摘要
在交通场景下进行多目标跟踪时,如何正确检测出车辆间的相互遮挡是影响车辆跟踪结果的关键。针对问题,运用投影理论分析交通场景的三维几何投影特征,用长方体投影轮廓模型对车辆进行建模,重构其三维投影轮廓,以进行遮挡的检测和分离。与以往的方法相比,它在估计出的车辆外形轮廓基础上进行遮挡检测,不需要匹配操作,计算量较小,并能解决基于匹配的方法无法对付的初始遮挡问题。用实验验证了该算法的有效性。
In multi - object tracking of traffic scene, how to detect the occlusion is a key problem for vehicle tracking. A novel vehicle contour based method is proposed to deal with this problem. This method firstly extracts the 3 D geometry character from the vehicle images according to the projecting theory, establishes the model of the vehicle with the cuboid project contour, and then reconstructs its 3D project contour, detects and segments the occlusion re- gions. Compared with traditional method, it deals with the problem with less computation because it generates the vehicle contour firstly and no match calculation is needed. Moreover, this method is able to solve the initial occlusion problem, which could not be solved with matching based method. The experimental results show that the proposed method is more efficient.
出处
《计算机仿真》
CSCD
北大核心
2009年第4期297-300,共4页
Computer Simulation
基金
中科院自动化所研究课题基金(A0602)
关键词
遮挡检测
投影轮廓模型
贝叶斯分类
Occlusion detection
Projection contour model
Bayes classification
作者简介
刘学亮（1981-），男（汉族），河北人，硕士研究生，主要研究方向为视频图像处理，智能交通系统。
严捷丰（1978-），男（汉族），江苏人，博士研究生，主要研究方向为视频图像处理，智能交通系统。
周荷琴（1946-），女（汉族），江苏人，教授，博士生导师，主要研究方向为医学图像处理，智能信息处理。
