Comparison of Median Filter and Gaussian Filter Performance in Removing Salt and Pepper Noise

Abstract

Image processing plays a critical role in various applications, from medical diagnostics to surveillance systems. However, one of the major challenges in digital image processing is the presence of noise, particularly salt and pepper noise, which significantly degrades image quality. This study aims to compare the effectiveness of two popular filtering techniques—Median Filter and Gaussian Filter—in removing salt and pepper noise from digital images. The evaluation is conducted both quantitatively, using Peak Signal-to-Noise Ratio (PSNR) and Mean Squared Error (MSE) metrics, and qualitatively, through visual analysis. The experimental results show that the Median Filter consistently outperforms the Gaussian Filter in terms of noise reduction performance. Median filtering yields higher PSNR and lower MSE values across various levels of noise intensity (5%, 10%, and 15%). Moreover, the visual assessment indicates that Median Filter preserves image edges and fine details more effectively, whereas Gaussian Filter tends to introduce blurring artifacts due to its smoothing nature. These findings suggest that for impulsive noise such as salt and pepper, Median Filter is a more appropriate and robust method, offering better restoration quality without compromising important image features.