用于步态识别的多层窗口图像矩

陈实; 马天骏; 黄万红; 高有行

doi:10.3724/SP.J.1146.2007.00247

用于步态识别的多层窗口图像矩

doi: 10.3724/SP.J.1146.2007.00247

基金项目:

宁波市自然科学基金(2008A61011)和陕西省自然科学基金 (2006F48)资助课题

计量
- 文章访问数: 3098
- HTML全文浏览量: 106
- PDF下载量: 1474
- 被引次数: 0
出版历程
- 收稿日期: 2007-08-08
- 修回日期: 2008-08-13
- 刊出日期: 2009-01-19

A Multi-layer Windows Method of Moments for Gait Recognition

摘要

摘要: 该文提出了一种以局部性矩统计量作为步态特征描述的步态识别方法。首先提取行人二值轮廓序列，构造一种基于直方图的轮廓点分布特征检测出步态周期；然后生成彩色步态运动历史图像CGHI描述步态的空间特征和时间信息；继而设计了多层同心矩形窗口分割CGHI，提取出一组矩形环窗口的矩特征量作为步态特征，在此基础上实现了步态识别。在Soton数据库上进行了实验，提出算法的正确识别率可达87.2%，优于现有方法。
- 步态识别;不变矩;多层窗口;历史图像
Abstract: A feature representation method based on multi-layer local moment invariants for gait analysis and recognition applications is developed. The method includes following steps: first, silhouette extraction is performed for each image sequence. Secondly, the gait cycle is detected by a histogram-based approach. Thirdly, a scalarvalued CGHI(Colored Gait History Images) is proposed to describe how human walking is evolved. To improve the recognition rate, the CGHI is decomposed into a sequnce of muti-layer rectangle windows. The moment invariants from the window are used as the gait features and finally used to recognize gait. The correct classification rate of 87.2% is achieved on Soton database, which show that the method outperforms the exist methods.

HTML全文

参考文献(1)

Sarkar S, Phillips P J, and Liu Z, et al.. The humanID gaitchallenge problem: Data sets, performance, and analysis[J].IEEE Trans. on Pattern Analysis and Machine Intelligence.2005, 27(2):162-177[2]Kale A, Sundaresan A, and Rajagopalan A N, et al..Identification of humans using gait. IEEE Trans. on ImageProcessing, 2004, 13(9): 1163-1173.[3]Wang L, Tan T, and Ning H, et al.. Silhouette analysis-basedgait recognition for human identification[J].IEEE Trans. onPattern Analysis and Machine Intelligence.2003, 25(12):1505-1518[4]Mowbray S D and Nixon M S. Automatic gait recognition viafourier descriptors of deformable objects. Proc. 4th Intl Conf.on Audio- and Video-Based Biometric Person Authentication,Guildford, UK, 2003: 566-573.[5]Boyd J. Synchronization of oscillations for machineperception of gaits[J].Computer Vision and ImageUnderstanding.2004, 96(1):35-59[6]He Q and Debrunner C. Individual recognition from periodicactivity using hidden markov models. Proc. IEEE Workshopon Human Motion, Austin, Tex, USA, 2000: 47-52.[7]Shutler J D and Nixon M S. Zernike velocity moments forsequence-based description of moving features[J].Image andVision Computing.2006, 24(4):343-356[8]Hu M-K. Visual pattern recognition by moment invariants.IRE Trans. on Information Theory, 1962, 8(2): 179-187.[9]Lee L and Grimson W E L. Gait analysis for recognition andclassification. Proc. 5th IEEE Intl Conf. on Automatic Faceand Gesture Recognition, Washington, DC, 2002: 148-155.[10]Bobick A F and Davis J W. The recognition of humanmovement using temporal templates[J].IEEE Trans. on PatternAnalysis and Machine Intelligence.2001, 23(3):257-267[11]Sluzek A. Identification and inspection of 2-d objects usingnew moment-based shape descriptors[J].Pattern RecognitionLetters.1995, 16(7):687-697[12]Shutler J D.[J].Grant M G, and Nixon M S, et al.. On a largesequence-based human gait database. Proc. 4th Intl Conf. onRecent Advances in Soft Computing, Nottingham (UK.2002,:-[13]Wagg D K and Nixon M S. On automated model-basedextraction and analysis of gait. Proc. 6th IEEE Intl Conf. onAutomatic Face and Gesture Recognition, Seoul, Korea, 2004:11-16.

施引文献

资源附件(0)

访问统计