CN110334637A - Human face in-vivo detection method, device and storage medium - Google Patents
Human face in-vivo detection method, device and storage medium Download PDFInfo
- Publication number
- CN110334637A CN110334637A CN201910575380.4A CN201910575380A CN110334637A CN 110334637 A CN110334637 A CN 110334637A CN 201910575380 A CN201910575380 A CN 201910575380A CN 110334637 A CN110334637 A CN 110334637A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- living body
- present
- present image
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention provides a kind of human face in-vivo detection method, device and storage medium, and wherein method includes: acquisition present image；Determine that the face in present image is matched with face registered in database；Obtain present image continuous N frame image later；Determine the target organ on the face in every frame image of N frame image；The target organ of N frame image is successively sent into machine learning model, obtains output result；Wherein, the movement in result including target organ is exported；According to the movement of the target organ in N frame image, determine whether the face in present image is living body.Human face in-vivo detection method, device and storage medium provided by the invention, can be improved the accuracy rate of face In vivo detection.
Description
Technical field
The present invention relates to electronic technology field more particularly to a kind of human face in-vivo detection methods, device and storage medium.
Background technique
It is generation in mobile phone, tablet computer and laptop with the continuous development of electronic technology and terminal technology
In some electronic equipments of table, all have the functions such as face unlock, payment.Those electronic equipments allow user to pass through face court
To the mode of camera of electronic equipment, it can be realized and electronic equipment is unlocked or is paid etc. using electronic equipment
Function.The face face recognition that face unlock, the function of paying are carried out based on electronic equipment, wherein electronic equipment is in addition to needing
It to be carried out in the face and face database that will test to whether including that face detects in camera institute acquired image
After comparison passes through, it is also necessary to whether be further that living body detects to the face in image, judge whether face is true
Face or the face forged, for example, malicious user can be used colored Paper Printing face figure, in electronic equipment screen
The modes such as face digital picture and mask forge face.
In the prior art, electronic equipment acquires continuous video image by camera during face unlocks, when
After determining that the face for including in a frame video image and face database compare successfully, each frame video in subsequent a period of time is extracted
In image, after the key point of the feature locations such as mouth, eyes and head of face, then according to those key points to face spy
Whether execution detects whether face is living body for sign position.
But by critical point detection face in video image whether be living body Detection accuracy it is lower, therefore how
The technical issues of accuracy rate for improving face In vivo detection is this field urgent need to resolve.
Summary of the invention
The present invention provides a kind of human face in-vivo detection method, device and storage medium, to improve the standard of face In vivo detection
True rate.
First aspect present invention provides a kind of human face in-vivo detection method, comprising:
Obtain present image；
Determine that the face in the present image is matched with face registered in database；
Obtain the present image continuous N frame image later；
Determine the target organ on the face in every frame image of the N frame image；
The target organ of the N frame image is successively sent into machine learning model, obtains output result；Wherein, described defeated
It out include the movement of the target organ in result；
According to the movement of the target organ in the N frame image, determine whether the face in the present image is living body.
In one embodiment of first aspect present invention, the movement according to the target organ in the N frame image is determined
Whether the face in the present image is living body, comprising:
According to the transformation that target organ in the N frame image acts, determine whether the target organ completes deliberate action；
When the target organ completes the deliberate action, determine that the face in the present image is living body；
When the target organ does not complete the prediction action, it is living for determining the face in the present image not
Body.
In one embodiment of first aspect present invention, the target organ is the eyes or mouth on the face；
When the target organ is eyes, the deliberate action is blink or closes one's eyes；
When the target organ is mouth, the deliberate action is to open one's mouth or shut up.
In one embodiment of first aspect present invention, further includes:
When the face in the present image is living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information is for prompting not
Electronic equipment can be unlocked.
In one embodiment of first aspect present invention, the machine learning model is convolutional neural networks CNN.
Second aspect of the present invention provides a kind of human face in-vivo detection method, comprising:
Obtain present image；
By single-frame images In vivo detection, determine that the face in the present image is living body；
Obtain the present image continuous N frame image later；
Target image is detected from the N frame image, the face in the target image meets the following conditions: the target
Face is identical as the face in the present image in image, and face completes deliberate action, Yi Jisuo in the target image
The result for stating the single frames In vivo detection of the face in target image is living body；
When detection termination condition meets, if there are at least one target images in the N frame image, it is determined that described
Face is living body in present image, if the target image is not present in the N frame image, it is determined that in the present image
Face be not living body.
In one embodiment of second aspect of the present invention, by single-frame images In vivo detection, determine in the present image
Before face is living body, further includes:
Determine that the face in the present image is matched with face registered in database.
In one embodiment of second aspect of the present invention, the detection termination condition reaches preset time for detection time, or
The number of person, detection image reach predetermined number.
In one embodiment of second aspect of the present invention, by single-frame images In vivo detection, determine in the present image
Face is living body, comprising:
The present image is inputted into machine learning model, exported as a result, the output result is used to indicate described in
Face in present image is living body or non-living body.
In one embodiment of second aspect of the present invention, further includes:
When the face in the present image is living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information is for prompting not
Electronic equipment can be unlocked.
Third aspect present invention provides a kind of face living body detection device, comprising:
Module is obtained, for obtaining present image；
Matching module, for determining that the face in the present image is matched with face registered in database；
The acquisition module is also used to, and obtains continuous N frame image after the present image；
Processing module, the target organ on the face in every frame image for determining the N frame image；
Machine learning module is exported for the target organ of the N frame image to be successively sent into machine learning model
As a result；It wherein, include the movement of the target organ in the output result；
Determining module determines the people in the present image for the movement according to the target organ in the N frame image
Whether face is living body.
