CN108470169A - Face identification system and method - Google Patents
Face identification system and method Download PDFInfo
- Publication number
- CN108470169A CN108470169A CN201810500469.XA CN201810500469A CN108470169A CN 108470169 A CN108470169 A CN 108470169A CN 201810500469 A CN201810500469 A CN 201810500469A CN 108470169 A CN108470169 A CN 108470169A
- Authority
- CN
- China
- Prior art keywords
- face
- identification
- facial image
- recognition
- measurand
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 230000001815 facial effect Effects 0.000 claims abstract description 120
- 238000004891 communication Methods 0.000 claims abstract description 10
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 19
- 210000003128 head Anatomy 0.000 claims description 11
- 210000001061 forehead Anatomy 0.000 claims description 7
- 210000001508 eye Anatomy 0.000 claims description 5
- 210000004709 eyebrow Anatomy 0.000 claims description 5
- 210000005252 bulbus oculi Anatomy 0.000 claims 1
- 238000000605 extraction Methods 0.000 description 10
- 239000011148 porous material Substances 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000012706 support-vector machine Methods 0.000 description 6
- 238000012795 verification Methods 0.000 description 6
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 210000000887 face Anatomy 0.000 description 5
- 230000009977 dual effect Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000009432 framing Methods 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 210000004209 hair Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/30—Scenes; Scene-specific elements in albums, collections or shared content, e.g. social network photos or video
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/24—Speech recognition using non-acoustical features
- G10L15/25—Speech recognition using non-acoustical features using position of the lips, movement of the lips or face analysis
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of face identification system and methods, the face identification system includes recognition of face terminal and server, the recognition of face terminal is connected with the server communication, the recognition of face terminal includes display panel, image capture module and identification module, the camera of described image acquisition module is mounted on the top of the display panel, the camera is used to acquire the facial image of measurand, and collected facial image is sent to the identification module, the identification module is used to carry out vivo identification to the measurand, and the facial image is compared with the facial image that prestores in face database in the case where identifying that measurand is live body, obtain face recognition result；The identification module includes the first recognition unit and the second recognition unit.Face skin characteristic is detected and is combined with speech detection by the present invention, be can effectively solve the problem that vivo identification fraud problem, is improved the accuracy rate of recognition of face.
Description
Technical field
The present invention relates to technical field of face recognition, more particularly to a kind of face identification system and method.
Background technology
With the rapid development of face recognition technology, recognition of face is in fields and law enforcement agencies such as business, education
It is used widely, such as recognition of face can help that bank is more acurrate, effectively verifies client identity.But concrete application
In, since the influence of the factors such as light can cause collected picture quality not high, influence face recognition accuracy rate.
In addition, measurand may use photo face or the face video segment prerecorded to carry out recognition of face,
Cause recognition of face efficiency low.In view of this, having also been proposed vivo identification technology in the prior art, i.e., in the process of recognition of face
In prove facial image it is corresponding be " living person ".Common vivo identification coordinates for random action in the market, generally requires quilt
It surveys object and makes the random actions progress vivo identification identification such as shake the head, blink, opening one's mouth.However, the vivo identification method there is also
Certain security threat, such as measurand can be imitated true man using three-dimensional face model and complete compulsory exercise to carry out
It forges and logs in, it is easy to falsely determine that non-living body for live body.
Invention content
The purpose of the present invention is at least solving one of drawbacks described above and deficiency, which is achieved through the following technical solutions
's.
The present invention provides a kind of face identification system, including recognition of face terminal and server, the recognition of face is whole
End is connected with the server communication, and the recognition of face terminal includes display panel, image capture module and identification module,
Described image acquisition module is electrically connected with the identification module, and the camera of described image acquisition module is mounted on the display surface
The top of plate, the camera are used to acquire the facial image of measurand, and collected facial image are sent to described
Identification module, the identification module is used to carry out vivo identification to the measurand, and is identifying that measurand is live body
In the case of the facial image is compared with the facial image that prestores in face database, obtain face recognition result；
The identification module includes the first recognition unit and the second recognition unit, and first recognition unit is skin-identification unit, is used
According to the skin characteristic of facial image progress vivo identification；Second recognition unit is voice recognition unit, is used for
Further vivo identification is carried out according to the lip reading information that measurand is sent out.
Further, the identification module further includes authentication unit, and the authentication unit is for identifying measurand
The facial image is compared with the facial image that prestores in face database in the case of for live body, obtains recognition of face
As a result.
Further, the camera includes visible image capturing head and/or black light camera.
Further, second recognition unit includes loud speaker and the Mike being arranged in the recognition of face terminal end surface
Wind is provided with 2 microphones.
The present invention also provides a kind of face identification method, the recognition of face detection method is according to above-mentioned recognition of face
Come what is implemented, specific steps include detecting system：
S1：It obtains the facial image of measurand and the facial image detected is marked；
S2：Vivo identification is carried out to the facial image that detects according to identification model, judge the facial image whether be
Live body, if so, thening follow the steps S3；If it is not, then sending vivo identification failure information to the display panel, terminates face and know
Not；
S3：The face characteristic of the facial image is compared with the facial image that prestores in standard database, is determined
Face recognition result.
Further, the step S2 includes：
Step 201：After the facial image for obtaining measurand, using the facial image detected as the first identification mould
The input of type carries out vivo identification to the facial image according to the first identification model, judges whether the facial image is living
Body executes the step S3 if so, vivo identification passes through；Otherwise, step 202 is executed；
Step 202：Using the facial image as the input of the second identification model, according to the second identification model to the people
Face image carries out vivo identification, judges whether the facial image is live body, if so, vivo identification passes through, executes the step
Rapid S3；Otherwise, vivo identification failure information is sent to the display panel, terminates recognition of face.
