基于MATLAB的散斑图像处理方法

Speckle image processing method based on MATLAB

利用MATLAB工具箱可对散斑图像进行图像处理和图像信息提取,包括散斑图像的叠加、对条纹图像进行增强、滤波、二值化、开闭操作和骨架提取等处理。在物体面内微小位移散斑图像的处理过程中,针对增强和滤波操作,分别采用了像素增加、对数变换、取反变换、分段性变换和中值滤波、均值滤波、高斯滤波等不同方法。通过对比和分析不同方法的处理结果发现,叠加后的条纹图依次经过线性变换增强、二值化、一定阈值闭操作和条纹细化处理的条纹图像较为连贯清晰,且条纹间距不同时,闭操作的阈值也需随之改变。这种图像处理方法可以为后续物体实际面内微小位移的测量和物体受热产生热形变的测量提供可靠的理论参考。

The speckle image can be processed and image information be extracted by using MATLAB,including speckle image superposition,fringe image enhancement,filtering,binarization,opening and closing operation and so on.In this paper,different methods such as pixel increase,logarithmic transform,inverse transform,piecewise transform,median filter,mean filter and Gaussian filter are used for fringe processing and extraction in the processing of the speckle image of object in-plane displacement.The research results show that,the fringe patterns obtained by different processing methods are slightly different in clarity and noise distribution.The fringe images obtained by linear transformation enhancement,binarization,closed operation with a certain threshold and fringe thinning are more coherent and clear.It is found that when the fringe spacing is different,the threshold of closed operation also needs to be changed.This processing method provides a reliable theoretical reference for the subsequent actual displacement measurement.