beam for the range of wavelengths answered passes through.For example, when infrared camera is for when the infrared light of 840nm is imaged, optical filter to be only Infrared light is allowed to pass through, without allowing visible light to pass through.
In one embodiment, optical filter 35 is another special optical filter, and this special optical filter is to visible The transmitance of light is lower, and higher to the transmitance of infrared light, for example is 10%~50% to the transmitance of visible light, and to red The transmitance of outer light is 60~99%, and external environmental light after this special optical filter by being irradiated in optics module, through light After learning the reflection of mould group, again through this special optical filter a possibility that substantially reduce, to reach below hiding display screen the One, the effect of two infrared cameras.
It is the situation of optical filter for optical filter 35, when a kind of preferable implementation, the first and second infrared camera only includes Imaging sensor (such as CCD or CMOS) and imaging len, without being reconfigured optical filter, with inside traditional infrared camera The mode of image sensor surface setting optical filter is compared, and cost and manufacture difficulty can be further reduced.
It is understood that above several optical filters are not limitation of the invention, any similar functions that may be implemented Optical filter can be used in the present invention.
It is understood that optical filter 35 can be independent optical device, be also possible to the first and second infrared camera or Person's cover board combines, for example when optical filter 35 is film, can be arranged in the form of plated film in cover board or surface optical device Optical filter.
Fig. 4 is the electronic equipment schematic cross-section of another embodiment according to the present invention, which includes first red Outer camera 43, the second infrared camera 44, display screen and cover board 41.Unlike embodiment illustrated in fig. 3, display screen includes extremely Few three parts: the first sub- display screen 421, the second sub- display screen 422 and the sub- display screen 423 of third, first camera 43 and the Two cameras 44 are separately positioned on the back side (or following) of the first sub- display screen 421 with the sub- display screen 423 of third, further to be promoted Shield accounting.
In view of display screen intercycle property structure will affect the image quality of camera, (for example beam diffraction will affect imaging Quality), the first sub- display screen 421 is configured to the performance different with the second sub- display screen 422 from the sub- display screen 423 of third.? In one embodiment, the first sub- display screen 421 and the sub- display screen 423 of third are transparent display screen, such as by all-transparent material Manufactured OLED display screen.At this point, the second sub- display screen 422 can be that transparent display screen is also possible to nontransparent display screen, Generally, the cost of transparent display screen therefore can in one embodiment higher than nontransparent display screen, the second sub- display screen 422 are configured to nontransparent display screen to reduce equipment cost.
In one embodiment, the first sub- display screen 421 is also arranged as different performances from the sub- display screen 423 of third, For example the high resolution of first infrared camera 43 corresponding with the first sub- display screen 421 is in the sub- display screen 423 of third corresponding The resolution ratio of two infrared cameras 44, generally, the image quality of the first infrared camera 43 are also higher than the second infrared camera 44, are This, is set below the sub- display screen 423 of third for the pixel resolution of the first sub- display screen 421, and/or, display aperture opening ratio is set It is set to and is higher than the sub- display screen 423 of third.
It is understood that the setting of the first, second and third sub- display screen is not limited to above embodiments, in corresponding camera Under the specific requirement of imaging, each sub- display screen can be configured accordingly.
In one embodiment, electronic equipment further includes optical filter (not shown), and setting is in first camera 43 and the Between one sub- display screen 421, and/or it is arranged between second camera 44 and the sub- display screen 423 of third, and be configured for subtracting Few visible light-transmissive promotes view so that external user cannot observe directly first camera 43 and/or second camera 44 Feel aesthetic feeling.Explanation in Fig. 3 about optical filter is equally applicable to the present embodiment.
For embodiment shown in Fig. 3,4, wherein infrared camera generally requires to carry out active infrared optical illumination in imaging, Therefore relevant electronic equipment also includes the infrared light supplies such as light emitting mould group, such as infrared floodlight, infrared projector.Light hair It penetrates mould group and is arranged below display screen and coupled with infrared camera, also be can be set in display screen aperture, or and display screen Independent to separate, positioned at other positions of electronic equipment, when infrared camera work, infrared light supply, which is synchronized, is opened.When camera or When light source is arranged below display screen, in order to promote imaging or lighting quality as far as possible, in display screen camera below or (such as the first and third son display of corresponding with camera or light source pixel region on display screen in imaging perhaps illumination of person's light source Screen) processing closed state.In one embodiment, the imaging of camera or light source or illumination frequencies just with the refresh frame of display screen Rate is identical, and the two mutually staggers, will not interfere with each other.