Further, the step 201 includes the skin characteristic of the acquisition facial image different parts, and by described in not
Skin characteristic with position is compared, if the similarity of obtained skin characteristic is less than preset similarity threshold, then, institute
It is live body to state facial image；If the similarity of obtained skin characteristic is more than preset similarity threshold, the facial image
For non-living body.
Further, the skin characteristic include in forehead, eyebrow, eyes, two cheeks, nose, lip, chin, ear extremely
The skin characteristic at few two positions.
Further, the step 202 sends out phonetic order, measurand root including the second identification model to measurand
Corresponding random digit combination password is read according to the phonetic order and carries out vivo identification, if the random digit that measurand is read
The standard lip characteristics of image that the lip characteristics of image of combination is combined with the random digit of second identification model matches, then
Determine that measurand is live body.
Further, first identification model and second identification model are obtained by training in advance.
Advantages of the present invention is as follows：
(1) present invention using Multiple recognition carry out recognition of face, and by face skin characteristic identification and speech recognition phase
In conjunction with vivo identification is carried out, it can effectively solve the problem that and carry out vivo identification using photo, video and three-dimensional face model etc. and hold
The fraud problem being also easy to produce reaches pinpoint accuracy vivo identification and the function of recognition of face, effectively improves the safety of recognition of face
Property.
(2) present invention is trained by using convolutional neural networks and obtains corresponding identification model, then will directly wait for
The facial image of identification, which is input in identification model, can be realized vivo identification, and user experience is good, highly practical.
(3) present invention can be acquired using the dual camera being made of visible light and black light under different light
Facial image obtains the image of high quality, improves face recognition accuracy rate.
Description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit are common for this field
Technical staff will become clear.Attached drawing only for the purpose of illustrating preferred embodiments, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.
Fig. 1 is the schematic diagram of face identification system provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram of recognition of face terminal provided in an embodiment of the present invention；
Fig. 3 is the concrete structure schematic diagram of recognition of face terminal provided in an embodiment of the present invention；
Fig. 4 is the flow chart of face identification method provided in an embodiment of the present invention；
Fig. 5 is the flow chart of the vivo identification of face identification method provided in an embodiment of the present invention；
Reference numeral is as follows in figure：
100- recognition of face terminal 200- servers
300- terminal devices
1- display panel 2- image capture modules
21- visible image capturing head 22- infrared cameras
3- identification modules the first recognition units of 31-
32- the second recognition unit 33- authentication units
321- loud speaker 322- microphones
Specific implementation mode
The illustrative embodiments of the disclosure are more fully described below with reference to accompanying drawings.Although showing this public affairs in attached drawing
The illustrative embodiments opened, it being understood, however, that may be realized in various forms the disclosure without the reality that should be illustrated here
The mode of applying is limited.It is to be able to be best understood from the disclosure on the contrary, providing these embodiments, and can be by this public affairs
The range opened completely is communicated to those skilled in the art.
Fig. 1 shows the schematic diagram of the face identification system provided according to the embodiment of the present invention.As shown in Figure 1, should
Face identification system includes recognition of face terminal 100 and server 200,200 communication link of recognition of face terminal 100 and server
It connects.Server 200 can be independent server, or can store the server set of high in the clouds data.
Fig. 2 to Fig. 3 shows the structural schematic diagram of the recognition of face terminal provided according to the embodiment of the present invention.Such as figure
2 to shown in Fig. 3, and recognition of face terminal 100 includes display panel 1, image capture module 2 and identification module 3, identification module 3
It is electrically connected respectively with display panel 1 and image capture module 2.
Image capture module 2 includes camera, and camera is arranged at the top of display panel 1, and camera is wide-angle imaging
Head, camera can acquire continuous facial image from the distant to the near, and are sent to identification module 3 and carry out recognition of face.Camera shooting
Head can automatically identify at least one of picture to be captured face, and then the people to detecting when carrying out personage's shooting
Face is automatically adjusted, such as automatic focusing, adjust automatically image zoom or diminution etc..
The light source of camera can be visible light source, such as feux rouges or blue light etc.；Can also be black light light source, example
Such as infrared light supply；It can also be that the dual camera being made of visible light and black light light source, the present invention do not limit specifically.Such as
In present embodiment, camera uses the dual camera being made of visible image capturing head 21 and infrared camera 22.
Infrared camera 22 can capture the infrared image that each subject diffusing reflection is formed in picture to be captured, due to
Human eye can not see infrared ray, therefore, can be to avoid the injury to measurand using infrared camera 22.And night uses
When infrared collecting, infrared camera 22 can adjust automatically exposure intensity, improve shooting quality.
When measurand station is when from recognition of face 100 a certain distance of terminal, camera carries out image to user and adopts
Collection, when face alignment camera, camera can capture, and clear, positive facial image obtains effective face figure
Picture.In preferred implementation, above-mentioned distance is at least 10cm.In image acquisition process, frame image buffer storage preservation is at least acquired.It is tested
Object can adjust oneself camera site, shooting distance etc. according to the preview screen shown on display panel 1.
Identification module 3 is used to according to collected facial image judge whether the personage in image is live body, and true
It is after the fixed facial image is live body, the face of the facial image stored in collected face characteristic and face database is special
Verification is compared in sign, judges whether to match, and obtains face recognition result and is shown in display panel 1.
Identification module 3 includes the first recognition unit 31 and the second recognition unit 32, and the first recognition unit 31 identifies for skin
Unit can judge whether the face in facial image is live body according to the skin characteristic of collected facial image frame.
Face characteristic includes the skin of face feature of face, and skin of face feature includes dermatoglyph feature and/or skin
Pore feature.Dermatoglyph feature includes the main features such as the depth, number of nodes and the texture thickness of dermatoglyph, and skin pore is special
Sign includes the quantity of pore and the size etc. of pore.
