摘要
针对现有方法在复杂多变环境下不能很好地检测出运动物体的问题,提出了改进的基于混合高斯模型的背景消减法,并对运动目标进行检测。模型初始化时,提出了一种能准确得到实际背景模型的方法;在模型更新中引入了加速因子和合理性反馈使得模型能更快、更准确地反应真实的背景。实验结果表明,同传统检测方法相比,改进的混合高斯模型方法能有效地消除物体发生运动时产生的拖影,并能很好地检测出运动物体。
In the view of the current complex environment , the detection of moving objects can not be satisfactory with traditional methods. The background of diminishing method based on the improved mixture Gaussian model was presented, and used to detect moving objects. In the initial model stage, a method which could access the accurately model was presented too. In updated model stage, the acceleration factor and reasonable feedback were introduced to reflect the true background. Experimental results show that comparing with the traditional methods, the improve mixture Gaussian model could eliminate moving objects＇ dragging shadow when they moving and detect moving objects well.
出处
《中国图象图形学报》
CSCD
北大核心
2007年第9期1585-1589,共5页
Journal of Image and Graphics
基金
国家自然科学基金项目(40674060)
关键词
智能视觉监控
混合高斯模型
运动目标检测
intelligent video monitoring system, mixture Gaussian model, moving object detecting and recognition