Electronic equipment described in the various embodiments described above contains at least two infrared cameras, the available target for having parallax Infrared image, is based on the electronic equipment, and the present invention also will be provided for the method for realizing recognition of face.
Shown in fig. 5 is face identification method schematic diagram according to an embodiment of the invention.Firstly, such as step S501, It include the first infrared (IR) image and the second infrared (IR) figure of face using the first infrared camera and the acquisition of the second infrared camera Picture, the first IR image and the 2nd IR image are then transferred to processor to execute recognition of face.Secondly, processor is to including people First IR image of face and the 2nd IR image carry out Face datection, and such as step 502, there are many modes for realizing Face datection, than The first IR image and IR are schemed such as eigenface, neural network et al. face detection algorithm as in one embodiment, being utilized respectively As progress Face datection, or Face datection, subsequent root only can be carried out to one of the first IR image and the 2nd IR image Determine that the face location in another image, the purpose of Face datection are found out in image according to the positional relationship between two cameras Human face region, according to different face recognition algorithms needs, the output of Face datection can also there are many, for example only include face The square pixels region of main feature (such as eyes, nose and mouth), or after the human face region that will test out is normalized The image of output, or be the curved profile area image etc. comprising face.It is understood that the face datection step The result of output is not limited to above several situations.
Again, such as step 503, the facial image in the first IR image that will test and the 2nd IR image is input to nerve For network model to carry out recognition of face, i.e. processor will be further to the people in the first IR image detected and the 2nd IR image Face image identified, recognition of face task includes to judge whether current face is consistent with the face pre-saved or find out to work as Preceding face is the tasks such as one in the face database pre-saved.Herein by housebroken neural network model to the first IR Image and the 2nd IR image directly carry out recognition of face, unlike traditional approach, do not need first to scheme the first IR here As carrying out disparity computation with the 2nd IR image to obtain the depth information of target.The neural network model can be to the first IR image Face characteristic extraction and recognition of face are carried out with the 2nd IR image, and needs training in advance.It such as in the training process, will be right (" false face " described herein is primarily referred to as the two-dimensional person in photo or picture for multiple real human faces and multiple false faces Face) the first IR image and the 2nd IR image are acquired respectively, and (real human face or false face) is labeled to these images; Then these images are input in neural network model, the nerve net to be trained to parameters in model, after training Network model will have three-dimensional judgement and two kinds of functions of recognition of face to the first IR image of input and the 2nd IR image.Then In the recognition of face stage, the facial image in the first IR image currently obtained and the 2nd IR image is input to neural network To carry out recognition of face in model.It is understood that due to the two images of the same target used in the training stage, for Pixel in real human face two images is distinguished in the presence of parallax relevant to 3D shape, and for the false people of planar picture etc. Face, there will be no parallaxes relevant to 3D shape to distinguish between the pixel in two images, therefore, with traditional single-view image Trained neural network model is compared, and the neural network model trained here will be provided with three-dimensional detection function.
In one embodiment, neural network model also may include the first sub-neural network model and the second sub- nerve net Network model, wherein whether the first sub-neural network model is for realizing being vertical to face in the first IR image and the 2nd IR image Body face is detected, and the second sub-neural network model is known for realizing to the face of first and/or the 2nd in IR image Not.First sub-neural network model the training stage input be the first IR image being collected in advance, the 2nd IR image and Whether be real human face markup information;And what the second sub-neural network model inputted in the training stage is the first IR image and At least one of two IR images.In cognitive phase, will first with the first sub-neural network model to the first IR image and Whether the face in the 2nd IR image is that real human face (whether being stereoscopic face) is detected, if testing result is true people Face just will do it the recognition of face of next step, i.e., using the second sub-neural network model in the first IR image and the 2nd IR image At least one carry out recognition of face.That is the face figure in the first IR image and the 2nd IR image that will test in step S503 As being input in neural network model to carry out recognition of face;The step is again the following steps are included: firstly, the first IR image of detection And whether the 2nd face in IR image is stereoscopic face;Then, in the first IR image and/or the 2nd IR image Face is identified.
