Medical Image Fusion Algorithm Based on Clustering Neural Network

This paper proposes a new image fusion algorithm based on clustering analysis for clinical image processing. According to the present image fusion algorithm, pixels of origin images are classified into clustering feature pixels and secondary pixels base on clustering analysis. Feature pixels have more useful medical information we need; secondary pixels have background information of the image. In the new algorithm, build different fusion rules on two types of pixels, rules of feature pixels base on partial gradient and rules of secondary pixels base on average gray. A great deal of experiments have done to testify feasibility of new algorithm, the result show fusion image has more information than origin images and improves the quality of the origin image, fusion image also protects characters of the image and heightens the visual impact, new algorithm is effective.