CN109961025A - A kind of true and false face recognition detection method and detection system based on image degree of skewness - Google Patents
A kind of true and false face recognition detection method and detection system based on image degree of skewness Download PDFInfo
- Publication number
- CN109961025A CN109961025A CN201910179174.1A CN201910179174A CN109961025A CN 109961025 A CN109961025 A CN 109961025A CN 201910179174 A CN201910179174 A CN 201910179174A CN 109961025 A CN109961025 A CN 109961025A
- Authority
- CN
- China
- Prior art keywords
- image
- skewness
- degree
- true
- block message
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of true and false face recognition detection method and detection system based on image degree of skewness, detection method include by image acquisition device obtain animal body face initial pictures block message, first by initial pictures block message be converted to specific format, again from the initial pictures block message extract obtain certain components, the certain components are arranged as corresponding column vector again, the deflection angle value of the column vector is obtained by calculating；The deflection angle value is compared with degree of skewness threshold value again, obtains judgement result.Detection system is by image collecting device and identification arithmetic unit, using technical solution of the present invention, by using the degree of skewness of the certain components under image block message specific format as the data target for determining true face and false face, true face and false the face significant difference in this index, to simplify recognition algorithms, the reliability for improving face recognition improves the safety of face identification system.
Description
Technical field
The present invention relates to technical field of image processing more particularly to a kind of true and false face recognition inspections based on image degree of skewness
Survey method and detection system.
Background technique
With the fast development of technologies such as artificial intelligence, semiconductor and internet in recent years, using biological human as object
The research of Classical correlation technology is the research hotspot of current scientific and technological circle, wherein especially with the with fastest developing speed of human face identification technology,
And have begun investment commercial applications, Chao Zhen automation, unmanned supervisionization trend development.For example, everybody is common day
Smart phone has begun universal face recognition comprehensively and unlocks；And the wechat payment and Alipay payment of Chinese daily
Brush face verifying identity is already supported to replace traditional fingerprint and password authentification；The place safety check such as airport, railway station has also been used
Newest face recognition identity validation technology, not only greatly accelerates passage speed, also reduces the heavy safety check of staff
Work, etc..
The targeted special object of face detection is human face, makees the features such as somatic fingerprint, iris with the past
To identify that the technology of object is compared, human face feature is easiest to obtain, thus face detection is most suitable for being widely popularized and answers
With.However, human face feature is also to be easiest to be utilized by spiteful personnel, therefore, if distinguished automatically and efficiently
Not Chu the true and false of human face image be institute's major issue to be solved in face detection wide popularization and application.Existing people
Body face detection mainly has following three kinds of implementation methods:
(1) movement cooperation class
Movement cooperation class face identification system is a kind of recognition detection method for biological human used earliest, this
Method acquired image information mostlys come from static photo, is usually accordingly such as blinked, shaken the head, opened by system sending
The action commands such as mouth, smile, user complete after acting accordingly under the guidance for the action command that system provides, could pass through
Verifying, when using this kind of detection system, when spiteful personnel hold the photo of forgery, since it could not be completed accordingly
Action command, thus detection system can identify the true and false of photo, however, referring to when spiteful personnel hold having for forgery
When the video acted surely, which can not just identify the true and false of the video, and safety is lower.In addition, this kind of detection system
System needs to issue corresponding action command guidance user and completes corresponding movement, the process thousand poor ten thousand of each user's execution
Not, the experience sense of user is affected, thus this kind of detection system is having been prepared for being eliminated in recent years.
(2) traditional images feature class
Traditional characteristic class method is one kind in silent detection method, such method target is very clear, exactly finds human body
Difference between face and non-human attack then carries out decision by classifier.For single frame detection method, picture quality is lost
Very, color characteristic, unity and coherence in writing feature etc. are common traditional characteristic characterizations.Also there is method discovery, in the more difficult differentiation people of rgb space
Body with it is non-human, but in HSV and YCbCr color space, human body and non-human unity and coherence in writing have notable difference.Some method is
Based on continuous multiple frames face image, by the difference between capture human body and non-human fine motion work come design feature.Such is more
Frame image detecting method is bad for imitative face's roll paper shake and video attack effect, because it assumes that human body and it is non-human it
Between non-rigid motion have significant difference, however in fact face's fine motion work be difficult description and study.Another multiple image detection
Method is to be based on carrying out heart rate extraction from face's video.Such method thinks that the Heart rate distribution extracted from photo is different,
And human body is the same.But itself is there is also robustness problem at present for image heart rhythm detection algorithms, therefore the result tested is not yet
It can guarantee.