Relevant instruction finally is executed further according to recognition result, for example face unlock application or recognition of face payment are answered With if identifying, current face is consistent with pre-stored face, will execute unlock or payment instruction.
Above is carried out in the method for recognition of face using the first IR image and the 2nd IR image, due to not needing to calculate Depth image, and disparity computation needs to consume biggish computing resource and causes speed slow, so as to save computing resource with Promote user experience.
Electronic equipment provided by the present invention and based on face identification method performed by the electronic equipment, and it is traditional Device and method are compared, and at least possess following the utility model has the advantages that firstly, screen accounting is lower.Due to by infrared camera, infrared light supply etc. Device is arranged under screen, as big as possible while realizing recognition of face to improve screen accounting.Then recognition efficiency is higher.Directly It connects and two width IR images is subjected to recognition of face, need not move through the biggish algorithmic procedure of this calculation amount of disparity computation, it can be substantially Degree promotes recognition of face efficiency, promotes user experience.
In the description of the present invention, it should be noted that term " center ", " longitudinal direction ", " transverse direction ", "upper", "lower", The orientation or positional relationship of the instructions such as "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom" "inner", "outside" is It is based on the orientation or positional relationship shown in the drawings, is merely for convenience of description of the present invention and simplification of the description, rather than instruction or dark Show that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore should not be understood as pair Limitation of the invention.In addition, term " first ", " second " are used for description purposes only, it is not understood to indicate or imply opposite Importance.
In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected;It can To be mechanical connection, it is also possible to be electrically connected;It can be directly connected, can also can be indirectly connected through an intermediary Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood as the case may be Concrete meaning in the present invention.
It should be noted that it can directly on the other element when element is referred to as " being fixed on " another element Or there may also be elements placed in the middle.When an element is considered as " connection " another element, it, which can be, is directly connected to To another element or it may be simultaneously present centering elements.Term as used herein " vertical ", " horizontal ", " left side ", " right side " and similar statement for illustrative purposes only, are not meant to be the only embodiment.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention The normally understood meaning of technical staff is identical.Term as used herein in the specification of the present invention is intended merely to description tool The purpose of the embodiment of body, it is not intended that in the limitation present invention.Each technical characteristic of embodiment described above can carry out arbitrarily Combination, for simplicity of description, it is not all possible to each technical characteristic in above-described embodiment combination be all described, so And as long as there is no contradiction in the combination of these technical features, it all should be considered as described in this specification.
It may be noted that each step/component described in this application can be split as more multistep according to the needs of implementation Two or more step/components or the part operation of step/component can also be combined into new step/component by suddenly/component, To achieve the object of the present invention.
Above-mentioned face identification method according to the present invention can be realized in hardware, firmware, or be implemented as being storable in Software or computer code in recording medium (such as CD ROM, RAM, floppy disk, hard disk or magneto-optic disk), or be implemented through The original storage of network downloading is in long-range recording medium or nonvolatile machine readable media and will be stored in local record Jie Computer code in matter, so that method described herein can be stored in using general purpose computer, application specific processor or can Such software processing in the recording medium of programming or specialized hardware (such as ASIC or FPGA).It is appreciated that computer, place Managing device, microprocessor controller or programmable hardware includes the storage assembly (example that can store or receive software or computer code Such as, RAM, ROM, flash memory etc.), it is real when the software or computer code are by computer, processor or hardware access and execute Existing processing method described herein.In addition, when general purpose computer accesses the code for realizing the processing being shown here, code Execution general purpose computer be converted to be used to execute the special purpose computer of processing being shown here.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that Specific implementation of the invention is only limited to these instructions.For those skilled in the art to which the present invention belongs, it is not taking off Under the premise of from present inventive concept, several equivalent substitute or obvious modifications can also be made, and performance or use is identical, all answered When being considered as belonging to protection scope of the present invention.