Since the skin of face feature of real human face different parts is different, for example, real human face forehead and chin position line
Reason is compared with thick, pore is larger, and texture is relatively thin at cheek, pore is smaller rather than real human face is by shooting image, does not have
Texture and pore of real skin etc..Therefore, it is possible to by the skin of face feature differentiation real skin of different parts with it is non-real
Real skin, to carry out vivo identification.
The skin characteristic of the facial image extracted includes：The dermatoglyph feature and skin pore feature of different parts
At least one of.Different parts include at least 2 extracted in forehead, eyebrow, eyes, two cheeks, nose, lip, chin, ear
A position.
Second recognition unit 32 is voice recognition unit, can carry out live body knowledge according to the lip reading information that facial image is sent out
Not, judge whether the face in facial image is live body.
Voice recognition unit includes the loud speaker 321 and microphone 322 being arranged on 100 surface of recognition of face terminal, is raised one's voice
For device 321 for playing voice, microphone 322 is used to receive extraneous voice signal, in present embodiment, if there are two microphones
322, form two wheat linear arrays, can it is significantly more efficient inhibit noise, echo interference, greatly improve the sensitive of phonetic incepting
The accuracy rate of degree and speech recognition.
Voice recognition unit can send out stochastic instruction, prompt user to read number combination providing with the machine, carry out voice
Detection identification.Number is combined as the number combination randomly generated in 1-9.Three random numbers are included at least in random digit combination
Word.
Second recognition unit 32 in 31 None- identified of the first recognition unit for going out when whether measurand is live body to start
Identification.The at most identification of second recognition unit 32 three times, if prompting vivo identification to fail after recognition failures three times, and by live body
Recognition result is sent to the display of display panel 1, and the prompt of such as " non-living body " is sent out by loud speaker 321, exits recognition of face
Program.
Loud speaker 321 can be also used for entire recognition of face terminal 100, sends out such as " please be directed at camera lens ", " verifies logical
Cross ", the prompt tone of " authentication failed " etc..
Vivo identification is carried out compared to requiring measurand to make the random actions such as shake the head, blink, open one's mouth, using actively matching
The random digit speech recognition of conjunction, it is ensured that the reliability of system, and anti-attack ability is improved, ensure the accurate of vivo identification
Property.
Identification module 3 further includes authentication unit 33, for determine measurand be live body after, will it is collected be tested pair
The face characteristic of elephant is compared with the face characteristic in standard database, and face verification is carried out to the facial image.
Standard database includes local data base and remote data base, and the data in local data base are stored in recognition of face
In the memory module of terminal 100, the data in remote data base are stored in server 200.
When recognition of face is verified, authentication unit 33 calls local data base first, if existing subscriber's data, according to guarantor
The user's facial feature information deposited, and the facial image feature of the measurand of input are compared, if similarity is higher than threshold
Value is then identified and is verified.If local useless user data, will be in the facial image feature of measurand by server 200
It reaches remote data base and verification is identified；Authentication unit 33 calls the identity stored in remote data base by server 200
The facial image feature of information and measurand is compared, and identification verification result is sent to display panel 1 and is shown, if
Similarity is then identified by more than threshold value, otherwise identifies authentication failed.
It in one embodiment, such as company's access control system, can be by brush face mode come the discrepancy of controllers, Yuan Gongxu
When entering company, it can stand apart from the position of 100 certain distance of recognition of face terminal, utilize the camera shooting of image capture module 2
Head acquisition user images, and then determine that user schemes according to the first recognition unit 31 of identification module 3 and the second recognition unit 32
Whether the user as in is live body, if it is live body, enters identification proving program in next step, verifies whether as the said firm person
Work allows access into if so, opening the door.
Recognition of face terminal 100 further includes communication module, by wired or wireless between communication module and miscellaneous equipment
Mode connects.Recognition of face terminal 100 can access the wireless network based on communication standard, for example, WiFi, 3G, 4G or it
Combination.In one embodiment, communication module further includes that near-field communication (NFC) unit is used for short range communication.NFC unit can
It is realized based on Bluetooth technology, radio frequency identification (RFID) technology, infrared technique, ultra wide band (UWB) technology and other technologies.
Recognition of face terminal 100 further includes the power supply (such as battery) powered for each component, it is preferable that power supply and each component
It is connected by circuit.Power supply may include one or more direct current or AC power, power failure detection circuit, power supply
The random components such as converter or inverter, power supply status indicator.
Above-mentioned face identification system further includes terminal device 300, terminal device 300 respectively with recognition of face terminal 100 and
Server 200 communicates to connect, and recognition of face can be carried out with remote control recognition of face terminal 100 by terminal device 300.Terminal
Equipment 300 can be one or more in smart mobile phone, tablet computer, PC computers, personal digital assistant (PDA) etc..
It is to be appreciated that recognition of face terminal 100 has a set of corresponding application program matched, in terminal device 300
Also there is a set of corresponding application program matched.
Fig. 4 shows the flow chart of the face identification method provided according to the embodiment of the present invention.As shown in figure 4, should
Face identification method is used for above-mentioned face identification system.Recognition of face terminal 100 can be used for executing the face identification method, face
Identification terminal 100 can also acquire the facial image of measurand in real time, and collected facial image is sent to server
200, which is executed by server 200.
The face identification method implemented according to above-mentioned face identification system, specific steps include：
Step S1, it obtains the facial image of measurand and the facial image detected is marked.
Wherein, it can also be non-living body facial image that facial image, which can be living body faces image,.Non-living body includes face
Existing image, such as the two dimensional image or identity document that are shown on human face photo, screen according to etc..
In specific implementation, facial image can be marked by indicia framing, indicia framing is typically to use rectangle frame, to people
Face is demarcated up to the crown, down toward neck, left and right to the region of ears.
If camera has collected the image for including people, animal and background, animal and background image are invalid images,
In order to get effective facial image, the information in image is detected, identifies label, to obtain in image about people
Or the image of the object of characterization people.In the present embodiment, facial image can only include the image of face facial area.
