Please wait a minute...
浙江大学学报(工学版)
计算机科学技术     
基于人体模型和超像素的黏连人群分割方法
蔡丹平, 于慧敏
浙江大学 信息与电子工程学系,浙江 杭州 310027
Overlapping crowd segmentation based on human model and superpixels
CAI Dan-ping, YU Hui-min
Department of  Information Science and Electronic  Engineering,Zhejiang University, Hangzhou 310027, China
 全文: PDF(2453 KB)   HTML
摘要:

 针对监控视频中拥挤人群的分割问题,提出一种对黏连人群进行分割的新方法.该方法利用投影法和Hough圆检测方法对目标的头部进行检测,利用卡尔曼滤波跟踪算法实现遮挡情况下的人群头部粗定位;采用人体模型实现行人的粗分割;对目标前景进行超像素分割,基于相邻像素块之间的颜色相似程度和与人体模型的形状匹配程度构建一个加权图模型,通过求解最优路径的方法得到黏连目标的最优分割边界.实验表明:该方法能有效解决黏连人群的分割问题,且能够精确地提取出完整的人体轮廓.

Abstract:

Aiming at the problem of crowd segmentation for the video surveillance, a new method of segmenting individual humans in overlapping situations was proposed. In this method, vertical projection histograms and Hough circle transformation were used to detect human’s heads, Kalman filter was used to help locate the positions of the heads when occlusion occurs, human shape models was used to segment pedestrian roughly, the foreground area was segmented into several superpixels, and the best segmentation boundary of the overlapping crowd is defined by the optimal path with a weighted graph model based on the dissimilarity between adjacent regions’ color and the degree of mismatching of a human model. Experimental results show that this method can segment the overlapping crowd effectively and extract the human body boundary precisely.

出版日期: 2015-04-01
:  TP 394.1  
基金资助:

国家“973”重点基础研究发展规划资助项目(2012CB316400).

通讯作者: 于慧敏,男,教授,博导.     E-mail: yhm2005@zju.edu.cn
作者简介: 蔡丹平(1988—),女,硕士生,从事视频/图像处理与分析研究.E-mail: cdp4520@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

蔡丹平, 于慧敏. 基于人体模型和超像素的黏连人群分割方法[J]. 浙江大学学报(工学版), 10.3785/j.issn.1008-973X.2014.06.006.

CAI Dan-ping, YU Hui-min. Overlapping crowd segmentation based on human model and superpixels. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 10.3785/j.issn.1008-973X.2014.06.006.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2014.06.006        http://www.zjujournals.com/eng/CN/Y2014/V48/I6/1004

[1] HU W, HU M, ZHOU X, et al. Principal axis-based correspondence between multiple cameras for people tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2006, 28(4): 663-671.
[2] ZHAO T, NEVATIA R. Bayesian human segmentation in crowded situations[C]∥IEEE Conference on Computer Vision and Pattern Recognition. \[S.l.\]: IEEE, 2003, 2: II45966.
[3] 邓颖娜,朱虹,刘薇.基于贝叶斯模型的相机间人群目标识别[J].中国图像图形学报,2009,14(9): 1750-1755.
DENG Ying-na, ZHU Hong, LIU Wei. Bayesian human recognition across multiple cameras in crowded situations[J]. Journal of Image and Graphics, 2009, 14(9): 17501755.[4] MITTAL A, DAVIS L. M2tracker: a multi-view approach to segmenting and tracking people in a cluttered scene[J]. International Journal of Computer Vision, 2003, 51(3): 189-203.
[5] GONEID A, GINDI S, SEWISY A. A method for the Hough transform detection of circles and ellipses using a one-dimensional array [C]∥ Proceedings of IEEE International Conference on Systems, Man, and Cybernetics. Piscataway, USA: IEEE, 1997: 3154-3157.
[6] KALMAN R. A new approach to linear filtering and prediction problems[J]. Journal of Basic Engineering, 1960, 82D: 35-45.
[7] FELZENSZWALB P, HUNTTENLOCHER D. Efficient graph-based image segmentation[J]. International Journal of Computer Vision, 2004, 59(2): 167-181.
[8] MORI G. Guiding model search using segmentation[C]∥  IEEE International Conference on Computer Vision. \[S.l.\]: IEEE, 2005: 1417-1423.
[9] ACHANTA R, SHAJI A, SMITH K, et al. Slic superpixels [R]. Lausanne: Ecole Polytechnique F_edrale de Lausanne(EPFL), 2010.
[10] LEVINSHTEIN A, STERE A, KUTULAKOS K, et al. Turbopixels: fast superpixels using geometric flows[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12): 2290-2297.
[11] BELONGIE S, MALIK J, PUZICHA J. Shape context: a new descriptor for shape matching and object recognition[C]∥ Proceedings. of NIPS’00. Denver, USA: [s.n], 2000.

[1] 贾松敏,卢迎彬,王丽佳,李秀智,徐涛. 分层特征移动机器人行人跟踪[J]. 浙江大学学报(工学版), 2016, 50(9): 1677-1683.
[2] 李迪, 陈向坚, 续志军, 白越. 模糊神经网络在机载相机稳像中的应用[J]. J4, 2012, 46(8): 1540-1545.