Enhanced Image Fusion Algorithm with Neural Networks for Sonar Images

Abstract: This paper proposes a simple neural network based image fusion algorithm. Image fusion is defined as a process where a new image is constructed by integrating complementary, multi-temporal or multi-view information from a set of source images. Particle swarm Optimization (PSO) is used to find out the optimal size of the blocks to be fused. A detailed experimentation is done with different performance metrics for different set of images. We have compared the results and the proposed method outperforms the existing methods visually as well as quantitatively