CN115953822A - Face video false distinguishing method and device based on rPPG physiological signal - Google Patents
Face video false distinguishing method and device based on rPPG physiological signal Download PDFInfo
- Publication number
- CN115953822A CN115953822A CN202310202394.8A CN202310202394A CN115953822A CN 115953822 A CN115953822 A CN 115953822A CN 202310202394 A CN202310202394 A CN 202310202394A CN 115953822 A CN115953822 A CN 115953822A
- Authority
- CN
- China
- Prior art keywords
- video
- rppg
- signal
- training
- data set
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a face video false distinguishing method and a device based on rPPG physiological signals, wherein the method comprises the following steps: the method comprises the following steps: collecting a face video and a finger PPG signal, and constructing a PPG signal video data set; collecting a real face video and a forged face video, and constructing a counterfeit identification data set; step two: training by using a PPG signal video data set as training data to obtain an rPPG signal extraction network; step three: extracting an rPPG signal by using the rPPG signal extraction network obtained by training in the step two by using the authentication data set, and then inputting the rPPG signal into a binary decision network for network training; step four: and (4) judging the authenticity of the video to be detected by using the rPPG signal extraction network obtained by the training in the second step and the binary decision network obtained by the training in the third step. The invention uses the rPPG physiological signal which is difficult to forge to distinguish the face forged composite video, thereby effectively improving the judgment accuracy.
Description
Technical Field
The invention belongs to the field of computer vision and video counterfeit discrimination, and relates to a human face video counterfeit discrimination method and device based on an rPPG physiological signal.
Background
In recent years, with the continuous development of deep learning, various face changing technologies are gradually emerging on the network. Most face changing technologies take the defakes as a basis, and modify and replace the faces in pictures and videos to achieve the effect of falseness and falseness, so that the technologies bring certain potential safety hazards to the society.
From the first defakes to the latest StyleGAN, the images forged by the face forging technology are difficult to identify by naked eyes, and in the face of the face forging technology which is continuously updated, the academic and industrial circles are actively exploring the face forging identification technology, from the traditional image evidence obtaining method to the deep learning method, from the edge contour, the blink, the mouth opening and the eyeball rotation behaviors of the face to the physiological signal method based on the pulse and the heart rate. For the problem of video counterfeit identification, the judgment of the authenticity of the video simply depending on the facial image characteristics is difficult to deal with the more and more advanced counterfeit means. The rPPG technology is a noninvasive physiological signal detection method for detecting blood volume change in living body tissues by means of a photoelectric means. rPPG physiological signal is a signal measured remotely, i.e. based on video, from which the heart rate can be calculated. During the heart beat cycle, the blood flowing through the arterioles, capillaries and venules in the peripheral blood vessels pulsates accordingly. The rPPG signal is widely applied to human face living body detection, mainly because for a real human face, when a light beam with a certain wavelength irradiates the surface of the skin of the face, the pulsatility change of the blood will cause the pulsatility change of the absorption of the light. The common camera captures the rPPG signal, so that the flowing characteristics of blood flow can be reflected, and a physiological signal value can be calculated. For a false face, since there is no real face skin, the rPPG signal reflected from the facial light back to the camera is significantly different from the real face. Because the generated false face video has no real rPPG signal, the existing counterfeiting technology can not carry out simulation counterfeiting on the rPPG physiological signal, and based on the characteristics, the rPPG signal can play an important role in the field of face video counterfeiting identification so as to solve the problem that the image characteristics are difficult to judge the authenticity of the forged face video.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a face video false distinguishing method and device based on an rPPG physiological signal, and the specific technical scheme is as follows:
a face video false distinguishing method based on rPPG physiological signals comprises the following steps:
the method comprises the following steps: collecting a face video and a finger PPG signal, and constructing a PPG signal video data set; collecting a real face video and a forged face video, and constructing a counterfeit identification data set;
step two: training by using a PPG signal video data set as training data to obtain an rPPG signal extraction network;
step three: extracting an rPPG signal by using the rPPG signal extraction network obtained by training in the step two by using the authentication data set, inputting the rPPG signal into a binary decision network, and performing network training to obtain a trained binary decision network;
step four: and (4) judging the authenticity of the video to be detected by using the rPPG signal extraction network obtained by the training in the second step and the binary decision network obtained by the training in the third step.