(3) deep learning method class
As deep learning achieves good effect in many detections, classification task, therefore, a collection of base has also been emerged in large numbers
In the human testing algorithm of deep learning.However due to human testing in terms of public data collection it is very little, be based on CNN even depth
The algorithm performance of habit can not surmount always above-mentioned traditional algorithm.Even if the algorithm of newest proposition, directly human testing and face
Detection is put into a frame, although speed is quickly, this recognition methods algorithm is sufficiently complex, is realized and is promoted and applied
Come relatively difficult.
Therefore, the safety of existing face detection is lower, designs a kind of simple, quick, efficient and safety
High recognition algorithms are most important.It using safe, prevents the spiteful third party from swarming into tool protection system
It is of great significance.
Summary of the invention
The embodiment of the present invention provides a kind of true and false face recognition detection method and detection system based on image degree of skewness, institute
State that detection method includes the following steps:
Step 1: the initial pictures of animal body face are obtained by the image acquisition device being deployed in around animal body
Block message, then the initial pictures block message is sent to the identification operation independently disposed relative to described image acquisition device and is filled
It sets；
Step 2: the initial pictures block message is converted to by specific format by identification arithmetic unit, then initial from this
It is extracted in image block message and obtains certain components, then the certain components are arranged as corresponding column vector, pass through characteristic function meter
Calculate the deflection angle value for obtaining the column vector；
Step 3: one degree of skewness threshold value of setting compares deflection angle value described in step 2 and the degree of skewness threshold value
Compared with, when deflection angle value be less than the degree of skewness threshold value when, initial pictures block message corresponding with the certain components is determined as very
Real face determines initial pictures block message corresponding with the certain components when deflection angle value is greater than the degree of skewness threshold value
For false face.
Specific format described in step 2 is HSV format, and the certain components are V component.
The true and false face recognition detection method based on image degree of skewness further include: after completing step 1 and into
Before row step 2, the initial pictures block message is handled, the black part on initial pictures block message periphery is filtered out
Point.
Initial pictures block message described in step 1 are rgb format.
Characteristic function described in step 2 are as follows:
Wherein, the meaning of γ (V) is image degree of skewness, and the meaning of μ is the mean value of V, and the meaning of σ is the standard deviation of V, E's
Meaning is expectation.
Degree of skewness threshold value described in step 3 is -0.5.
In addition, the present invention also provides a kind of true and false face recognition detection system based on image degree of skewness, including deployment
Image collecting device and the identification arithmetic unit with the relatively independent arrangement of described image acquisition device, image are adopted around animal body
Establishing between acquisition means and identification arithmetic unit has communication connection；
Image collecting device: for obtaining the initial pictures block message of animal body face；
Identify arithmetic unit: for carrying out face recognition to face image acquired in image collecting device；Identify operation
Device includes image conversion module, computing module, comparison module and memory module, and image conversion module, compares mould at computing module
Block is sequentially connected in series, and memory module links together with computing module, comparison module respectively；
Image conversion module is for being converted to the initial pictures block message that image collecting device obtains in hsv color space
The HSV image block message of lower expression；
Characteristic function and degree of skewness threshold value are stored within memory module；
Computing module, which is used to extract from HSV image block message, obtains multiple images block V component, then by multiple images block V
Component is arranged as column vector V, recalls the characteristic function in memory module, by characteristic function calculate obtain step 3 in arrange to
Measure the deflection angle value of V；
The deflection angle value of column vector V is compared by comparison module with the degree of skewness threshold value being stored within memory module,
And comparison result is shown to user.
