结合小波分析和分水岭分割法的微观表面形貌分析方法

An Analysis Method of Micro Surfaces Using Wavelet Transform and Watershed Segmentation

为了实现微观表面粗糙峰特征参数的获取,提出了一种结合小波分析和分水岭分割法的微观表面形貌分析方法。该方法用小波分析从原始表面形貌数据中提取反映表面粗糙度的数据,然后用分水岭分割法对粗糙度数据进行粗糙峰分割,再通过曲面拟合获得表征每个粗糙峰的参数,最后通过统计得到整个粗糙表面的特征参数。将该方法用于分析玻璃微球的微观表面粗糙峰特征参数,获得了与传统方法接近的结果。

Such as the mean radius of asperity peaks, the mean asperity distance, the density of asperities and the standard deviation of asperity height. A method combining wavelet transform and watershed segmentation was put forward to obtain all these feature parameters. The major idea of the method was： first, the data of surface roughness were picked up from the profile data by wavelet transform; then every surface asperity was separated by watershed segmentation） the third step was to gain the parameters of each asperity by curve fitting; finally the whole surface feature parameters were calculated by statistics. The surfaces of micro glass balls were analyzed using the above method, and the similar results with previous research are achieved.