Current face recognition algorithms use hand-crafted features or extract features by deep learning.This paper presents a face recognition algorithm based on improved deep networks that can automatically extract the discriminative features of the target more accurately.Firstly,this algorithm uses ZCA(Zero-mean Component Analysis)whitening to preprocess the input images in order to reduce the correlation between features and the complexity of the training networks.Then,it organically combines convolution,pooling and stacked sparse autoencoder to get a deep network feature extractor.The convolution kernels are achieved through a separate unsupervised learning model.The improved deep networks get an automatic deep feature extractor through preliminary training and fine-tuning.Finally,the softmax regression model is used to classify the extracted features.This algorithm is tested on several commonly used face databases.It is indicated that the performance is better than the traditional methods and common deep learning methods.
[1] Deng W,Hu J,Lu J,et al.Transform-Invariant PCA:A unified approach to fully automatic face alignment,representationand recognition[J].IEEE Trans onPattern Analysis and Machine Intelligence,2014,36(6):1275-1284.
[2] 周家锐,纪震,等.基于Gabor小波与Memetic算法的人脸识别方法[J].电子学报,2012,40(4):642-646. Zhou Jia-rui,Ji Zhen,et al.Face recognitionusingGabor wavelets and Memetic algorithm[J].Acta Electronica Sinica,2012,40(4):642-646.(in Chinese)
[3] 杨飞,苏剑波.人脸显性特征的融合构造方法及识别[J].电子学报,2012,40(3):466-471. Yang Fei,Su Jian-bo.Face recognition based on explicit facial features by fusion construction method[J].Acta Electronica Sinica,2012,40(3):466-471.(in Chinese)
[4] Braverman M.Poly-logarithmic independence fools bounded-depthbooleancircuits[J].Communications of the ACM,2011,54(4):108-115.
[5] Zhang H,Zhou F,Zhang W,et al.Real-time action recognition based on a modified Deep Belief Network model[A].International Conference on Information and Automation[C].Hailar,HulunBuir,Inner Mongolia,China:IEEE,2014.225-228.
[6] Ouyang W,Wang X,Zeng X,et al.Deepid-net:Deformable deep convolutional neural networks for object detection[A].Conference on Computer Vision and Pattern Recognition[C].Boston,MA,USA:IEEE,2015.2403-2412.
[7] BENGIO Y.Learning deep architectures for AI[J].Foundations and Trends in Machine Learning,2009,2(1):1-12.
[8] Bengio Y,Le Cun Y.Scaling learning algorithms towards AI[J].Large-Scale Kernel Machines,2007,34(5):321-359.
[9] Sun Y,Wang X,Tang X.Deep convolutional network cascade for facial point detection[A].Conference on Computer Vision and Pattern Recognition[C].Portland,OR,USA:IEEE,2013.3476-3483.
[10] Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2006,18(7):1527-1554.
[11] Taigman Y,Yang M,Ranzato M A,et al.Deepface:Closing the gap to human-level performance in face verification[A].Conference on Computer Vision and Pattern Recognition[C].Columbus,OH,USA:IEEE,2014.1701-1708.
[12] Deng J,Zhang Z,Marchi E,et al.Sparse autoencoder-based feature transfer learning for speech emotion recognition[A].Humaine Association Conference on Affective Computing andIntelligent Interaction[C].Geneva,Switzerland:IEEE,2013.511-516.
[13] Phan H T,Duong A T,Le N D H,et al.Hierarchical sparse autoencoder using linear regression-based features in clustering for handwritten digit recognition[A].International Symposium on Image and Signal Processing and Analysis[C].Melbourne,Australia:IEEE,2013.183-188.
[14] Zhang C,Zhang Z.Improving multiview face detection with multi-task deep convolutional neural networks[A].IEEE Winter Conference on Applications of Computer Vision[C].Steamboat Springs,CO,USA:IEEE,2014.1036-1041.
[15] Dahl G E,Yu D,Deng L,et al.Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition[J].IEEE Transactions on Audio,Speech,and Language Processing,2012,20(1):30-42.
[16] Hinton G E,Salakhutdinov R R.Replicated softmax:an undirected topic model[A].Advances in Neural Information Processing Systems[C].Vancouver,BC,Canada:NIPS,2009.1607-1614.
[17] Heisele B,Ho P,Poggio T.Face recognition with support vector machines:Global versus component-based approach[A].International Conference on Computer Vision[C].Vancouver,Canada:IEEE,2001.688-694.
2017, Vol. 45 