In one embodiment of third aspect present invention, the determining module is specifically used for,
According to the transformation that target organ in the N frame image acts, determine whether the target organ completes deliberate action；
When the target organ completes the deliberate action, determine that the face in the present image is living body；
When the target organ does not complete the prediction action, it is living for determining the face in the present image not
Body.
In one embodiment of third aspect present invention, the target organ is the eyes or mouth on the face；
When the target organ is eyes, the deliberate action is blink or closes one's eyes；
When the target organ is mouth, the deliberate action is to open one's mouth or shut up.
In one embodiment of third aspect present invention, further includes: unlocked state is used for:
When the face in the present image is living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information is for prompting not
Electronic equipment can be unlocked.
In one embodiment of third aspect present invention, the machine learning model is convolutional neural networks CNN.
Fourth aspect present invention provides a kind of face living body detection device, comprising:
Module is obtained, for obtaining present image；
Detection module, the face for by single-frame images In vivo detection, determining in the present image are living body；
The acquisition module is also used to, and obtains continuous N frame image after the present image；
Processing module, for detecting target image from the N frame image, the face in the target image meets following
Condition: face is identical as the face in the present image in the target image, and face completes pre- in the target image
If the result of the single frames In vivo detection of movement and the face in the target image is living body；
Determining module, for when detect termination condition meet when, if there are at least one target figures in the N frame image
Picture, it is determined that face is living body in the present image, if the target image is not present in the N frame image, it is determined that
Face in the present image is not living body.
In one embodiment of fourth aspect present invention, further includes: matching module, for determining the people in the present image
Face is matched with face registered in database.
In one embodiment of fourth aspect present invention, the detection termination condition reaches preset time for detection time, or
The number of person, detection image reach predetermined number.
In one embodiment of fourth aspect present invention, the detection module is specifically used for, and the present image is inputted machine
Device learning model is exported as a result, the face that the output result is used to indicate in the present image is living body or non-
Living body.
In one embodiment of fourth aspect present invention, further includes: unlocked state, for when the face in the present image
When for living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information is for prompting not
Electronic equipment can be unlocked.
Fifth aspect present invention provides a kind of electronic equipment, comprising: processor, memory and computer program；Wherein,
The computer program is stored in the memory, and is configured as being executed by the processor, the computer journey
Sequence includes for executing the instruction such as any one of above-mentioned first aspect method as described in the examples.
Sixth aspect present invention provides a kind of electronic equipment, comprising: processor, memory and computer program；Wherein,
The computer program is stored in the memory, and is configured as being executed by the processor, the computer journey
Sequence includes for executing the instruction such as any one of above-mentioned second aspect method as described in the examples.
Seventh aspect present invention provides a kind of computer readable storage medium, and the computer-readable recording medium storage has
Computer program, the computer program are performed, and realize the method as described in any one of above-mentioned first aspect embodiment.
Eighth aspect present invention provides a kind of computer readable storage medium, and the computer-readable recording medium storage has
Computer program, the computer program are performed, and realize the method as described in any one of above-mentioned second aspect embodiment.
To sum up, the present invention provides a kind of human face in-vivo detection method, device and storage medium, and wherein method includes: to obtain
Present image；Determine that the face in present image is matched with face registered in database；It is continuous after acquisition present image
N frame image；Determine the target organ on the face in every frame image of N frame image；Successively the target organ of N frame image is sent
Enter machine learning model, obtains output result；Wherein, the movement in result including target organ is exported；According in N frame image
The movement of target organ determines whether the face in present image is living body.Human face in-vivo detection method provided by the invention, dress
Set and storage medium, can when carrying out In vivo detection to the face in present image, after the face matching in present image,
It is handled by target organ of the machine learning model to the face in continuous N frame image later, and according to machine learning mould
The movement of target organ, determines whether the face of present image is living body in the subsequent N frame video image of type output.Therefore, originally
Human face in-vivo detection method, device and the storage medium that invention provides can be by the output results of machine learning model to face
Whether it is that living body is detected, thus the detection when carrying out face In vivo detection independent of the key point of face in image,
And then improve the accuracy rate of face In vivo detection.
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 without any creative labor, may be used also for those of ordinary skill in the art
To obtain other drawings based on these drawings.
Fig. 1 is the schematic diagram of application scenarios of the present invention；
Fig. 2 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention；
Fig. 3 is the schematic diagram of target organ in human face in-vivo detection method provided by the invention；
Fig. 4 is the action schematic diagram of target organ in human body biopsy method embodiment provided by the invention；
Fig. 5 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention；
Fig. 6 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention；
Fig. 7 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention；
Fig. 8 is the structural schematic diagram of one embodiment of face living body detection device provided by the invention；
Fig. 9 is the structural schematic diagram of one embodiment of face living body detection device provided by the invention；
Figure 10 is the structural schematic diagram of one embodiment of electronic equipment provided by the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first ", " second ", " third ", " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiment of the present invention described herein for example can be to remove
Sequence other than those of illustrating or describe herein is implemented.In addition, term " includes " and " having " and theirs is any
Deformation, it is intended that cover it is non-exclusive include, for example, containing the process, method of a series of steps or units, system, production
Product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for this
A little process, methods, the other step or units of product or equipment inherently.
Fig. 1 is the schematic diagram of application scenarios of the present invention, is carried out as shown in Figure 1 for electronic equipment 2 by the face to user 1
It identifies the application scenarios to enable electronic equipment correlation function, in this scenario, is said so that electronic equipment 2 is mobile phone as an example
It is bright, rather than it is defined.The electronic equipment 2 can also be that tablet computer, laptop etc. have the electricity of camera
Sub- equipment.In application scenarios as shown in Figure 1, electronic equipment 2 can be carried out after identifying successfully by the face to user 1,
Realize the functions such as face unlock, face payment.For example, electronic equipment 2 may be implemented after the recognition of face success to user 1
The unlock of electronic equipment 2 simultaneously wakes up display interface 22, then shows on its display interface 22 and welcomes interface.