Claims (10)

1. a kind of efficient face identification method, which comprises the following steps:
S1: the first IR image and the 2nd IR image comprising face are received;
S2: the face in detection the first IR image and the 2nd IR image;
S3: the facial image in the first IR image that will test and the 2nd IR image is input to the nerve trained in advance Network model is to carry out recognition of face;
S4: relevant instruction is executed according to recognition result.
2. face identification method according to claim 1, which is characterized in that the step S3 includes following sub-step:
S31: whether the face in the first IR image of detection and the 2nd IR image is stereoscopic face;
S32: the face in the first IR image and/or the 2nd IR image is identified.
3. face identification method according to claim 1, which is characterized in that mistake trained in advance described in the step S3 Journey are as follows: to more than two real human faces and more than two false faces acquire the first IR image respectively and the 2nd IR schemes Picture, and these images are labeled;Then these images are input in neural network model, to parameters in model It is trained, so that the neural network model after training will have solid and sentence to the first IR image of input and the 2nd IR image Disconnected and two kinds of functions of recognition of face.
4. face identification method according to claim 3, which is characterized in that the neural network model includes the first son mind Through network model and the second sub-neural network model, in the training stage, the first IR image that is collected in advance, the 2nd IR image with It and whether is that the markup information of real human face is input to the first sub-neural network model;The first IR being collected in advance is schemed Picture and/or the 2nd IR image are input to the second sub-neural network model;The first sub-neural network model is for detecting Whether facial image to be measured is stereoscopic face, and the second sub-neural network model is used to carry out people to the facial image to be measured Face identification.
5. a kind of electronic equipment characterized by comprising
Shell;
It is arranged in positive display screen, first camera, second camera, light emitting mould group, memory and processor;
The first camera and second camera are used to acquire infrared image of the external people or object under Infrared irradiation, described Light emitting mould group is for providing infrared illumination;
The first camera and second camera form binocular depth camera, and the first camera, second camera, light emitting mould group are set It sets at the back side of the display screen;The memory storage has program;The processor runs described program, described in control First camera and the second camera acquire the first IR image and the 2nd IR image comprising face respectively, and described in control Electronic equipment executes the face identification method as described in claim 1-4 is any.
6. electronic equipment according to claim 5, which is characterized in that the field angle of the first camera and second camera, Resolution ratio or focal length are different.
7. electronic equipment according to claim 5, which is characterized in that the display screen includes the first sub- display screen, second Sub- display screen and the sub- display screen of third, the first camera and second camera are separately positioned on the described first sub- display screen and third The back side of sub- display screen.
8. electronic equipment according to claim 7, which is characterized in that the first sub- display screen and the sub- display screen of third are equal For transparent display screen, the resolution ratio of the sub- display screen of third described in the resolution ratio of the first sub- display screen is low, first son The display aperture opening ratio of the display aperture opening ratio display screen than the third of display screen is high.
9. electronic equipment according to claim 5, which is characterized in that between the first camera and display screen, the second phase Optical filter is additionally provided between machine and display screen, for reducing the transmission of visible light.
10. electronic equipment according to claim 9, which is characterized in that the optical filter is optical switch, is located when power is off In non-transparent state, light can not pass through, and pellucidity is in when being powered, and light can pass through;Alternatively, the optical filter is Single-way perspective membrane, a side plane of the single-way perspective membrane towards optics module are less than reflectivity for the transmitance of visible light, A side plane towards display screen is greater than reflectivity for the transmitance of infrared light, so that it is aobvious to allow external infrared light to pass through Display screen enters optics module and internal light is prevented to pass through display screen;Alternatively, the optical filter is optical filter, it is visible for preventing Light is passed through by making with the light beam in second camera corresponding wavelength section;And when the optical filter is optical filter, The second camera only includes imaging sensor and imaging len.
CN201811628411.XA 2018-12-28 2018-12-28 Efficient face identification method and electronic equipment Pending CN109766806A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811628411.XA CN109766806A (en) 2018-12-28 2018-12-28 Efficient face identification method and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811628411.XA CN109766806A (en) 2018-12-28 2018-12-28 Efficient face identification method and electronic equipment

