CN108596082A - Human face in-vivo detection method based on image diffusion velocity model and color character - Google Patents
Human face in-vivo detection method based on image diffusion velocity model and color character Download PDFInfo
- Publication number
- CN108596082A CN108596082A CN201810365304.6A CN201810365304A CN108596082A CN 108596082 A CN108596082 A CN 108596082A CN 201810365304 A CN201810365304 A CN 201810365304A CN 108596082 A CN108596082 A CN 108596082A
- Authority
- CN
- China
- Prior art keywords
- image
- convolutional neural
- neural networks
- diffusion velocity
- color character
- 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
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The human face in-vivo detection method based on image diffusion velocity model and color character that the present invention relates to a kind of, belongs to image procossing and technical field of computer vision.Including：S1 does normalized to face images, and is classified as training set and test set；Diffusion velocity image is sent into convolutional neural networks by S2, using training set training convolutional neural networks, and uses the class probability of network acquisition test set；S3 extracts facial image color character, training SVM models, and uses the class probability of model acquisition test set；Weighting network is respectively trained according to class probability of the training set in convolutional neural networks and SVM models in S4, obtains respective optimal weights；Class probability of the test set in convolutional neural networks and SVM is done Weighted Fusion by S5, obtains final detection result.Image information is more fully utilized in combination with image diffusion velocity feature and color character in the present invention, improves the accuracy rate of face In vivo detection.
Description
Technical field
The invention belongs to image procossing and technical field of computer vision, be related to it is a kind of based on image diffusion velocity model and
The human face in-vivo detection method of color character.
Background technology
Recognition of face is always the research hotspot of computer vision field in recent years, in fields such as security protection, traffic, finance
It is widely used.It is welcomed by the people since the technology is that authentication procedures bring prodigious convenience.With this
Meanwhile higher requirements are also raised for safety of the people to recognition of face.Current face recognition technology is based primarily upon image
Information, therefore, the technology are easy under attack.There are mainly two types of attack types：The photo attack of validated user, validated user
Video is attacked.
Real human face image is obtained by shooting three-dimensional face, and false facial image passes through secondary acquisition validated user
Photo or video obtain.Difference based on real human face and false face, current research method are broadly divided into three kinds：The first
It is the human face in-vivo detection method based on texture information, for such methods under specific light environment, performance is preferable, but extensive
Ability is weak.Second is the human face in-vivo detection method based on movable information.Such methods utilize the natural reaction of human body, such as
Blink, head movement, proposes corresponding face In vivo detection algorithm at mouth movement respectively.Face live body based on movable information
Detection method is easy to be influenced by natural environments such as illumination, and usually requires processing sequence image, and it is longer to calculate the time.Third
Kind is the human face in-vivo detection method based on picture quality.Real human face image is obtained by shooting three-dimensional face, shooting process
In be easy to focus, image is more clear, and false face is obtained by the photo or video of secondary acquisition validated user, it is not easy to
It focuses, it may lost part information during secondary acquisition.
Real human face is three-dimensional structure, and diffusing reflection can be instead given birth to after light irradiates, and profile ambient light changes greatly, and image expands
It is fast to dissipate speed.Photo and video are two dimensional surface, and mirror-reflection can occur after light irradiates, and the variation of profile ambient light is small,
It is slow to spread express delivery.Diffusion velocity model can be good at assessing this species diversity.In terms of color, due to physics is imaged, very
Real facial image will present out abundanter color information.Therefore, color information is difference real human face image and false people
The important textural characteristics of face image.
Invention content
In view of this, the purpose of the present invention is to provide a kind of face based on image diffusion velocity model and color character
Biopsy method, image diffusion velocity model can describe the space characteristics of image well, and color character information is difference
The important textural characteristics of real human face image and false facial image；It is promoted in conjunction with image diffusion velocity model and color character
Face recognition accuracy rate.
In order to achieve the above objectives, the present invention provides the following technical solutions：
Human face in-vivo detection method based on image diffusion velocity model and color character, includes the following steps：
S1：Facial image is extracted, normalized is done to face images, and be classified as training set and test set；
S2：Anisotropic diffusion is carried out to facial image, and diffusion velocity image is sent into convolutional neural networks, uses instruction
Practice collection training convolutional neural networks, and obtains the class probability of test set using the network；
S3：Facial image color character is extracted, trains SVM models using training set, and test set is obtained using the model
Class probability；
S4：Weighting network is trained according to output probability of the training set in convolutional neural networks and SVM models, is obtained optimal
Weight；
S5：According to optimal weights, classification results of the test set in convolutional neural networks and SVM are done into Weighted Fusion, are obtained
To final detection result.
