基于小波变换的零件图像融合与边缘检测

Parts image data fusion and edge detection based on wavelet transform

提出基于小波变换的零件图像数据融合和边缘检测的方法,对图像进行分解,将高频区域中的绝对值较大的系数作为重要小波系数;在低频区域,对逼近系数进行加权平均得到新的逼近系数,然后进行小波重构实现图像数据融合。应用小波变换对融合图像进行多尺度边缘检测,获取图像边缘,或对图像进行小波多尺度边缘检测,然后融合边缘。

A mothed for parts image data fusion and edge detection besed on wavelet transform is presented, by image analysis, to regard the coefficients with larger absolute value in high frequency range as the important wavelet coefficients and the weighted mean value of approximation coefficients in low frequency range as approximation coeffwients, the image data fasion can be obtained using the inverse wavelet transform. The edges from fused image are detected or the edges from parts image are fused using wavelet multiscale edge detection.