摘要
提出了一种红外图像序列中运动弱小目标检测的新方法。该方法在时间剖面上采用数学形态学滤波和中值滤波来抑制杂波背景干扰,并对去除背景后的图像进行分割,然后利用序列图像中目标运动的连续性和轨迹的一致性来检测出真正的目标。利用实测数据进行了仿真,实验结果表明了该方法的有效性。
A new detection method for moving dim small targets in infrared image sequences is presented. Firstly, the proposed method suppresses the background clutter with mathematical morphology and median filtering in temporal profile of image sequences, then segments the candidate target from background. Finally, the continuity of moving and the consistency of track of the target in sequential images are used to identify the real target. The experimental result shows the effectiveness of the proposed method.
出处
《红外与激光工程》
EI
CSCD
北大核心
2004年第3期307-310,共4页
Infrared and Laser Engineering
关键词
弱小目标检测
红外图像序列
数学形态学
Dim small target detection
Infrared image sequence
Mathematical morphology
