Medical Image Fusion using PCNN Optimized by Whale Optimization Algorithm

This paper proposes a novel algorithm for the fusion of medical images using a modified Pulse Coupled Neural Network(PCNN) optimized by the Whale Optimization Algorithm(WOA). In this proposed algorithm the two images are passed through the WOA which computes the necessary parameters for the PCNN feature extractor through a multi-criteria fitness function. The features thus extracted are then used to fuse the images. The resultant fused image is judged by evaluating the entropy, Mutual Information (MI), Structural Similarity (SSIM), etc. as well as a specialized performance metric called feature mutual information.

I. Introduction

Image fusion is an imperative operation in the field of Image Processing. It involves the integration of complementing multi-sum, multi-temporal, multi-view information, into a new image of novel quality [1]. One such application of image fusion lies in medical imaging, wherein simultaneous valuation of Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron Emission Tomography (PET) images are procured [2]. This can also be referred to as multi-modal image fusion since the images are coming from different sensors.