Further, the step one specifically includes: respectively acquiring hyperspectral face video and a finger PPG signal section by using hyperspectral camera equipment and an oximeter, constructing a PPG signal video data set, and then performing segmentation processing on the PPG signal video data set to obtain a plurality of groups of PPG signal video data; the method comprises the steps of collecting a real face video and a forged face video by utilizing an open source data set FaceForencics and ForgeryNet, selecting a real face video segment and a forged face video segment to construct a pseudo-identification data set, and dividing the pseudo-identification data set into a training set and a testing set.
Further, the second step specifically includes: inputting each frame image of a hyperspectral human face video in a PPG signal video data set into a ResNet50 feature extraction network to extract facial features, inputting the facial features extracted from continuous frames into an LSTM long-short term memory artificial neural network according to a time sequence to perform time sequence modeling, and using fast Fourier transform to obtain an rPPG signal on a frequency domain; finger PPG signals collected in PPG signal video data set are used as real labels, loss function calculation is carried out on the finger PPG signals and the rPPG signals, parameters of a ResNet50 feature extraction network and parameters of an LSTM long-short term memory artificial neural network are updated through an SGD random gradient descent method, iterative training is carried out by using all PPG signal video data, and the parameters after training are stored to obtain the rPPG signal extraction network.
Further, the third step specifically includes: inputting videos in a training set of a pseudo-identification data set frame by frame into an rPPG signal extraction network obtained by training in the second step, extracting rPPG signals of the video data, inputting the extracted rPPG signals into a ResNet18 binary decision network, obtaining a binary result, setting a real video label as 0, setting a fake video label as 1, performing loss function calculation, updating parameters of the ResNet18 binary decision network through an SGD random gradient descent method, performing iterative training, and storing the parameters after the training is finished to obtain the binary decision network for judging the authenticity of the videos.
Further, the loss functions in the second step and the third step are an MSE mean square error loss function and a cross entropy loss function, respectively.
Further, the fourth step specifically includes: inputting the video to be detected into the rPPG signal extraction network obtained by training in the second step frame by frame, extracting the rPPG signal of the whole video, inputting the obtained rPPG signal into the binary decision network obtained by training in the third step to obtain a label judged by the network, wherein if the label is 0, the video to be detected is judged to be a real video, and if the label is 1, the video to be detected is judged to be a fake video.
A human face video false distinguishing device based on an rPPG physiological signal comprises one or more processors and is used for realizing the human face video false distinguishing method based on the rPPG physiological signal.
A computer readable storage medium, on which a program is stored, which when executed by a processor, implements the rPPG physiological signal-based human face video authentication method.
Has the beneficial effects that: the invention uses the rPPG physiological signal which is difficult to forge to distinguish the face forged composite video, can effectively improve the judgment accuracy, is suitable for various forging methods, and solves the problem that the authenticity of the forged face video is difficult to judge by the image characteristics.
Drawings
FIG. 1 is a schematic flow chart of a human face video identification method based on rPPG physiological signals according to the present invention;
fig. 2 is a schematic diagram of rPPG signal extraction network training flow used in the present invention;
FIG. 3 is a schematic diagram of a binary decision network training process used in the present invention;
fig. 4 is a schematic structural diagram of a face video counterfeit discrimination device based on rPPG physiological signals according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and technical effects of the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples.
As shown in fig. 1, a face video identification method based on rPPG physiological signals according to an embodiment of the present invention specifically includes the following steps:
the method comprises the following steps: collecting a face video and a finger PPG signal, and constructing a PPG signal video data set; and (4) collecting the real face video and the fake face video, and constructing a fake identification data set.
Specifically, the present embodiment uses 316 pairs of hyper-spectral face video and finger PPG signal segments, which are respectively acquired by using hyper-spectral camera equipment and oximeter, to construct a PPG signal video data set. The acquisition frame rate of the hyperspectral camera is 30 Hz, the wave band is 400nm to 1000nm, and the acquisition period is 2 minutes; the oximeter collects the PPG signal of the right fingertip of the person, the collection frequency is 30 Hz, and the collection period is 2 minutes. The acquired video data set of PPG signals is divided into small segments of 15 seconds, resulting in 2528 sets of video data of PPG signals.
