基于多纹理CS-LBP特征的多视角人脸检测算法
|
摘要
提出一种多纹理中心对称局部二值模式(CS-LBP)特征,实现复杂环境下的多视角人脸检测.该特征保留Haar特征的序数关系,借鉴局部二值模式(LBP)的组合方式,从水平、垂直、+45°和-45°这4个纹理方向进行特征提取,以保证人脸检测在方向、光照、旋转等方面的鲁棒性.算法采用级联架构,首先针对人脸图像中的不同视角进行分区,分别进行多纹理特征的提取,然后设计独立的分类器,逐级剔除非人脸窗口,最后采用多层感知器(MLP)综合各视角的检测效果,输出检测结果.在数据集FDDB和CMU PIE上进行验证性实验的结果表明,该方法对复杂环境下的多视角人脸检测有效,与传统的卷积神经网络人脸检测方法相比,该方法具有更高的精度.
We proposed a multi-texture centrosymmetric local binary pattern(CS-LBP)feature to realize multi-view face detection in complex environments.The feature retained the characteristics of Haar ordinal relations,so we drew on the experience of the combination of local binary pattern(LBP)to extract features from four texture directions,such as horizontal,vertical,+45°and-45°,so as to ensure the robustness of face detection in direction,illumination,rotation and so on.The algorithm adopted the cascade architecture.First,face images were partitioned according to different angles of view,and multi-texture features were extracted respectively.Then some independent classifiers were designed to eliminate the non-face window step by step.Finally,the multilayer perceptron(MLP)was used to synthesize the detection effect of each angle of view to output the detection results.The results of verification experiments on data sets FDDB and CMU PIE show that this method is effective for multi-view face detection in complex environment.Compared with traditional convolution neural network face detection method,this method has higher accuracy.
We proposed a multi-texture centrosymmetric local binary pattern(CS-LBP)feature to realize multi-view face detection in complex environments.The feature retained the characteristics of Haar ordinal relations,so we drew on the experience of the combination of local binary pattern(LBP)to extract features from four texture directions,such as horizontal,vertical,+45°and-45°,so as to ensure the robustness of face detection in direction,illumination,rotation and so on.The algorithm adopted the cascade architecture.First,face images were partitioned according to different angles of view,and multi-texture features were extracted respectively.Then some independent classifiers were designed to eliminate the non-face window step by step.Finally,the multilayer perceptron(MLP)was used to synthesize the detection effect of each angle of view to output the detection results.The results of verification experiments on data sets FDDB and CMU PIE show that this method is effective for multi-view face detection in complex environment.Compared with traditional convolution neural network face detection method,this method has higher accuracy.
引文
[1]王庆伟,应自炉.一种基于Haar-Like T特征的人脸检测算法[J].模式识别与人工智能,2015,28(1):35-41.(WANG Qingwei,YING Zilu.A Face Detection Algorithm Based on Haar-Like T Features[J].PR&AI,2015,28(1):35-41.)
[2]黄艳国,赵书玲,许伦辉.基于纹理特征和颜色匹配的车牌定位方法[J].微电子学与计算机,2011,28(9):123-126.(HUANG Yanguo,ZHAO Shuling,XU Lunhui.Vehicle Locense Plate Location Technique Based on Texture Feature and Color Matching[J].Microelectronics&Computer,2011,28(9):123-126.)
[3]Ojala T,Harwood I.A Comparative Study of Texture Measures with Classification Based on Feature Distributions[J].Pattern Recognition,1996,29(1):51-59.
[4]HeikkilaM,Pietikanen M,Schmid C.Description of Interest Regions with Local Binary Patterns[J].Pattern Recognition,2009,42(3):425-436.
[5]Viola P,Jones M.Rapid Object Detection Using a Boosted Cascade of Simple Features[C]//IEEE Society Conference on Computer Vision&Pattern Recognition.Piscataway,NJ:IEEE,2003:511-518.
[6]Yan S Y,Shan S G,Chen X L,et al.Locally Assembled Binary(LAB)Feature with Feature-Centric Cascade for Fast and Accurate Face Detection[C]//IEEE Conference on Computer Vision&Pattern Recognition.Piscataway,NJ:IEEE,2008:1-7.
[7]Felzenszwalb P,Girshick R,Mcallester D.Visual Object Detection with Deformable Part Models[J].Communications of the ACM,2010,56(9):2241-2248.
