光流与纹理特征融合的人脸活体检测算法

doi:10.3778/j.issn.1002-8331.2108-0046

摘要/Abstract

摘要： 针对照片与视频重放这一常见人脸识别欺诈手段，利用人脸攻击图像的语义信息提出一种基于光流与纹理特征融合的人脸活体检测算法：采集连续两帧待检测人脸图像，通过光流法及人脸检测方法生成人脸区域光流场变化图，将其与原始RGB图像输入至2通道卷积神经网络提取并融合得到人脸动-静态特征，基于融合特征实现真实人脸与欺诈人脸分类。此外，为了更好地捕捉人脸区域的光流场变化，应用影像动作放大技术将待检测视频帧中0.04~0.4?Hz信号放大20倍。使用IDIAP的Replay Attack人脸活体检测数据集中1?300段视频数据进行模型训练、验证和测试。实验表明，提出的人脸活体检测算法在Replay Attack数据集测试表现良好并取得了1.04%半错误率，能够有效识别照片、视频攻击。

关键词: 计算机视觉, 神经网络, 活体检测, 光流法

Abstract: Aiming at video replaying, the common face recognition system fraud method, this paper utilizes the semantic information of face attack images to propose a face spoofing detection algorithm based on optical flow and texture features. The optical flow field map of the face area is generated by the optical flow method and face detection method from two consecutive frames of the captured face video. Then, the original RGB face area image and optical flow field map are input to a 2-channels convolutional neural network to extract and fuse the features of the face. Finally, based on the optical flow and texture features, it realizes classification of real and fake faces. In addition, in order to better generate the optical flow field map containing liveness information, a motion amplification algorithm is applied to enhance the 0.04-0.4 Hz signal in the video frame by 20 times. This paper uses the Replay Attack spoofing dataset from IDIAP consisting of 1,300 videos for model training, verification and testing. Experiments show that the proposed algorithm performs well on the Replay Attack data set and achieves half total error rate of 1.04%.

Key words: computer vision, neural network, spoofing detection, optical flow

王宏飞, 程鑫, 赵祥模, 周经美. 光流与纹理特征融合的人脸活体检测算法[J]. 计算机工程与应用, 2022, 58(6): 170-176.

WANG Hongfei, CHENG Xin, ZHAO Xiangmo, ZHOU Jingmei. Face Liveness Detection Based on Fusional Optical Flow and Texture Features[J]. Computer Engineering and Applications, 2022, 58(6): 170-176.