The collected real face video and forged face video are mainly from open source data sets faceforces and ForgeryNet, and 10000 real face videos and 10000 forged face videos are selected from the open source data sets faceforces and ForgeryNet to be used for constructing a counterfeit identification data set. 9000 sections of real face videos and 9000 sections of forged face videos are used as training sets, and 1000 sections of real face videos and 1000 sections of forged face videos are used as test sets.
Step two: and training by using the PPG signal video data set as training data to obtain an rPPG signal extraction network, wherein the rPPG signal extraction network is used for extracting the rPPG signal from the face video.
Specifically, as shown in fig. 2, each frame of image of a hyperspectral human face video recorded in a PPG signal video data set is input into a ResNet50 feature extraction network to extract facial features, wherein the size of the input image is 112 × 112 pixels, the dimension of the facial features extracted by the ResNet50 feature extraction network is 512 dimensions, the facial features extracted by continuous frames are sent into an LSTM long-short term memory artificial neural network for time sequence modeling according to a time sequence, a fast fourier transform is used to obtain rPPG signals in a frequency domain, the PPG signals collected in the PPG signal video data set are used as real labels, loss function calculation is performed on the rPPG signals calculated by the network by using an MSE mean square error loss function, parameters of the ResNet50 feature extraction network and the LSTM long-term memory artificial neural network are updated by an SGD random gradient descent method, all PPG signal video data are used for iterative training, 50 times are trained, and the parameters after the training are stored to obtain the rpg signal extraction network.
Step three: and (3) extracting an rPPG signal by using the pseudo-identification data set and using the rPPG signal extraction network obtained by training in the step two, inputting the rPPG signal into a binary decision network, and carrying out network training to obtain the trained binary decision network, wherein the binary decision network is used for judging the authenticity of the video.
Specifically, as shown in fig. 3, a video in a training set of the authentication data set is input into the rPPG signal extraction network obtained by training in step two frame by frame to extract an rPPG signal of the video data, the extracted rPPG signal is input into the ResNet18 binary decision network to obtain a binary result, a real video label is set to 0, a counterfeit video label is set to 1, a cross entropy loss function is used for performing loss function calculation, parameters of the ResNet18 binary decision network are updated by an SGD random gradient descent method, training is performed for 20 times in total, and the parameters after training are stored to obtain the trained binary decision network for judging the authenticity of the video.
Step four: and (4) judging the authenticity of the video to be detected by using the rPPG signal extraction network obtained by the training in the second step and the binary decision network obtained by the training in the third step.
Specifically, the video segment to be detected is input into the rPPG signal extraction network obtained through training in the second step frame by frame, the rPPG signal of the whole video segment is extracted, the obtained rPPG signal is input into the binary decision network obtained through training in the third step, a network judgment label is obtained, if the label is 0, the video to be detected is judged to be a real video, and if the label is 1, the video to be detected is judged to be a fake video.
The following table 1 is a table showing the performance of the method provided by the above embodiment of the present invention on 1000 segments of real face videos and 1000 segments of forged face videos, and sequentially displays other reference methods for comparison and the results of the present embodiment from top to bottom, and uses the accuracy as an evaluation index, and the accuracy is defined as follows:
p (accuracy) = determine correct video number/total number of videos in test set;
specific table 1 is as follows:
table 1 false identification accuracy table of the method of the embodiment of the present invention and other reference methods.
Corresponding to the embodiment of the human face video false distinguishing method based on the rPPG physiological signal, the invention also provides an embodiment of a human face video false distinguishing device based on the rPPG physiological signal.
Referring to fig. 4, the face video counterfeit discrimination device based on rPPG physiological signals provided in the embodiment of the present invention includes one or more processors, and is configured to implement the face video counterfeit discrimination method based on rPPG physiological signals in the above embodiment.
