Detection of Dim Trajectory Based on Multiscale Morphological Reconstruction

摘要：

A method for ballistic trajectory detection based on multiscale morphological reconstruction is presented. The morphological gray reconstruction algorithm is used for background estimation, and the image is binarized to extract suspicious objects in single frame firstly. Then according to the shells agglomeration characteristics, the multiscalemorphological binary reconstruction could be applied for eliminating false targets. Finally, ballistic trajectory is identified based on the shells dynamic characteristics by multi-frame data fusion. Our simulation results show that the multiscale morphological reconstruction has excellent practical value for ballistic trajectory detection in image sequences.

作者: Zhou Ning Zai-ming Li

作者单位: 102 Lab.UESTC Chengdu 610054, China

会议类型: 国际会议

会议名称: 2007年通信、电路与系统国际会议(2007 International Conference on Communications,Circuits and Systems Proceedings)

会议地点: 日本福冈

会议语种:英文

在线出版日期: 2007-07-11（万方平台首次上网日期，不代表论文的发表时间）