Please wait a minute...
浙江大学学报(工学版)
J4  2010, Vol. 44 Issue (3): 458-462    DOI: 10.3785/j.issn.1008973X.2010.03.008
自动化技术、计算机技术     
视频目标定位的减法聚类改进算法
孙志海1, 孔万增2, 朱善安1
1. 浙江大学 电气工程学院, 浙江 杭州 310027；2. 杭州电子科技大学 计算机学院, 浙江 杭州 310018）
Improved algorithm of subtractive clustering for object location  in video sequences
 全文: PDF 
摘要：

针对减法聚类算法用于视频运动目标定位时存在的不足,提出了6点优化技术,即采用不同维度的邻域半径改进了原算法采用固定邻域半径的不足；修正目标邻域半径的取值,改进了原算法无法准确描述不同尺度目标定位效果的问题；引入下采样技术、改变密度值函数及构造网格重定义数据集3种方法以提高算法的定位效率；引入模糊隶属度对视频运动目标做进一步精确定位,解决了原算法无法精确定位的问题.实验结果表明,改进后的定位算法可获得更好的定位效果.
关键词： 减法聚类;  视频目标定位;  密度值函数;  下采样    
Abstract:

 For the weakness appeared when using subtractive clustering for moving object location in video sequences, six optimization techniques of subtractive clustering were proposed. Different effective radius in different dimensions was considered and corresponding extended method was proposed. Techniques of downsampling, choosing density function and redefining clustering data set based on certain grid method were put forward. A much more accurate and further segmentation method based on fuzzy membership was also presented. Comparison with the conventional method showed the superiority of the proposed subtractive clustering method to different data sets.
Key words: subtractive clustering    video object location    density function    downsampling
出版日期: 2010-04-01
:  TP391.41  
通讯作者: 朱善安,男,教授,博导.     E-mail: zsa@cee.zju.edu.cn
作者简介: 孙志海(1981—), 男, 福建漳州人, 博士生, 主要从事视频运动目标检测研究.
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　　
孙志海
孔万增
朱善安

引用本文:

孙志海, 孔万增, 朱善安. 视频目标定位的减法聚类改进算法[J]. J4, 2010, 44(3): 458-462.

SUN Zhi-Hai, KONG Mo-Ceng, SHU Shan-An. Improved algorithm of subtractive clustering for object location  in video sequences. J4, 2010, 44(3): 458-462.

链接本文:

http://www.zjujournals.com/xueshu/eng/CN/10.3785/j.issn.1008973X.2010.03.008    或    http://www.zjujournals.com/xueshu/eng/CN/Y2010/V44/I3/458

［1］ YANG T, Li S, PAN Q, et al. Realtime and accurate segmentation of moving objects in dynamic scene［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2004, 14(6): 796812.［2］ 明英,蒋晶珏.视觉监视中基于柯西分布的统计变化检测［J］.中国图象图形学报,2008,13(2): 328334.
MING Ying, JIANG Jingjue. Cauchy distribution based on statistical change detection for visual surveillance［J］. Journal of Image and Graphics,2008,13(2): 328334.
［3］ 徐东彬,刘昌平,黄磊.基于概率统计自适应背景模型的运动目标检测方法［J］.中国图象图形学报, 2008,13(2): 351358.
XU Dongbin, LIU Changping, HUANG Lei. Adaptive backgroudn model for motion detection based on statistic of probability［J］. Journal of Image and Graphics,2008,13(2): 351358.
［4］ CHIU S. Fuzzy model identification based on cluster estimation［J］. Journal of Intelligent and Fuzzy Systems, 1994, 2(3): 267278.
［5］ SATHIT N, PERAPHON S, WILLIAM R. Fuzzy subtractive clustering based indexing approach for software components classification［J］. Journal of Computer and Information Science, 2004, 5(1): 6372.
［6］ PEREIRA C, DOURADO A. Application of a neurofuzzy network with support vector learning to a solar power plant［M］. Almeria, Espanha: 2nd IHP Workshop,2002.
［7］ KIM D, LEE K, LEE D, et al. A kernelbased subtractive clustering method［J］. Pattern Recognition Letters, 2005, 26(7): 879891.
［8］ KONG Wanzeng, ZHU Shanan. Multiface detection based on downsampling and modified subtractive clustering for color images［J］. Journal of Zhejiang University: SCIENCE A, 2007,8(1): 7278.
［9］ YAGER R, FILEV D. Approximate clustering via the mountain method［J］. IEEE Transactions on Systems, Man, and Cybernetics, 1994, 24: 12791284.
［10］ JANG J, SUN C, MIZUTANI E. Neurofuzzy and soft computing ［M］. ［S.l.］: Prentice Hall, 1997.
[1] 尤海辉, 马增益, 唐义军, 王月兰, 郑林, 俞钟, 吉澄军. 循环流化床入炉垃圾热值软测量[J]. 浙江大学学报(工学版), 2017, 51(6): 1163-1172.
[2] 焦竹青, 邵金涛, 徐保国. 非下采样Contourlet变换域多聚焦图像融合方法[J]. J4, 2010, 44(7): 1333-1337.