摘要
当今世界纷繁复杂,各种场合对安全的需求也越来越高。安全、有效、唯一等等得天独厚的优势使得生物特征识别技术逐渐被人们所重视。而根据人行走方式的不同来进行身份识别的步态识别则是一种新兴生物特征识别技术。由于具有其它生物特征识别所无可比拟的隐蔽性、非侵犯性和对图像分辨率要求不高等优点,在视觉监控领域有着广泛的应用前景。
步态识别过程主要由步态图像预处理、特征提取和分类识别三个部分组成。其中步态图像预处理就是通过背景建模、前景检测和形态学后处理等工作,从输入序列图像中检测运动目标,这是后续特征提取和目标分类工作的基础,有很重要的意义。而特征提取是决定识别效果的决定因素,是重中之重,也是本文研究的重点。
本文针对侧面视角条件下提出了两种提取特征的步态识别方法。其中之一基于Radon变换,将用于直线检测的Radon变换推广到特征空间的建立上。根据人体运动特性,将运动人体时空不变的身体结构参数和随时空变化的动态参数结合起来,构造一个周期的特征向量模板。其后运用主成分分析(PCA)降降低特征空间维数,提取特征主向量。另一种方法利用反映动态信息的各个对象步态能量图(GEI )的标准差为动态权值掩模(DWM),通过校准后与GEI做Hadamard积,从而得到动态信息增强的步态能量图(EGEI),这幅图像不仅保留了轮廓、频率、相位等步态信息,而其一定程度上解决了遮挡问题。再运用结合了行方向与列方向的二维主成分分析((2D)2PCA)来计算训练和测试样本的主元分量,得到对识别贡献最大的特征向量矩阵。
单一的步态特征对步态的描述存在局限性,本文抓住不同步态特征在识别时能够提供互补信息这一关键,采用基于计分规则的方法,在决策层对两种特征的信息进行了融合。
实验采用CASIA步态数据库中包含124个对象、3种行走条件的Dataset B,从以上算法的识别性能进行测试评估。结果表明本文提出的通过增强能量图来提取特征的算法对衣着变化和携带物的影响具有很好的鲁棒性。将两种特征信息通过合理的规则进行融合对提高识别性能更具效果。
The world we live in is so diverse and complicated that we need higher and higher security guarantee in every situation. Its advantageous characteristics such as safety, efficiency and unique etc. make biometrics identification technology (biometrics) valued by more and more people gradually. Gait recognition, which is to identify one's identity according to the different ways one walks, is a rising biometrics. And because of its incomparable crypticity, noninvasive and low demand of image resolution ratio which other biometrics doesn't have, it has an extensive prospect of application in the vision monitoring area.
The process of gait recognition is mainly made up of three parts, that is, pretreatment of gait image , feature extraction and discrimination. Pretreatment of gait image is to input detected moving targets of the image sequence via background modeling, future check and the morphology reprocessing. It is the foundation of following extracting features and classifying objective subjects, which is significant. Of all the work, extracting features is the most important, and it is the decisive factor in an effect of discrimination. And because of this, it is a key point this paper deals with. In view of side vision, this paper comes up with two ways of extracting features.
One is based on Radon transformation. That is, Radon transformation used in line detection is extended to the foundation of feature space. In line with characteristics of human motion, we can combine space-time invariant physical structure parameters of moving human with Time-varying motion parameters, making up a periodic eigen vector template. Then we can use Principal Component Analysis(PCA) to reduce the feature space dimension to extract main vctor of characteristics. The other way is to use standard deviation of each object’s gait energy image(GEI) which can reflect the dynamic information as dynamic weight mask(DWM). After correcting, we can use Hadamard product with GEI to get enhanced gait energy image(EGEI) whose dynamic information is strengthen. Not only does it continue to have the gait information of contour, frequency, phase etc. but also it solves occlusion issue to a degree. Then, we can use Two Directional Two Dimensional Princdipal Component Analysis((2D)2PCA) to calculate prineipal components of training and testing samples, and we get eigenvector matrix which contributes most to identification.
