就如何检测复杂背景下低信噪比的运动小目标展开讨论,提出了用空间高通滤波方法改善图像质量,达到抑制背景噪声,增强小目标的效果,随后用似然比检测理论进行目标的初步分离,接着采用邻域判决的方法实现运动目标的进一步分离,最后用图像流分析法进行目标的最终检测。实验结果表明,该算法能够对小目标甚至是点目标的运动进行可靠的检测 This paper discusses how to detect moving targets with low signal-to-noise ratio in complex background, and proposes the idea of ​​using spatial high-pass filtering to improve the image quality, to suppress the background noise and enhance the small target, and then to use the likelihood ratio detection theory Separation, and then use the method of neighborhood decision-making to achieve the further separation of moving targets, and finally the image flow analysis method for the final target detection. The experimental results show that the algorithm can detect the movement of small targets or even point targets reliably