摘要：

针对传统目标检测算法实时性较差且在面对复杂云层干扰时虚警率高的问题,提出一种基于人类视觉系统的小目标快速检测算法。利用局部对比度测量方法计算候选目标,根据拉普拉斯高斯尺度空间理论,计算候选目标处的多尺度滤波响应,进而通过自适应阈值分割获取真实目标。实验结果表明,该算法的检测率高达97%,虚警率低于3%,且能够在5 ms内完成目标位置计算。

关键词: 人类视觉系统, 局部对比度测量, 尺度空间, 多尺度响应, 自适应阈值, 红外小目标

Abstract:

Aiming at the problems that the traditional target detection algorithm has poor real-time performance and high false alarm rate in the face of complex cloud interference,a small target fast detection algorithm based on Human Visual System(HVS) is proposed.The candidate target is calculated using the Local Contrast Measure(LCM) method.According to the Laplace Gaussian scale space theory,the multi-scale filter response at the candidate target is calculated,and then the real target is obtained by adaptive threshold segmentation.Experimental results show that the proposed algorithm has a detection rate of up to 97%,a false alarm rate of less than 3%,and the calculation of the target position can be completed within 5 ms.

Key words: Human Visual System(HVS), Local Contrast Measure(LCM), scale space, multi-scale response, adaptive threshold, infrared small target

中图分类号: 

• TP391