An algorithm for image restoration with mixed noise using total variation regularization

Authors: CONG THANG PHAM, GUILHEM GAMARD, ANDREI KOPYLOV, THI THU THAO TRAN

Abstract: We present here an effective scheme for image denoising based on total variation regularization. The proposed scheme allows to efficiently remove Poisson noise as well as Gaussian noise simultaneously with the help of a new kind of data fidelity term, suitable for the mixed Poisson - Gaussian noise model. The results show that the algorithm corresponding to our new scheme outperforms the existing methods for mixed Poisson - Gaussian noise removal.

Keywords: Image denoising, Gaussian noise, Poisson noise, total variation regularization, mixed noise distribution, gradient flow

Full Text: PDF