Described image acquisition device includes the image filters module to interconnect and the camera shooting of at least one near-infrared
Head；
Near-infrared camera: for acquiring the initial pictures block message for obtaining animal body face；
Image filtering module: for filtering out the black portions on initial pictures block message periphery, described image filter module
Block has communication connection as foundation between the output end and the identification arithmetic unit of described image acquisition device.
Described image acquisition device further includes the infrared light compensating lamp connecting with described image filtering module, and infrared light compensating lamp is pressed
The near-infrared camera shooting both sides of head is arranged according to its uniform amount.
The quantity of the infrared light compensating lamp is at least 2.
Above-mentioned technical proposal has the following beneficial effects: using technical solution of the present invention, and institute is obtained by calculation first
The degree of skewness of acquired image block message V component in hsv color space, then by reading the corresponding deflection of the degree of skewness
Threshold value is compared, and can be obtained the judgement true and false to face image, and decision process and method are simple, are easy to by corresponding
Computer program realize, in hsv color space, V component represents illuminance information, according to the comparison to true and false face image
It can be found that false face image face illumination is low and difference is smaller, thus the V component value of false face image is largely concentrated
In low value region, long streaking form is showed in histogram；And true face image illumination hierarchy is enriched, true face's figure
The V component Distribution value of picture is more uniform and span is big, and the form of distribution uniform is showed in histogram, it is seen then that true face
There are significant differences for portion's V component degree of skewness corresponding with false face, therefore, by the way that institute's acquired image block message exists
The degree of skewness of V component is compared with corresponding degree of skewness threshold value in hsv color space, so as to simplify the work of face recognition
Skill, entire identification process only need to calculate the degree of skewness of V component, without carrying out cumbersome operation, make face recognition process more
Increase effect, and since true face V component degree of skewness corresponding with false face has very significant difference, is obtained
Face recognition the result is that true and reliable, confidence level is high, and compared with prior art, user does not need complete under the guidance of system
At corresponding movement, holds the photo of forgery, dynamic video when malicious persons or wear the attempts such as 3D statue mask and pass through and be
When system verifying, since there are significant differences for true face V component degree of skewness corresponding with false face, make malicious persons can not
By verifying, to ensure that the safety of system.In addition, the present invention also provides accordingly based on the true and false face of image degree of skewness
Portion's recognition detection system, the image collecting device in the detection system combined using infrared camera and infrared light compensating lamp extract it is dynamic
Object face initial pictures block message, is not easily susceptible to the influence of external environment illumination, has good robustness, can comprehensive 24
Hour longtime running, also effectively prevents the invasion of malicious persons in time-domain.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow chart of detection method；
Fig. 2 is the connection schematic diagram of detection system of the present invention.
In figure: 1- near-infrared camera, 2- infrared light compensating lamp.
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.
As shown in Figure 1, the present invention provides a kind of true and false face recognition detection method based on image degree of skewness, it is suitable for pair
Human body or other animal faces with life cycle characteristic carry out recognition detection, comprising the following steps:
Step 1: the initial pictures of animal body face are obtained by the image acquisition device being deployed in around animal body
Block message, then the initial pictures block message is sent to the identification operation independently disposed relative to described image acquisition device and is filled
It sets；It is preferred that the initial pictures block message is rgb format.
In addition, the detection method further include: after completing step 1 and before carrying out step 2, to described initial
Image block message is handled, and the black portions on initial pictures block message periphery are filtered out.
Step 2: the initial pictures block message is converted to by specific format by identification arithmetic unit, then initial from this
It is extracted in image block message and obtains certain components, then the certain components are arranged as corresponding column vector, pass through characteristic function meter
Calculate the deflection angle value for obtaining the column vector；Further, the specific format is HSV format, and the certain components are V component.
The characteristic function are as follows:
Wherein, the meaning of γ (V) is image degree of skewness, and the meaning of μ is the mean value of V, and the meaning of σ is the standard deviation of V, E's
Meaning is expectation.