Specifically, when electronic equipment 2 is when identifying the face of user 1, being acquired first by camera 21 includes user's 1
Facial image 211.Then, electronic equipment 2 executes S01, detects to the face in image 211.It, will be in S01 and in S02
It detects the registered face stored in obtained face and the database of electronic equipment 2 to compare, if comparing successfully, explanation
User 1 has registered, and can carry out subsequent operation according to the face of user 1.
And in order to determine that electronic equipment 2 by the face in the facial image 211 collected of camera 21 is really to use
The face at family 1, rather than other people use photo, use the face number in colored Paper Printing face figure, electronic equipment screen
The face for the user 1 that the modes such as word image and mask are forged, therefore, electronic equipment 2 is after S02 is compared successfully, it is also necessary to
It whether is further that living body detects to the face in facial image 211 by S03.
In the prior art, electronic equipment 2 will be after the face alignment success in facial image 211 in S02, it is also necessary into one
Step acquires the multi-frame video image in subsequent a period of time including user 1 by camera 21 first, then extracts in S03
In each frame video image of multi-frame video image, after the key point of the feature locations such as mouth, eyes and head of face, with
According to those key points, to face characteristic position, whether execution detects whether face is living body afterwards.Wherein, if multiframe regards
The variation pattern of face key point meets preset condition in frequency image, it is determined that face is living body；If people in multi-frame video image
The variation pattern of face key point does not meet preset condition, it is determined that face is not living body.
To sum up, in the prior art, although electronic equipment can pass through the side of face critical point detection in above-mentioned video image
Whether formula is that living body detects to face.But since electronic equipment needs the key point to face in every frame video image
It is detected, the result for resulting in face In vivo detection also relies on electronic equipment to the result of face critical point detection.And
Under the adverse circumstances such as illumination condition is poor, once electronic equipment can not detect face key point, it is living to will lead to face
Accuracy rate when physical examination is surveyed is difficult to ensure, reduces the accuracy rate that electronic equipment carries out face In vivo detection.And how to improve electricity
The technical issues of sub- equipment carries out accuracy rate when face In vivo detection, is this field urgent need to resolve.
Therefore, the embodiment of the present application provides a kind of human face in-vivo detection method, to pass through the output knot of machine learning model
Whether fruit is that living body detects to face, thus the key point when carrying out face In vivo detection independent of face in image
Detection, and then improve face In vivo detection accuracy rate.
Technical solution of the present invention is described in detail with specifically embodiment below.These specific implementations below
Example can be combined with each other, and the same or similar concept or process may be repeated no more in some embodiments.
Fig. 1 is a kind of flow diagram of one embodiment of human face in-vivo detection method provided by the present application, as shown in Figure 1,
The executing subject of method provided by the embodiment can be electronic equipment 2 as shown in Figure 1.Specifically, people provided in this embodiment
Face biopsy method includes:
S101: present image is obtained.
Specifically, electronic equipment is in order to carry out In vivo detection to face, first in the video flowing acquired by camera
Every frame video image detected, detect video flowing in every frame video image whether include face.If detecting a certain frame
In video image include face, then using the frame video image be used as the present image in the present embodiment, and execution after afterflow
Journey.And if face is not detected in video image, continue to test in video flowing and whether wrapped in next frame continuous videos image
Include face.
Wherein, in one possible implementation, electronic equipment is used for in the scene that face is identified and is unlocked,
Then present image acquired in electronic equipment is the video image in video flowing that camera acquires in real time.And it is possible at other
In implementation, the present embodiment executing subject be can also be dedicated for carrying out the electronic equipment of face In vivo detection, then the electricity
Sub- equipment obtains video flowing transmitted by other equipment, and in acquired video flowing to the face for including in present image into
After row detection, subsequent face In vivo detection step is executed.
S102: determine that the face in present image is matched with face registered in database.
Then, in S102, electronic equipment verifies the face in present image identified in S101.Specifically may be used
By by by face and face registered in database carry out it is matched in a manner of verify.Wherein, the database can be with
It is stored in the storage equipment of electronic equipment, alternatively, database can also be stored in cloud network server, electronic equipment exists
The database is obtained when needing from server.
For example, user A and user B are registered in electronic equipment using its face, then the database of electronic equipment
The facial image of middle storage registered user A and user B.Then in S102, if the people that the face in present image is user A
Face, then after the face of user A registered in the face and database in present image is carried out successful match by electronic equipment, i.e.,
Executable subsequent step；And if the face in present image is the face of user C, electronic equipment is by the face in present image
Face in C and database is unable to successful match, then will not continue to execute subsequent step.
S103: continuous N frame image after present image is obtained.
Wherein, when electronic equipment carries out Face datection and present image to present image respectively by S101 and S102
After middle face successful match, that is, it can determine that the corresponding user of the face in present image is used electronic equipment, and it is desirable that
Electronic equipment executes the function such as unlock, payment.In order to further be verified to user, in S103 and later the step of
In, whether electronic equipment also needs to be that living body is verified to the face in present image.
In the present embodiment, in present image after face successful match, electronic equipment in order to face whether be living body into
Row verifying, it is also necessary to continue to obtain the continuous N frame video image in video flowing after present image, and be regarded according to continuous N frame
Frequency image carries out face In vivo detection.The N is the integer more than or equal to 2.For example, being wrapped in video flowing acquired in electronic equipment
Include ten frame video images continuously marked as 1-10, if carried out in the current video image marked as 1 Face datection with
After matching, if N is 5, acquired N frame image can be the continuous video image of 5 frames marked as 2-6 at this time.
