一种基于深度学习的移动端人脸验证系统
|
摘要
作为一种新兴的身份验证技术,人脸验证广泛应用于门禁、考勤等需要身份验证的场合。本文综合考虑移动端人脸验证的需求与生产环境,以及现有人脸验证算法的效率和可移植性,设计并实现一种基于安卓系统平台的人脸验证系统。该系统能够离线部署于搭载安卓系统的移动端设备,通过摄像头获取人脸图像并在本地进行图像的数据处理完成人脸验证工作。在算法上,该系统采用深度卷积神经网络进行图像处理与人脸特征向量提取以提高人脸验证的准确率。在实现上,通过联合编译Java和C++代码提高算法运行效率以适应深度学习算法在移动端的应用。实验表明,本系统能够在确保准确率高达97.16%的前提下快速完成人脸验证流程,基本满足工业化应用需求。
As a new authentication technology,face verification is widely used in access control,attendance and other needs of the occasion of authentication. This paper designs and implements a face verification system based on the Android system platform,taking into account the requirements and production environment of mobile face verification and the efficiency and portability of existing face verification algorithms. The system can be deployed off-line in the mobile device equipped with Android system,through the camera obtaining a face image and the local image processing data to complete face verification work. In the algorithm,the system uses deep convolution neural network for image processing and face feature vector extraction to improve the accuracy of face verification. In the implementation,through the joint compiler Java and C + + codes improving the efficiency of the algorithm to adapt to the depth learning algorithms in the mobile side of the application. Experiments show that the system can quickly ensure the accuracy rate of 97. 16% under the premise of rapid completion of the face verification process,basically meet the needs of industrial applications.
As a new authentication technology,face verification is widely used in access control,attendance and other needs of the occasion of authentication. This paper designs and implements a face verification system based on the Android system platform,taking into account the requirements and production environment of mobile face verification and the efficiency and portability of existing face verification algorithms. The system can be deployed off-line in the mobile device equipped with Android system,through the camera obtaining a face image and the local image processing data to complete face verification work. In the algorithm,the system uses deep convolution neural network for image processing and face feature vector extraction to improve the accuracy of face verification. In the implementation,through the joint compiler Java and C + + codes improving the efficiency of the algorithm to adapt to the depth learning algorithms in the mobile side of the application. Experiments show that the system can quickly ensure the accuracy rate of 97. 16% under the premise of rapid completion of the face verification process,basically meet the needs of industrial applications.
引文
[1]Ahonen T,Hadid A,Pietikainen M.Face description with local binary patterns:Application to face recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(12):2037-2041.
[2]Cao Xudong,Wipf D,Wen Fang,et al.A practical transfer learning algorithm for face verification[C]//Proceedings of the IEEE International Conference on Computer Vision.2013:3208-3215.
[3]Berg T,Belhumeur P N.Tom-vs-Pete classifiers and identity-preserving alignment for face verification[C]//Proceedings of the British Machine Vision Conference.2012:129.1-129.11.
[4]Song Fengxi,Zhang D,Wang Jizhong,et al.A parameterized direct LDA and its application to face recognition[J].Neurocomputing,2007,71(1-3):191-196.
[5]Rodriguez C S,Fischer G.The Linux Kernel Primer:A Top-down Approach for x86 and Power PC Architectures[M].India:Pearson Education,2006.
[6]Cheng Kwang-Ting,Wang Yi-Chu.Using mobile GPU for general-purpose computing-A case study of face recognition on smartphones[C]//2011 International Symposium on VLSI Design,Automation and Test(VLSI-DAT).2011:1-4.
[7]韩超,梁泉.Android系统原理及开发要点详解[M].北京:电子工业出版社,2010.
[8]Li Haoxiang,Lin Zhe,Shen Xiaohui,et al.A convolutional neural network cascade for face detection[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.2015:5325-5334.
[9]Xiong Xuehan,De la Torre F.Supervised descent method and its applications to face alignment[C]//Proceedings of2013 IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2013:532-539.
[10]Zhang Zhanpeng,Luo Ping,Chen C L,et al.Facial landmark detection by deep multi-task learning[C]//Proceedings of 2014 13th European Conference on Computer Vision.2014:94-108.
[11]Sun Yi,Wang Xiaogang,Tang Xiaoou.Deep convolutional network cascade for facial point Detection[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition.2013:3476-3483.
[12]Zapata B C.Android Studio Application Development[M].Packt Publishing,2013.
[13]Stone J E,Gohara D,Shi G.Open CL:A parallel programming standard for heterogeneous computing systems[J].Computing in Science&Engineering,2010,12(3):66-73.
