视频序列图像中运动目标检测与阴影去除

被引:4
作者
明英 [1 ,2 ]
蒋晶珏 [3 ]
机构
[1] 武警武汉指挥学院
[2] 清华大学电子工程系
[3] 武汉大学计算机学院
关键词
运动目标检测; 背景建模; 阴影; 柯西分布;
D O I
10.13203/j.whugis2008.12.020
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
提出了一种新的基于柯西分布的光照模型(shading model,SM)变化检测方法。在一种快速动态背景图像初始化的基础上,建立了Gaussian统计背景模型;基于使用统计假设检验方法检测变化区域的结果,利用YCbCr颜色空间的亮度、颜色信息,识别和消除视频序列图像中的阴影和反光等。试验表明,该文所提出的方法可以承受整体或局部的、缓慢或突然的背景光线变化,以及由场景背景中小的背景扰动、阴影或反光导致的噪声,可以较精确地检测背景目标,改善了SM方法在较暗区域的目标检测效果。
引用
收藏
页码:1216 / 1220
页数:5
相关论文
共 15 条
[1]  
Change detection and background extraction by linear algebra. Durucan E,Ebrahimi T. Proceedings of the IEEE . 2001
[2]  
Statisticalmodeling of complex backgrounds for foreground object detec-tion. L.Li,W.M.Huang,I.Y.H.Gu,,and Q.Tian. IEEE Trans.on Image Processing . 2004
[3]  
Illumination independent change detection for real world image sequences. Skifstad,K.,Jain,R. Computer Vision . 1989
[4]  
Detection and Tracking of SemanticMoving Objects Based on Cauchy Distribution. Ming Ying. . 2004
[5]  
Bounds on the k-neighborhood for locally uniformly sampled surfaces. Andersson M,Giesen J,Pauly M,et al. Proceedings of the 1st Symposium on Point Based Graphics(SPBG) . 2004
[6]  
Detecting Moving Shadows: Algorithms and Evaluation. Prati,,I Mikic,M Trivedi, et al. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2003
[7]  
Pattern Classification. Duda R O,Hart P E,Stork D G. . 2000
[8]  
Tracking of non-Gaussian clusters in the PETS2001image sequences. Pece A E C. Proceedings of the Second IEEE Workshopon Performance Evaluation in Tracking and Surveillance(PETS’2001) . 2001
[9]  
Non-parametric Model for Background Subtraction. Elgammal A,Harwood D,and Davis L. The 6th European Conference on Computer Vision . 2000
[10]  
Non-parametric Model for Background Subtraction. Elgammal A,Harwood D,and Davis L. The 6th European Conference on Computer Vision . 2000