Open AccessOptimized Hybrid Model for Gaussian Noise Reduction images
Author(s) -
Lubna Farhi,
Agha Yasir Ali,
Syed Muslim Jamel,
Farhan Ur Rehman,
Baqar Ali Zardari,
Ramsha Shakeel,
Samia Shakeel
Publication year - 2021
Publication title -
quaid-e-awam university research journal of engineering science and technology
Language(s) - English
Resource type - Journals
eISSN - 2523-0379
pISSN - 1605-8607
DOI - 10.52584/qrj.1901.02
Subject(s) - wiener filter , gaussian noise , noise reduction , peak signal to noise ratio , smoothness , algorithm , mathematics , salt and pepper noise , reduction (mathematics) , noise (video) , filter (signal processing) , mean squared error , median filter , image noise , gaussian , block (permutation group theory) , image (mathematics) , computer science , artificial intelligence , computer vision , image processing , statistics , mathematical analysis , geometry , physics , quantum mechanics
In this paper, image noise is removed by using a hybrid model of wiener and fuzzy filters. It is a challenging task to remove Gaussian noise (GN) from an image and to protect the image’s edges. The Fuzzy-Wiener filter (FWF) hybrid model is used for optimizing the image smoothness and efficiency at a high level of GN. The efficiency is measured by using Structural Similarity (SSIM), Mean Square Error (MSE), and Peak Signal to Noise Ratio (PSNR). The proposed algorithm substitutes a mean value of the matrix for a non-overlapping block and replaces the total pixel number with each direction. In the proposed model, overall results proved that the optimized hybrid model FWF has an enormous computational speed and impulsive noise reduction, which enables efficient filtering as compared to the existing techniques.