Specifically, recognition of face terminal 100 is by external camera, under the current visual field of camera, acquisition camera shooting
Image (for example, frame, picture etc.) in head range, by taking frame as an example.Recognition of face terminal 100 can be examined after collecting picture frame
It surveys in the picture frame and if the facial image is marked there are facial image with the presence or absence of facial image, and cache guarantor
It deposits.
The picture frame of acquisition also can be sent to server 200 by recognition of face terminal 100 after collecting picture frame, by
If server 200 is detected in the picture frame and is marked to the facial image there are facial image with the presence or absence of facial image again
Note.
For example, real-name authentication or account complaint, Bank Account Number open an account etc. and to need to carry out authentication in social software
Scene in, need recognition of face terminal 100 and server 200 to coordinate, when measurand is close to recognition of face terminal 100,
The camera of recognition of face terminal 100 acquires the picture frame of real-time scene, obtains measurand under current field range
Facial image, and the facial image is sent to server 200, the facial image is marked by server 200.
Step S2：Vivo identification is carried out to the facial image detected according to identification model, judges that the facial image is
It is no to then follow the steps S3 if live body for live body；If not live body, then send vivo identification failure information to the display surface
Plate terminates recognition of face.
Specifically, recognition of face terminal 100 or server 200 are after obtaining by the facial image of altimetric image, described in extraction
The face characteristic of facial image, and the face characteristic is inputted in the first identification model, whether to identify the facial image
For living body faces image.
First identification model can classify to facial image, if the face characteristic of extraction meets living body faces image
When face characteristic, living body faces image class is classified to the facial image of picture by tested.If the face characteristic of extraction meets non-live
When the face characteristic of body facial image, non-living body facial image class is classified to the facial image of picture by tested.
As shown in figure 5, extracting face characteristic according to identification model and including the step of carrying out vivo identification：
Step 201：After the facial image for obtaining measurand, lived to the facial image according to the first identification model
Body identifies, judges whether the facial image is live body.If the facial image is live body, vivo identification passes through, and executes step
Rapid S3；If the facial image is not live body, thens follow the steps 202 and carry out vivo identification using the second identification model.
First identification model is the depth convolutional neural networks model that training obtains in advance, and convolutional neural networks are artificial god
One kind through network can allow image avoid feature complicated in tional identification algorithm directly as the input of network
Extraction and data reconstruction processes.
Convolutional neural networks include convolutional layer and output layer, and convolutional layer is mainly using trainable convolution kernel come to inputting number
According to progress convolution operation, and by result, form exports in some combination, and essence is the feature extraction to input data.Output layer
It is converted using nonlinear function, so that model is obtained nonlinear characteristic and export-restriction in given range with this, become
Activation primitive.I.e. output layer is used for carrying out specific vivo identification.
After obtaining training sample, disaggregated model is trained, the first identification model can be obtained.Passing through instruction
It gets to after the first identification model, just can be classified using first identification model, judge the facial image of measurand
In skin characteristic be real skin be still non-genuine skin, to realize the vivo identification of facial image.
The acquisition of training sample includes obtaining known real skin image and non-genuine skin image.In training sample
Include the face sample set of multiple classifications, includes multiple facial image samples in the face sample set of each classification, i.e.,
Include the set of multigroup samples such as real human face, photo face, video human face, 3D models face, certificate face in training sample.
Using convolutional layer carry out feature extraction when, it is first determined the skin area in acquired facial image, then from
The skin characteristic of facial image is extracted in identified skin area.
Wherein, the skin characteristic of the facial image extracted includes：The dermatoglyph feature and skin hair of different parts
At least one of hole characteristic.Different parts include in extraction forehead, eyebrow, eyes, two cheeks, nose, lip, chin, ear
It is at least two kinds of.
In one embodiment, the dermatoglyph feature of two cheeks and forehead is extracted as training sample；In another embodiment, example
Such as face head portrait is there are when mask shelter, and the dermatoglyph feature of extraction forehead, place between the eyebrows and canthus is as training sample.
In specific implementation, the quantity of facial image is as more as possible, and the skin characteristic as much as possible for obtaining different parts, training sample
The first identification model that the more how final training of this quantity obtains is more accurate.
Machine learning training is carried out using above-mentioned training sample, obtains the skin characteristic of different parts in each facial image
Likelihood probability the classification of facial image sample ownership is determined according to the likelihood probability, obtain and final be used for vivo identification
Standard identification model.The skin characteristic of different parts in same facial image is compared, if obtained skin characteristic
Similarity be less than preset similarity threshold, then, judge the facial image currently inputted as live body, so that it is determined that passing through work
Body identifies, into next step face verification step；If the similarity of obtained skin characteristic is more than preset similarity threshold, that
, judge that the facial image currently inputted is not live body, enter step 202, carry out further vivo identification.
Step 202：Using the facial image as the input of the second identification model, according to the second identification model to the people
Face image carries out vivo identification, judges whether the facial image is live body.If live body thens follow the steps S3, to the people
Face image carries out verification identification, determines face recognition result；If not live body, then send vivo identification failure information and shown to described
Show panel, terminates recognition of face.
In order to improve the accuracy of vivo identification, when live body being mistaken for non-living body using the first identification model or will be non-
When live body is mistaken for live body, speech recognition is carried out using the second identification model, prevents disabled user from attacking.
Second identification model is speech recognition modeling, and speech recognition modeling sends out phonetic order to measurand, prompts quilt
It surveys object and reads number combination providing with the machine, measurand reads corresponding random digit group according to the phonetic order and heals up
It enables and carries out vivo identification.Wherein number is combined as the number combination randomly generated in 1-9.Random digit combination includes at least three
A random digit.
Second identification model is obtained again by advance training, and the lip image for the random digit that measurand is read is special
The input as the second identification model is levied, is matched with the standard lip characteristics of image in the second identification model of training in advance
It compares, obtains vivo identification result.If the lip characteristics of image and random digit of each random digit that measurand is read
Standard lip characteristics of image matches, it is determined that measurand is live body, and otherwise measurand is non-living body.
