Enhancement and colorized a noisy image using MATLAB and AI libraries

Abstract

The main reason of images corruption is noise. The process of removing noise from distorted images is a challenge. In this paper, the effect of some filters used in filtering some distorted images, such as mean and median, was studied depending on their ability to reconstruct images affected by noise. After that, artificial intelligence and deep learning libraries are used to colorize the images. The noise generated within these images will be removed using these algorithms.The results of this experiment will be shown which algorithm is most preferable for impulse noise removal in digital gray images. To evaluate performance, the quality of the improved images is measured by means of statistical quantitative measures: root mean square error (RMSE) and peak signal to noise ratio (PSNR).