Further, facial image is extracted described in step S1, specifically included：
It is used as original from randomly selecting a frame image in video sequence or randomly selecting an image from a series of images
Beginning image；User's face detection algorithm detects the position of face, and cuts out facial image.
Further, step S2 training convolutional neural networks, specifically include：
Anisotropy parameter is carried out to facial image, enhances the marginal information of image；By facial image anisotropy parameter
Diffusion velocity of the front and back difference as image builds diffusion velocity model；And diffusion velocity image is sent into convolutional Neural net
Network using training set training convolutional neural networks, and uses the class probability of network acquisition test set；
Further, step S3 extracts the color character of facial image, specifically includes：
The color space of facial image is converted into tone H, saturation degree S, lightness V from the RGB patterns of red R, green G, indigo plant B
HSV patterns；Mean value, variance, the degree of bias of HSV color space facial images are calculated, while the direction gradient for calculating facial image is straight
Fang Tu；Using the minimum value of the mean value of facial image, variance, the degree of bias and histograms of oriented gradients and maximum value as face figure
The color character of picture.
Further, training weighting network, obtains optimal weights, step S4 is specifically included：
1) random initializtion output probability weight w1、w2And the weight w of network nodeq；
2) utilize direction of error propagation algorithm with new weight w1、w2；
3) fixed w1、w2, update w using error backpropagation algorithmq；
2) and the 3) step 4) iteration carries out the, until loss function is restrained, obtains optimal weights.
Further, step S5 is specifically included：
Class probability of the test set in convolutional neural networks and SVM models is obtained, according to optimal weights by two kinds of probability
It is worth weighted sum, the final probability value expression of certain classification：
P=P1w1+P2w2
Wherein：P1For the probability of the category under convolutional neural networks；w1For the optimal weights under convolutional neural networks；P2For
The probability of the category under SVM models；w2For the optimal weights under SVM；Take the maximum classification of probability value as final detection knot
Fruit.
The beneficial effects of the present invention are：1) present invention proposes a kind of based on image diffusion velocity model and color character
Face In vivo detection algorithm, this method can accurately detect false facial image.2) expansion of this method Simultaneous Extracting Image
Velocity characteristic and color character are dissipated, image information is more comprehensively utilized, improves the accuracy rate of face In vivo detection.
Description of the drawings
In order to keep the purpose of the present invention, technical solution and advantageous effect clearer, the present invention provides following attached drawing and carries out
Explanation：
Fig. 1 is overall flow schematic diagram of the present invention；
Fig. 2 is convolutional neural networks structural schematic diagram；
Fig. 3 is convolutional neural networks assorting process schematic diagram；
Fig. 4 is ROC curve figure of the present invention on test set.
Specific implementation mode
It is a kind of to the present invention with reference to the accompanying drawings of the specification to be lived based on the face of image diffusion velocity model and color character
Body detecting method is further detailed.
In the present invention, user's face detection algorithm detects the position of face in original image first, and to face figure
As doing normalized.Then the color character of facial image after extraction normalizes, and training SVM models.Speed is spread in image
It spends in model, anisotropy parameter is carried out to the facial image after normalization, by the difference before and after facial image anisotropy parameter
It is worth the diffusion velocity as image, builds diffusion velocity model；And diffusion velocity image is sent into convolutional neural networks, use instruction
Practice collection training convolutional neural networks, and obtains the class probability of test set using the network.It is instructed using error backpropagation algorithm
Practice weighting network, obtains optimal weights.Weighted Fusion is done to classification results according to optimal weights, obtains final classification result.
This data set used shares 50 testers, covers the ethnic group of the different colours of skin, shares 1200 short-sighted frequencies,
Positive sample (real human face video) 200, negative sample (false face video) 1000.
Fig. 1 be the present invention is based on the schematic diagram of the human face in-vivo detection method of image diffusion velocity model and color character,
As shown, the method for the present invention specifically includes following steps：
S1：Facial image is extracted, normalized is done to face images, and be classified as training set and test set.
The extraction facial image refer to from a frame image is randomly selected in video sequence or from a series of images with
Machine extracts an image as original image；User's face detection algorithm detects the position of face, and cuts out facial image.
