The GFIS is a multi-layered neuro-diffuse structure that combines both the Mamdani model and the fuzzy TS model to form a fuzzy hybrid system. The GFIS can not only preserve the interpretability property of the Mamdani model but also maintain the solid local stability criteria of the TS model. The results of the simulation indicate that the proposed model shows a high quality restoration of filtered images for the noise model than those using medium filters or wiener filters in terms of peak signal to noise ratio (PSNR).
The image corrupted by different types of noise is a problem frequently encountered in the acquisition and transmission of images. The noise comes from noisy sensors or channel transmission errors. Several types of noise are discussed here. Impulse noise (or salt and pepper noise) is caused by sharp and sudden disturbances in the image signal; Its appearance is randomly scattered white or black (or both) pixels on the image. Gaussian noise is an idealized form of white noise, which is caused by random fluctuations in the signal. Spotted (or simply simply mottled) noise can be modeled by random values multiplied by pixel values, hence also called multiplicative noise. If the image signal is subjected to a periodic disturbance, instead of random, we could obtain an image damaged by periodic noise. Usually, periodic noise requires the use of frequency domain filtering. This is because while other forms of noise can be modeled as local degradations, periodic noise is a global effect.