正在加载图片...
.424. 智能系统学报 第8卷 tation for human identification:mass vector[C]//Proceed- 结束语 ings of the 2nd IEEE Conference on Industrial Electronics 本文提出了一种基于线性插值的矩阵步态识别 and Applications.Harbin,China.2007:669-673. 算法框架,并将投影特征、Hough变换特征、Trace变 [9]NANDINI C,RAVIKUMAR C N.An approach to gait rec- 换特征和Fan-Beam映射特征在CASIA(B)步态库 ognition[C]//International Symposium on Biometrics and 上验证了该框架在步态识别问题中的有效性.Trace Security Technologies.Islamabad,Pakistan,2008:1-3. [10]BENABDELKADER C,CUTLER R,DAVIS L.Motion- 变换在Radon变换的基础上加上欧氏距离的度量作 based recognition of people in EigenGait space[C]//Pro- 用,识别率略有提高,特征维数相差不多:角度投影 ceedings of the Fifth IEEE International Conference on Au- 特征表达形式简单,但识别率不高:在连续的情况 tomatic Face and Gesture Recognition.Washington,DC, 下,Hough变换可被看作是Radon变换的特例,但识 USA,2002:267-272. 别率不如Radon变换:Fan-Beam映射的识别率最 [11]YU Shiqi,WANG Liang,HUANG Kaiqi,et al.Gait anal- 高,但是以特征维数多作为代价.最后还得出该框架 ysis for human identification in frequency domain[C]// 的本质就是一种权值不同的能量形式的结论.下一 Proceedings of the 3rd International Conference on Image 步的工作重点将考虑步态的张量表达形式,寻找更 and Graphics.Hong Kong,China,2004:282-285. 有效的特征提取与选择方法 [12]XU Junhong,CONG Wang,LI Jin,et al.Gait recognition based on key frame and elliptical model[C]//Proceedings 参考文献: of 2010 IEEE International Conference on Information and Automation.Harbin,China,2010:2483-2487. [1]贲晛烨基于人体运动分析的步态识别算法研究[D].哈 [13]WANG Kejun,BEN Xianye,ZHAO Yue.Gait period de- 尔滨:哈尔滨工程大学,2010 tection based on regional characteristics analysis[C]// [2]NIXON M S,CARTER J N,SHUTLER J D,et al.New ad- Proceedings of the 2009 Chinese Conference on Pattern vances in automatic gait recognitionJ.Information Securi- Recognition,and the CJK Joint Workshop on Pattern Rec- ty Technical Report,2002,7(4):23-35. ognition.Nanjing,China,2009:542-547. [3]BOULGOURIS N V,HATZINAKOS D,PLATANIOTIS K [l4]徐旦华,幸嘉,李松毅,等.Zernike矩的快速算法[J].东 N.Gait recognition:a challenging signal processing technol- 南大学学报:自然科学版,2002,32(2):189-192. ogy for biometric identification[J].IEEE Signal Processing XU Danhua,GU Jia,LI Songyi,et al.Fast algorithm for Magazine,2005,22(6):78-90. computation of Zernike moments[J].Journal of Southeast [4]贲晛烨,徐森,王科俊.行人步态的特征表达及识别综述 University:Natural Science Edition,2002,32(2):189- [J].模式识别与人工智能,2012,25(1):71-81 192. BEN Xianye,XU Sen,WANG Kejun.Review on pedestrian [l5]叶斌,彭嘉雄.伪Zernike矩不变性分析及其改进研究 gait feature expression and recognition[J].Pattern Recogni- [J].