S2：Anisotropic diffusion is carried out to facial image, and diffusion velocity image is sent into convolutional neural networks, uses instruction
Practice collection training convolutional neural networks, and obtains the class probability of test set using the network.
Anisotropic diffusion is mainly used to smoothed image in image processing field, overcomes Gaussian noise, uses items herein
Anisotropic diffusion enhances the marginal information of image.Real human face is three-dimensional structure, and diffusing reflection, profile week can be instead given birth to after light irradiates
Enclose light variation greatly, image diffusion velocity is fast.Photo and video are two dimensional surface, and mirror-reflection can occur after light irradiates,
The variation of facial contour ambient light is small, and diffusion express delivery is slow.Diffusion velocity model can be good at assessing this species diversity, anisotropy
Diffusion formula：
I is original image, and k indicates iterations,For gradient operator, div is divergence, controls diffusion rate.
Convolutional neural networks model that the present invention uses is as shown in Fig. 2, share 5 convolutional layers, 2 full articulamentums, and preceding two
A convolutional layer and first full articulamentum also include a pond layer.Convolutional neural networks can extract diffusion velocity image well
Feature.Convolutional neural networks to the assorting process of facial image as shown in figure 3, input is diffusion velocity image, export for point
Class probability.
The present invention first to facial image carry out anisotropy parameter, using difference before and after facial image anisotropy parameter as
The diffusion velocity of image；Then training set training convolutional neural networks are used, and general using the classification of network acquisition test set
Rate.
S3：Facial image color character is extracted, trains SVM models using training set, and test set is obtained using the model
Class probability；
The color space of facial image is converted into tone H, saturation degree S, lightness V from the RGB patterns of red R, green G, indigo plant B
HSV patterns；Mean value, variance, the degree of bias of HSV color space facial images are calculated, while the direction gradient for calculating facial image is straight
Fang Tu；Using the minimum value of the mean value of facial image, variance, the degree of bias and histograms of oriented gradients and maximum value as face figure
The color character of picture.
Wherein, rgb color pattern is a kind of color standard of industrial quarters, is by red (R), green (G), three, indigo plant (B)
The variation of Color Channel and their mutual superpositions obtain miscellaneous color, RGB be represent it is red, green,
The color in blue three channels.HSV is a kind of color sky created in 1978 by A.R.Smith according to the intuitive nature of color
Between, also referred to as hexagonal pyramid model.The parameter of color is respectively in this model：Tone (H), saturation degree (S), lightness (V).
S4：Weighting network is trained according to output probability of the training set in convolutional neural networks and SVM models, is obtained optimal
Weight；
It specifically includes using error backpropagation algorithm training weight network, obtains optimal weights, weight network was trained
Journey is as follows：
1) random initializtion output probability weight w1、w2And the weight w of network nodeq。
2) utilize direction of error propagation algorithm with new weight w1、w2。
3) fixed w1、w2, update w using error backpropagation algorithmq。
2) and the 3) step 4) iteration carries out the, until loss function is restrained, obtains optimal weights.
S5：According to optimal weights, classification results of the test set in convolutional neural networks and SVM are done into Weighted Fusion, are obtained
To final detection result.
Class probability of the test set in convolutional neural networks and SVM models is obtained, according to optimal weights by two kinds of probability
It is worth weighted sum, the final probability value expression of certain classification：
P=P1w1+P2w2
Wherein：P1For the probability of the category under convolutional neural networks；w1For the optimal weights under convolutional neural networks；P2For
The probability of the category under SVM models；w2For the optimal weights under SVM；Take the maximum classification of probability value as final detection knot
Fruit.
Fig. 4 is ROC curve figure of the present invention on test set.Probability (the False Acceptance that mistake receives
Rate, FAR) and the probability (False Rejection Rate, FRR) of False Rejects be defined as follows：
FP is false positive example, and TN is true counter-example, and FN is false counter-example, and TP is real example.The present invention is testing as can be seen from Figure 4
The probability that the probability of False Rejects and mistake receive on collection has all reached very low level, and waits error rates (Half Total
Error Rate, HTER) probability of happening be no more than 0.05, this also illustrates the present invention can accurately detect false people
Face.Etc. error rates be defined as follows：
Wherein, D is test set, and threshold value when τ is test, FAR is wrong acceptance probability, and FRR is False Rejects probability.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include：ROM, RAM, disk or CD etc..
