A nonlinear filter for salt and pepper noise removal from images using mathematical morphology and median filter / Siti Nabihah Othman

• Main Author: Othman, Siti Nabihah
• Format: Thesis
• Language: English
• Published: 2018
• Subjects: Electronic Computers. Computer Science; Computer software; Electricity and magnetism
• Online Access: http://ir.uitm.edu.my/id/eprint/39988/1/39988.pdf; http://ir.uitm.edu.my/id/eprint/39988/

Around 1964 mathematical morphology has been developed by Georges Matheron and Jean Serra at Ecole Des Mines de Paris, France. Mathematical morphology is one of theory and technique in analysis and processing for describing shape using sets. This technique widely applied in digital images. Median filter is a technique that often used to remove noise from images. This technique has been widely used because it preserves edges while removing noise. Both techniques are effective to remove salt and pepper noise. This research will investigate the effectiveness of mathematical morphology and median filter in removing salt and pepper noise measured by peak signal to noise ratio (PSNR) and mean square error (MSE).The results are analyzed using computer software matlab . From the investigate, the effectiveness and strength of this technique can be shown by images that have been corrupted by impulse noise from 0.01% until 0.1% and filtered image.