中国图象图形学报,2003,8(3):246-252 tion and Artificial Intelligence,2012,25(1):71-81. YE BIN,PENG Jiaxiong.Improvement and invariance a- [5]BOULGOURIS N V,PLATANIOTIS K N,HATZINAKOS nalysis of pseudo-Zernike moments[J].Journal of Image D.Gait recognition using dynamic time warping[C]//Pro- and Graphics,2003,8(3):246-252. ceedings of the Fourth IEEE International Symposium on [16]秦开怀,王海颖,郑辑涛.一种基于Hough变换的圆和矩 Multimedia Signal Processing.Siena,Italy,2004:263-266. 形的快速检测方法[J].中国图象图形学报,2010,15 [6]BOULGOURIS N V,PLATANIOTIS K N,HATZINAKOS (1):109-115. D.Gait recognition using linear time normalization[J].Pat- QIN Kaihuai,WANG Haiying,ZHENG Jitao.A unified tern Recognition,2006,39(5):969-979. approach based on Hough transform for quick detection of [7]TIAN Guangjian,HU Fuyuan,ZHAO Rongchun.Gait rec- circles and rectangles[J].Journal of Image and Graphics, ognition based on Fourier descriptors[C]//Proceedings of 2010,15(1):109-115. 2004 International Symposium on Intelligent Multimedia, [17]KADYROV A,PETROU M.The trace transform and its Video and Speech Processing.Hong Kong,China,2004: applications[J].IEEE Transactions on Pattern Analysis 29-32. and Machine Intelligence,2001,23(8):811-828. [8 HONG S,LEE H,NIZAMI I F,et al.A new gait represen- [18]高上凯.医学成像系统[M].北京:清华大学出版社5 结束语 本文提出了一种基于线性插值的矩阵步态识别 算法框架ꎬ并将投影特征、Hough 变换特征、Trace 变 换特征和 Fan ̄Beam 映射特征在 CASIA(B) 步态库 上验证了该框架在步态识别问题中的有效性.Trace 变换在 Radon 变换的基础上加上欧氏距离的度量作 用ꎬ识别率略有提高ꎬ特征维数相差不多ꎻ角度投影 特征表达形式简单ꎬ但识别率不高ꎻ在连续的情况 下ꎬHough 变换可被看作是 Radon 变换的特例ꎬ但识 别率不如 Radon 变换ꎻFan ̄Beam 映射的识别率最 高ꎬ但是以特征维数多作为代价.最后还得出该框架 的本质就是一种权值不同的能量形式的结论.下一 步的工作重点将考虑步态的张量表达形式ꎬ寻找更 有效的特征提取与选择方法. 参考文献: [1]贲晛烨.基于人体运动分析的步态识别算法研究[D]. 哈 尔滨:哈尔滨工程大学ꎬ 2010. [2]NIXON M Sꎬ CARTER J Nꎬ SHUTLER J Dꎬ et al. New ad ̄ vances in automatic gait recognition[J]. Information Securi ̄ ty Technical Reportꎬ 2002ꎬ 7(4): 23 ̄35. [3] BOULGOURIS N Vꎬ HATZINAKOS Dꎬ PLATANIOTIS K N. Gait recognition: a challenging signal processing technol ̄ ogy for biometric identification[ J]. IEEE Signal Processing Magazineꎬ 2005ꎬ 22(6): 78 ̄90. [4]贲晛烨ꎬ徐森ꎬ王科俊.行人步态的特征表达及识别综述 [J].模式识别与人工智能ꎬ 2012ꎬ 25(1): 71 ̄81. BEN Xianyeꎬ XU Senꎬ WANG Kejun. Review on pedestrian gait feature expression and recognition[J]. Pattern Recogni ̄ tion and Artificial Intelligenceꎬ 2012ꎬ 25(1): 71 ̄81. [5] BOULGOURIS N Vꎬ PLATANIOTIS K Nꎬ HATZINAKOS D. Gait recognition using dynamic time warping[C] / / Pro ̄ ceedings of the Fourth IEEE International Symposium on Multimedia Signal Processing. Sienaꎬ Italyꎬ 2004: 263 ̄266. [6] BOULGOURIS N Vꎬ PLATANIOTIS K Nꎬ HATZINAKOS D. Gait recognition using linear time normalization[J]. Pat ̄ tern Recognitionꎬ 2006ꎬ 39(5): 969 ̄979. [7]TIAN Guangjianꎬ HU Fuyuanꎬ ZHAO Rongchun. Gait rec ̄ ognition based on Fourier descriptors [ C] / / Proceedings of 2004 International Symposium on Intelligent Multimediaꎬ Video and Speech Processing. Hong Kongꎬ Chinaꎬ 2004: 29 ̄32. [8]HONG Sꎬ LEE Hꎬ NIZAMI I Fꎬ et al. A new gait represen ̄ tation for human identification: mass vector[C] / / Proceed ̄ ings of the 2nd IEEE Conference on Industrial Electronics and Applications. Harbinꎬ China. 2007: 669 ̄673. [9]NANDINI Cꎬ RAVIKUMAR C N. An approach to gait rec ̄ ognition[ C] / / International Symposium on Biometrics and Security Technologies. Islamabadꎬ Pakistanꎬ 2008: 1 ̄3. [10] BENABDELKADER Cꎬ CUTLER Rꎬ DAVIS L. Motion ̄ based recognition of people in EigenGait space[C] / / Pro ̄ ceedings of the Fifth IEEE International Conference on Au ̄ tomatic Face and Gesture Recognition. Washingtonꎬ DCꎬ USAꎬ 2002: 267 ̄272. [11]YU Shiqiꎬ WANG Liangꎬ HUANG Kaiqiꎬ et al. Gait anal ̄ ysis for human identification in frequency domain [ C] / / Proceedings of the 3rd International Conference on Image and Graphics. Hong Kongꎬ Chinaꎬ 2004: 282 ̄285. [12]XU Junhongꎬ CONG Wangꎬ LI Jinꎬ et al. Gait recognition based on key frame and elliptical model[C] / / Proceedings of 2010 IEEE International Conference on Information and Automation. Harbinꎬ Chinaꎬ 2010: 2483 ̄2487. [13]WANG Kejunꎬ BEN Xianyeꎬ ZHAO Yue. Gait period de ̄ tection based on regional characteristics analysis [ C] / / Proceedings of the 2009 Chinese Conference on Pattern Recognitionꎬ and the CJK Joint Workshop on Pattern Rec ̄ ognition. Nanjingꎬ Chinaꎬ 2009: 542 ̄547. [14]徐旦华ꎬ辜嘉ꎬ李松毅ꎬ等.Zernike 矩的快速算法[ J].东 南大学学报:自然科学版ꎬ 2002ꎬ 32(2): 189 ̄192. XU Danhuaꎬ GU Jiaꎬ LI Songyiꎬ et al. Fast algorithm for computation of Zernike moments[ J]. Journal of Southeast University: Natural Science Editionꎬ 2002ꎬ 32( 2): 189 ̄ 192. [15]叶斌ꎬ彭嘉雄.伪 Zernike 矩不变性分析及其改进研究 [J].中国图象图形学报ꎬ 2003ꎬ 8(3): 246 ̄252. YE BINꎬ PENG Jiaxiong. Improvement and invariance a ̄ nalysis of pseudo ̄Zernike moments [ J]. Journal of Image and Graphicsꎬ 2003ꎬ 8(3): 246 ̄252. [16]秦开怀ꎬ王海颍ꎬ郑辑涛.一种基于 Hough 变换的圆和矩 形的快速检测方法[ J].中国图象图形学报ꎬ 2010ꎬ 15 (1): 109 ̄115. QIN Kaihuaiꎬ WANG Haiyingꎬ ZHENG Jitao. A unified approach based on Hough transform for quick detection of circles and rectangles[J]. Journal of Image and Graphicsꎬ 2010ꎬ 15(1): 109 ̄115. [17]KADYROV Aꎬ PETROU M. The trace transform and its applications [ J]. IEEE Transactions on Pattern Analysis and Machine Intelligenceꎬ 2001ꎬ 23(8): 811 ̄828. [18]高上凯. 医学成像系统[ M]. 北京:清华大学出版社ꎬ 􀅰424􀅰 智 能 系 统 学 报 第 8 卷
<<向上翻页向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有