Publications (1)

Publication Number Publication Date
CN109766806A true CN109766806A (en) 2019-05-17

Family

ID=66452213

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811628411.XA Pending CN109766806A (en) 2018-12-28 2018-12-28 Efficient face identification method and electronic equipment

Country Status (1)

Country Link
CN (1) CN109766806A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111079673A (en) * 2019-12-20 2020-04-28 四川长虹电器股份有限公司 Near-infrared face recognition method based on naked eye three-dimension
CN114419718A (en) * 2022-03-10 2022-04-29 荣耀终端有限公司 Electronic equipment and face recognition method
WO2023124237A1 (en) * 2021-12-29 2023-07-06 荣耀终端有限公司 Image processing method and apparatus based on under-screen image, and storage medium

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101231450A (en) * 2008-02-25 2008-07-30 陈伟山 Multipoint and object touch panel arrangement as well as multipoint touch orientation method
CN102346991A (en) * 2010-07-30 2012-02-08 鸿富锦精密工业(深圳)有限公司 Display device with photographing function
WO2017049923A1 (en) * 2015-06-17 2017-03-30 广州市巽腾信息科技有限公司 Multi-function mobile image processing apparatus, processing method, and application
CN106599797A (en) * 2016-11-24 2017-04-26 北京航空航天大学 Infrared face identification method based on local parallel nerve network
CN106921767A (en) * 2017-03-07 2017-07-04 捷开通讯(深圳)有限公司 A kind of mobile terminal of screen accounting high
CN107066942A (en) * 2017-03-03 2017-08-18 上海斐讯数据通信技术有限公司 A kind of living body faces recognition methods and system
CN107609383A (en) * 2017-10-26 2018-01-19 深圳奥比中光科技有限公司 3D face identity authentications and device
CN107911554A (en) * 2017-11-22 2018-04-13 广东欧珀移动通信有限公司 The control method of electronic device and electronic device
CN108108704A (en) * 2017-12-28 2018-06-01 努比亚技术有限公司 Face identification method and mobile terminal
CN108667968A (en) * 2018-05-16 2018-10-16 维沃移动通信有限公司 A kind of display module and mobile terminal
CN108810201A (en) * 2018-06-04 2018-11-13 Oppo广东移动通信有限公司 Electronic device and the method for shooting photo using electronic device
CN109068043A (en) * 2018-09-27 2018-12-21 维沃移动通信有限公司 A kind of image imaging method and device of mobile terminal

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101231450A (en) * 2008-02-25 2008-07-30 陈伟山 Multipoint and object touch panel arrangement as well as multipoint touch orientation method
CN102346991A (en) * 2010-07-30 2012-02-08 鸿富锦精密工业(深圳)有限公司 Display device with photographing function
WO2017049923A1 (en) * 2015-06-17 2017-03-30 广州市巽腾信息科技有限公司 Multi-function mobile image processing apparatus, processing method, and application
CN106599797A (en) * 2016-11-24 2017-04-26 北京航空航天大学 Infrared face identification method based on local parallel nerve network
CN107066942A (en) * 2017-03-03 2017-08-18 上海斐讯数据通信技术有限公司 A kind of living body faces recognition methods and system
CN106921767A (en) * 2017-03-07 2017-07-04 捷开通讯(深圳)有限公司 A kind of mobile terminal of screen accounting high
CN107609383A (en) * 2017-10-26 2018-01-19 深圳奥比中光科技有限公司 3D face identity authentications and device
CN107911554A (en) * 2017-11-22 2018-04-13 广东欧珀移动通信有限公司 The control method of electronic device and electronic device
CN108108704A (en) * 2017-12-28 2018-06-01 努比亚技术有限公司 Face identification method and mobile terminal
CN108667968A (en) * 2018-05-16 2018-10-16 维沃移动通信有限公司 A kind of display module and mobile terminal
CN108810201A (en) * 2018-06-04 2018-11-13 Oppo广东移动通信有限公司 Electronic device and the method for shooting photo using electronic device
CN109068043A (en) * 2018-09-27 2018-12-21 维沃移动通信有限公司 A kind of image imaging method and device of mobile terminal