Embodiment provided above has carried out further detailed description, institute to the object, technical solutions and advantages of the present invention
It should be understood that embodiment provided above is only the preferred embodiment of the present invention, be not intended to limit the invention, it is all
Any modification, equivalent substitution, improvement and etc. made for the present invention, should be included in the present invention within the spirit and principles in the present invention
Protection domain within.
Claims (6)
1. the human face in-vivo detection method based on image diffusion velocity model and color character, which is characterized in that including following step
Suddenly：
S1：Facial image is extracted, normalized is done to face images, and be classified as training set and test set；
S2：Anisotropic diffusion is carried out to facial image, and diffusion velocity image is sent into convolutional neural networks, uses training set
Training convolutional neural networks, and use the class probability of network acquisition test set；
S3：Facial image color character is extracted, trains SVM models using training set, and obtain point of test set using the model
Class probability；
S4：Weighting network is respectively trained according to class probability of the training set in convolutional neural networks and SVM models, obtains respectively
Optimal weights；
S5：According to convolutional neural networks and the respective optimal weights of SVM models, by test set in convolutional neural networks and SVM
Class probability do Weighted Fusion, obtain final detection result.
2. the human face in-vivo detection method according to claim 1 based on image diffusion velocity model and color character,
It is characterized in that：Facial image is extracted described in step S1, is specifically included：
From randomly selecting a frame image in video sequence or randomly select an image from a series of images as original graph
Picture；User's face detection algorithm detects the position of face, and cuts out facial image.
3. the human face in-vivo detection method according to claim 1 based on image diffusion velocity model and color character,
It is characterized in that：Step S2 is specifically included：
Anisotropy parameter is carried out to facial image, enhances the marginal information of image；Before and after facial image anisotropy parameter
Diffusion velocity of the difference as image, build diffusion velocity model；Diffusion velocity image is sent into convolutional neural networks, is used
Training set training convolutional neural networks, and use the class probability of network acquisition test set.
4. the human face in-vivo detection method according to claim 1 based on image diffusion velocity model and color character,
It is characterized in that：The color character of step S3 extraction facial images, specifically includes：
The color space of facial image is converted into the HSV of tone H, saturation degree S, lightness V from the RGB patterns of red R, green G, indigo plant B
Pattern；Mean value, variance, the degree of bias of HSV color space facial images are calculated, while calculating the direction gradient histogram of facial image
Figure；Using the minimum value of the mean value of facial image, variance, the degree of bias and histograms of oriented gradients and maximum value as facial image
Color character.
5. the human face in-vivo detection method according to claim 1 based on image diffusion velocity model and color character,
It is characterized in that：Step S4 is specifically included：
1) the weight w of random initializtion output probability1、w2And the weight w of network nodeq；
2) utilize direction of error propagation algorithm with new weight w1、w2；
3) fixed w1、w2, update w using error backpropagation algorithmq；
2) and the 3) step 4) iteration carries out the, until loss function is restrained, obtains optimal weights.
6. the human face in-vivo detection method according to claim 1 based on image diffusion velocity model and color character,
It is characterized in that：Step S5 is specifically included：
Class probability of the test set in convolutional neural networks and SVM models is obtained, is added two kinds of probability values according to optimal weights
Power summation, the final probability value expression of certain classification：
P=P1w1+P2w2
Wherein：P1For the probability of the category under convolutional neural networks；w1For the optimal weights under convolutional neural networks；P2For SVM moulds
