光流分量分解的步态识别

doi:10.11896/j.issn.1002-137X.2016.09.059

摘要/Abstract

摘要： 步态识别因其远距离和难于伪装等特点在生物识别技术中颇受关注。针对目前特征提取方法信息量不足的现状,提出一种基于光流分量分解的步态识别方法,对步态光流图中横向分量和纵向分量为正的部分按行、列进行分解,求出每行和每列的光流横、纵向分量,得到4个特征向量。根据训练得出的每个特征向量在识别过程中的权重进行特征融合。将主成分分析和线性判别分析相结合,用动态时间规整算法进行匹配,最后采用最近邻分类算法分类。在CASIA Database B和C上的实验表明,该方法在正常、背包和穿大衣的条件下分别得到了97%,90%和64%的识别率,在慢速和快速行走的条件下分别得到了88%和87%的识别率。

关键词: 步态识别,光流,主成分分析,线性判别分析,动态时间规整,特征融合

Abstract: Gait recognition has gained tremendous attention for its characteristics of long distance and hard-to-disguise.Aiming at the problem of insufficient information of the existing feature extraction method,this paper proposed a novel gait recognition method based on the decomposition of optical flow components.The positive transverse or longitudinal components in gait optical flow image are decomposed by rows and columns,then the transverse and longitudinal components of optical flow for each row or colum are calculated,and four feature vectors are obtained.The four vectors are fused according to their weight in recognition.Principle components analysis and linear discriminant analysis techniques-are combined,and dynamic time warping algorithm is used to match.Finally,K-nearest neighbor algorithm is used for classification.Experiments on CASIA Database B and C show that,the proposed method achieves recognition accuracy of 97%,90% and 64% respectively under the conditions of normal,backpack-wearing and coat-wearing,88% and 87% under conditions of slow walking and fast walking.

Key words: Gait recognition,Optical flow,Principle components analysis,Linear discriminant analysis,Dynamic time warping,Feature fusion

罗正平,刘延钧,杨天奇. 光流分量分解的步态识别[J]. 计算机科学, 2016, 43(9): 295-300. https://doi.org/10.11896/j.issn.1002-137X.2016.09.059

LUO Zheng-ping, LIU Yan-jun and YANG Tian-qi. Gait Recognition Based on Decomposition of Optical Flow Components[J]. Computer Science, 2016, 43(9): 295-300. https://doi.org/10.11896/j.issn.1002-137X.2016.09.059

参考文献

[1] Lee T K M,Belkhatir M,Sanei S.A comprehensive review of past and present vision-based techniques for gait recognition[J].Multimed Tools Appl,2014,2(3):2833-2869
[2] Zhang Er-hu,Zhao Yong-wei,Xiong Wei.Active energy image plus 2DLPP for gait recognition[J].Signal Processing,2010,0(7):2295-2302
[3] Bouchrika I,Nixon M S.Model-based feature extraction for gait analysis and recognition[J].Lect Notes Comput Sc,2007,8:150-160
[4] Tafazzoli F,Safabakhsh R.Model-based human gait recognition using leg and arm movements[J].Eng Appl Artif Intel,2010,3(8):1237-1246
[5] Chen C H,Liang J M,Zhao H,et al.Frame difference energy image for gait recognition with incomplete silhouettes[J].Pattern Recognition Letters,2009,0(11):977-984
[6] Muramatsu D,Shiraishi A,Makihara Y,et al.Gait-Based Person Recognition Using Arbitrary View Transformation Model[J].IEEE Transactions on Image Processing,2015,4(1):140-154
[7] Chen X,Yang T Q,Xu J M.Cross-view gait recognition based on human walking trajectory[J].J Vis Commun Image R,2014,5(8):1842-1855
[8] Horn B,Schunck B.Determining Optical Flow[J].Artificial Intelligence,1981,7(2):185-203
[9] Lam T H W,Cheung K H,Liu J N K.Gait flow image:A silhouette-based gait representation for human identification[J].Pattern Recognition,2011,4(4):973-987
[10] Yu C C,Cheng C H,Fan K C.A Gait Classification Systemusing Optical Flow Features[J].J Inf Sci Eng,2014,0(1):179-193
[11] Yang Ya-zhou,Tu Dan,Li Guo-hui.Gait recognition using flow histogram energy image[C]∥22nd International Conference on Pattern Recognition.2014:444-449
[12] Pers J,Sulic V,Kristan M,et al.Histograms of optical flow for efficient representation of body motion[J].Pattern Recognition Letters,2010,1(11):1369-1376
[13] Yang Yang,Guo Ji-chang.Gait recognition method based onspatial distribution of optical flow[J].Application Research of Computers,2013,0(7):2206-2209(in Chinese) 杨阳,郭继昌.基于光流空间分布的步态识别方法[J].计算机应用研究,2013,0(7):2206-2209
[14] Xu Yan-qun,Zhang Bin.Application of Segmentation Based on Optical Flow in Gait Recognition[J].Computer Science,2012,9(4):275-277,2(in Chinese) 徐艳群,张斌.一种基于光流的多区域分割在步态识别中的应用[J].计算机科学,2012,9(4):275-277,2
[15] Lucas Bruce D,Takeo K.An Iterative Image Registration Technique with an Application to Stereo Vision[C]∥Procedings of Imaging Understanding Workshop.1981:121-130
[16] Harold H.Analysis of a complex of statistical variables intoprincipal components[M].Baltimore:Warwick & York,1933
[17] Wang Cheng-liang,Lan Li-bin,Zhang Yu-wei,et al.Face recognition based on principle component analysis and support vector machine[C]∥2011 3rd International Workshop on Intelligent Systems and Applications.2011
[18] Zhou C J,Wang L,Zhang Q,et al.Face recognition based on PCA image reconstruction and LDA[J].Optik,2013,4(22):5599-5603
[19] Balakrishnama S,Ganapathiraju A,Picone J.Linear Discrimi-nant Analysis for signal processing problems[C]∥IEEE Southeastcon ’99.1999:78-81
[20] Hernandez-Vela A,Bautista M A,Perez-Sala X,et al.Probability-based Dynamic Time Warping and Bag-of-Visual-and-Depth-Words for Human Gesture Recognition in RGB-D[J].Pattern Recognition Letters,2014,0:112-121
[21] Sheikhan M,Gharavian D,Ashoftedel F.Using DTW neural-based MFCC warping to improve emotional speech recognition[J].Neural Comput Appl,2012,1(7):1765-1773
[22] Chinese Academy of Sciences.Institute of Automation Gait Database.http://www.sinobiometrics.com
[23] Liu Zhi-yong, Feng Guo-can, Chen Wei-fu.Gait RecognitionBased on Local Binary Pattern and Discriminant Common Vector[J].Computer Science,2013,0(9):262-265(in Chinese) 刘志勇,冯国灿,陈伟福.基于局部二值模式和辨识共同向量的步态识别[J].计算机科学,2013,0(9):262-265
[24] Zeng W, Wang C.Gait recognition across different walkingspeeds via deterministic learning[J].Neurocomputing,2015,2:139-150

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed