Remove Noise from Grayscale Digital Images by Traditional and Fuzzy Filters | ||
AL-Rafidain Journal of Computer Sciences and Mathematics | ||
Article 23, Volume 6, Issue 2, August 2009, Pages 47-100 PDF (3.49 M) | ||
Document Type: Research Paper | ||
DOI: 10.33899/csmj.2009.163815 | ||
Author | ||
Baydaa I. Khalil | ||
College of Computer Science and Mathematics University of Mosul, Iraq | ||
Abstract | ||
Image denoising and enhancement is an important field and it is used very much in image processing, where images are corrupted by many kinds of noise, therefore, methods and techniques must be used to remove these noises. In this research seven traditional filters are used to remove noise from digital images corrupted with salt&pepper noise and Gaussian noise. And also adopting principle of fuzzy logic to hybrid between traditional filters and fuzzy logic using double bell shaped membership function and also hybrid with double adjusted sigmoid membership function to create seven fuzzy sigmoidal filters. After applying seven traditional methods, seven fuzzy bell methods and seven fuzzy sigmoidal methods , by using measures of restored image PSNR, MSE, MAE , shows the final methods i.e. seven fuzzy sigmoidal methods are better than other filters i.e. methods that are used in this research after comparing the results. | ||
Keywords | ||
Image Processing; Noise; Fuzzy Logic; sigmoid membership function | ||
Statistics Article View: 139 PDF Download: 656 |