There are limitations in single gait description to gait characteristics. The text keeps to the key point that different gait characteristics can provide complementary information when identifying, adopts methods based on score and fuses together information of two kinds of character at decision layer.
The experiment adopts CASIA which contains 124 objects and Dataset B of 3 walking conditions. Using above ways of algorithm, we test and assess its distinctions. The results show that the algorithm referred to in the paper which extracts feature by enhanced gait energy image, has a perfect robustness to clothes varying and carried materials. And it's better to merge together two sorts of characteristic information by reasonable rules to improve its distincions.
引文
[1] Jain, R. Bolle, S. Pankanti, Biometrics: personal identification in networked society[M], Kluwer Academic Publishers, Boston, 1999.
[2]何杰.生物特征识别技术综述[J].中国安防产品信息, 2003, 9(6): 38-42.
[3] Niyogi S, Adelson E. Analyzing and recognizing walking figures in XYT[C]. In Proceeding of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Seattle, USA: 1994: 469-474.
[4] Dawson MR.Gait Recognition Final Report[R]. Technology and Medieine. London: 2002:1-15.
[5]薛召军,万柏坤等.基于多参数模型结合的步态特征提取新方法[J].中国生物医学工程学报, 2009, 28(1): 22-26.
[6] Kale, A.N. Rajagopalan, N. Cuntoor, and V. Kruger. Human Identification Using Gait[C]. International Conference on Automatic Face and Gesture Recognition, 2002: 336-341.
[7] Collins R.T, Gross R, and Shi J.B. Silhouette-based Human Identification from Body Shape and Gait[C]. In proceedings of the 5th IEEE International Conference on Automatic Face and Gesture Recognition, 2002:366-371.
[8] L. Lee and W. Grimson. Gait Analysis for Recognition and Classification[C]. In proceedings of the Fifth IEEE International Conference on Automatic Face and Gesture Recognition, 2002:112-115.
[9] Cunado D, Nixon M, Carter J. Using Gait as a Biometric, via Phase-wighted Magnitude Specter[C]. In Proceeding of International Conference on Audio and Video Based Biometric Person Authentication. Crans-Montana, Switzerland:[s. n], 1997:95-102.
[10] Yam C.Y, Nixon M.S, Carter J.N. Automated Person Recognition byWalking and Running via Model-Based Approaches[J]. Pattern Recognition, 2004, 37(5): 1057-1072.
[11] Yoo J.H. and Nixon M.S. On Laboratory Gait Analysis via Computer Vision[C]. In proceeding of AISB’03 Symposium on Biologically-Inspired Machine Vsion,Theory and Application. Aberystwyth, UK, 2003, 109-113
[12] Lee Lily. Gait Analysis for Classification[R]. AI Technical Report 2003-014, Cambridge, Massachusetts, USA: Massachusetts Institute of Technology, 2003.
[13] Ning Hua-zhong, Tan Tie-niu, Wang Liang and Hu Wei-ming. Kinematics-based tracking of human walking in monocular video sequence[J]. Image and Vision Computing, 2004, 22(5): 429-441
[14] Urtasun R, Fua P. 3D Ttracking for Gait Characterization and Recognition[C]. In proceeding ofthe 6th IEEE International Conference on Automatic Face and Gesture Recognition, 2004: 17-22.
[15] Han J, Bhanu B. Performance Prediction for Individual Recognition by Gait[J]. Pattern Recognition Letters, 200526(5): 615-624.
[16] Dockstader S.L, Berq M.J, Tekalp A.M. Stochastic Kinematic Modeling and Feature Extraction for Gait Analysis[C]. IEEE Transactions on Image Processing, 2003, 12(8): 962-976.
[17] Murase H, Sakai R. Moving Object Recognition in Eigenspace Representation: Gait Analysis and Lip Reading[J]. Pattern Recognition Letters, 1996, 17(2):155-162.
[18] Huang P.S, Harris C.J, Nixon M.S. Recognizing Humans by Gait via Parametric Canonical Space[J]. Artificial Intelligence in Engineering,1999,13(4):359-366.
[19]王亮,胡卫明,谭铁牛.基于步态的身份识别[J].计算机学报, 2003, 26(3): 353-360.
