Particle Swarm Optimization for Image Deblurring
Abida Toumi, Abdelmalik Taleb-Ahmed, K. Benmahammed, N Rechid, Hichem Arioui, Rochdi Merzouki, Hadj Ahmed Abbassi · AIP conference proceedings · 2008
Within the framework of this first study we suggest the use of the Particle Swarm Optimization technique (PSO) in the image restoration field. In our knowledge, we did not still find works concerning the image deblurring (restoration) using the PSO. So, in this paper, we present the use of the PSO in two manners: (i) by taking as hypothesis the degraded image as the entire swarm and pixels as particles; (ii) the degraded image is taken as particle and we generate a population of images to buildup a swarm of variable size. The first results which we give were validated on real images degraded by a Gaussian blur only, and degraded by a Gaussian blur and an additive Gaussian noise. We finish with the comparison of our results with some classical restoration methods.