S104: by Face datection, the target organ in N frame image in every frame image on face is determined.
Specifically, electronic equipment is further handled N frame image acquired in S103, determines N frame video figure
It include the target area of target organ in target organ and every frame image as on every frame image face.Wherein, this implementation
Target organ described in example can be eyes or mouth on face.
For example, Fig. 3 is the schematic diagram of target organ in human face in-vivo detection method provided by the invention, it is with target organ
It is illustrated for eyes on face.Then in video image A as shown in Figure 3, the eyes of face all in opening state,
Electronic equipment by include in the target area A1 that can determine to video image using Face datection face eyes, it is subsequent can
Target organ is handled by the target area A1 in video image A including eyes.Similarly, face in video image B
Eyes be in blink state, include in the target area B1 that electronic equipment can determine by video image using Face datection
The eyes of face, it is subsequent target organ to be handled by the target area B1 in video image B including eyes.
In addition, it should be noted that, since the human face detection tech based on image procossing is highly developed, the application for
The concrete mode of human face target organ is determined without limitation from video image.
S105: successively sending the target organ of N frame image into machine learning model, is exported as a result, the output is tied
It include the movement of target organ in fruit.
Then, machine learning model is used in the present embodiment, and target organ obtained in S104 is judged, and by
Machine learning model output is as a result, include the movement of target organ in the output result.
For example, Fig. 4 is the action schematic diagram of target organ in human body biopsy method embodiment provided by the invention, such as
In embodiment shown in Fig. 4, in the continuous video image of N frame between time T1-T2, by the detection of machine learning model,
User can be obtained in N frame continuous videos image, target organ (eyes) completes the movement of blink.
Optionally, in the present embodiment, the movement of machine learning model target organ of user in determining N frame video image
When, can pass sequentially through machine learning model to include each frame video image in, the target area including ownership goal organ
After being identified, the state of target organ in each frame video image is obtained.Then further according to mesh in continuous N frame video image
The state of organ is marked, if according to the common movement for determining target organ and being completed in N frame video image of default rule variation.Then
At this point, machine learning model can be used to identify the state of target organ, which can be in advance using multiple including mesh
The image of mark organ is trained to obtain.
Particularly, in the present embodiment, the target organ for including in video image is only inputted machine learning mould by electronic equipment
In type, rather than whole image.That is, by shown in Fig. 3 include eye target area A1 and B1 input machine learning model into
Row calculates.So that machine learning model does not need to calculate whole video image A and B, so that machine learning can also be improved
The computational efficiency of model.
In addition, machine learning model described in the present embodiment includes but is not limited to for example: convolutional neural networks
(Convolutional Neural Networks, referred to as: CNN), K- nearest neighbor algorithm (k-Nearest Neighbor, referred to as:
KNN), support vector machines (Support Vector Machine, referred to as: SVM) or other engineerings based on deep learning
Practise model etc..
S106: according to the movement of target organ in N frame image, determine whether face is living body in present image.
Specifically, in S106, electronic equipment is according to the transformation in S105 to the movement of target organ in N frame image, really
It is scheduled on whether target organ in N frame image completes deliberate action, whether the target organ to judge in N frame video image is completed
Deliberate action.Wherein, when target organ completes deliberate action, determine that the face in present image is living body；Work as target organ
When not completing institute's survey movement, determine that the face in present image is not living body.
For example, the deliberate action can be blink or eye closing if target organ is behaved when eyes on the face.Then exist
In example as shown in Figure 4, moved again to the blink opened when being completed according to the eyes of face in N frame image by opening closure
After work, determine that the face in present image is living body.Similarly, described default if target organ is behaved when mouth on the face
Movement, which can be, opens one's mouth or shuts up.
Finally, in the present embodiment S106, target organ of the electronic equipment in passing through N frame video image on face, really
After the organ that sets the goal completes deliberate action, determine that face is living body, to complete face In vivo detection.And if determining N in S106
Target organ in frame video image does not complete deliberate action, it is determined that face is non-living body, it may be possible to shown by still photo
Face.
To sum up, it in human face in-vivo detection method provided in this embodiment, can be carried out to the face in present image
When In vivo detection, after the face matching in present image, by machine learning model to the face in continuous N frame image later
Target organ handled, and according to machine learning model output subsequent N frame video image in target organ movement, really
Whether the face for determining present image is living body.Therefore, the present embodiment can be by the output result of machine learning model to face
Whether it is that living body is detected, thus the detection when carrying out face In vivo detection independent of the key point of face in image,
And then improve the accuracy rate of face In vivo detection.
Further, in a kind of specific application of above-described embodiment, if being applied in application scenarios as shown in Figure 1
When, electronic equipment 2 can also further execute phase according to the face in S106 after carrying out face In vivo detection to user 1
Close function.For example, unlocking electronic equipment 2, and on display circle when the face in present image identified in S106 is living body
It is shown on face 22 and welcomes interface；And when the face in present image identified in S106 is not living body, prompt information is exported,
The prompt information fails to unlock electronic equipment 2 for prompting.The prompt information can be electronic equipment 2 in its display interface
" unlock failure " printed words shown on 22, alternatively, the prompt information can also be jingle bell, the vibration etc. that electronic equipment 2 issues
Information.
Fig. 5 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention, implementation as shown in Figure 5
It is illustrated a kind of specific implementation logic of above-mentioned method as shown in Figure 2, wherein present frame face is the present image, after
A few frame faces are the N frame image in video flowing after present frame face.Then, in those available images, face mouth and/
Or behind the position of the approximate location of eye, it is sent into CNN and is identified, and determined in present frame according to the output result of CNN
Face is living body or is non-living body.
