The improved algorithm of image fusion based segmentation region

In the pixel level image fusion methods, anglicizing the wavelet transform, this paper proposes an segmentation region based image fusion method. First, the source images are decomposed by wavelet to get the approximate and detailed sub-images, and the segmentation by watershed method for these sub-images are used to get the regions of each level, these regions are used to guide fusion process. Then, measure the match degree and activity level of the wavelet coefficients of source images are implemented in these region segmentation, and the maximum value rule and the weighted average norm are used to combine the coefficients of detailed sub-images and approximate sub-image respectively. At last, the combination coefficients are transformed by wavelet inversely to obtain the fused image. The experimental results show that the fusion effect is superior to the multiscale image data fusion method.