Since when carrying out vivo identification, recognition failures may be led to because of odjective cause.Therefore voice is being carried out
When identification, still further comprise：If recognition result be non-living body, prompt measurand to re-start speech recognition, that is, make by
It surveys object and corresponding digital combining random number combination password is read according to the stochastic instruction that the second identification model is sent out；If by
After the speech recognition of preset times, obtained recognition result is still non-living body, then judges that measurand for non-living body, exits whole
The recognition of face program of body.
In specific implementation, the number of speech recognition is at most 3 times.By multiple speech recognition, work can be further promoted
The accuracy of body identification, reduces the interference of enchancement factor.
In another embodiment, when carrying out vivo identification using the first identification model, first to the face figure of measurand
Skin characteristic extraction as carrying out different parts, recycles support vector machines to carry out vivo identification.Specifically include following steps：
Step 2011：After obtaining the facial image that measurand arrives, the skin characteristic in the facial image is extracted.
In specific implementation, the skin characteristic extracted is subjected to related operation, feature vector is calculated；Small echo can be used
Packet and two-dimensional Fourier transform analysis method and mathematics morphological analysis method carry out skin characteristic extraction.
Step 2012：Using the skin characteristic as the input of the first identification model, whether to identify the facial image
For living body faces image S3 is thened follow the steps if live body；If not live body, then carry out live body knowledge using the second identification model
Not.
First identification model is the standard identification model trained using SVM (SVM) algorithm, for example, by using core letter
Number is that support vector machines (SVM) training of Gaussian radial basis function obtains the standard identification model.
It will be sent into branch from the corresponding data scaling of all feature vectors in the skin characteristic extracted in training sample
Vector machine (SVM) is supportted, includes real human face, photo face, video human face, 3D models face, certificate people in the training sample
Whether face etc. is live body according to the skin characteristic different instructions measurand of live subject different parts, and training generates classification and knows
Other model.
Step S3：The face characteristic of the facial image is compared with the facial image that prestores in standard database,
Determine face recognition result.
Face characteristic number of the face characteristic including geometric properties (such as Euclidean distance), algebraic characteristic (eigenmatrix)
According to.By the known face comparison in face and standard database to be identified, matching result is obtained.If successful match, face
Identification is verified；If matching is unsuccessful, recognition of face authentication failed.
The facial image that prestores in standard database can be stored in recognition of face terminal 100, can also storage service
It is called by recognition of face terminal 100 in device 200.
The present invention carries out recognition of face using Multiple recognition, and face skin characteristic is identified and is combined with speech recognition
Vivo identification is carried out, be can effectively solve the problem that and carried out vivo identification using photo, video and three-dimensional face model etc. and be easy production
Raw fraud problem reaches pinpoint accuracy vivo identification and the function of recognition of face, effectively improves the safety of recognition of face.Separately
Outside, the present invention is trained by using convolutional neural networks and obtains corresponding identification model, then directly by people to be identified
Face image, which is input in identification model, can be realized vivo identification, and user experience is good, highly practical.In addition, the present invention use by
The dual camera of visible light and black light composition, can acquire facial image under different light, obtain the figure of high quality
Picture improves face recognition accuracy rate.
It should be pointed out that in the description of the present invention, term " first ", " second " are only used for an entity or behaviour
Make with another entity or operate distinguish, without necessarily requiring or implying between these entities or operation there are it is any this
The actual relationship of kind or sequence.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Any one skilled in the art in the technical scope disclosed by the present invention, the change or replacement that can be readily occurred in,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of the claim
Subject to enclosing.
Claims (10)
1. a kind of face identification system, which is characterized in that including recognition of face terminal and server, the recognition of face terminal and
Server communication connection, the recognition of face terminal include display panel, image capture module and identification module, described
Image capture module is electrically connected with the identification module, and the camera of described image acquisition module is mounted on the display panel
Top, the camera are used to acquire the facial image of measurand, and collected facial image is sent to the identification
Module, the identification module is used to carry out vivo identification to the measurand, and is identifying that measurand is the feelings of live body
The facial image is compared with the facial image that prestores in face database under condition, obtains face recognition result；It is described
Identification module includes the first recognition unit and the second recognition unit, and first recognition unit is skin-identification unit, is used for root
Vivo identification is carried out according to the skin characteristic of the facial image；Second recognition unit is voice recognition unit, is used for basis
The lip reading information that measurand is sent out carries out further vivo identification.
2. face identification system according to claim 1, which is characterized in that the identification module further includes authentication unit,
The authentication unit is used for will be in the facial image and face database in the case where identifying that measurand is live body
The facial image that prestores is compared, and obtains face recognition result.
3. face identification system according to claim 1, which is characterized in that the camera includes visible image capturing head
And/or black light camera.
4. face identification system according to claim 1, which is characterized in that second recognition unit includes being arranged in institute
The loud speaker and microphone for stating recognition of face terminal end surface are provided with 2 microphones.
5. a kind of face identification method, the recognition of face detection method is people according to any one of claim 1 to 4
Face recognition detection system is implemented, which is characterized in that is identified including front end face vivo identification and rear end face alignment, specifically
Step includes：
S1：It obtains the facial image of measurand and the facial image detected is marked；
S2：Vivo identification is carried out to the facial image detected according to identification model, judges whether the facial image is live body,
If so, thening follow the steps S3；If it is not, then sending vivo identification failure information to the display panel, terminate recognition of face；
S3：The face characteristic of the facial image is compared with the facial image that prestores in standard database, determines face
Recognition result.