Fig. 6 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention.In reality as shown in FIG. 6
It applies in example, shows another human face in-vivo detection method that application provides.This method and method as shown in Figure 2 can be by same
One electronic equipment executes, or is executed respectively by different electronic equipments.Specifically, face In vivo detection side as shown in FIG. 6
Method includes:
S201: present image is obtained.
Wherein, S201 shown in the present embodiment can refer to the description to S101 as shown in Figure 2, implementation and principle
It is identical, it repeats no more.
S202: by single-frame images In vivo detection, determine that the face in present image is living body.
Specifically, in the present embodiment S202, electronic equipment need to the face for including in present image in S201 whether be
Living body is detected.Wherein, the In vivo detection mode based on the single-frame images, detect the face in current single-frame images whether be
Living body.
For example, in a kind of concrete implementation mode, electronic equipment by present image input be used to determine face whether be
The machine learning model of living body obtains machine learning model output as a result, the output result is used to indicate the current figure
Face as in is living body or non-living body.The present embodiment pair and single frames carry out the mode of face In vivo detection without limitation, can
Referring to the prior art.
Optionally, before S202, further includes: determine registered people in the face and database in the present image
Face matching.Specifically, electronic equipment is it needs to be determined that the face in S201 in identified present image is verified.It specifically can be with
It is verified by the way that face is carried out matched mode with face registered in database.Matched retouch is carried out to face in detail
The description that can refer in S102 is stated, implementation is identical as principle, repeats no more.
S203: continuous N frame image after present image is obtained.
Wherein, S203 shown in the present embodiment can refer to the description to S103 as shown in Figure 2, implementation and principle
It is identical, it repeats no more.
S204: target image is detected from N frame image.
Specifically, electronic equipment detects whether that there are targets from N frame video image acquired in S203 in S204
Way is high.Wherein, the face in the target image meets the following conditions: identified current in face and S201 in target image
Face in image is identical, and face completes the single frames living body of the face in deliberate action and target image in target image
The result of detection is living body.
In a kind of concrete implementation mode, such as electronic equipment obtains continuous ten frame video image marked as 1-10
Afterwards, face is detected in the current video marked as 1, if N is 5.Then then successively examined since the video image of label 2
Whether the video image surveyed marked as 2-6 is target image.Wherein, for any frame image, detect the frame image whether be
When target image, after can first being detected to the face in image, carried out with the face of present image acquired in S201
It compares.After both comparisons are identical, then judge whether the face in the frame image completes deliberate action, for example, deliberate action
It can be eye closing, open one's mouth.If the face in the image completes default work, living body inspection finally is carried out to the single-frame images again
After survey, if detecting the face in the frame image is living body, it is determined that the frame image is target image.It is understood that if should
Frame image does not meet above-mentioned condition simultaneously, then the frame image is not target image.
S205: when detection termination condition meets, if there are at least one target images in the N frame image, really
Face is living body in the fixed present image, if the target image is not present in the N frame image, it is determined that described current
Face in image is not living body.
Then in S205, electronic equipment is it needs to be determined that in N frame video image acquired in S203, if has at least one
A target image.Optionally, the detection termination condition is that detection time reaches preset time, alternatively, the number of detection image
Reach predetermined number.
For example, if detection termination condition is 2 seconds preset times after present image.Then electronic equipment is obtaining video flowing
It afterwards, does not include target in the N frame video image acquired in 2 seconds after S201 determining present image preset times
Image, it is determined that the face in present image is not living body, and if including extremely in N frame video image acquired in preset time
A few target image, it is determined that the face in present image is living body.In another example if testing conditions are predetermined number, it is described
Predetermined number can be N, then similarly, if electronic equipment S201 determine present image after N frame video image in, not
Including target image, it is determined that the face in present image is not living body；And N frame video image acquired after present image
It include at least one target image in if, it is determined that the face in present image is living body.
To sum up, it in human face in-vivo detection method provided in this embodiment, can be carried out to the face in present image
Detection, matching and single phase In vivo detection after, continue to after present image in N frame image whether include target image examine
It surveys, only when detection termination condition satisfaction, and in N frame video image, there are the faces at least one face and present image
When identical, face completes target image of the result of the single frames In vivo detection of deliberate action and face for the condition of living body,
The face that can determine present image is living body.To the pass when carrying out face In vivo detection independent of face in image
The detection of key point, while also not dependent on the testing result of a certain single image, and then improve electronic equipment and carrying out face
Accuracy rate when In vivo detection.
Further, in a kind of specific application of above-described embodiment, if being applied in application scenarios as shown in Figure 1
When, electronic equipment 2 can also further execute phase according to the face in S205 after carrying out face In vivo detection to user 1
Close function.For example, unlocking electronic equipment 2, and on display circle when the face in present image identified in S205 is living body
It is shown on face 22 and welcomes interface；And when the face in present image identified in S205 is not living body, prompt information is exported,
The prompt information fails to unlock electronic equipment 2 for prompting.The prompt information can be electronic equipment 2 in its display interface
" unlock failure " printed words shown on 22, alternatively, the prompt information can also be jingle bell, the vibration etc. that electronic equipment 2 issues
Information.
