摘要
介绍了一种新的基于背景预测的空中红外弱小目标检测方法,以提高对复杂背景预测的准确性,减小云层边缘预测不准确形成的虚警。该方法对云层边缘处的点根据其不同尺度邻域上的亮暗点分布特点进行预测;对非边缘点采用基本背景预测法进行预测,最后经过背景对消,将弱小目标检测出来。实验结果表明,与已报导的其它方法比较,该方法能够更有效地抑制云层边缘引起的虚警。
A novel method to detect aerial weak and small infrared targets is proposed in this paper, which aims to improve the prediction accuracy of complicated background and decrease the false alarm probability caused by the cloud edge. Edge points are predicted with the distribution of bright and dim points in their multi-scale neighborhood; non-edge points are predicted using the basic background prediction method. Finally targets are detected through eliminating background. Experimental results indicate that the algorithm performs better in inhibiting false alarm caused by the cloud edge.
出处
《红外技术》
CSCD
北大核心
2007年第7期404-408,共5页
Infrared Technology
基金
教育部留学回国人员科研启动基金资助项目
西北工业大学"英才计划"资助项目
关键词
背景预测
红外弱小目标
目标检测
background prediction： infrared weak and small targets
target detection
