Abstract

With the explosion in the number of digital images taken every day, there is a growing demand for more precise and visually appealing images. Images captured by modern cameras, on the other hand, are eventually ruined by noise, which contributes to a reduction in visual image quality. Impulse noise is one of noise as white and black dispersed pixels that can be found in both gray and color images. The impulse noise model is comprised of Salt and Pepper (SAP) noise and random valued impulse noise (RVIN). So far, a lot of impulse denoising methods have been developed for the images (both gray and color). This article provides a comparative study of impulse noise reduction methods applied to color images wherein impulse noise reduction methods are studied with regard to their performance on color images and a thorough comparison is also carried out to cover all of the denoising methods in detail as well as the results they produce. These methods are contrasted with their functionality, relative performance and time complexity. Extensive simulations have been conducted on a set of standard images for performance evaluation of various denoising methods with regard to PSNR, SSIM and NMSE quality metrics.