低分辨率目标的检测与实现

提出以柯西分布作为背景剔除时图像像素比值的统计分布模型，并融合单个像素点和邻近像素点所蕴涵的时空信息，实现了对场景变化自适应的背景图像比值的建模，应用假设检验方法，通过背景剔除(background subtraction)实现了对低分辨率目标具有鲁棒性的检测。最后的实验表明，该文提供的算法可以抗背景中全局或局部光照的渐变和突变，可以有效地抑制背景中活动物体和阴影的杂波干扰，能够适应下雨的恶劣天气。

A Detecting Algorithm for Low Resolution Objects

An algorithm based on a Cauchy distribution statistical model for the purpose of background modeling and subtracting is presented. This approach takes the advantages of the spatial-temporal information on both each single pixel and the region around a pixel to distinguish the changing pixels corresponding to moving objects from one corresponding to background image. The paper also discusses the changes in background scene in detail. At last, a robust background subtracting based moving object detecting approach being invariant or adapting to the changes in background scene is acquired by hypothesis test. Experimental results demonstrate the proposed algorithms can tolerate the whole or local sudden or slow change in illumination, filter clutter noises caused by small motion in background scene, and adapt to rain.