针对机器人运动物体检测中差图像的噪声问题 ,提出了基于噪声概率分布的通用阈值选择方法。在该算法中 ,先判断相应区域灰度分布的一致与否。若分布不一致 ,再取其差图像。实验结果证明 ,这种方法能有效地消除差图像的噪声 Aiming at the problem of noise in poor images of moving objects in robots, a general threshold selection method based on noise probability distribution is proposed. In the algorithm, first determine whether the gray level distribution of the corresponding region or not. If the distribution is inconsistent, then take the difference image. Experimental results show that this method can effectively eliminate the bad image noise