在分析了目标、背景与噪声的基本特性后,提出了一种时空域累积差分融合的红外小目标检测算法。在空域上,对原始图像采用高通与低通滤波并差分突出目标信息。在时域上利用目标的运动特性进行多帧累积和差分处理,并对空域和时域处理的图像进行自适应分段线性变换增强图像,最后将增强处理的图像根据相应的准则融合采取自适应阈值分割处理。实验结果表明文中提出的算法能有效检测出低信噪比环境下的运动小目标。

A fusion detection algorithm of small infrared target based on spatial-temporal accumulative difference was proposed. As for air-space, combined HPF and LPF difference is applied to original infrared image to extrude information of target. As for time domain, based on target characters, frame accumulation and difference treatment are carried out, then adaptive subsection liner transform is applied to the spatial-temporal images for image enhancement. Last, according to the rule to fuse the images and take adaptive threshold to segment them. Simulation results indicate the algorithm presented in this paper can effectively detect target with low SNR.