IMAGE DENOISING USING IMPROVED FUZZY BASED APPROACH
Jenita Subash, Dr.Kalaivani .S · Indian Journal of Computer Science and Engineering · 2021
Digital image gets corrupted due to unlimited and number of unrecognized sources.The denoising of image in algorithmic approaches uses a sliding window that first detects the noisy or corrupted pixel and denoises by comparing with the nearby pixels of same window.The limitations of conventional filters for image denoising and their not-so-efficient accuracy are some major flaws that seek attention.In this paper, the improved fuzzy logic is implemented for image filtering.The medical images are sourced for experimentation and the design is tested on three noises i.e., Poisson, Gaussian and salt & pepper noises.Four parameters are selected to evaluate the performance of proposed system (PSNR, SSIM, IQI and IEF).Proposed architecture works well with salt & pepper and poisson noises.As noise density increases the performance of system is much better compared to performance of traditional median filter.