The embodiment of the human face video false distinguishing device based on the rPPG physiological signal can be applied to any equipment with data processing capability, such as computers and other equipment or devices. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for running through the processor of any device with data processing capability. In terms of hardware, as shown in fig. 4, the present invention is a hardware structure diagram of an arbitrary device with data processing capability where a human face video false distinguishing apparatus based on rPPG physiological signals is located, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 4, in an embodiment, the arbitrary device with data processing capability where the apparatus is located may also include other hardware according to the actual function of the arbitrary device with data processing capability, which is not described again.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present invention. One of ordinary skill in the art can understand and implement it without inventive effort.
The embodiment of the invention also provides a computer readable storage medium, which stores a program, and when the program is executed by a processor, the method for identifying the fake face video based on the rPPG physiological signal in the above embodiment is realized.
The computer readable storage medium may be an internal storage unit, such as a hard disk or a memory, of any data processing capability device described in any of the foregoing embodiments. The computer readable storage medium may also be an external storage device such as a plug-in hard disk, a Smart Media Card (SMC), an SD Card, a Flash memory Card (Flash Card), etc. provided on the device. Further, the computer readable storage medium may include both internal storage units and external storage devices of any data processing capable device. The computer-readable storage medium is used for storing the computer program and other programs and data required by the arbitrary data processing-capable device, and may also be used for temporarily storing data that has been output or is to be output.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way. Although the foregoing has described in detail the practice of the invention, it will be appreciated by those skilled in the art that variations may be applied to the embodiments described in the foregoing examples, or equivalents may be substituted for elements thereof. All changes, equivalents and the like which come within the spirit and principles of the invention are desired to be protected.
Claims (8)
1. A face video false distinguishing method based on rPPG physiological signals is characterized by comprising the following steps:
the method comprises the following steps: collecting a face video and a finger PPG signal, and constructing a PPG signal video data set; collecting a real face video and a forged face video, and constructing a counterfeit identification data set;
step two: training by using a PPG signal video data set as training data to obtain an rPPG signal extraction network;
step three: extracting an rPPG signal by using the rPPG signal extraction network obtained by training in the step two by using the authentication data set, inputting the rPPG signal into a binary decision network, and carrying out network training to obtain a trained binary decision network;
step four: and (4) judging the authenticity of the video to be detected by using the rPPG signal extraction network obtained by the training in the second step and the binary decision network obtained by the training in the third step.
2. The method for identifying the face video based on the rPPG physiological signal as claimed in claim 1, wherein the step one specifically comprises: respectively acquiring hyperspectral face video and a finger PPG signal section by using hyperspectral camera equipment and an oximeter, constructing a PPG signal video data set, and then performing segmentation processing on the PPG signal video data set to obtain a plurality of groups of PPG signal video data; the method comprises the steps of utilizing an open source data set FaceForencics and ForgeryNet to collect a real face video and a forged face video, selecting a real face video segment and a forged face video segment to construct a counterfeit identification data set, and dividing the counterfeit identification data set into a training set and a testing set.
3. The method for identifying the face video based on the rPPG physiological signal as claimed in claim 2, wherein the second step specifically comprises: inputting each frame image of a hyperspectral human face video in a PPG signal video data set into a ResNet50 feature extraction network to extract facial features, inputting the facial features extracted from continuous frames into an LSTM long-short term memory artificial neural network according to a time sequence to perform time sequence modeling, and using fast Fourier transform to obtain an rPPG signal on a frequency domain; finger PPG signals collected in PPG signal video data sets are used as real labels, loss function calculation is carried out on the finger PPG signals and the rPPG signals, parameters of a ResNet50 feature extraction network and parameters of an LSTM long-short term memory artificial neural network are updated through an SGD random gradient descent method, iterative training is carried out through all PPG signal video data, and the parameters after training are stored to obtain the rPPG signal extraction network.
4. The method for identifying the face video based on the rPPG physiological signal as claimed in claim 3, wherein the step three specifically comprises: inputting the video in the training set of the authentication data set frame by frame into the rPPG signal extraction network obtained by training in the second step to extract the rPPG signal of the video data, inputting the extracted rPPG signal into a ResNet18 binary decision network, obtaining a binary classification result, setting a real video label as 0, setting a forged video label as 1, then performing loss function calculation, updating parameters of the ResNet18 binary decision network through an SGD random gradient descent method, performing iterative training, and storing the parameters after the training is finished to obtain the binary decision network for judging the authenticity of the video.
