A New Method on Vehicle Behavior Analysis of Intelligent Transportation Monitoring
This paper proposes a new method on the vehicle behavior analysis. Firstly, complete vehicle object detection based from the background of road and achieve object tracking by the Cam-Shift algorithm, and then get the target trajectory of the vehicle. According to the basic behavior of vehicle trajectory features and speed features based on the analysis of the vehicle trajectory, we propose a method for extracting vehicle trajectory features and obtaining the motion feature vector. Finally, the improved voting strategy in the SVM algorithm is proposed to recognize the vehicle movement behavior. Experimental results show that the proposed method could complete the target vehicle behavior analysis, which functions more quickly and accurately compared with existing behavior-based vehicle trajectory analysis.
KeywordsBehavior analysis Cam-shift Trajectory feature representation SVM
- Rodríguez AS, González Castolo JC, Suárez ÓD (2013) Time viewer, a tool for visualizing the problems of the background subtraction. Image and video technology-6th pacific-rim symposium, vol 8333, pp 372–384, PSIVT Guanajuato, MexicoGoogle Scholar
- Wang L, Wang L, Wen M et al (2007) Background subtraction using incremental subspace learning. In: 2007 IEEE international conference on image processing, ICIP 2007 Proceedings, vol 5, pp 45–48, San Antonio, USAGoogle Scholar