Fig. 7 is the flow diagram of one embodiment of human face in-vivo detection method provided by the invention, implementation as shown in Figure 7
It is illustrated a kind of specific implementation logic of above-mentioned method as shown in Figure 6, wherein the present frame is embodiment as shown in Figure 6
In present image, then electronic equipment obtain present frame after, the face in present frame can be detected, and judge present frame
In face be single phase living body after, then obtain the next frame image after present frame carry out subsequent processing；Otherwise present frame is obtained
Next frame image afterwards is used as present frame to carry out Face datection again.And after obtaining next frame, continue to next frame image into
Row Face datection, and face comparison is carried out with the face in present frame, it, can be to the next frame only in the identical situation of the two
In face carry out the detection of motion detection and single phase living body.And the face in the next frame meet face it is identical,
After the requirement for completing deliberate action and single phase living body, just judging face is living body.Any one condition is unsatisfactory for all
Judge that the face in the next frame for non-living body, and before testing conditions termination condition satisfaction, continues to obtain next frame image
It is detected.
In above-mentioned embodiment provided by the present application, trigger from the angle of electronic equipment to method provided by the embodiments of the present application
It is described.In order to realize each function in above-mentioned method provided by the embodiments of the present application, electronic equipment may include hardware
Structure and/or software module are realized above-mentioned each in the form of hardware configuration, software module or hardware configuration add software module
Function.Some function in above-mentioned each function is come in such a way that hardware configuration, software module or hardware configuration add software module
It executes, specific application and design constraint depending on technical solution.
For example, Fig. 8 is the structural schematic diagram of one embodiment of face living body detection device provided by the invention, as shown in Figure 8
Face living body detection device can be used for executing the method as described in Fig. 2-5.Face living body detection device packet as described in Figure 8
It includes: obtaining module 801, matching module 802, processing module 803, machine learning module 804 and determining module 805.Wherein, it obtains
Module 801 is for obtaining present image；Matching module 802 is used to determine registered in the face and database in present image
Face matching；It obtains module 801 to be also used to, obtains continuous N frame image after present image；Processing module 803 is for determining N
The target organ on face in every frame image of frame image；Machine learning module 804 is used for successively by the object machine of N frame image
Official is sent into machine learning model, obtains output result；Wherein, the movement in result including target organ is exported；Determining module 805
For the movement according to the target organ in N frame image, determine whether the face in present image is living body.
Optionally it is determined that module 805 is specifically used for, according to the transformation that target organ in N frame image acts, object machine is determined
Whether official completes deliberate action；When target organ completes deliberate action, determine that the face in present image is living body；Work as target
When organ does not complete prediction action, determine that the face in present image is not living body.
Optionally, eyes or mouth that target organ is behaved on the face；When target organ is eyes, deliberate action is to blink
Eye is closed one's eyes；When target organ is mouth, deliberate action is to open one's mouth or shut up.
Optionally, face living body detection device provided in this embodiment further include: unlocked state 806；Unlocked state 806 is used
In: when the face in present image is living body, unlock electronic equipment；When the face in present image is not living body, output
Prompt information, prompt information cannot unlock electronic equipment for prompting.
Optionally, machine learning model is convolutional neural networks CNN.
The face living body detection device provided in the present embodiment can be used for executing the method as described in Fig. 2-5, in fact
Existing mode is identical as principle, can refer to the description of preceding method, repeats no more.
Fig. 9 is the structural schematic diagram of one embodiment of face living body detection device provided by the invention.Face as shown in Figure 9
Living body detection device can be used for executing the method as described in Fig. 6-7.Face living body detection device as described in Figure 9 includes: to obtain
Modulus block 901, detection module 902, processing module 903 and determining module 904.Wherein, module 901 is obtained for obtaining current figure
Picture；Detection module 902 is used to determine that the face in present image is living body by single-frame images In vivo detection；Obtain module 901
It is also used to, obtains continuous N frame image after present image；Processing module 903 is used to detect target image from N frame image,
Face in target image meets the following conditions: face is identical as the face in present image in target image, in target image
The result that face completes the single frames In vivo detection of the face in deliberate action and target image is living body；Determining module 904
For when detect termination condition meet when, if there are at least one target images in N frame image, it is determined that people in present image
Face is living body, if target image is not present in N frame image, it is determined that the face in present image is not living body.
Optionally, face living body detection device provided in this embodiment further include: matching module 905；Matching module 905 is used
In determining that the face in present image matches with face registered in database.
Optionally, detection termination condition is that detection time reaches preset time, alternatively, the number of detection image reaches default
Number.
Optionally, detection module 902 is specifically used for, and present image is inputted machine learning model, is exported as a result, defeated
It is living body or non-living body that result, which is used to indicate the face in present image, out.
Optionally, face living body detection device provided in this embodiment further include: unlocked state 906；Unlocked state 906 is used
In when the face in present image is living body, electronic equipment is unlocked；When the face in present image is not living body, output is mentioned
Show that information, prompt information cannot unlock electronic equipment for prompting.
The face living body detection device provided in the present embodiment can be used for executing the method as described in Fig. 6-7, realize
Mode is identical as principle, can refer to the description of preceding method, repeats no more.
It is schematical, only a kind of logical function partition to the division of module in the embodiment of the present application, it is practical to realize
When there may be another division manner, in addition, each functional module in each embodiment of the application can integrate at one
It manages in device, is also possible to physically exist alone, can also be integrated in two or more modules in a module.It is above-mentioned integrated
Module both can take the form of hardware realization, can also be realized in the form of software function module.
Figure 10 is the structural schematic diagram of one embodiment of electronic equipment provided by the invention；It is set as Figure 10 shows a kind of electronics
Standby structural schematic diagram, the electronic equipment can be used for executing any one of the application previous embodiment the method.
As shown in Figure 10, electronic equipment 1000 provided in this embodiment includes: transceiver 1010, memory 1030 and processing
Device 1020.Wherein, memory 1030 can be independent physical unit, can be connect by bus 1040 with processor 1020.
Memory 1030, processor 1020 also can integrate together, pass through hardware realization etc..Memory 1030 for store realize with
The computer program of upper embodiment of the method, wherein the computer program is stored in the memory 1030, and is matched
It is set to and is executed by the processor 1020, the computer program includes for executing as described in any one of previous embodiment
The instruction of method.
