Removing the Impulse Noise from Grayscaled and Colored Digital Images Using Fuzzy Image Filtering

Digital image is an important information source for various applications. Impulse noise may cause a blur to the image information that lead to lose its importance. Thus, the removal of this degradation is one of the areas that has faced a growing interest by researchers. This paper concerns to reduce Fixed and Random impulsive noises from a grayscaled and colored digital image. The fuzzy logic (FL) methods are applied to reduce these impulsive noises from the digital images. This method tries to detect the corrupted pixels before starting the filtration process and processes only the pixels which have been classified as a corrupted pixel. This is carried out by applying two algorithms: Impulse Noise Reduction for Colored Digital Images (INRC) and Fuzzy Random Impulse Noise Reduction (FRINR) algorithms in order to filter the grayscaled and colored images. To restore a digital image, a number of phases were performed where impulsive noise was added to the set of images in several ratios for each image (5%, 10%, ..., 80%), afterwards the previously mentioned algorithms were applied through a series of experiments and monitoring results and observations of each results of each experiment. The fuzzy methods gave densities of up to 50% of the noise. These obtained results are comparable to the results rates that accessed in other research papers in the world. Also, Peak Signal to Noise Ratio (PSNR), Mean Squared Error (MSE) and universal image quality index (UIQ) measurements are used in this paper.