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
周羽: "红外图像人脸识别研究", 《信息科技辑》, no. 2008, pages 138 - 720 *
都伊林,: "智能安防新发展与应用", 《华中科技大学出版社》, pages: 52 - 54 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111079673A (en) * 2019-12-20 2020-04-28 四川长虹电器股份有限公司 Near-infrared face recognition method based on naked eye three-dimension
WO2023124237A1 (en) * 2021-12-29 2023-07-06 荣耀终端有限公司 Image processing method and apparatus based on under-screen image, and storage medium
CN114419718A (en) * 2022-03-10 2022-04-29 荣耀终端有限公司 Electronic equipment and face recognition method
CN114419718B (en) * 2022-03-10 2022-08-02 荣耀终端有限公司 Electronic equipment and face recognition method

Similar Documents

Publication Publication Date Title
CN109840475A (en) Face identification method and electronic equipment
CN209328043U (en) With the electronic equipment shielded comprehensively
CN209170442U (en) Electronic equipment with high screen accounting
CN109544618B (en) Method for obtaining depth information and electronic equipment
EP3872658B1 (en) Face recognition method and electronic device
CN103106401B (en) Mobile terminal iris recognition device with human-computer interaction mechanism
WO2019109768A1 (en) Task execution method, terminal device and computer readable storage medium
CN107113415B (en) The method and apparatus for obtaining and merging for more technology depth maps
CN105122121B (en) Pairs of contact lenses, contact lenses and storage medium
WO2017049922A1 (en) Image information collection apparatus, image collection method, and application thereof
CN109766806A (en) Efficient face identification method and electronic equipment
CN207802143U (en) depth imaging mobile terminal
CN106843491A (en) Smart machine and electronic equipment with augmented reality
CN109068043A (en) A kind of image imaging method and device of mobile terminal
CN103049744A (en) User recognition device for access control
CN103839054A (en) Multi-functional mobile intelligent terminal sensor supporting iris recognition
CN109788174A (en) A kind of light compensation method and terminal
CN109471263A (en) A kind of Gait Recognition intelligent glasses
CN103150565A (en) Portable two-eye iris image acquisition and processing equipment
CN105320941A (en) Biometric identification apparatus based on fusion of iris and human face and biometric identification method using apparatus
CN113260949A (en) Method for reducing power consumption and electronic equipment
CN108415875A (en) The method of Depth Imaging mobile terminal and face recognition application
CN206921118U (en) Double-wavelength images acquisition system
WO2022068193A1 (en) Wearable device, intelligent guidance method and apparatus, guidance system and storage medium
CN106408695A (en) Apparatus and method for iris identification of intelligent lock

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
CB02 Change of applicant information
CB02 Change of applicant information

Address after: 11-13 / F, joint headquarters building, high tech Zone, 63 Xuefu Road, Yuehai street, Nanshan District, Shenzhen, Guangdong 518000

Applicant after: Obi Zhongguang Technology Group Co., Ltd

Address before: 12 / F, joint headquarters building, high tech Zone, 63 Xuefu Road, Nanshan District, Shenzhen, Guangdong 518000

Applicant before: SHENZHEN ORBBEC Co.,Ltd.