Estimating Image Enhancement Factor (IEF) using a Fuzzy Set & Noise Level of Salt-Pepper Noise for Gray Level Imaging

Abstract

Improving a noisy image is a necessary task when processing digital images. To correct the noise content in the natural image, adding known noise to the image before processing. Therefore, the simulated noise is added to the image just to understand the noise elimination process. A filtering technique that can be applied to eliminate noise from images. After observing the results of the quality measurement values, it is concluded that the filter works best to eliminate image noise in all chosen noise models. Eliminating noise from the image is one of the deep challenges in the area of image processing and computer vision, where the core objective is to estimate the experimental image, smoothing noise from a noise-impure version of the image. Image noise can be caused by unlike intrinsic and extrinsic conditions that are repeated not possible to avoid in realistic state.. Therefore, denoising image plays an vital role in a ample range of aim such as image restoration, visual tracking, image registration, image segmentation and classification, where to obtain image content The original is crucial for performance solid. Noise reduction is the process of eliminating noise from images; Each pixel in the image will change from the original values in a small amount. A noise elimination algorithm is to achieve noise reduction and resource preservation, but due to the limitations of the methods, it is blurred. The noise in different pixels can be correlated or not, because noise modeling is a very difficult task. We observed that the performance of the proposed study's diffuse set and the 3x3, 3x5, 2x3 size filter windows, the adaptive weighted median filter and the median filters and also adaptive fuzzy filter were used to reduce the salt and pepper noise filters and the elimination context noise, the most relevant value Accuracy is recovered. Finally, our results are compared with the image improvement factor (IEF), the mean square error (MSE) and the peak signal-to-noise ratio (PSNR)