岩土材料微结构图像预处理方法

In order to acquire clear images of geomaterial’s microstructure, clear parts of the same image array are extracted and recombined by use of wavelet multi-resolution technology, thus complete and clear images are acquired; In order to extract completely characteristic parameters of geomaterial’s microstructure, it is done a comparative analysis whether three kind of frequently-used threshold image segmentation are applicable to process images of geomaterial microstructure. The study’s findings indicate: recombining images of geomaterial’s microstructure by use of wavelet multi-resolution technology can improve image signal-to-noise ratio, let up gray scale variance and optimize images of geomaterial’s microstructure; if image segmentation of average gray scale and threshold automatically extracted according to the maximum deviation are used to transform gray image of geomaterial’s microstructure, images can be segmented with high quality and characteristic parameters of geomaterial’s microstructure can be extracted perfectly.