摘要
研究了红外图像运动小目标检测及跟踪问题 .首先利用向量小波的优良性质对运动图像进行预处理 ,然后经Fisher算法分割出可疑目标 ,再根据红外小目标的运动特性 ,采取邻域分析并结合对运动目标在速度平面上的拟合直线分析的方法 ,可以准确提取和跟踪运动小目标 .模拟实验表明了算法是有效的 。
An algorithm for detecting and tracking the small infrared moving target was suggested. By utilizing the excellent quality of multi-wavelet, the infrared image was preprocessed with the proper pre-filter, and segmented by the Fisher algorithm and possible targets were segmented from the image. According to the moving characteristics of small infrared target, a method for extracting and tracking small moving target was adopted, using the neighboring region analysis and analyzing line fitting of moving target in the speed plane. The simulation experiments indicated that the whole algorithm was quite effective and especially suitable for the cases in which a target was partly occluded or missing occasionally.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2004年第1期67-69,共3页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国防科技预研基金资助项目
关键词
向量小波变换
图像分割
红外运动小目标
邻域分析
目标检测和跟踪
multi-wavelet transform
image segmentation
small infrared moving target
neighboring region analysis
target detection and tracking