The probability of the category under type；w2For the optimal weights under SVM；Take the maximum classification of probability value as final testing result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810365304.6A CN108596082A (en) | 2018-04-20 | 2018-04-20 | Human face in-vivo detection method based on image diffusion velocity model and color character |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810365304.6A CN108596082A (en) | 2018-04-20 | 2018-04-20 | Human face in-vivo detection method based on image diffusion velocity model and color character |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108596082A true CN108596082A (en) | 2018-09-28 |
Family
ID=63614664
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810365304.6A Pending CN108596082A (en) | 2018-04-20 | 2018-04-20 | Human face in-vivo detection method based on image diffusion velocity model and color character |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108596082A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583342A (en) * | 2018-11-21 | 2019-04-05 | 重庆邮电大学 | Human face in-vivo detection method based on transfer learning |
| CN109685018A (en) * | 2018-12-26 | 2019-04-26 | 深圳市捷顺科技实业股份有限公司 | A kind of testimony of a witness method of calibration, system and relevant device |
| CN109858375A (en) * | 2018-12-29 | 2019-06-07 | 深圳市软数科技有限公司 | Living body faces detection method, terminal and computer readable storage medium |
| CN109902613A (en) * | 2019-02-23 | 2019-06-18 | 重庆邮电大学 | A kind of human body feature extraction method based on transfer learning and image enhancement |
| CN110175655A (en) * | 2019-06-03 | 2019-08-27 | 中国科学技术大学 | Data identification method and device, storage medium and electronic equipment |
| CN110189307A (en) * | 2019-05-14 | 2019-08-30 | 慧影医疗科技（北京）有限公司 | A kind of pulmonary nodule detection method and system based on multi-model fusion |
| CN110348322A (en) * | 2019-06-19 | 2019-10-18 | 西华师范大学 | Human face in-vivo detection method and equipment based on multi-feature fusion |
| CN110705392A (en) * | 2019-09-17 | 2020-01-17 | Oppo广东移动通信有限公司 | Face image detection method and device and storage medium |
| CN111191521A (en) * | 2019-12-11 | 2020-05-22 | 智慧眼科技股份有限公司 | Face living body detection method and device, computer equipment and storage medium |
| CN111444364A (en) * | 2020-03-04 | 2020-07-24 | 中国建设银行股份有限公司 | Image detection method and device |
| CN111539910A (en) * | 2020-03-13 | 2020-08-14 | 石家庄铁道大学 | Corrosion region detection method and terminal equipment |
| CN111582045A (en) * | 2020-04-15 | 2020-08-25 | 深圳市爱深盈通信息技术有限公司 | Living body detection method and device and electronic equipment |
| CN111967289A (en) * | 2019-05-20 | 2020-11-20 | 高新兴科技集团股份有限公司 | Uncooperative human face in-vivo detection method and computer storage medium |
| CN112070041A (en) * | 2020-09-14 | 2020-12-11 | 北京印刷学院 | Living body face detection method and device based on CNN deep learning model |
| CN112927164A (en) * | 2021-03-22 | 2021-06-08 | 重庆邮电大学 | No-reference low-illumination image enhancement method based on deep convolutional neural network |
| CN113792701A (en) * | 2021-09-24 | 2021-12-14 | 北京市商汤科技开发有限公司 | Living body detection method and device, computer equipment and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069400A (en) * | 2015-07-16 | 2015-11-18 | 北京工业大学 | Face image gender recognition system based on stack type sparse self-coding |
| CN105389554A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Face-identification-based living body determination method and equipment |
| CN105787458A (en) * | 2016-03-11 | 2016-07-20 | 重庆邮电大学 | Infrared behavior identification method based on adaptive fusion of artificial design feature and depth learning feature |
| US20170039418A1 (en) * | 2013-12-31 | 2017-02-09 | Beijing Techshino Technology Co., Ltd. | Face authentication method and device |
| CN106485230A (en) * | 2016-10-18 | 2017-03-08 | 中国科学院重庆绿色智能技术研究院 | Based on the training of the Face datection model of neutral net, method for detecting human face and system |
| CN106599797A (en) * | 2016-11-24 | 2017-04-26 | 北京航空航天大学 | Infrared face identification method based on local parallel nerve network |
| CN106778590A (en) * | 2016-12-09 | 2017-05-31 | 厦门大学 | It is a kind of that video detecting method is feared based on convolutional neural networks model cruelly |
| CN106971161A (en) * | 2017-03-27 | 2017-07-21 | 深圳大图科创技术开发有限公司 | Face In vivo detection system based on color and singular value features |
