|Moving Object Refining in Traffic Monitoring Applications|
|Kunfeng Wang; Qingming Yao; Xin Qiao; Shuming Tangand Fei-Yue Wang
|会议名称||The Proceedings of the 10th IEEE International Conference on Intelligent Transportation Systems
|会议日期||Sept. 30 - Oct. 3, 2007
|会议地点||Seattle, WA, USA
|摘要||Moving object segmentation is an important task in vision-based traﬃc monitoring applications. In traﬃc scenes, various outliers such as sudden illumination changes, moving cast shadows, camera jitter, etc., often cause serious errors in image analysis due to misclassiﬁcation of moving objects. An eﬃcient moving object reﬁning approach is thus expected. In this paper, we address the problem of moving object reﬁning by processing the background subtraction results. In an analytical multi-stage procedure, we remove sudden illumination changesandlocalreﬂectedregionsemployingphotometriccolor invariants, remove moving cast shadows based on a single Gaussian shadow model, uniform-region classiﬁcation, and spatial analysis, and further remove other types of outliers in a postprocessing stage of area ﬁltering and area-to-perimeter test. Experimental results on actual video sequences representative of diﬀerent traﬃc scenes and illuminations are presented. The results illustrate that our approach is eﬃcient when handling widely diﬀerent conditions that can occur.|
Image Colour Analysis
Kunfeng Wang,Qingming Yao,Xin Qiao,et al. Moving Object Refining in Traffic Monitoring Applications[C],2007.