小波变换和形态学相结合的图像融合新算法

针对边缘模糊导致图像融合质量下降,结合小波变换方法和数学形态学方法各自的优势,提出了一种将小波变换和数学形态学理论相结合的图像融合算法.针对高频系数融合,选用绝对值取大法,并适当地增强边缘；针对低频系数融合,对低频子图像运用形态学方法获得边缘灰度图像,若边缘灰度图像中该点像素值在153～255之间,标记该点为边缘点,选用区域空间频率取大法融合边缘点处对应的低频系数；若该点像素值在O～152之间,标记该点为非边缘点,选用平均法融合非边缘点处对应的低频系数.实验结果证明,该算法的融合质量得到了进一步改善和提升.

New Image Fusion Algorithm Based on Combination of Wavelet Transformation and Morphology

For the decline in image fusion quality because of edge blur, combining respective advantages of wavelet transformation method and mathematical morphology method, an image fusion algorithm based on combination of wavelet transformation and mathematical morphology theory is proposed. Maximal absolute value method is adopted for the fusion of high-frequency coefficients and edges are enhanced properly. For low-frequency sub-images, mathematical morphology method is used to get edge gray images. If the pixel value of points from edge gray images ranges between 153 and 255, the points are called as the edge points and the maximal region spatial frequency method is adopted for the fusion of low-frequency coefficients in the edge point positions. If the pixel value of points from edge gray images ranges between 0 and 152, the points are called as the non-edge points and the average method is adopted for the fusion of low-frequency coefficients in the non-edge point positions. The experimental results show that the proposed algorithm improves image fusion quality further.