摘要
提出一种用于运动目标的分割,基于最大统计概率的自适应背景模型,采用简单的背景重建方法,用于维护背景以适应场景的动态变化.利用阴影区域亮度和色调的特点,在HSV(Hue Sataration Value)空间消除运动阴影,使得运动目标的分割更为准确.为了客观的评价所提出的阴影检测算法的性能,引入一种量化的方法,对不同光照和环境条件视频的实验结果及量化分析表明,方法是有效的.
A method to divide the motion target by using an adaptive background model based on ,maximum statistical probability is proposed in this paper. Using a simple method of background recomstruetion and background maintenanee to adapt the scene changed. Moving cast shadows mostly exhibit a challenge for accurate moving targets detection; the problem is addressed in this paper by exploiting HSV (Hue Sataration Value) color information. Furthermore, a quantitative method is introduced to evaluate the algorithm on a benchmark suite of indoor and outdoor video sequences. The experimental results are given and the performance of the algorithm is effect.
出处
《华侨大学学报(自然科学版)》
CAS
北大核心
2007年第1期30-33,共4页
Journal of Huaqiao University(Natural Science)
关键词
运动检测
背景减除
HSV彩色空间
阴影消除
motion detection, background subtraction, HSV color space, shadow elimination