Step 3: one degree of skewness threshold value of setting compares deflection angle value described in step 2 and the degree of skewness threshold value
Compared with, when deflection angle value be less than the degree of skewness threshold value when, initial pictures block message corresponding with the certain components is determined as very
Real face determines initial pictures block message corresponding with the certain components when deflection angle value is greater than the degree of skewness threshold value
For false face.It is preferred that the degree of skewness threshold value is -0.5.The selection needs of degree of skewness threshold value are summarized from many experiments to be obtained
One empirical value, the sample of statistics is more, and the threshold value of setting is more accurate, both can detecte out all false face's targets, can also
Increase true face's percent of pass to the maximum extent.Threshold value is set as -0.5 in the present invention, and degree of skewness is less than threshold value, then it is assumed that the face
Image is true face, otherwise regards as false face image, and system, which is given, to be refused.
The degree of skewness of stochastic variable is a common probability statistics feature, can be used to describe a stochastic variable statistics
The asymmetric distribution deflection of probability distribution.The degree of skewness that standard is just being distributed very much is 0.Degree of skewness is bigger, and corresponding distribution is got over
Toward left side deflection, right side hangover is longer；On the contrary, degree of skewness is smaller, the corresponding more past the right deflection of distribution, left side is trailed longer.
By to really face V component corresponding with false face in the distribution in histogram it can be found that between true and false face's V component
Difference can distinguish characterization well by degree of skewness.The degree of skewness of false face V vector is nearly all greater than 0, very few
The case where negative skewness occur, which is also due to, detects that face is imperfect, the reason of the non-face's content in part, face recognition occurs
Algorithm can't pass, and the degree of skewness of true face V vector is all -0.5 hereinafter, it is therefore preferable that the degree of skewness threshold value is set
It is set to optimal when -0.5.
Using technical solution of the present invention, institute's acquired image block message is obtained by calculation first in hsv color sky
Between middle V component degree of skewness, be then compared by the way that the corresponding deflection of the degree of skewness is read threshold value, can be obtained and face is schemed
As true and false judgement, decision process and method are simple, are easy to realize by corresponding computer program, in hsv color sky
Between in, H component represents hue information, and S component represents saturation infromation, and V component represents illuminance information, according to true and false face
The comparison of image is it can be found that false face image face illumination is low and difference is smaller, thus the V component value of false face image
It is largely focused on low value region, long streaking form is showed in histogram；And true face image illumination hierarchy is enriched,
The V component Distribution value of true face image is more uniform and span is big, and the form of distribution uniform is showed in histogram, can
See, there are significant differences for true face V component degree of skewness corresponding with false face, therefore, by by the collected figure of institute
As the degree of skewness of V component is compared with corresponding degree of skewness threshold value block message in hsv color space, so as to simplify face
The technique of portion's identification, entire identification process only need to calculate the degree of skewness of V component, without carrying out cumbersome operation, make face
Identification process is more efficient, and since the V component degree of skewness presence corresponding with false face of true face is very significant
Difference, the result is that true and reliable, confidence level is high for face recognition obtained, compared with prior art, user do not need be
Corresponding movement is completed under the guidance of system, when malicious persons hold the photo, dynamic video or wearing 3D statue mask of forgery
When equal attempts are verified by system, since there are significant differences for true face V component degree of skewness corresponding with false face, make
Malicious persons can not be by verifying, to ensure that the safety of system.In addition, the present invention also provides inclined based on image accordingly
The true and false face recognition detection system of gradient, the image collecting device in the detection system utilize infrared camera and infrared light filling
Lamp, which combines, extracts animal body face initial pictures block message, is not easily susceptible to the influence of external environment illumination, has good robust
Property, can comprehensive 24 hours longtime runnings, the invasion of malicious persons is also effectively prevented in time-domain.
In addition, as shown in Fig. 2, the present invention also provides a kind of true and false face recognition detection system based on image degree of skewness
System, suitable for carrying out recognition detection to human body or other animal faces with life cycle characteristic, the detection system includes
It is deployed in image collecting device and the identification arithmetic unit with the relatively independent arrangement of described image acquisition device around animal body, is schemed
There is communication connection as establishing between acquisition device and identification arithmetic unit；
Image collecting device: for obtaining the initial pictures block message of animal body face；It is preferred that described image acquisition device
Including the image filters module to interconnect and at least one near-infrared camera；
Near-infrared camera: for acquiring the initial pictures block message for obtaining animal body face；
Image filtering module: for filtering out the black portions on initial pictures block message periphery, described image filter module
Block has communication connection as foundation between the output end and the identification arithmetic unit of described image acquisition device.Also, into one
Walk it is outstanding to select described image acquisition device further include the infrared light compensating lamp connecting with described image filtering module, infrared light compensating lamp is pressed
The near-infrared camera shooting both sides of head is arranged according to its uniform amount.The quantity of the infrared light compensating lamp is at least 2.
