A Traffic Parameter Detection Based on Computer Vision

摘要：

Aiming at the characteristics of the road video traffic flow, mainly study the light change on the impact of the moving vechicle recognition system, and analysis the changes of the spatial information, which raised a moving object extraction algorithm based on background adaptive updating. The algorithm model fully considered the graded influence which produced by light field to video image sequences, realized self-adapted extraction to background, and overcomed the problem which was difficult to extracte vehicle target by using fixed background. With the algorithm of projection principle and two-dimensional combined statistics, the problem of traffic parameter extraction in complex background was solved. Experimental results show that the algorithm is effective for traffic parameters extraction, and have very strong robustness in the conditions of the big space, complex background and different illumination.These algorithms can truely reflect the prevailing traffic condition, which also provide a new detection method for intelligent transportation systems.

关键词： projection principle two-dimensional combined statistics vehicle detection traffic parameters extraction

作者: Jianxin Chen Qing Liu Junqin Huang

作者单位: Key Laboratory of Special Area Highway Engineering of Ministry Education Changan University Xian, Engineering Training Center Xian University of Technology Xian, China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Advanced Computer Control(2011年IEEE第三届高端计算机控制国际会议 ICACC2011)

会议地点: 哈尔滨

会议语种:英文

页码: 591-594

在线出版日期: 2011-01-18（万方平台首次上网日期，不代表论文的发表时间）