基于颜色差分直方图的运动目标提取  

Moving object extraction based on color difference histogram

作　　者：康登飞 刘锋[1] 曾连荪[1] 

机构地区：[1]上海海事大学信息工程学院,上海201306

出　　处：《微型机与应用》2017年第22期53-57,共5页Microcomputer & Its Applications

摘　　要：为了降低动态变化的背景对运动目标提取的干扰,使用颜色差分直方的方法来降低由于背景不稳定、光照强度变化造成的错误。首先计算一个位置点所在的小邻域内像素之间的差值。其次,使用高斯隶属度函数的模糊关系做颜色差分。最后,提出一种新的模糊颜色差分直方图,即采用模糊C-means聚类和颜色差分结合的方法。模糊C-means聚类聚类算法在颜色差分直方图中的使用可以降低计算的直方图的维数较大的复杂性,也降低了背景因强度变化而产生的移动假象或背景照明度的变化而产生的影响。随机选择两个视频帧集合来测试该方法。仿真结果表明,该算法在一些复杂的视频序列场景中能够表现出较好的性能和结果。In order to reduce the background of dynamic changes on the moving target extraction,this paper uses color difference histogram（ CDH） method to reduce the error,which is caused by the instability background,and changes in light intensity. Firstly,it calculates the difference between apixel and its neighbors in a small neighborhood of a location point. Secondly,the color difference is done fuizzified with a Gaussian membership function. Finally,a new method named fuzzy color difference histogram is proposed,which is based on the combination of fuzzy C-means clustering and color difference. The use of FCM clustering algorithm can reduce the complexity of the computation of the dimension of the histogram,and also reduce the impact of changes in the intensity of the background caused by changes in the intensity of the moving artifacts or changes in the background lighting. Two video frame sets are selected randomly to test the method. Simulation results show that this algorithm can show better performance and results in some complex video sequences.

关 键 词：背景减除法 颜色差分直方图 模糊C-均值聚类 目标提取 

分 类 号：TN911.4[电子电信—通信与信息系统]