提出了一种基于数字信号处理器(DSP)的红外弱小目标搜索高速实时算法。算法主要针对天空中不规则云团在预处理过程中残留的边缘而设计，是在背景预测算法基础上改进的一种基于陷波滤波器的搜索算法。在DSP实时处理平台上实现并经实际场景试验验证，该算法能有效削弱残留边缘的影响，使最远探测距离指标上升至前向迎头26 km，实时处理速度达到75 Hz，满足系统要求。采用信噪比增益指标对算法进行了评价，结果显示，该算法能够显著提高图像信噪比，利于检测出目标。

A high efficiency algorithm for detecting infrared dim target based on digital signal processor (DSP) is proposed. It is designed especially for the cloud edge left after the image preprocessing of irregular clouds, and it is a searching algorithm based on improved notch filter after the background predicting. A DSP-based real-time image processing platform has been used to implement and test the algorithm in real scenarios. It is proved that the algorithm can restrain the negative effect of the residual cloud edges effectively. The detection distance reaches 26 km and the real-time processing speed can reach 75 Hz. The algorithm improves the signal-to-noise ratio of images significantly, which is beneficial to target detection.