High Impulse Noise Removal and Edge Detection in MRI Images Using Advanced Filtering Techniques

Abstract

Medical imaging plays a critical role in disease diagnosis and treatment planning, with Magnetic Resonance Imaging (MRI) being a cornerstone in capturing detailed anatomical structures. However, MRI images often suffer from high impulse noise, such as salt and pepper noise, introduced during acquisition and transmission, which compromises image quality and diagnostic accuracy. Additionally, precise edge detection is essential to delineate anatomical boundaries and regions of interest. This research proposes a novel two stage approach combining advanced denoising and edge detection techniques. The first stage employs an Adaptive Modified Weighted Median Filter (AMWMF) to efficiently remove high impulse noise while preserving image details. The second stage integrates Rough Set Theory (RST) for accurate edge detection by leveraging intensity based local variations and approximations. Extensive experiments were conducted to evaluate the proposed method against existing techniques, using metrics such as Peak Signal to Noise Ratio (PSNR) and Mean Squared Error (MSE). Results demonstrate significant improvements in noise suppression, edge preservation, and feature enhancement, offering a robust framework for enhanced MRI image analysis. The proposed method provides a reliable solution for medical imaging applications, aiding in improved visualization, classification, and diagnosis of complex medical conditions.