[20] BenAbdelkader C, Cutler R.G, Davis L.S. Motion-based Recognition of People in Eigen Gait Space[C]. In proceeding of the 5th IEEE International Conference on Automatic Face and Gesture Recognition, 2002:267-272.
[21] Little J.J, Boyd J.E. Recognizing People by Their Gait: The Shape of Motion[J]. Journal of Computer Vision Research, Videre,1998,1(2).
[22] Hayfron-Acquah J.B, Nixon M.S, Cater J.N. Automatic Gait Recognition by Symmetry Analysis[J]. Pattern Recognition Letters,2003,24(13):2175-2183.
[23] Kale A, Rajagopalan A, Cuntoor N, et al. Gait Based Recognition of Humans Using Continuous Hmms[C]. In proceeding of IEEE International Conference on Automatic Face and Gesture Recognition, 2002:321-326.
[24] Johnson A.Y, Bobick A.F. A Mullti-View Method for Gait Recognition Using Static Body Parameters[C]. In proceeding of the 3rd International Conference on Audio-and Video-Based Biometric Person Authentication, 2001:301-305.
[25] Lee Chan-Su, Elgammal A. Gait Style and Gait Content: Bilinear Models for Gait Recognition Using Gait Re-Sampling[C]. In proceeding of the 6th IEEE International Conference on Automatic Face and Gesture Recognition, 2004:147-152.
[26] Sarkar S, Phillips P.J, Liu Zong-yi, et al. The HumanID Gait Challenge Problem:Data Sets, Performance, and Analysis[C]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005,27(2):162-177.
[27] Shakhnarovich G, Lee L, Darrell T. Integrated Face and Gait Recognition from Multiple Views[C]. In proceeding of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2001:439-446.
[28]王科俊,刘丽丽,贲晛烨.基于步态识别的智能监控系统研究[J].计算机应用, 2009, 29(2):386-388.
[29]阮雨慧.步态识别的算法研究和系统实现[D].北京:北京交通大学, 2008,6.
[30] Mech R, Wolborn M. A Noise Robust Method for 2D Shape Estimation of Moving Objects in Video Sequences Considering a Moving Camera[J]. Signal Processing, 1998,66(2):203-217.
[31] Meyer D. Human gait classification based on Hidden Markov Models[J]. 3D Image Analysis and Synthesis. Erlangen: 1997:39-146.
[32] Lipton A, Fujiyoshi H, Patil R. Moving Target Classification and Tacking from Real-time Video[C]. In proceedings of IEEE Workshop on Applications of Computer Vision, Princeton, NJ, 1998: 8-14.
[33] Collins R. A System forVideo Surveillance Anti monitoring: VSAM Final Report. Carnegie Mellon University, Technical Report: CMU-RI-TR-00-12, 2000: 85-103.
[34] Magee D.R. Tracking Multiple Vehicle Using Foreground, Background and Motion Models[J]. Image and Vision Computing,2004,22(2): 143-155.
[35] Haritaoglu I, Harwood D, Davis L.s. W4:Real-Time Surveillance of People and Their Activities[J]. IEEE Trans Pattern Analysis and Machine Intelligence, 2000,22(8): 809-830.
[36] Gonzalez R.C, Woods R.E, Eddins S.L. Digital Image Processing[M]. Beijing: Publishing House of Electronics Industry, 2007: 446-450.
[37]侯本博.基于步态的身份识别研究[D].哈尔滨:哈尔滨工程大学, 2007,1.
[38] Nixon M.S, Carter J.N, Cunado D, et al. Automatic Gait Recognition[C]. Biometrics: Personal Identification in Network Society. Netherlands: Kluwer Academic Publishers, 1999:231-250.
[39] Cutler R, Davis L. Robust Rea-Time Periodic Motion Detection, Analysis, and Applications[C]. IEEE Transactions Pattern Analysis and Machine Intelligence, 2000, 22(8):778-796.
[40] Kale A, Sundaresan A, Rajagopalan A.N, et al. Identification of Human Using Gait[C]. IEEE Transactions on Image Processing, 2004, 13(9): 1163-1173.
