Optimization of bilateral filter parameters using a whale optimization algorithm

Noise removal and restoration are important topics in image processing. The filter-based denoising techniques can effectively reduce the image noise but sometimes lose image quality and information, such as blurring the edges of the image. In this paper, at first, a denoising filter is proposed by combining the whale optimization algorithm (WOA) and bilateral filter. Then, a WOA-based Richardson-Lucy (R-L) algorithm is applied to restore the image. The bilateral filter performance is largely dependent on the proper parameter selection. The bilateral filter parameters are optimized by using the weighted sum of the peak signal-to-noise ratio (PSNR) and the structural similarity index measure (SSIM) as a fitness function of the WOA algorithm to design the proposed filter. For the restoration purpose, the point spread function (PSF) of the Richardson-Lucy (R-L) algorithm is optimized by using the weighted sum of the PSNR and the second derivative like measure enhancement (SDME) as a fitness function of the WOA. The performance of the proposed denoising technique is compared with classical image denoising filters, PSO-based bilateral filters on various images and the performance of the suggested restoration technique are also compared with the blind deconvolution technique (BD) and PSO-based restoration method (BDPSO) on denoised images, with the experimental findings demonstrating that the suggested method outperforms the others. The proposed technique in the current paper can be implemented in different image applications such as medical and satellite images, etc.