In view of narrowband NIR camera is influenced by environment solar irradiation smaller, shoot to obtain with near-infrared camera
Face image illumination it is relatively uniform.Consider the low situation of night-environment illuminance simultaneously, two are evenly arranged around camera
The infrared light compensating lamp of a some strength is able to carry out light filling, thus make entirely to detect identifying system no matter daytime or night all
Can operate normally and work, which can filter out a large amount of attack tools with smooth surface medium, as mobile phone,
Tablet computer, smooth photographic paper, 3D printing mask etc..It can be generated when infrared light filling is irradiated to above-mentioned smooth dielectric surface strong
It is reflective, camera acquired image almost one it is black, comprising seldom useful information, can not detect face, exclude false
Face cheating, false face is due to being other objects forged, and surface roughness is larger, and light diffusing reflection occurs on its surface,
To enable camera is more accurate reliably to obtain face's initial pictures block message.
Identify arithmetic unit: for carrying out face recognition to face image acquired in image collecting device；Identify operation
Device includes image conversion module, computing module, comparison module and memory module, and image conversion module, compares mould at computing module
Block is sequentially connected in series, and memory module links together with computing module, comparison module respectively；
Image conversion module is for being converted to the initial pictures block message that image collecting device obtains in hsv color space
The HSV image block message of lower expression；
Characteristic function and degree of skewness threshold value are stored within memory module；
Computing module, which is used to extract from HSV image block message, obtains multiple images block V component, then by multiple images block V
Component is arranged as column vector V, recalls the characteristic function in memory module, by characteristic function calculate obtain step 3 in arrange to
Measure the deflection angle value of V；
The deflection angle value of column vector V is compared by comparison module with the degree of skewness threshold value being stored within memory module,
And comparison result is shown to user.
Above specific embodiment has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Illustrate, it should be understood that the above is only a specific embodiment of the invention, the protection model that is not intended to limit the present invention
It encloses, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention
Protection scope within.
Claims (10)
1. a kind of true and false face recognition detection method based on image degree of skewness, it is characterised in that: the following steps are included:
Step 1: believed by the initial pictures block that the image acquisition device being deployed in around animal body obtains animal body face
Breath, then the initial pictures block message is sent to the identification arithmetic unit independently disposed relative to described image acquisition device；
Step 2: being converted to specific format for the initial pictures block message by identification arithmetic unit, then from the initial pictures
It is extracted in block message and obtains certain components, then the certain components are arranged as corresponding column vector, obtained by characteristic function calculating
Obtain the deflection angle value of the column vector；
Step 3: deflection angle value described in step 2 is compared by one degree of skewness threshold value of setting with the degree of skewness threshold value, when
When deflection angle value is less than the degree of skewness threshold value, initial pictures block message corresponding with the certain components is determined as true face
Initial pictures block message corresponding with the certain components is determined as void when deflection angle value is greater than the degree of skewness threshold value by portion
False face.
2. the true and false face recognition detection method based on image degree of skewness as described in claim 1, it is characterised in that: step 2
Described in specific format be HSV format, the certain components be V component.
3. the true and false face recognition detection method based on image degree of skewness as described in claim 1, it is characterised in that: the base
In the true and false face recognition detection method of image degree of skewness further include: after completing step 1 and before carrying out step 2,
The initial pictures block message is handled, the black portions on initial pictures block message periphery are filtered out.
4. the true and false face recognition detection method based on image degree of skewness as described in claim 1, it is characterised in that: step 1
Described in initial pictures block message be rgb format.