5. The method as claimed in claim 4, wherein the loss functions in the second step and the third step are MSE mean square error loss function and cross entropy loss function, respectively.
6. The method for identifying the face video based on the rPPG physiological signal as claimed in claim 4, wherein the fourth step is specifically as follows: inputting the video to be detected into the rPPG signal extraction network obtained by training in the second step frame by frame, extracting the rPPG signal of the whole video, inputting the obtained rPPG signal into the binary decision network obtained by training in the third step to obtain a label judged by the network, wherein if the label is 0, the video to be detected is judged to be a real video, and if the label is 1, the video to be detected is judged to be a fake video.
7. A face video false distinguishing device based on an rPPG physiological signal is characterized by comprising one or more processors and being used for realizing the face video false distinguishing method based on the rPPG physiological signal in any one of claims 1 to 6.
8. A computer-readable storage medium, on which a program is stored, which when executed by a processor, implements the rPPG physiological signal-based face video authentication method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310202394.8A CN115953822B (en) | 2023-03-06 | 2023-03-06 | Human face video fake identification method and device based on rPPG physiological signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310202394.8A CN115953822B (en) | 2023-03-06 | 2023-03-06 | Human face video fake identification method and device based on rPPG physiological signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115953822A true CN115953822A (en) | 2023-04-11 |
| CN115953822B CN115953822B (en) | 2023-07-11 |
Family
ID=85891136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310202394.8A Active CN115953822B (en) | 2023-03-06 | 2023-03-06 | Human face video fake identification method and device based on rPPG physiological signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115953822B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116524612A (en) * | 2023-06-21 | 2023-08-01 | 长春理工大学 | rPPG-based human face living body detection system and method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815794A (en) * | 2018-12-14 | 2019-05-28 | 北京飞搜科技有限公司 | Recognition of face is counter to cheat method, apparatus and electronic equipment |
| CN111860079A (en) * | 2019-04-30 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Living body image detection method and device and electronic equipment |
| CN111967427A (en) * | 2020-08-28 | 2020-11-20 | 广东工业大学 | Fake face video identification method, system and readable storage medium |
| CN113435292A (en) * | 2021-06-22 | 2021-09-24 | 北京交通大学 | AI counterfeit face detection method based on inherent feature mining |
| CN113449657A (en) * | 2021-07-05 | 2021-09-28 | 中山大学 | Method, system and medium for detecting depth-forged face video based on face key points |
| CN113827234A (en) * | 2021-11-25 | 2021-12-24 | 之江实验室 | Hyperspectral face video-based remote pulse wave reconstruction method |
| WO2022073414A1 (en) * | 2020-10-09 | 2022-04-14 | 腾讯科技（深圳）有限公司 | Image recognition method and apparatus, computing device and computer-readable storage medium |
| WO2022167996A1 (en) * | 2021-02-08 | 2022-08-11 | Euronovate Sa | System for the verification of the identity of a person by facial recognition |
| CN115019370A (en) * | 2022-06-21 | 2022-09-06 | 深圳大学 | Depth counterfeit video detection method based on double fine-grained artifacts |
| WO2023004572A1 (en) * | 2021-07-27 | 2023-02-02 | 京东方科技集团股份有限公司 | Model training method, signal recognition method and apparatus, computing processing device, computer program, and computer-readable medium |
-
2023
- 2023-03-06 CN CN202310202394.8A patent/CN115953822B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815794A (en) * | 2018-12-14 | 2019-05-28 | 北京飞搜科技有限公司 | Recognition of face is counter to cheat method, apparatus and electronic equipment |
| CN111860079A (en) * | 2019-04-30 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Living body image detection method and device and electronic equipment |
| CN111967427A (en) * | 2020-08-28 | 2020-11-20 | 广东工业大学 | Fake face video identification method, system and readable storage medium |