| CN107220635A (en) * | 2017-06-21 | 2017-09-29 | 北京市威富安防科技有限公司 | Human face in-vivo detection method based on many fraud modes |
| CN107292887A (en) * | 2017-06-20 | 2017-10-24 | 电子科技大学 | A kind of Segmentation Method of Retinal Blood Vessels based on deep learning adaptive weighting |
| CN107480611A (en) * | 2017-07-31 | 2017-12-15 | 浙江大学 | A kind of crack identification method based on deep learning convolutional neural networks |
| CN107679525A (en) * | 2017-11-01 | 2018-02-09 | 腾讯科技（深圳）有限公司 | Image classification method, device and computer-readable recording medium |
| CN107862299A (en) * | 2017-11-28 | 2018-03-30 | 电子科技大学 | A kind of living body faces detection method based on near-infrared Yu visible ray binocular camera |
-
2018
- 2018-04-20 CN CN201810365304.6A patent/CN108596082A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170039418A1 (en) * | 2013-12-31 | 2017-02-09 | Beijing Techshino Technology Co., Ltd. | Face authentication method and device |
| CN105069400A (en) * | 2015-07-16 | 2015-11-18 | 北京工业大学 | Face image gender recognition system based on stack type sparse self-coding |
| CN105389554A (en) * | 2015-11-06 | 2016-03-09 | 北京汉王智远科技有限公司 | Face-identification-based living body determination method and equipment |
| CN105787458A (en) * | 2016-03-11 | 2016-07-20 | 重庆邮电大学 | Infrared behavior identification method based on adaptive fusion of artificial design feature and depth learning feature |
| CN106485230A (en) * | 2016-10-18 | 2017-03-08 | 中国科学院重庆绿色智能技术研究院 | Based on the training of the Face datection model of neutral net, method for detecting human face and system |
| CN106599797A (en) * | 2016-11-24 | 2017-04-26 | 北京航空航天大学 | Infrared face identification method based on local parallel nerve network |
| CN106778590A (en) * | 2016-12-09 | 2017-05-31 | 厦门大学 | It is a kind of that video detecting method is feared based on convolutional neural networks model cruelly |
| CN106971161A (en) * | 2017-03-27 | 2017-07-21 | 深圳大图科创技术开发有限公司 | Face In vivo detection system based on color and singular value features |
| CN107292887A (en) * | 2017-06-20 | 2017-10-24 | 电子科技大学 | A kind of Segmentation Method of Retinal Blood Vessels based on deep learning adaptive weighting |
| CN107220635A (en) * | 2017-06-21 | 2017-09-29 | 北京市威富安防科技有限公司 | Human face in-vivo detection method based on many fraud modes |
| CN107480611A (en) * | 2017-07-31 | 2017-12-15 | 浙江大学 | A kind of crack identification method based on deep learning convolutional neural networks |
| CN107679525A (en) * | 2017-11-01 | 2018-02-09 | 腾讯科技（深圳）有限公司 | Image classification method, device and computer-readable recording medium |
| CN107862299A (en) * | 2017-11-28 | 2018-03-30 | 电子科技大学 | A kind of living body faces detection method based on near-infrared Yu visible ray binocular camera |
Non-Patent Citations (1)
| Title |
|---|
| WONJUN KIM ET AL.: "Face Liveness Detection From a Single Image via Diffusion Speed Model", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583342A (en) * | 2018-11-21 | 2019-04-05 | 重庆邮电大学 | Human face in-vivo detection method based on transfer learning |
| CN109583342B (en) * | 2018-11-21 | 2022-11-04 | 重庆邮电大学 | Human face living body detection method based on transfer learning |
| CN109685018A (en) * | 2018-12-26 | 2019-04-26 | 深圳市捷顺科技实业股份有限公司 | A kind of testimony of a witness method of calibration, system and relevant device |
| CN109858375A (en) * | 2018-12-29 | 2019-06-07 | 深圳市软数科技有限公司 | Living body faces detection method, terminal and computer readable storage medium |
| CN109858375B (en) * | 2018-12-29 | 2023-09-26 | 简图创智(深圳)科技有限公司 | Living body face detection method, terminal and computer readable storage medium |
| CN109902613A (en) * | 2019-02-23 | 2019-06-18 | 重庆邮电大学 | A kind of human body feature extraction method based on transfer learning and image enhancement |
| CN110189307A (en) * | 2019-05-14 | 2019-08-30 | 慧影医疗科技（北京）有限公司 | A kind of pulmonary nodule detection method and system based on multi-model fusion |
| CN110189307B (en) * | 2019-05-14 | 2021-11-23 | 慧影医疗科技（北京）有限公司 | Pulmonary nodule detection method and system based on multi-model fusion |
| CN111967289A (en) * | 2019-05-20 | 2020-11-20 | 高新兴科技集团股份有限公司 | Uncooperative human face in-vivo detection method and computer storage medium |
| CN110175655A (en) * | 2019-06-03 | 2019-08-27 | 中国科学技术大学 | Data identification method and device, storage medium and electronic equipment |
| CN110348322A (en) * | 2019-06-19 | 2019-10-18 | 西华师范大学 | Human face in-vivo detection method and equipment based on multi-feature fusion |
| CN110705392A (en) * | 2019-09-17 | 2020-01-17 | Oppo广东移动通信有限公司 | Face image detection method and device and storage medium |
| CN111191521A (en) * | 2019-12-11 | 2020-05-22 | 智慧眼科技股份有限公司 | Face living body detection method and device, computer equipment and storage medium |
| CN111191521B (en) * | 2019-12-11 | 2022-08-12 | 智慧眼科技股份有限公司 | Face living body detection method and device, computer equipment and storage medium |
| CN111444364A (en) * | 2020-03-04 | 2020-07-24 | 中国建设银行股份有限公司 | Image detection method and device |
| CN111444364B (en) * | 2020-03-04 | 2024-01-30 | 中国建设银行股份有限公司 | Image detection method and device |
| CN111539910A (en) * | 2020-03-13 | 2020-08-14 | 石家庄铁道大学 | Corrosion region detection method and terminal equipment |
| CN111539910B (en) * | 2020-03-13 | 2024-03-19 | 石家庄铁道大学 | Rust area detection method and terminal equipment |
| CN111582045A (en) * | 2020-04-15 | 2020-08-25 | 深圳市爱深盈通信息技术有限公司 | Living body detection method and device and electronic equipment |
| CN111582045B (en) * | 2020-04-15 | 2024-05-10 | 芯算一体(深圳)科技有限公司 | Living body detection method and device and electronic equipment |
| CN112070041A (en) * | 2020-09-14 | 2020-12-11 | 北京印刷学院 | Living body face detection method and device based on CNN deep learning model |
| CN112927164A (en) * | 2021-03-22 | 2021-06-08 | 重庆邮电大学 | No-reference low-illumination image enhancement method based on deep convolutional neural network |
| CN112927164B (en) * | 2021-03-22 | 2023-04-07 | 重庆邮电大学 | No-reference low-illumination image enhancement method based on deep convolutional neural network |
| CN113792701A (en) * | 2021-09-24 | 2021-12-14 | 北京市商汤科技开发有限公司 | Living body detection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108596082A (en) | Human face in-vivo detection method based on image diffusion velocity model and color character | |
| CN107886064B (en) | Face recognition scene adaptation method based on convolutional neural network | |
| CN110348319B (en) | Face anti-counterfeiting method based on face depth information and edge image fusion | |
| WO2020151489A1 (en) | Living body detection method based on facial recognition, and electronic device and storage medium | |
| CN112801057B (en) | Image processing method, image processing device, computer equipment and storage medium | |
| CN107463920A (en) | A kind of face identification method for eliminating partial occlusion thing and influenceing | |
| CN108182409A (en) | Biopsy method, device, equipment and storage medium | |
| CN109858375A (en) | Living body faces detection method, terminal and computer readable storage medium | |
| CN102567734B (en) | Specific value based retina thin blood vessel segmentation method | |
| CN109977887A (en) | A kind of face identification method of anti-age interference | |
| CN108492301A (en) | A kind of Scene Segmentation, terminal and storage medium | |
| CN112200075B (en) | Human face anti-counterfeiting method based on anomaly detection | |
| WO2022156214A1 (en) | Liveness detection method and apparatus | |
| CN108734145A (en) | A kind of face identification method based on degree adaptive face characterization model | |
| CN116229528A (en) | Living body palm vein detection method, device, equipment and storage medium | |
| CN113807237B (en) | Training of in vivo detection model, in vivo detection method, computer device, and medium | |
| CN113436735A (en) | Body weight index prediction method, device and storage medium based on face structure measurement | |
| CN105069396B (en) | Dynamic percentage feature cuts AdaBoost Face datection algorithms | |
| Pathak et al. | Multimodal eye biometric system based on contour based E-CNN and multi algorithmic feature extraction using SVBF matching | |
| Jha et al. | Ubsegnet: Unified biometric region of interest segmentation network | |
| a Taha et al. | Face spoofing detection using deep CNN | |
| Méndez-Llanes et al. | On the use of local fixations and quality measures for deep face recognition | |
| Poosarala | Uniform classifier for biometric ear and retina authentication using smartphone application | |
| Narlagiri et al. | Biometric authentication system based on face recognition | |
| de Souza et al. | Efficient width-extended convolutional neural network for robust face spoofing detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |
|
| RJ01 | Rejection of invention patent application after publication |