[41]马勤勇,王申康,聂栋栋等.基于瞬时步态能量图的远距离身份识别[J].电子学报,2007, 35(11): 2078-2082.
[42] BenAbdelkader C, Cutler R.G, Davis L.S. Stride and Cadence as a Biometric in Automatic Person Identification and Verification[C]. In Proceedings of the 5th IEEE International Conference on Automatic Face Gesture Recognition, 2002:372-377.
[43] Liang Wang, Tieniu Tan, Weiming Hu and Huazhong Ning. Automatic Gait Recognition based on Statistical Shape Analysis[J]. IEEE. Transactions on Image Proeessing, 2003, 12(9): 1120-1131.
[44] Chiraz Benabdelkadery, Ross Cutlerz, Harsh Nanday and Larry Davisy. Gait: Motion-based Recognition of People using Image Self-Similarity[C]. AVBPA’01, 2001: 312-317.
[45]张浩,刘志镜.基于动态时间规整的不太自动识别[J].图像图形技术研究与应用, 2009: 139-145.
[46] Liang W, Tieniu T, Huazhong N, Weiming H. Silouette Analysis-based Gait Recognition for Human Identification[A]. IEEE Transaction on Pattern Analysisi and Machine Intelligence[C], 2003,25(12):1505-1517..
[47]赵子健,吴晓娟.基于近似时空切片向量的步态识别方法研究阴[J].模式识别与人工智能, 2005, 18(5): 608-614.
[48]黄凤岗,韩雪花.基于Radon变换的特征提取在步态识别中的应用[J].哈尔滨工程大学学报, 2007,28(3):301-304.
[49] Ju H, Bhanu B. Statistical Feature Fusion for Gait-based Human Recognition[A]. IEEE Conference on Computer Vision and Pattern Recognition [C]. 2004,2: 842 -847.
[50]王利巧,裴亚南.基于步态能量图和不变矩的身份识别算法[J].计算机与数字工程, 2009, 37(10): 128-131.
[51] Chen L, Liao H, Lin J, et al. Why Recognition in A Statistics-based Face Recognition System Should be Based on The Pure Face Portion A Probabilistic Decision-based Proof[J]. Pattern Recognition, 2001,34(7):1393-1403.
[52] Pankanti S, Prabhakar S, Jain S, et al. On the Individuality of Fingerprints[A]. IEEE TPAMI[C], 2002, 24(8):1010-1025.
[53] Jian Yang, David Zhang. Two-dimensional PCA: A New Approach to Appearance-based Face Representation and Recognition [A]. IEEE Transactions on Pattern Analysis and Machine Intelligence[C], 2004, 26(1): 131-137.
[54]何国辉,甘俊英.二维主元分析在人脸识别中的应用研究[J].计算机工程与设计, 2006, 27(24): 4667-4673.
[55] Dao-qiang Zhang, Zhi-hua Zhou. Two-directional Two-dimensional PCA for Efficient Face Representation and Recognition[J]. Neuron Computing, 2005,69: 224-231.
[56]曾莹,杨路明,曾庆冬.基于特征融合的步态识别算法[J].计算机工程与应, 2008, 44(3): 220-222.
[57]韩宗昭,朱洪艳,段战胜.多元信息融合[M].北京:清华大学出版社, 2006.
[58]刘映杰,冯晓兰,马义德等.多生物特征识别系统的关键技术[J].舰船电子工程, 2006, 1(15):36-39.
[59]鲁继文,张二虎,薛延学.基于独立成分分析和信息融合的步态识别[J].模式识别与人工智能, 2007, 20(3):365-370.
[60]张开银.模式识别中的信息融合方法一二次决策[J].仪器仪表学报, 2004, 25(4):561-564.
[61]潘泉,于昕,程咏梅等.信息融合理论的基本方法与进展[J].自动化学报, 2003,29(4):599-615.
[62] Nikolaos V, Boulgouris, Konstantios N, Plataniotis, et al. Gait Recognition using Linear Time Normalization[J]. Pattern Recognition, 2006, 39:969-979.