| WO2022073414A1 (en) * | 2020-10-09 | 2022-04-14 | 腾讯科技（深圳）有限公司 | Image recognition method and apparatus, computing device and computer-readable storage medium |
| WO2022167996A1 (en) * | 2021-02-08 | 2022-08-11 | Euronovate Sa | System for the verification of the identity of a person by facial recognition |
| CN113435292A (en) * | 2021-06-22 | 2021-09-24 | 北京交通大学 | AI counterfeit face detection method based on inherent feature mining |
| CN113449657A (en) * | 2021-07-05 | 2021-09-28 | 中山大学 | Method, system and medium for detecting depth-forged face video based on face key points |
| WO2023004572A1 (en) * | 2021-07-27 | 2023-02-02 | 京东方科技集团股份有限公司 | Model training method, signal recognition method and apparatus, computing processing device, computer program, and computer-readable medium |
| CN113827234A (en) * | 2021-11-25 | 2021-12-24 | 之江实验室 | Hyperspectral face video-based remote pulse wave reconstruction method |
| CN115019370A (en) * | 2022-06-21 | 2022-09-06 | 深圳大学 | Depth counterfeit video detection method based on double fine-grained artifacts |
Non-Patent Citations (4)
| Title |
|---|
| MEHMET FATIH GÜNDOĞAR等: "Detection of Presentation Attacks in Face Recognition Systems with Non-contact Photoplethysmography", 2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU) * |
| 暴雨轩;芦天亮;杜彦辉;: "深度伪造视频检测技术综述", 计算机科学 * |
| 李晓双: "基于人脸的活体检测技术研究", 中国优秀硕士学位论文全文数据库 * |
| 黄义妨等: "面向不同传感器与复杂场景的人脸识别系统防伪方法综述", 计算机工程 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116524612A (en) * | 2023-06-21 | 2023-08-01 | 长春理工大学 | rPPG-based human face living body detection system and method |
| CN116524612B (en) * | 2023-06-21 | 2023-09-12 | 长春理工大学 | rPPG-based human face living body detection system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115953822B (en) | 2023-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yuan et al. | Fingerprint liveness detection using an improved CNN with image scale equalization | |
| CN105631439B (en) | Face image processing process and device | |
| CN109543526B (en) | True and false facial paralysis recognition system based on depth difference characteristics | |
| CN108140123A (en) | Face living body detection method, electronic device and computer program product | |
| CN107392177A (en) | A kind of human body identification verification method and its device | |
| CN109063572A (en) | It is a kind of based on multiple dimensioned and multireel lamination Fusion Features fingerprint activity test methods | |
| CN108021892A (en) | A kind of human face in-vivo detection method based on extremely short video | |
| Liu et al. | One-class fingerprint presentation attack detection using auto-encoder network | |
| CN103593648B (en) | Face recognition method for open environment | |
| CN109034016A (en) | A kind of hand back vein image-recognizing method based on S-CNN model of universality | |
| CN108846269A (en) | One kind is towards manifold identity identifying method and identification authentication system | |
| Awad et al. | Impact of some biometric modalities on forensic science | |
| CN108875509A (en) | Biopsy method, device and system and storage medium | |
| CN109063643A (en) | A kind of facial expression pain degree recognition methods under the hidden conditional for facial information part | |
| CN113469143A (en) | Finger vein image identification method based on neural network learning | |
| CN111178130A (en) | Face recognition method, system and readable storage medium based on deep learning | |
| CN115953822B (en) | Human face video fake identification method and device based on rPPG physiological signals | |
| Das et al. | A framework for liveness detection for direct attacks in the visible spectrum for multimodal ocular biometrics | |
| CN115937953A (en) | Psychological change detection method, device, equipment and storage medium | |
| Nikitin et al. | Face anti-spoofing with joint spoofing medium detection and eye blinking analysis | |
| CN108875549A (en) | Image-recognizing method, device, system and computer storage medium | |
| Jingade et al. | DOG-ADTCP: A new feature descriptor for protection of face identification system | |
| CN111814738A (en) | Human face recognition method, human face recognition device, computer equipment and medium based on artificial intelligence | |
| CN110175522A (en) | Work attendance method, system and Related product | |
| CN106897657A (en) | A kind of human face in-vivo detection method and device |
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 |