6. face identification method according to claim 5, which is characterized in that the step S2 includes：
Step 201：After the facial image for obtaining measurand, using the facial image detected as the first identification model
Input carries out vivo identification to the facial image according to the first identification model, judges whether the facial image is live body, if
It is that then vivo identification passes through, executes the step S3；Otherwise, step 202 is executed；
Step 202：Using the facial image as the input of the second identification model, according to the second identification model to the face figure
As carrying out vivo identification, judges whether the facial image is live body, if so, vivo identification passes through, execute the step S3；
Otherwise, vivo identification failure information is sent to the display panel, terminates recognition of face.
7. face identification method according to claim 6, which is characterized in that the step 201 includes obtaining the face
The skin characteristic of image different parts, and the skin characteristic of the different parts is compared, if obtained skin characteristic
Similarity is less than preset similarity threshold, then, the facial image is live body；If the similarity of obtained skin characteristic is big
In preset similarity threshold, then the facial image is non-living body.
8. face identification method according to claim 7, which is characterized in that the skin characteristic includes forehead, eyebrow, eye
The skin characteristic at least two positions in eyeball, two cheeks, nose, lip, chin, ear.
9. face identification method according to claim 6, which is characterized in that the step 202 includes the second identification model
Phonetic order is sent out to measurand, measurand reads corresponding random digit combination password according to the phonetic order and carries out
Vivo identification, if the random number of the lip characteristics of image and second identification model for the random digit combination that measurand is read
The standard lip characteristics of image of word combination matches, it is determined that measurand is live body.
10. face identification method according to claim 6, which is characterized in that first identification model and described second
Identification model is obtained by training in advance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810500469.XA CN108470169A (en) | 2018-05-23 | 2018-05-23 | Face identification system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810500469.XA CN108470169A (en) | 2018-05-23 | 2018-05-23 | Face identification system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108470169A true CN108470169A (en) | 2018-08-31 |
Family
ID=63260445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810500469.XA Pending CN108470169A (en) | 2018-05-23 | 2018-05-23 | Face identification system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108470169A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299677A (en) * | 2018-09-07 | 2019-02-01 | 西安知微传感技术有限公司 | A kind of recognition of face living body judgment method and system |
| CN109359995A (en) * | 2018-11-29 | 2019-02-19 | 中国科学院重庆绿色智能技术研究院 | A kind of dining room payment system based on recognition of face |
| CN109615387A (en) * | 2018-11-29 | 2019-04-12 | 中国科学院重庆绿色智能技术研究院 | A kind of consumption and payment system and method based on recognition of face |
| CN109635021A (en) * | 2018-10-30 | 2019-04-16 | 平安科技（深圳）有限公司 | A kind of data information input method, device and equipment based on human testing |
| CN110082723A (en) * | 2019-05-16 | 2019-08-02 | 浙江大华技术股份有限公司 | A kind of sound localization method, device, equipment and storage medium |
| CN110110582A (en) * | 2019-03-14 | 2019-08-09 | 广州市金其利信息科技有限公司 | In conjunction with the face identification method and system of 3D structure light, infrared light and visible light |
| CN110298230A (en) * | 2019-05-06 | 2019-10-01 | 深圳市华付信息技术有限公司 | Silent biopsy method, device, computer equipment and storage medium |
| CN110427887A (en) * | 2019-08-02 | 2019-11-08 | 腾讯科技（深圳）有限公司 | A kind of membership's recognition methods and device based on intelligence |
| CN110458062A (en) * | 2019-07-30 | 2019-11-15 | 深圳市商汤科技有限公司 | Face identification method and device, electronic equipment and storage medium |
| CN110619294A (en) * | 2019-09-06 | 2019-12-27 | 中南大学 | Personalized mouth shape identification method based on RFID system customization |
| CN110688946A (en) * | 2019-09-26 | 2020-01-14 | 上海依图信息技术有限公司 | Public cloud silence in-vivo detection device and method based on picture identification |
| CN111242029A (en) * | 2020-01-13 | 2020-06-05 | 湖南世优电气股份有限公司 | Device control method, device, computer device and storage medium |
| CN111310541A (en) * | 2019-11-27 | 2020-06-19 | Oppo广东移动通信有限公司 | Scene prediction method, terminal and storage medium |
| CN111325139A (en) * | 2020-02-18 | 2020-06-23 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| CN111368721A (en) * | 2020-03-03 | 2020-07-03 | 深圳市腾讯计算机系统有限公司 | Identity recognition processing method and device, electronic equipment and storage medium |
| CN111464519A (en) * | 2020-03-26 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Account registration method and system based on voice interaction |
| CN111488756A (en) * | 2019-01-25 | 2020-08-04 | 杭州海康威视数字技术股份有限公司 | Face recognition-based living body detection method, electronic device, and storage medium |
| CN111508125A (en) * | 2020-03-13 | 2020-08-07 | 北京澎思科技有限公司 | Access control equipment with integrated temperature measurement function and temperature measurement method |
| CN111539249A (en) * | 2020-03-11 | 2020-08-14 | 西安电子科技大学 | Multi-factor human face in-vivo detection system and method |
| CN112000072A (en) * | 2020-08-25 | 2020-11-27 | 唐山惠米智能家居科技有限公司 | Intelligent bathroom system with face recognition function and control method |
| CN112183167A (en) * | 2019-07-04 | 2021-01-05 | 钉钉控股（开曼）有限公司 | Attendance checking method, authentication method, living body detection method, device and equipment |
