Revolutionizing nonrigid demons registration with the whale optimization algorithm

Abstract

Image registration is one of the popular image transformation models in satellite and medical imaging currently. Image registration refers to image mapping with two or more than images. The ground-breaking fusion of the whale optimization algorithm and nonrigid demons registration (WOA-NDR) is applied in the current work to improve image registration's precision and effectiveness. NDR is an effective method for aligning images that have pliable structures. Nevertheless, it frequently runs into issues with local minima and massive deformations. To address these issues, WOA-which draws inspiration from whale hunting behavior-is integrated into the NDR architecture. The WOA-NDR approach intelligently explores the solution space, enhancing convergence and avoiding premature convergence. With the innovative WOA and NDR integration, the nonrigid image registration process is revolutionized and yields superior outcomes in terms of robustness and accuracy. The efficiency of the suggested strategy is demonstrated by experimental findings on a dataset of monomodal images. The obtained results are also compared with particle swarm optimization (PSO) based framework.