计算机微视觉微运动测量系统图像噪声分析

Image noise analysis of micro-motion measurement based on computer micro-vision

对计算机微视觉微运动测量系统采集的图像噪声及处理方法进行了研究。分析了各种在微运动测量时引起测量误差的图像噪声。在此基础上,通过对一幅典型的由计算机微视觉微运动测量系统采集的MEMS图像进行分析,给出了减少这些噪声的常用方法。分析表明了对于这种微结构图像以下噪声处理方法的有效性:首先采用中值滤波器减少散粒噪声以及采用一个低通滤波器减少高频噪声,然后采用两点修正法可以减少固定图像噪声;此外,在采集图像时适当增加曝光时间可以减少机械噪声的影响。

Image noises and the processing approaches are studied, which acquired by computer micro-vision system for micro-motion measurement. The various image noises that can cause measurement error are analyzed, On the basis of the analysis, the methods of reducing noises are presented by analyzing the typical MEMS image acquired by computer micro-vision system for micro-motion measurement. The analysis shows the validity of the processing approach as follow for microstructure images acquired by this system.＂ firstly, the shot noise can be reduced using a median filter, and the high frequency noise can be decreased using a low passfilter; secondly, the effect of fixed-pattern noise can be reduced using a two-point correction technique; additionally, the approach of increasing the exposure time can be used to minimize the effect of the mechanical noise.