一种基于运动检测的时域降噪算法及实现

A Temporal Noise Reduction Algorithm and Implementation Based on Motion Detection

基于运动估计技术在时域方向上计算当前帧和参考帧之间的运动轨迹,提出了一种基于"运动检测"的时域视频降噪算法,可根据物体运动强度动态地调整时域滤波的强度.Matlab仿真环境的实验结果表明:采用该降噪算法能有效地去除视频序列中的噪声,并有效避免了运动物体"拖尾"现象的产生.

Based on the rule of motion estimation technology by which motion track is exacted between the current frame and the reference frame in temporal domain,a temporal video noise reduction algorithm based on motion detection was proposed,the filtering strength was dynamically adjusted according to the motion intensity. The experimental result in Matlab simulation environment shows that the noise reduction algorithm can effectively reduce the noise in video session and avoid trailing phenomenon of moving object.