| CN112395965A (en) * | 2020-11-06 | 2021-02-23 | 山东鲁能软件技术有限公司 | Mobile terminal face recognition system and method based on power intranet |
| CN113420613A (en) * | 2021-06-03 | 2021-09-21 | 中国联合网络通信集团有限公司 | Identity verification method based on face recognition, face recognition system and associated equipment |
| CN115100726A (en) * | 2022-08-25 | 2022-09-23 | 中国中金财富证券有限公司 | Intelligent one-way video witness method and related products |
| CN115352792A (en) * | 2022-10-20 | 2022-11-18 | 江苏菲达宝开电气股份有限公司 | Efficient intensive intelligent vertical warehouse automatic storing and taking system |
| CN115995115A (en) * | 2023-03-22 | 2023-04-21 | 成都华安视讯科技有限公司 | Face recognition system and method based on bifocal transformation |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622588A (en) * | 2012-03-08 | 2012-08-01 | 无锡数字奥森科技有限公司 | Dual-certification face anti-counterfeit method and device |
| CN102789572A (en) * | 2012-06-26 | 2012-11-21 | 五邑大学 | Living body face safety certification device and living body face safety certification method |
| CN105389554A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Face-identification-based living body determination method and equipment |
| CN105389553A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Living body detection method and apparatus |
| CN105518708A (en) * | 2015-04-29 | 2016-04-20 | 北京旷视科技有限公司 | Method and equipment for verifying living human face, and computer program product |
| CN105512535A (en) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | User authentication method and user authentication device |
| CN106096519A (en) * | 2016-06-01 | 2016-11-09 | 腾讯科技（深圳）有限公司 | Live body discrimination method and device |
| CN106778496A (en) * | 2016-11-22 | 2017-05-31 | 重庆中科云丛科技有限公司 | Biopsy method and device |
| 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 |
| CN107133608A (en) * | 2017-05-31 | 2017-09-05 | 天津中科智能识别产业技术研究院有限公司 | Identity authorization system based on In vivo detection and face verification |
| CN107563289A (en) * | 2017-07-31 | 2018-01-09 | 百度在线网络技术（北京）有限公司 | A kind of method, apparatus of In vivo detection, equipment and computer-readable storage medium |
| CN107977559A (en) * | 2017-11-22 | 2018-05-01 | 杨晓艳 | A kind of identity identifying method, device, equipment and computer-readable recording medium |
| WO2018086543A1 (en) * | 2016-11-10 | 2018-05-17 | 腾讯科技（深圳）有限公司 | Living body identification method, identity authentication method, terminal, server and storage medium |
-
2018
- 2018-05-23 CN CN201810500469.XA patent/CN108470169A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102622588A (en) * | 2012-03-08 | 2012-08-01 | 无锡数字奥森科技有限公司 | Dual-certification face anti-counterfeit method and device |
| CN102789572A (en) * | 2012-06-26 | 2012-11-21 | 五邑大学 | Living body face safety certification device and living body face safety certification method |
| CN105518708A (en) * | 2015-04-29 | 2016-04-20 | 北京旷视科技有限公司 | Method and equipment for verifying living human face, and computer program product |
| CN105389554A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Face-identification-based living body determination method and equipment |
| CN105389553A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Living body detection method and apparatus |
| CN105512535A (en) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | User authentication method and user authentication device |
| CN106096519A (en) * | 2016-06-01 | 2016-11-09 | 腾讯科技（深圳）有限公司 | Live body discrimination method and device |
| WO2018086543A1 (en) * | 2016-11-10 | 2018-05-17 | 腾讯科技（深圳）有限公司 | Living body identification method, identity authentication method, terminal, server and storage medium |
| CN106778496A (en) * | 2016-11-22 | 2017-05-31 | 重庆中科云丛科技有限公司 | Biopsy method and device |
| 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 |
| CN107133608A (en) * | 2017-05-31 | 2017-09-05 | 天津中科智能识别产业技术研究院有限公司 | Identity authorization system based on In vivo detection and face verification |
| CN107563289A (en) * | 2017-07-31 | 2018-01-09 | 百度在线网络技术（北京）有限公司 | A kind of method, apparatus of In vivo detection, equipment and computer-readable storage medium |
| CN107977559A (en) * | 2017-11-22 | 2018-05-01 | 杨晓艳 | A kind of identity identifying method, device, equipment and computer-readable recording medium |
Non-Patent Citations (1)
| Title |
|---|
| 温喆: "基于物联网的四网融合技术研究及其应用", 30 April 2016, 长春：吉林人民出版社, pages: 192 * |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299677A (en) * | 2018-09-07 | 2019-02-01 | 西安知微传感技术有限公司 | A kind of recognition of face living body judgment method and system |
| CN109635021A (en) * | 2018-10-30 | 2019-04-16 | 平安科技（深圳）有限公司 | A kind of data information input method, device and equipment based on human testing |
| CN109359995A (en) * | 2018-11-29 | 2019-02-19 | 中国科学院重庆绿色智能技术研究院 | A kind of dining room payment system based on recognition of face |
| CN109615387A (en) * | 2018-11-29 | 2019-04-12 | 中国科学院重庆绿色智能技术研究院 | A kind of consumption and payment system and method based on recognition of face |
| CN111488756A (en) * | 2019-01-25 | 2020-08-04 | 杭州海康威视数字技术股份有限公司 | Face recognition-based living body detection method, electronic device, and storage medium |
| US11830230B2 (en) | 2019-01-25 | 2023-11-28 | Hangzhou Hikvision Digital Technology Co., Ltd. | Living body detection method based on facial recognition, and electronic device and storage medium |
| CN111488756B (en) * | 2019-01-25 | 2023-10-03 | 杭州海康威视数字技术股份有限公司 | Face recognition-based living body detection method, electronic device, and storage medium |
| CN110110582A (en) * | 2019-03-14 | 2019-08-09 | 广州市金其利信息科技有限公司 | In conjunction with the face identification method and system of 3D structure light, infrared light and visible light |
| CN110298230A (en) * | 2019-05-06 | 2019-10-01 | 深圳市华付信息技术有限公司 | Silent biopsy method, device, computer equipment and storage medium |
| CN110082723A (en) * | 2019-05-16 | 2019-08-02 | 浙江大华技术股份有限公司 | A kind of sound localization method, device, equipment and storage medium |