5. the true and false face recognition detection method based on image degree of skewness as described in claim 1, it is characterised in that: step 2
Described in characteristic function are as follows:
Wherein, the meaning of γ (V) is image degree of skewness, and the meaning of μ is the mean value of V, and the meaning of σ is the standard deviation of V, the meaning of E
For expectation.
6. the true and false face recognition detection method based on image degree of skewness as described in claim 1, it is characterised in that: step 3
Described in degree of skewness threshold value be -0.5.
7. a kind of true and false face recognition detection system based on image degree of skewness, it is characterised in that: including being deployed in animal body week
Enclose image collecting device and the identification arithmetic unit with the relatively independent arrangement of described image acquisition device, image collecting device and knowledge
Establishing between other arithmetic unit has communication connection；
Image collecting device: for obtaining the initial pictures block message of animal body face；
Identify arithmetic unit: for carrying out face recognition to face image acquired in image collecting device；Identify arithmetic unit
Including image conversion module, computing module, comparison module and memory module, image conversion module, computing module, comparison module according to
Secondary series connection, memory module link together with computing module, comparison module respectively；
Image conversion module is for being converted to the initial pictures block message that image collecting device obtains in hsv color space following table
The HSV image block message shown；
Characteristic function and degree of skewness threshold value are stored within memory module；
Computing module, which is used to extract from HSV image block message, obtains multiple images block V component, then by multiple images block V component
It is arranged as column vector V, recalls the characteristic function in memory module, is calculated by characteristic function and obtains column vector V in step 3
Deflection angle value；
The deflection angle value of column vector V is compared by comparison module with the degree of skewness threshold value being stored within memory module, and will
Comparison result is shown to user.
8. the true and false face recognition detection method based on image degree of skewness as claimed in claim 7, it is characterised in that: the figure
As acquisition device includes the image filters module to interconnect and at least one near-infrared camera；
Near-infrared camera: for acquiring the initial pictures block message for obtaining animal body face；
Image filtering module: for filtering out the black portions on initial pictures block message periphery, described image filtering module is made
Establishing between the output end and the identification arithmetic unit of described image acquisition device has communication connection.
9. the true and false face recognition detection method based on image degree of skewness as claimed in claim 8, it is characterised in that: the figure
As acquisition device further includes the infrared light compensating lamp connecting with described image filtering module, infrared light compensating lamp is arranged according to its uniform amount
Cloth images both sides of head in the near-infrared.
10. the true and false face recognition detection method based on image degree of skewness as claimed in claim 9, it is characterised in that: described
The quantity of infrared light compensating lamp is at least 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910179174.1A CN109961025B (en) | 2019-03-11 | 2019-03-11 | True and false face identification and detection method and detection system based on image skewness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910179174.1A CN109961025B (en) | 2019-03-11 | 2019-03-11 | True and false face identification and detection method and detection system based on image skewness |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109961025A true CN109961025A (en) | 2019-07-02 |
| CN109961025B CN109961025B (en) | 2020-01-24 |
Family
ID=67024210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910179174.1A Active CN109961025B (en) | 2019-03-11 | 2019-03-11 | True and false face identification and detection method and detection system based on image skewness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961025B (en) |
Citations (9)
| 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 |
| CN101976352A (en) * | 2010-10-29 | 2011-02-16 | 上海交通大学 | Various illumination face identification method based on small sample emulating and sparse expression |
| CN103218615A (en) * | 2013-04-17 | 2013-07-24 | 哈尔滨工业大学深圳研究生院 | Face judgment method |
