基于自适应背景估计的视频运动对象提取
Moving Video Object Extraction Based on Adaptive Background Estimation
-
摘要: 提出了一种结合帧间差和减背景的运动目标提取算法,在使用差值图像的高阶统计量检测运动对象的同时估计可靠的背景模型,并通过背景选择性替换机制解决局部背景区域变化的问题。算法需调整的参数少,易于实现。实验结果表明算法具有良好的性能。Abstract: A moving object extraction algorithm that incorporates frame difference with background subtraction is proposed.The algorithm constructs a reliable background model while detecting moving object using higher order statistics,and accommodates background changes such as background objects displacement via selective background substitution.Experiment result is given to show the performance of the algorithm.