登录 | 注册 | 退出 | 公司首页 | 繁体中文 | 满意度调查
综合馆
一种基于Boosting判别模型的运动阴影检测方法
  • 摘要

    在视频处理中,由于运动阴影具有与运动前景相同的特性,当在提取前景时,会误把阴影检测为前景.特别是当阴影和其它前景发生粘连时,这可能会严重地影响跟踪、识别等后续处理.该文提出了一种用于运动阴影检测的Boosting判别模型.这种方法先利用Boosting在不同的特征空间来区分前景和阴影,然后在判别随机场(DRFs)中结合前景和阴影的时空一致性,实现对前景和阴影的分割.首先,差分前图像与背景图像得到颜色不变子空间和纹理不变子空间;然后在这两个子空间上应用Boosting来区分前景和阴影;最后利用前景和阴影的时空一致性,在判别随机场中通过图分割的方法准确地分割前景和阴影.实验结果表明,无论是在室内场景,还是在室外场景,该文的方法要好于传统的方法.

  • 作者

    查宇飞  楚瀛  王勋  马时平  毕笃彦  ZHA Yu-Fei  CHU Ying  WANG Xun  MA Shi-Ping  BI Du-Yan 

  • 作者单位

    空军工程大学工程学院信号与信息处理实验室,西安,710038/华中科技大学图像识别与人工智能研究所图像信息处理与智能控制教育部重点实验室,武汉,410074

  • 刊期

    2007年8期 ISTIC EI PKU

  • 关键词

    阴影检测  Boosting  判别随机场  图分割 

参考文献
  • [1] Stander J.;Mech R.. Detection of moving cast shadows for object segmentation. IEEE transactions on multimedia, 1999,1
  • [2] Kolmogorov V.;Zabin R.. What energy functions can be minimized via graph cuts?. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004,2
  • [3] Yang Wang;Kia-Fock Loe;Jian-Kang Wu. A dynamic conditional random field model for foreground and shadow segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006,2
  • [4] Ahmed Elgammal;Ramani Duraiswami;David Harwood;Larry D. Davis. Background and foreground modeling using nonparametric kernel density estimation for visual surveillance. Proceedings of the IEEE, 2002,7
  • [5] Horprasert T;Harwood D;Davis L. A statistical approach for real-time robust background subtraction and shadow detection. Kerkyra,Greece, 1999
  • [6] Zhao T;Nevatia R. Tracking multiple humans in complex situations. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004,09
  • [7] Boykov Y;Veksler O;Zabih R. Fast approximate energy minimization via graph cuts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001,11
  • [8] Boykov Y;Kolmogorov V. An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004,09
  • [9] Klinker G;Shafer A;Kanada T. A physical approach to color image understanding. International Journal of Computer Vision, 1990,01
  • [10] Kumar S;Hebert M. Discriminative random fields:A discriminative framework for contextual interaction in classification. Nice,France, 2003
  • [11] Schapire R. The Boosting approach to machine learning:An overview. Berkeley,California, 2001
  • [12] Freund Y;Schapire R. A decision-theoretic generalization of on-line learning and an application to Boosting. Journal of Computer & System Sciences, 1997,01
  • [13] Mikic I;Cosman P C;Kogut G T;Trivedi M M. Moving shadow and object detection in traffic scenes. Los Alamitos,California, 2000
  • [14] Porikli F;Thornton J. Shadow flow:A recursive method to learn moving cast shadow. 北京, 2005
  • [15] Cucchiara R;Grana C;Piccardi M. Detecting moving objects,ghosts and shadow in video streams. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,10
  • [16] Salvador E;Cavallaro A;Ebrahimi T. Cast shadow segmenration using invariant color features. COMPUTER VISION AND IMAGE UNDERSTANDING, 2004,02
  • [17] Nadimi S;Bhanu B. Physical models for moving shadow and object detection in video. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004,08
查看更多︾
相似文献 查看更多>>
3.95.131.208