| CN112183167B (en) * | 2019-07-04 | 2023-09-22 | 钉钉控股（开曼）有限公司 | Attendance checking method, authentication method, living body detection method, device and equipment |
| CN112183167A (en) * | 2019-07-04 | 2021-01-05 | 钉钉控股（开曼）有限公司 | Attendance checking method, authentication method, living body detection method, device and equipment |
| CN110458062A (en) * | 2019-07-30 | 2019-11-15 | 深圳市商汤科技有限公司 | Face identification method and device, electronic equipment and storage medium |
| CN110427887A (en) * | 2019-08-02 | 2019-11-08 | 腾讯科技（深圳）有限公司 | A kind of membership's recognition methods and device based on intelligence |
| CN110427887B (en) * | 2019-08-02 | 2023-03-10 | 腾讯科技（深圳）有限公司 | Member identity identification method and device based on intelligence |
| CN110619294A (en) * | 2019-09-06 | 2019-12-27 | 中南大学 | Personalized mouth shape identification method based on RFID system customization |
| CN110688946A (en) * | 2019-09-26 | 2020-01-14 | 上海依图信息技术有限公司 | Public cloud silence in-vivo detection device and method based on picture identification |
| CN111310541A (en) * | 2019-11-27 | 2020-06-19 | Oppo广东移动通信有限公司 | Scene prediction method, terminal and storage medium |
| CN111310541B (en) * | 2019-11-27 | 2023-09-29 | Oppo广东移动通信有限公司 | Scene prediction method, terminal and storage medium |
| CN111242029A (en) * | 2020-01-13 | 2020-06-05 | 湖南世优电气股份有限公司 | Device control method, device, computer device and storage medium |
| CN111325139A (en) * | 2020-02-18 | 2020-06-23 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| CN111325139B (en) * | 2020-02-18 | 2023-08-04 | 浙江大华技术股份有限公司 | Lip language identification method and device |
| CN111368721A (en) * | 2020-03-03 | 2020-07-03 | 深圳市腾讯计算机系统有限公司 | Identity recognition processing method and device, electronic equipment and storage medium |
| CN111539249A (en) * | 2020-03-11 | 2020-08-14 | 西安电子科技大学 | Multi-factor human face in-vivo detection system and method |
| CN111508125A (en) * | 2020-03-13 | 2020-08-07 | 北京澎思科技有限公司 | Access control equipment with integrated temperature measurement function and temperature measurement method |
| CN111464519A (en) * | 2020-03-26 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Account registration method and system based on voice interaction |
| CN112000072A (en) * | 2020-08-25 | 2020-11-27 | 唐山惠米智能家居科技有限公司 | Intelligent bathroom system with face recognition function and control method |
| CN112395965A (en) * | 2020-11-06 | 2021-02-23 | 山东鲁能软件技术有限公司 | Mobile terminal face recognition system and method based on power intranet |
| CN113420613B (en) * | 2021-06-03 | 2023-09-05 | 中国联合网络通信集团有限公司 | Identity verification method based on face recognition, face recognition system and associated equipment |
| CN113420613A (en) * | 2021-06-03 | 2021-09-21 | 中国联合网络通信集团有限公司 | Identity verification method based on face recognition, face recognition system and associated equipment |
| CN115100726A (en) * | 2022-08-25 | 2022-09-23 | 中国中金财富证券有限公司 | Intelligent one-way video witness method and related products |
| CN115352792A (en) * | 2022-10-20 | 2022-11-18 | 江苏菲达宝开电气股份有限公司 | Efficient intensive intelligent vertical warehouse automatic storing and taking system |
| CN115995115A (en) * | 2023-03-22 | 2023-04-21 | 成都华安视讯科技有限公司 | Face recognition system and method based on bifocal transformation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108470169A (en) | Face identification system and method | |
| US10650261B2 (en) | System and method for identifying re-photographed images | |
| US20190138807A1 (en) | Iris liveness detection for mobile devices | |
| WO2018188453A1 (en) | Method for determining human face area, storage medium, and computer device | |
| CN106850648B (en) | Identity verification method, client and service platform | |
| CN104537292B (en) | The method and system detected for the electronic deception of biological characteristic validation | |
| CN109871883A (en) | Neural network training method and device, electronic equipment and storage medium | |
| WO2016172872A1 (en) | Method and device for verifying real human face, and computer program product | |
| CN208351494U (en) | Face identification system | |
| CN107590430A (en) | Biopsy method, device, equipment and storage medium | |
| CN109359548A (en) | Plurality of human faces identifies monitoring method and device, electronic equipment and storage medium | |
| CN108280418A (en) | The deception recognition methods of face image and device | |
| CN110503023A (en) | Biopsy method and device, electronic equipment and storage medium | |
| CN108573202A (en) | Identity identifying method, device and system and terminal, server and storage medium | |
| KR102593624B1 (en) | Online Test System using face contour recognition AI to prevent the cheating behaviour and method thereof | |
| CN104143086A (en) | Application technology of portrait comparison to mobile terminal operating system | |
| CN109508706B (en) | Silence living body detection method based on micro-expression recognition and non-sensory face recognition | |
| CN110287671A (en) | Verification method and device, electronic equipment and storage medium | |
| CN105022999A (en) | Man code company real-time acquisition system | |
| CN110287672A (en) | Verification method and device, electronic equipment and storage medium | |
| KR102711511B1 (en) | Online Test System using face contour recognition AI to prevent the cheating behaviour by using a front camera of examinee terminal and an auxiliary camera and method thereof | |
| KR102581415B1 (en) | UBT system using face contour recognition AI to prevent the cheating behaviour and method thereof | |
| KR102615709B1 (en) | Online Test System using face contour recognition AI to prevent the cheating behavior by using a front camera of examinee terminal installed audible video recording program and an auxiliary camera and method thereof | |
| KR20230013236A (en) | Online Test System using face contour recognition AI to prevent the cheating behaviour by using speech recognition and method thereof | |
| KR100795360B1 (en) | A Method Of Face Recognizing |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180831 |