根据目标、背景干扰和噪声在红外序列图像中的差异,提出了一种基于空间高通滤波和时间域上N帧轨迹积累的运动小目标检测方法。该方法可以在低信噪比的情况下消除红外起伏和随机噪声的影响,有效检测出弱小目标。实验结果表明,采用这种时空混合处理的方法可以得到满意的结果。 According to the difference of target, background interference and noise in infrared sequence images, a small moving target detection method based on spatial high-pass filtering and N-track trajectory accumulation in time domain is proposed. The method can eliminate the influence of infrared undulation and random noise under the condition of low signal-to-noise ratio, and effectively detect weak targets. The experimental results show that satisfactory results can be obtained by using this method of time-space mixing.