Total variation based parameter-free model for impulse noise removal

In this talk, based on total variation (TV), we propose a method for reconstructing blurred images corrupted by impulse noise. Following the idea of the two-phase method, we utilize a noise detector to identify image pixels contaminated by noise, and then we reconstruct the noisy pixels by solving an equality constraint total variation minimization problem that preserves the exact values of the noise-free pixels. The proposed method improves the computational efficiency and has the advantage of being parameter-free. Numerical results suggest that the method is competitive in terms of its restoration capabilities and speed.