基于加权局部判别CCA的多视角步态识别方法
|
摘要
为了解决步态特征提取难题和克服单一视觉的步态进行身份识别方法的局限性,提出一种加权局部判别典型相关分析(WLDCCA)算法。在此基础上,提出一种基于WLDCCA的多视角步态识别方法。该方法通过在WLDCCA中引入样本的类信息和局部信息,将不同视觉的步态特征有机地融合起来,提取的融合特征能够最小化同类样本之间的距离,同时最大化异类样本之间的距离,提高了步态识别的识别率和鲁棒性。在CASIA步态数据库上的实验结果验证了该算法的有效性和可行性。
To solve the problem of gait feature extraction and overcome the influence of single observation angle on gait recognition rate, a Weighted Local Discriminent Canonical Correlation Analysis(WLDCCA)algorithm is proposed and then a multi-view gait recognition method is proposed. The main characteristic of the proposed method is that the class information and local information is introduced into WLDCCA algorithm to effectively fuse the gait features of different views. The extracted fusion features can maximize the distance between intra-class samples and minimize the distance between inter-class samples at the same time, and then the recognition rate is greatly improved. The experimental results on the CASIA gait database verify the effectiveness and feasibility of the proposed algorithm.
To solve the problem of gait feature extraction and overcome the influence of single observation angle on gait recognition rate, a Weighted Local Discriminent Canonical Correlation Analysis(WLDCCA)algorithm is proposed and then a multi-view gait recognition method is proposed. The main characteristic of the proposed method is that the class information and local information is introduced into WLDCCA algorithm to effectively fuse the gait features of different views. The extracted fusion features can maximize the distance between intra-class samples and minimize the distance between inter-class samples at the same time, and then the recognition rate is greatly improved. The experimental results on the CASIA gait database verify the effectiveness and feasibility of the proposed algorithm.
引文
[1]赵秀萍.生物特征识别技术发展综述[J].刑事技术,2011(6):44-48.
[2]毛巨勇.生物特征识别技术的突破与创新-生物识别技术的发展与现状[J].中国安防,2010(8):35-39.
[3]钟兴志,王晨升,刘丰,等.步态识别综述[J].软件,2013,34(4):160-164.
[4]Chen J,Liu J.Average gait differential image based human recognition[J].Scientific World Journal,2014(6):262-398.
[5]Zhang Shanwen,Zhang Chuanlei.Orthogonal margin maximization projection for gait recognition[J].Informatica,2015,26(2):357-367.
[6]张全贵,王炳超,李凡,等.基于SVM的步态识别方法综述[J].测控技术,2016,35(8):1-5.
[7]陈欣,杨天奇.不受服饰携带物影响的步态识别方法[J].计算机工程与应用,2016,52(5):141-146.
[8]Deepak N A,Sinha U N.Topic model for person identification using gait sequence analysis[J].International Journal of Computer Applications,2016,133(7):1-6.
[9]刘若水,关羽,杨薇.人脸与步态识别技术在人物识别与追踪领域的应用[J].河南科学,2017,35(8):1232-1236.
[10]Zhang S,Zhang L.Combining weighted adaptive CS-LBPand local linear discriminant projection for gait recognition[J].Multimedia Tools&Applications,2017(12):1-17.
[11]王金甲,刘青玉,陈浩.基于深度卷积神经网络的帕金森步态识别[J].中国生物医学工程学报,2017,36(4):418-425.
[12]于仕琪.多视角步态分析与识别[D].北京:中国科学院自动化研究所,2007.
[13]吕卓纹.基于类能量图与耦合度量的步态识别方法研究[D].哈尔滨:哈尔滨工程大学,2015.
[14]Hardoon D R,Szedmak S,Shawe-Taylor J.Canonical correlation analysis:an overview with application to learning methods[J].Neural Computation,2004,16(12):2639-2664.
[15]Sun T,Chen S.Locality preserving CCA with applications to data visualization and pose estimation[J].Image Vision Comput,2007,25(5):531-543.
[16]Xing X,Wang K,Yan T,et al.Complete canonical correlation analysis with application to multi-view gait recognition[J].Pattern Recognition,2016,50:107-117.
[17]Li B,Wang C,Huang D S.Supervised feature extraction based on orthogonal discriminant projection[J].Neurocomputing,2009,73(1):191-196.
[18]Wang G,Cui W,Shao Y.Discriminant locality preserving projection[J].Research Journal of Applied Sciences Engineering&Technology,2012,4(24):5572-5577.
[19]Zhang S,Lei Y K,Wu Y H,et al.Modified orthogonal discriminant projection for classification[J].Neurocomputing,2011,74(17):3690-3694.
[2]毛巨勇.生物特征识别技术的突破与创新-生物识别技术的发展与现状[J].中国安防,2010(8):35-39.
[3]钟兴志,王晨升,刘丰,等.步态识别综述[J].软件,2013,34(4):160-164.
[4]Chen J,Liu J.Average gait differential image based human recognition[J].Scientific World Journal,2014(6):262-398.
[5]Zhang Shanwen,Zhang Chuanlei.Orthogonal margin maximization projection for gait recognition[J].Informatica,2015,26(2):357-367.
[6]张全贵,王炳超,李凡,等.基于SVM的步态识别方法综述[J].测控技术,2016,35(8):1-5.
[7]陈欣,杨天奇.不受服饰携带物影响的步态识别方法[J].计算机工程与应用,2016,52(5):141-146.
[8]Deepak N A,Sinha U N.Topic model for person identification using gait sequence analysis[J].International Journal of Computer Applications,2016,133(7):1-6.
[9]刘若水,关羽,杨薇.人脸与步态识别技术在人物识别与追踪领域的应用[J].河南科学,2017,35(8):1232-1236.
[10]Zhang S,Zhang L.Combining weighted adaptive CS-LBPand local linear discriminant projection for gait recognition[J].Multimedia Tools&Applications,2017(12):1-17.
[11]王金甲,刘青玉,陈浩.基于深度卷积神经网络的帕金森步态识别[J].中国生物医学工程学报,2017,36(4):418-425.
[12]于仕琪.多视角步态分析与识别[D].北京:中国科学院自动化研究所,2007.
[13]吕卓纹.基于类能量图与耦合度量的步态识别方法研究[D].哈尔滨:哈尔滨工程大学,2015.
[14]Hardoon D R,Szedmak S,Shawe-Taylor J.Canonical correlation analysis:an overview with application to learning methods[J].Neural Computation,2004,16(12):2639-2664.
[15]Sun T,Chen S.Locality preserving CCA with applications to data visualization and pose estimation[J].Image Vision Comput,2007,25(5):531-543.
[16]Xing X,Wang K,Yan T,et al.Complete canonical correlation analysis with application to multi-view gait recognition[J].Pattern Recognition,2016,50:107-117.
[17]Li B,Wang C,Huang D S.Supervised feature extraction based on orthogonal discriminant projection[J].Neurocomputing,2009,73(1):191-196.
[18]Wang G,Cui W,Shao Y.Discriminant locality preserving projection[J].Research Journal of Applied Sciences Engineering&Technology,2012,4(24):5572-5577.
[19]Zhang S,Lei Y K,Wu Y H,et al.Modified orthogonal discriminant projection for classification[J].Neurocomputing,2011,74(17):3690-3694.