[8]Zhang C,Zhang Z.A Survey of Recent Advances in Face Detection[R].Redmond:Microsoft Research,2010.
[9]王亚,朱明.基于卷积神经网络的人脸检测方法[J].网络新媒体技术,2016,5(6):24-29.(WANG Ya,ZHU Ming.Approach on Convolutional Neural Network Face Detection[J].Network New Media Technology,2016,5(6):24-29.)
[10]叶学义,陈雪婷,陈华华.级联型P-RBM神经网络的人脸检测[J].中国图象图形学报,2016,21(7):875-885.(YE Xueyi,CHEN Xueting,CHEN Huahua.Cascaded Probability State-Restricted Boltzmann Machine for Face Detection[J].Journal of Image and Graphics,2016,21(7):875-885.)
[11]Wu S Z,Kan M N,He Z L.Funnel-Structured Cascade for Multi-view Face Detection with Alignment-Awareness[J].Neurocomputing,2017,221(C):138-145.
[12]Jain V,Learned-Miller E.FDDB:A Benchmark for Face Detection in Unconstrained Settings[R].Amherst:University of Massachusetts,2010.
[13]Heo J,Savvides M.Face Recognition across Pose Using View Based Active Appearance Models(VBAAMs)on CMU Multi-PIE Dataset[C]//International Conference on Computer Vision Systems.Berlin:Springer,2008:527-535.
[14]方承志,苏腾云.基于改进型LBP的Floatboost人脸检测[J].南京邮电大学学报(自然科学版),2014,34(6):75-79.(FANG Chengzhi,SU Tengyun.Floatboost Algorithm for Face Detection Based on Improved LBP Features[J].Journal of Nanjing University of Posts and Telecommunications(Natural Science),2014,34(6):75-79.)
[15]Ojala T,Pietikainen M,Maenpaa T.Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns[J].European Conference on Computer Vision,2000,24(7):404-420.
[16]Lowe D G.Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[17]贾伟,王正勇,张杰,等.一种基于改进的CS-LBP算子纹理图像自适应检索方法[J].微电子学与计算机,2013,30(9):75-78.(JIA Wei,WANG Zhengyong,ZHANG Jie,et al.A Self-adaptive Texture Image Retrieval Method Based on the Improved CS-LBP Operator[J].Microelectronics&Computer,2013,30(9):75-78.)
[18]Mirjalili S.How Effective Is the Grey Wolf Optimizer in Training Multi-layer Perceptrons[J].Applied Intelligence,2015,43(1):150-161.
[19]Müller H,Müller W,Squire D M,et al.Performance Evaluation in Content-Based Image Retrieval:Overview and Proposals[J].Pattern Recognition Letters,2001,22(5):593-601.
[20]Lecun Y,Bottou L,Bengio Y,et al.Gradient-Based Learning Applied to Document Recognition[J].Proceeding of the IEEE,1998,86(11):2278-2324.
[21]李丹,沈夏炯,张海香,等.基于Lenet-5的卷积神经网络改进算法[J].计算机时代,2016(8):4-6.(LI Dan,SHEN Xiajiong,ZHANG Haixiang,et al.Improved Convolutional Neural Network Based on Lenet-5[J].Computer Era,2016(8):4-6.)
[22]Krizhevsky A,Sutskever I,Hinton G E.Imagenet Classification with Deep Convolutional Neural Networks[J].Communications of the ACM,2017,60(6):84-90.
[23]YUAN Zhengwu,ZHANG Jun.Feature Extraction and Image Retrieval Based on AlexNet[C]//Eighth International Conference on Digital Image Processing.[S.l.]:SPIE,2016:10033E.
[2]黄艳国,赵书玲,许伦辉.基于纹理特征和颜色匹配的车牌定位方法[J].微电子学与计算机,2011,28(9):123-126.(HUANG Yanguo,ZHAO Shuling,XU Lunhui.Vehicle Locense Plate Location Technique Based on Texture Feature and Color Matching[J].Microelectronics&Computer,2011,28(9):123-126.)
[3]Ojala T,Harwood I.A Comparative Study of Texture Measures with Classification Based on Feature Distributions[J].Pattern Recognition,1996,29(1):51-59.
[4]HeikkilaM,Pietikanen M,Schmid C.Description of Interest Regions with Local Binary Patterns[J].Pattern Recognition,2009,42(3):425-436.
[5]Viola P,Jones M.Rapid Object Detection Using a Boosted Cascade of Simple Features[C]//IEEE Society Conference on Computer Vision&Pattern Recognition.Piscataway,NJ:IEEE,2003:511-518.
[6]Yan S Y,Shan S G,Chen X L,et al.Locally Assembled Binary(LAB)Feature with Feature-Centric Cascade for Fast and Accurate Face Detection[C]//IEEE Conference on Computer Vision&Pattern Recognition.Piscataway,NJ:IEEE,2008:1-7.
[7]Felzenszwalb P,Girshick R,Mcallester D.Visual Object Detection with Deformable Part Models[J].Communications of the ACM,2010,56(9):2241-2248.
[8]Zhang C,Zhang Z.A Survey of Recent Advances in Face Detection[R].Redmond:Microsoft Research,2010.
[9]王亚,朱明.基于卷积神经网络的人脸检测方法[J].网络新媒体技术,2016,5(6):24-29.(WANG Ya,ZHU Ming.Approach on Convolutional Neural Network Face Detection[J].Network New Media Technology,2016,5(6):24-29.)
[10]叶学义,陈雪婷,陈华华.级联型P-RBM神经网络的人脸检测[J].中国图象图形学报,2016,21(7):875-885.(YE Xueyi,CHEN Xueting,CHEN Huahua.Cascaded Probability State-Restricted Boltzmann Machine for Face Detection[J].Journal of Image and Graphics,2016,21(7):875-885.)
[11]Wu S Z,Kan M N,He Z L.Funnel-Structured Cascade for Multi-view Face Detection with Alignment-Awareness[J].Neurocomputing,2017,221(C):138-145.
[12]Jain V,Learned-Miller E.FDDB:A Benchmark for Face Detection in Unconstrained Settings[R].Amherst:University of Massachusetts,2010.
[13]Heo J,Savvides M.Face Recognition across Pose Using View Based Active Appearance Models(VBAAMs)on CMU Multi-PIE Dataset[C]//International Conference on Computer Vision Systems.Berlin:Springer,2008:527-535.
[14]方承志,苏腾云.基于改进型LBP的Floatboost人脸检测[J].南京邮电大学学报(自然科学版),2014,34(6):75-79.(FANG Chengzhi,SU Tengyun.Floatboost Algorithm for Face Detection Based on Improved LBP Features[J].Journal of Nanjing University of Posts and Telecommunications(Natural Science),2014,34(6):75-79.)
[15]Ojala T,Pietikainen M,Maenpaa T.Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns[J].European Conference on Computer Vision,2000,24(7):404-420.
[16]Lowe D G.Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[17]贾伟,王正勇,张杰,等.一种基于改进的CS-LBP算子纹理图像自适应检索方法[J].微电子学与计算机,2013,30(9):75-78.(JIA Wei,WANG Zhengyong,ZHANG Jie,et al.A Self-adaptive Texture Image Retrieval Method Based on the Improved CS-LBP Operator[J].Microelectronics&Computer,2013,30(9):75-78.)
[18]Mirjalili S.How Effective Is the Grey Wolf Optimizer in Training Multi-layer Perceptrons[J].Applied Intelligence,2015,43(1):150-161.
[19]Müller H,Müller W,Squire D M,et al.Performance Evaluation in Content-Based Image Retrieval:Overview and Proposals[J].Pattern Recognition Letters,2001,22(5):593-601.
[20]Lecun Y,Bottou L,Bengio Y,et al.Gradient-Based Learning Applied to Document Recognition[J].Proceeding of the IEEE,1998,86(11):2278-2324.
[21]李丹,沈夏炯,张海香,等.基于Lenet-5的卷积神经网络改进算法[J].计算机时代,2016(8):4-6.(LI Dan,SHEN Xiajiong,ZHANG Haixiang,et al.Improved Convolutional Neural Network Based on Lenet-5[J].Computer Era,2016(8):4-6.)
[22]Krizhevsky A,Sutskever I,Hinton G E.Imagenet Classification with Deep Convolutional Neural Networks[J].Communications of the ACM,2017,60(6):84-90.
[23]YUAN Zhengwu,ZHANG Jun.Feature Extraction and Image Retrieval Based on AlexNet[C]//Eighth International Conference on Digital Image Processing.[S.l.]:SPIE,2016:10033E.