参考文献

[1] CHINGOVSKA I，ANJOS A，MARCEL S.On the effectiveness of local binary patterns in face anti-spoo?ng[C]//International Conference of the Biometrics Special Interes Group，Darmstadt，Sep 6-7，2012.Germany：IEEE，2012：1-7.
[2] BOULKENAFET Z，KOMULAINEN J，HADID A.Face anti-spoofing based on color texture analysis[C]//International Conference on Image Processing，Quebec，Sep 27-30，2015.New York：IEEE，2015：2636-2640.
[3] GALBALLY J，MARCEL S.Face anti-spoofing based on general image quality assessment[C]//International Conference on Pattern Recognition，Stockholm，Aug 24-28，2014.Los Alamitos：IEEE，2014：1173-1178.
[4] BOULKENAFET Z，KOMULAINEN J，HADID A.On the generalization of color texture-based face anti-spoofing[J].Image and Vision Computing，2018，77（9）：1-9.
[5] 施新岚，黄俊，黄洋，等.基于Gabor纹理增强的人脸活体检测算法[J].重庆邮电大学学报（自然科学版），2020，32（6）：1023-1030.
SHI X L，HUANG J，HUANG Y，et al.Face liveness detection using Gabor texture enhancement[J].Journal of Chongqing University of Posts and Telecommunications（Natural Science Edition），2020，32（6）：1023-1030.
[6] TIRUNAGARI S，POH N，WINDRIDGE D，et al.Detection of face spoofing using visual dynamics[J].IEEE Transactions on Information Forensics and Security，2015，10（4）：762-777.
[7] KIM W，SUH S，HAN J.Face liveness detection from a single image via diffusion speed model[J].IEEE Transactions on Image Processing，2015，24（8）：2456-2465.
[8] BHARADWAJ S，DHAMECHA T，VATSA M，et al.Computationally efficient face spoofing detection with motion magnification[C]//IEEE Conference on Computer Vision and Pattern Recognition，Portland，Jun 23-28，2013.New York：IEEE，2013：105-110.
[9] FREITAS T，KOMULAIEN J，ANJOS A，et al.Face liveness detection using dynamic texture[J].EURASIP Journal on Image and Video Processing，2014（1）：2.
[10] GE H，TU X G，AI W J，et al.Face anti-spoofing by the enhancement of temporal motion[C]//International Conference on Advances in Computer Technology，Information Science and Communications，Suzhou，Mar 20-22，2020.Piscataway：IEEE，2020：106-111.
[11] STEINER R，KOLB A，JUNG N.Reliable face anti-spoofing using multispectral SWIR imaging[C]//International Conference on Biometrics，Halmstad，Jun 13-16，2016.NJ：IEEE，2016：1-8.
[12] ALOTAIBI A，MAHMOOD A.Deep face liveness detection based on nonlinear diffusion using convolution neural network[J].Signal Image & Video Processing，2017，11（4）：713-720.
[13] ZHANG S F，WANG X B.A dataset and benchmark for largescale multimodal face anti-spoofing[C]//Conference on Computer Vision and Pattern Recognition，CA，Jun 16-20，2019.Los Alamitos：IEEE，2019：919-928.
[14] YAO F，WU F，SHAO X H，et al.Joint 3D face reconstruction and dense alignment with position map regression network[C]//European Conference on Computer Vision，Munich，Sep 08-14，2018.Switzerland：Springer，2018：557-574.
[15] HE K M，ZHANG X Y，REN S Q，et al.Deep residual learning for image recognition[C]//Conference on Computer Vision and Pattern Recognition，Seattle，Jun 27-30，2016.New York：IEEE，2016：770-778.
[16] 邓茜文，冯子亮，邱晨鹏.基于近红外与可见光双目视觉的活体人脸检测方法[J].计算机应用，2020，40（7）：2096-2103.
DENG X W，FENG Z L，QIU C P.Face liveness detection method based on near-infrared and visible binocular vision[J].Journal of Computer Applications，2020，40（7）：2096-2103.
[17] WANG Z Z，YU Z T，ZHAO C X，et al.Deep spatial gradient and temporal depth learning for face antispoofing[C]//IEEE Conference on Computer Vision and Pattern Recognition，Electr Network，Jun 14-19，2020.New York：IEEE，2020：5041-5050.
[18] LIU S Q，LAN X Y，YUEN P C.Multi-channel remote photoplethysmography correspondence feature for 3D mask face presentation attack detection[J].IEEE Transactions on Information Forensics and Security，2021，16：2683-2696.
[19] GUNDOGAR M F，ERDEM C E.Detection of presentation attacks in face recognition systems with non-contact photoplethysmography[C]//Signal Processing and Communications Applications Conference，Electr Network，Oct 05-07，2020.New York：IEEE，2020.
[20] WU H Y，RUBINSTEIN M，SHIH E，et al.Eulerian video magnification for revealing subtle changes in the world[J].ACM Transactions on Graphics，2012，31（4）：65.
[21] OH T H，JAROENSRI R，KIM C，et al.Learning-based video motion magnification[C]//European Conference on Computer Vision，Munich，Sep 08-14，2018.Munich：Springer，2018：663-679.
[22] FARNEBACK G.Two-frame motion estimation based on polynomial expansion[C]//Scandinavian Conference on Image Analysis，Halmstad，Jun 29-Jul 02，2003.Halmstad：Springer，2003：363-370.
[23] HU J，LI S，ALBANIE S，et al.Squeeze-and-excitation networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2020，42（8）：2011-2023.
[24] REHMAN Y A U，PO L M，LIU M Y.Livenet：improving features generalization for face liveness detection using convolution neural networks[J].Expert Systems with Applications，2018，108：159-169.
[25] BOULKENAFET Z，KOMULAINEN J，HADID A.Face antispoofing using speeded-up robust features and fisher vector encoding[J].IEEE Signal Processing Letters，2017，24（2）：141-145.
[26] BOULKENAFET Z，KOMULAINEN J，HADID A.Face spoofing detection using colour texture analysis[J].IEEE Transactions on Information Forensics and Security，2016，11（8）：1818-1830.
[27] BENLAMOUDI A，AIADI K E，OUAFI A，et al.Face antispoofing based on frame difference and multilevel representation[J].Journal of Electronic Imaging，2017，26（4）：043007.