Pixel-Level Image Fusion Using Particle Swarm Optimization with Local Search

Pixel-level image fusion is widely used in many fields. We proposed a pixel-level image fusion algorithm based on particle swarm optimization with local search, that is, PSO-LS, which improves performance further. PSO-LS integrated the self-improvement mechanisms from memetic algorithms and can avoid local minimum in PSO. Experiments were carried out on two real world images. It is shown that fusion algorithm based on PSO-LS outperforms that based on PSO, and the former obtained optimal solution rapidly using fewer particles.