在有云层背景的红外序列图像中,检测和跟踪微弱小目标一直是研究的重点。介绍了一种简单可行的在云层背景下检测小目标的方案。算法中,首先进行背景抑制以获得更高的信噪比,接着是用门限分割以得到二值图像,最后依照目标在时间和空间运动的连续性,用一种管线法进行目标识别。实验结果表明算法十分有效。 In infrared sequence images with cloud backgrounds, detecting and tracking weak targets has been the focus of research. A simple and feasible scheme for detecting small targets in the background of clouds is introduced. In the algorithm, the background suppression is first used to obtain a higher signal-to-noise ratio, then the threshold is used to obtain the binary image. Finally, a pipeline method is used to identify the target according to the continuity of the target in time and space. Experimental results show that the algorithm is very effective.