[14]Phillips P J,Wechsler H,Huang J,et al.The FERET database and evaluation procedure for face-recognition algorithms[J].Image and Vision Computing,1998,16(5):295-306.
[15]Yan Shengye,Shan Shiguang,Chen Xilin,et al.Locally assembled binary(LAB)feature with feature-centric cascade for fast and accurate face detection[C]//2008 IEEE Conference on Computer Vision and Pattern Recognition.2008:1-7.
[16]Bay H,Tuytelaars T,Gool L V.SURF:Speeded up robust features[J].Computer Vision&Image Understanding,2006,110(3):404-417.
[17]Zhang Jie,Shan Shiguang,Kan Meina,et al.Coarse-tofine auto-encoder networks(cfan)for real-time face alignment[C]//European Conference on Computer Vision.2014:1-16.
[18]Zhang Baochang,Gao Yongsheng,Zhao Sanqiang,et al.Local derivative pattern versus local binary pattern:face recognition with high-order local pattern descriptor[J].IEEE Transactions on Image Processing,2010,19(2):533-544.
[19]Dalal N,Triggs B.Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2005,1:886-893.
[20]Krizhevsky A,Sutskever I,Hinton G E.Imagenet classification with deep convolutional neural networks[C]//Advances in Neural Information Processing Systems.2012:1097-1105.
[2]Cao Xudong,Wipf D,Wen Fang,et al.A practical transfer learning algorithm for face verification[C]//Proceedings of the IEEE International Conference on Computer Vision.2013:3208-3215.
[3]Berg T,Belhumeur P N.Tom-vs-Pete classifiers and identity-preserving alignment for face verification[C]//Proceedings of the British Machine Vision Conference.2012:129.1-129.11.
[4]Song Fengxi,Zhang D,Wang Jizhong,et al.A parameterized direct LDA and its application to face recognition[J].Neurocomputing,2007,71(1-3):191-196.
[5]Rodriguez C S,Fischer G.The Linux Kernel Primer:A Top-down Approach for x86 and Power PC Architectures[M].India:Pearson Education,2006.
[6]Cheng Kwang-Ting,Wang Yi-Chu.Using mobile GPU for general-purpose computing-A case study of face recognition on smartphones[C]//2011 International Symposium on VLSI Design,Automation and Test(VLSI-DAT).2011:1-4.
[7]韩超,梁泉.Android系统原理及开发要点详解[M].北京:电子工业出版社,2010.
[8]Li Haoxiang,Lin Zhe,Shen Xiaohui,et al.A convolutional neural network cascade for face detection[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.2015:5325-5334.
[9]Xiong Xuehan,De la Torre F.Supervised descent method and its applications to face alignment[C]//Proceedings of2013 IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2013:532-539.
[10]Zhang Zhanpeng,Luo Ping,Chen C L,et al.Facial landmark detection by deep multi-task learning[C]//Proceedings of 2014 13th European Conference on Computer Vision.2014:94-108.
[11]Sun Yi,Wang Xiaogang,Tang Xiaoou.Deep convolutional network cascade for facial point Detection[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition.2013:3476-3483.
[12]Zapata B C.Android Studio Application Development[M].Packt Publishing,2013.
[13]Stone J E,Gohara D,Shi G.Open CL:A parallel programming standard for heterogeneous computing systems[J].Computing in Science&Engineering,2010,12(3):66-73.
[14]Phillips P J,Wechsler H,Huang J,et al.The FERET database and evaluation procedure for face-recognition algorithms[J].Image and Vision Computing,1998,16(5):295-306.
[15]Yan Shengye,Shan Shiguang,Chen Xilin,et al.Locally assembled binary(LAB)feature with feature-centric cascade for fast and accurate face detection[C]//2008 IEEE Conference on Computer Vision and Pattern Recognition.2008:1-7.
[16]Bay H,Tuytelaars T,Gool L V.SURF:Speeded up robust features[J].Computer Vision&Image Understanding,2006,110(3):404-417.
[17]Zhang Jie,Shan Shiguang,Kan Meina,et al.Coarse-tofine auto-encoder networks(cfan)for real-time face alignment[C]//European Conference on Computer Vision.2014:1-16.
[18]Zhang Baochang,Gao Yongsheng,Zhao Sanqiang,et al.Local derivative pattern versus local binary pattern:face recognition with high-order local pattern descriptor[J].IEEE Transactions on Image Processing,2010,19(2):533-544.
[19]Dalal N,Triggs B.Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2005,1:886-893.
[20]Krizhevsky A,Sutskever I,Hinton G E.Imagenet classification with deep convolutional neural networks[C]//Advances in Neural Information Processing Systems.2012:1097-1105.