Superior Formula for Gradient Impulse Noise Reduction from Images

In this work, we present a novel approach for effectively reducing impulse noise from images using Conjugate Gradient Method. Our Proposed method used the iterative property of the conjugate gradient method to accurately identify and restore corrupted pixel values while preserving the underlying image details. In the first part, we propose a new conjugate gradient parameter, and proved the global convergence using some assumptions. Then numerical experiment is conducted on some benchmark images contaminated with impulse noise to validate the efficacy of the method. The numerical results demonstrate a significant noise reduction in comparison with an existing method showing that our proposed method is more effective in denoising degraded images.