| CN105654028A (en) * | 2015-09-29 | 2016-06-08 | 厦门中控生物识别信息技术有限公司 | True and false face identification method and apparatus thereof |
| JP2016112762A (en) * | 2014-12-15 | 2016-06-23 | シヤチハタ株式会社 | Processing method of printing surface |
| CN105718925A (en) * | 2016-04-14 | 2016-06-29 | 苏州优化智能科技有限公司 | Real person living body authentication terminal equipment based on near infrared and facial micro expression |
| CN106372615A (en) * | 2016-09-19 | 2017-02-01 | 厦门中控生物识别信息技术有限公司 | Face anti-counterfeiting identification method and apparatus |
| CN106650669A (en) * | 2016-12-27 | 2017-05-10 | 重庆邮电大学 | Face recognition method for identifying counterfeit photo deception |
| CN107992794A (en) * | 2016-12-30 | 2018-05-04 | 腾讯科技（深圳）有限公司 | A kind of biopsy method, device and storage medium |
-
2019
- 2019-03-11 CN CN201910179174.1A patent/CN109961025B/en active Active
Patent Citations (9)
| 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 |
| CN101976352A (en) * | 2010-10-29 | 2011-02-16 | 上海交通大学 | Various illumination face identification method based on small sample emulating and sparse expression |
| CN103218615A (en) * | 2013-04-17 | 2013-07-24 | 哈尔滨工业大学深圳研究生院 | Face judgment method |
| JP2016112762A (en) * | 2014-12-15 | 2016-06-23 | シヤチハタ株式会社 | Processing method of printing surface |
| CN105654028A (en) * | 2015-09-29 | 2016-06-08 | 厦门中控生物识别信息技术有限公司 | True and false face identification method and apparatus thereof |
| CN105718925A (en) * | 2016-04-14 | 2016-06-29 | 苏州优化智能科技有限公司 | Real person living body authentication terminal equipment based on near infrared and facial micro expression |
| CN106372615A (en) * | 2016-09-19 | 2017-02-01 | 厦门中控生物识别信息技术有限公司 | Face anti-counterfeiting identification method and apparatus |
| CN106650669A (en) * | 2016-12-27 | 2017-05-10 | 重庆邮电大学 | Face recognition method for identifying counterfeit photo deception |
| CN107992794A (en) * | 2016-12-30 | 2018-05-04 | 腾讯科技（深圳）有限公司 | A kind of biopsy method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 刘华成: "人脸活体检测关键技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109961025B (en) | 2020-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shao et al. | Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing | |
| US9922238B2 (en) | Apparatuses, systems, and methods for confirming identity | |
| CN105930709B (en) | Face recognition technology is applied to the method and device of testimony of a witness consistency check | |
| CN106874871B (en) | Living body face double-camera identification method and identification device | |
| Schwartz et al. | Face spoofing detection through partial least squares and low-level descriptors | |
| WO2019134536A1 (en) | Neural network model-based human face living body detection | |
| CN100361135C (en) | Method for acquiring human-face image, human-face discrimination and discriminating system | |
| CN107798281B (en) | Face living body detection method and device based on LBP (local binary pattern) characteristics | |
| WO2019137178A1 (en) | Face liveness detection | |
| CN109101871A (en) | A kind of living body detection device based on depth and Near Infrared Information, detection method and its application | |
| CN107423703B (en) | Multi-mode recognition device and method based on face, fingerprint and finger vein features | |
| CN103678984A (en) | Method for achieving user authentication by utilizing camera | |
| US9449217B1 (en) | Image authentication | |
| CN110688901A (en) | Face recognition method and device | |
| CN110443102B (en) | Living body face detection method and device | |
| CN105447443A (en) | Method and device for improving identification accuracy of iris identification device | |
| CN109902604B (en) | High-safety face comparison system and method based on Feiteng platform | |
| US11315360B2 (en) | Live facial recognition system and method | |
| CN105335691A (en) | Smiling face identification and encouragement system | |
| CN112818722A (en) | Modular dynamically configurable living body face recognition system | |
| CN109657627A (en) | Auth method, device and electronic equipment | |
| Jiang et al. | Multilevel fusing paired visible light and near-infrared spectral images for face anti-spoofing | |
| Hadiprakoso et al. | Face anti-spoofing using CNN classifier & face liveness detection | |
| CN108446687A (en) | A kind of adaptive face vision authentication method based on mobile terminal and backstage interconnection | |
| CN111860394A (en) | Gesture estimation and gesture detection-based action living body recognition method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wang Shuigen Inventor after: Cui Dongshun Inventor after: Qian Xing Inventor before: Wang Shuigen Inventor before: Cui Dongshun Inventor before: Qian Xing Inventor before: Huang Guangbin |
|
| CB03 | Change of inventor or designer information |