Image Restoration Using Hybrid Technique Based on Noise Detection and Uncetainty
Abdurahman E. Saleh, Farag J. Zbeda, Salem A. Salem, Said O. Albasheer · 2021 IEEE 1st International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering MI-STA · 2021
Image enhancement in the digital image processing area is one of the most important techniques, to restore images that have been corrupted with some types of noise. In this paper the restoration method includes suppression of impulse noise from grayscale images using a hybrid technique which is Morphological Opening and Modified Enhanced Recursive Median Filter based on noise detection and uncertainty. The first step of the proposed technique is detecting whether the pixel is corrupted or not, then a morphological opening is applied on corrupted pixels only. A second detection step and uncertainty process will be applied to the resulting image from the first step, then the only corrupted pixel will be manipulated using the modified enhanced recursive median filter to suppress a very high density of noise. The processing time is very short for the proposed technique compared to modified enhanced recursive median filter. The performance of these techniques have been measured using Mean Square Error and Peak signal to Noise Ratio.