基于小波变换的孔探图像边缘粗糙度分析

针对航空发动机内部结构裂纹损伤的孔探图像识别问题,提出一种描述孔探图像边缘曲线粗糙度特征的方法——粗糙度系数法.该方法利用二进小波变换对边缘曲线在所有可能尺度上进行小波分解,得到尺度系数和小波系数.对于这些小波分解系数,构造一种对数熵算法来计算尺度系数和小波系数的能量强度.小波系数与尺度系数的对数熵的比值被定义为粗糙度系数,用粗糙度系数来评估曲线的粗糙程度,粗糙度系数越大,曲线越粗糙.实验表明以粗糙度系数作为孔探图像边缘曲线的识别特征,能够有效地识别孔探图像中存在的裂纹损伤. 

Analysis of coarseness of edges extracted from borescope images based on wavelet transform

To recognize whether there existed crack damages in the inner structure of aircraft engine based on the borescope images,a method(that was called coarseness factor) was presented to describe the coarseness of edge curves extracted from the borescope images.The method first decomposed an edge curve into the scale coefficients and wavelet coefficients using dynastic wavelet transform at all of the possible scales.For the coefficients of the wavelet transform,the logarithm entropy was presented to calculate the respective strength of the scale coefficients and wavelet coefficients.And the ratio of the logarithm entropies of the wavelet coefficients and ones of the scale coefficients was defined as coarseness factor.Using the coarseness factor to estimate the coarseness of a curve,the greater the coarseness factor is,the coarser the curve is.Experiments show that using the coarseness factor as recognition feature of borescope images,the crack damages of borescope images can be recognized successfully.