Genetic Algorithm Based Approach to Image Denoising Problem

Abstract

Digital images can be degraded by noise during the process of acquisition, transmission, storage or compression. It is necessary to remove the noise in the image before the image is suitable for different processing operations. Image denoising is a process which is deployed to remove the noise through the manipulation of image data to recover quality image from the noisy image. The image denoising process should be such that the original image can be recovered without losing important features such as edges, corners and textures. One of the powerful and perspective approaches in this area is image denoising using discrete wavelet transform. This work combines genetic algorithm with wavelet based denoising methods. During the evolutionary process, wavelet based denoising methods are applied as local search operators and filtering techniques are applied as mutation operators. A set of digital images, commonly used by the scientific community as benchmarks, is contaminated by different level of additive Gaussian noise and the proposed algorithm is used to reduce the noise level in the image. The results in terms of PSNR & SSIM values obtained by the proposed method shows that application of genetic algorithm can improve the result obtained from wavelet based denoising methods. Also the proposed method is compared against denoising methods in the literature. On average it outperforms the compared methods in terms of PSNR & SSIM values