摘要
采用三帧差分法检测静止背景下视频序列中的运动目标,得到运动目标的二值图像。并利用矩方法对运动目标的几何特征进行了研究,在此基础上提出了一种运动目标的质心和最小外接矩形的估算算法,并对该算法进行了改进,使之适合噪声干扰比较大的情况。该算法可实现运动目标进行快速准确定位和区域估算,实验表明该算法的时间复杂度和空间复杂度低,效果良好,且具有很好的鲁棒性。
The moving object in video sequences of still background is detected by using three-frame-difference algorithm,thus to achieve the two-values image of moving object.The geometrical characteristics of the moving object based on moment method are studied,a centroid mer-minimum enclosing rectangle estimation algorithm is proposed,and then this algorithm is improved for adapting to the situation with much interfering noise.This algorithm could implement quick positioning of the moving object and object region estimation,and the experiment proves that the algorithm is fairly low in time and space complexity,excellent in experiment result,and strong in robustness.
出处
《通信技术》
2011年第5期119-121,145,共4页
Communications Technology
关键词
运动目标检测
三帧差分
质心
最小外接矩形
moving object detection
three-frame-difference
centroid
mer-minimum enclosing rectangle