Optionally, when passing through software realization some or all of in the method for above-described embodiment, above-mentioned electronic equipment
1000 can also only include processor.Memory for storing program is located at except electronic equipment 1000, and processor passes through electricity
Road/electric wire is connect with memory, for reading and executing the computer program stored in memory.During processor 1020 can be
Central processor (Central Processing Unit, CPU), network processing unit (Network Processor, NP) or CPU
With the combination of NP.Processor 1020 can further include hardware chip.Above-mentioned hardware chip can be specific integrated circuit
(Application-Specific Integrated Circuit, ASIC), programmable logic device (Programmable
Logic Device, PLD) or combinations thereof.Above-mentioned PLD can be Complex Programmable Logic Devices (Complex
Programmable Logic Device, CPLD), field programmable gate array (Field-Programmable Gate
Array, FPGA), Universal Array Logic (Generic Array Logic, GAL) or any combination thereof.Memory 1030 can be with
Including volatile memory (Volatile Memory), such as random access memory (Random-Access Memory,
RAM)；Memory also may include nonvolatile memory (Non-volatile Memory), such as flash memory (Flash
Memory), hard disk (Hard Disk Drive, HDD) or solid state hard disk (Solid-state Drive, SSD)；Memory may be used also
With include mentioned kind memory combination.
In addition, the present invention also provides a kind of program product, for example, computer readable storage medium, comprising: computer journey
Sequence, computer program are used to execute when executed any of the above item the method for the present invention.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence；And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (13)
1. a kind of human face in-vivo detection method characterized by comprising
Obtain present image；
Determine that the face in the present image is matched with face registered in database；
Obtain the present image continuous N frame image later；
Determine the target organ on the face in every frame image of the N frame image；
The target organ of the N frame image is successively sent into machine learning model, obtains output result；Wherein, the output knot
It include the movement of the target organ in fruit；
According to the movement of the target organ in the N frame image, determine whether the face in the present image is living body.
2. the method according to claim 1, wherein the target organ according in the N frame image is dynamic
Make, determine whether the face in the present image is living body, comprising:
According to the transformation that target organ in the N frame image acts, determine whether the target organ completes deliberate action；
When the target organ completes the deliberate action, determine that the face in the present image is living body；
When the target organ does not complete the prediction action, determine that the face in the present image is not living body.
3. according to the method described in claim 2, it is characterized in that, the target organ is the eyes or mouth on the face
Bar；
When the target organ is eyes, the deliberate action is blink or closes one's eyes；
When the target organ is mouth, the deliberate action is to open one's mouth or shut up.
4. method according to claim 1-3, which is characterized in that further include:
When the face in the present image is living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information fails to solve for prompting
Lock electronic equipment.
5. the method according to claim 1, wherein
The machine learning model is convolutional neural networks CNN.
6. a kind of human face in-vivo detection method characterized by comprising
Obtain present image；
By single-frame images In vivo detection, determine that the face in the present image is living body；
Obtain the present image continuous N frame image later；
Target image is detected from the N frame image, the face in the target image meets the following conditions: the target image
Middle face is identical as the face in the present image, and face completes deliberate action and the mesh in the target image
The result of the single frames In vivo detection of face in logo image is living body；
When detection termination condition meets, if there are at least one target images in the N frame image, it is determined that described current
Face is living body in image, if the target image is not present in the N frame image, it is determined that the people in the present image
Face is not living body.
7. according to the method described in claim 6, it is characterized in that, determining the current figure by single-frame images In vivo detection
Before face as in is living body, further includes:
Determine that the face in the present image is matched with face registered in database.
8. method according to claim 6 or 7, which is characterized in that the detection termination condition is that detection time reaches pre-
If the time, alternatively, the number of detection image reaches predetermined number.
9. according to the described in any item methods of claim 6-8, which is characterized in that by single-frame images In vivo detection, determine institute
Stating the face in present image is living body, comprising:
By the present image input machine learning model, exported as a result, the output result be used to indicate it is described currently
Face in image is living body or non-living body.
10. the method according to claim 6, which is characterized in that further include:
When the face in the present image is living body, electronic equipment is unlocked；
When the face in the present image is not living body, prompt information is exported, the prompt information fails to solve for prompting
Lock electronic equipment.
11. a kind of face living body detection device, which is characterized in that for executing such as the described in any item sides of claim 1-10
Method.
12. a kind of electronic equipment characterized by comprising processor, memory and computer program；Wherein, the calculating
Machine program is stored in the memory, and is configured as being executed by the processor, and the computer program includes using
In the instruction for executing such as described in any item methods of claim 1-10.
13. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program are performed, and realize such as the described in any item methods of claim 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910575380.4A CN110334637A (en) | 2019-06-28 | 2019-06-28 | Human face in-vivo detection method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910575380.4A CN110334637A (en) | 2019-06-28 | 2019-06-28 | Human face in-vivo detection method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334637A true CN110334637A (en) | 2019-10-15 |
Family
ID=68143611
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910575380.4A Pending CN110334637A (en) | 2019-06-28 | 2019-06-28 | Human face in-vivo detection method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334637A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783644A (en) * | 2020-06-30 | 2020-10-16 | 百度在线网络技术（北京）有限公司 | Detection method, device, equipment and computer storage medium |
| CN112926355A (en) * | 2019-12-05 | 2021-06-08 | 深圳云天励飞技术有限公司 | Method and device for detecting living body based on human face |
| CN113609959A (en) * | 2021-04-16 | 2021-11-05 | 六度云计算有限公司 | Face living body detection method and device |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100158319A1 (en) * | 2008-12-22 | 2010-06-24 | Electronics And Telecommunications Research Institute | Method and apparatus for fake-face detection using range information |
| WO2012146823A1 (en) * | 2011-04-29 | 2012-11-01 | Nokia Corporation | Method, apparatus and computer program product for blink detection in media content |
| US20140270412A1 (en) * | 2012-01-20 | 2014-09-18 | Cyberlink Corp. | Liveness detection system based on face behavior |
| CN104751110A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Bio-assay detection method and device |
| CN105426815A (en) * | 2015-10-29 | 2016-03-23 | 北京汉王智远科技有限公司 | Living body detection method and device |
| CN106682578A (en) * | 2016-11-21 | 2017-05-17 | 北京交通大学 | Human face recognition method based on blink detection |
| CN107992842A (en) * | 2017-12-13 | 2018-05-04 | 深圳云天励飞技术有限公司 | Biopsy method, computer installation and computer-readable recording medium |
| CN108182409A (en) * | 2017-12-29 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Biopsy method, device, equipment and storage medium |
| CN108319901A (en) * | 2018-01-17 | 2018-07-24 | 百度在线网络技术（北京）有限公司 | Biopsy method, device, computer equipment and the readable medium of face |
| CN108875333A (en) * | 2017-09-22 | 2018-11-23 | 北京旷视科技有限公司 | Terminal unlock method, terminal and computer readable storage medium |
| CN109684974A (en) * | 2018-12-18 | 2019-04-26 | 北京字节跳动网络技术有限公司 | Biopsy method, device, electronic equipment and storage medium |
-
2019
- 2019-06-28 CN CN201910575380.4A patent/CN110334637A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100158319A1 (en) * | 2008-12-22 | 2010-06-24 | Electronics And Telecommunications Research Institute | Method and apparatus for fake-face detection using range information |
| WO2012146823A1 (en) * | 2011-04-29 | 2012-11-01 | Nokia Corporation | Method, apparatus and computer program product for blink detection in media content |
| US20140270412A1 (en) * | 2012-01-20 | 2014-09-18 | Cyberlink Corp. | Liveness detection system based on face behavior |
| CN104751110A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Bio-assay detection method and device |
| CN105426815A (en) * | 2015-10-29 | 2016-03-23 | 北京汉王智远科技有限公司 | Living body detection method and device |
| CN106682578A (en) * | 2016-11-21 | 2017-05-17 | 北京交通大学 | Human face recognition method based on blink detection |
| CN108875333A (en) * | 2017-09-22 | 2018-11-23 | 北京旷视科技有限公司 | Terminal unlock method, terminal and computer readable storage medium |
| CN107992842A (en) * | 2017-12-13 | 2018-05-04 | 深圳云天励飞技术有限公司 | Biopsy method, computer installation and computer-readable recording medium |
| CN108182409A (en) * | 2017-12-29 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Biopsy method, device, equipment and storage medium |
| CN108319901A (en) * | 2018-01-17 | 2018-07-24 | 百度在线网络技术（北京）有限公司 | Biopsy method, device, computer equipment and the readable medium of face |
| CN109684974A (en) * | 2018-12-18 | 2019-04-26 | 北京字节跳动网络技术有限公司 | Biopsy method, device, electronic equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926355A (en) * | 2019-12-05 | 2021-06-08 | 深圳云天励飞技术有限公司 | Method and device for detecting living body based on human face |
| CN111783644A (en) * | 2020-06-30 | 2020-10-16 | 百度在线网络技术（北京）有限公司 | Detection method, device, equipment and computer storage medium |
| CN113609959A (en) * | 2021-04-16 | 2021-11-05 | 六度云计算有限公司 | Face living body detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106897658B (en) | Method and device for identifying human face living body | |
| CN109858381A (en) | Biopsy method, device, computer equipment and storage medium | |
| US20180034852A1 (en) | Anti-spoofing system and methods useful in conjunction therewith | |
| CN109325964A (en) | A kind of face tracking methods, device and terminal | |
| CN108171207A (en) | Face identification method and device based on video sequence | |
| WO2019023606A1 (en) | System and method for identifying re-photographed images | |
| US11423702B2 (en) | Object recognition method and apparatus | |
| CN110334637A (en) | Human face in-vivo detection method, device and storage medium | |
| KR20190001066A (en) | Face verifying method and apparatus | |
| CN107133608A (en) | Identity authorization system based on In vivo detection and face verification | |
| CN110348331B (en) | Face recognition method and electronic equipment | |
| CN106127865B (en) | Ticket checking method and channel management equipment | |
| CN109389153A (en) | A kind of holographic false proof code check method and device | |
| US8103090B2 (en) | Behavior and pattern analysis using multiple category learning | |
| JP7071991B2 (en) | Methods and equipment for inspecting certificates and bearers | |
| CN109376592A (en) | Biopsy method, device and computer readable storage medium | |
| CN101167086A (en) | Human detection and tracking for security applications | |
| CN109389589A (en) | Method and apparatus for statistical number of person | |
| CN108494778A (en) | Identity identifying method and device | |
| CN109325413A (en) | A kind of face identification method, device and terminal | |
| CN108491794A (en) | The method and apparatus of face recognition | |
| CN109815884A (en) | Unsafe driving behavioral value method and device based on deep learning | |
| US11501482B2 (en) | Anonymization apparatus, surveillance device, method, computer program and storage medium | |
| CN112052815A (en) | Behavior detection method and device and electronic equipment | |
| CN109003347A (en) | A kind of anti-cheating movable attendance checking terminal |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210518 Address after: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Applicant after: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. Applicant after: Shanghai Xiaodu Technology Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Applicant before: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191015 |
|
| RJ01 | Rejection of invention patent application after publication |