Application Of An Improved Image Fusion Algorithm In Medical Images

Image fusion is an important application of image processing,its purpose is to integrate the image information of different modalities through computer technology,so that the fused image contains as much information as possible.However,the images in practical problems often contain unnecessary noise,so it is necessary to conduct image denoising pre-processing before image fusion.Based on the existing image fusion algorithms,this paper proposes a class of improved image fusion algorithm and applies it to medical images.The main work is as follows:variational model based on data-driven tight frame(DDTF),this paper firstly proposes a new initial tight frame system for fusing clear medical images.Secondly,for medical images containing noise,combined with the new tight frame system,an improved variational model is proposed.The model adds the regularization term and fidelity term to remove noise on the basis of DDTF,which can remove noise in the process of image fusion,it is solved by a split Bregman iterative algorithm with better convergence.In this paper,the convergence analysis of the improved algorithm is carried out to verify its theoretical feasibility.In the numerical experiments,we use MATLAB software for programming,for clear and different noise-containing images,we use the improved algorithm to carry out experiments and compare it with the other two algorithms.Numerical experimental results show that the improved algorithm has better fusion effect.