NOISE REMOVAL ON COLOR IMAGE SEQUENCES USING COUPLED ANISOTROPIC DIFFUSIONS AND NOISE-ROBUST MOTION DETECTION (TuePmPO4)

Author(s) : Pascal Bourdon (SIC - FRE CNRS 2731, University of Poitiers, France) Bertrand Augereau (SIC - FRE CNRS 2731, University of Poitiers, France) Christian Olivier (SIC - FRE CNRS 2731, University of Poitiers, France) Christian Chatellier (SIC - FRE CNRS 2731, University of Poitiers, France)

Abstract : In this paper, we propose a new approach for noise removal on color image sequences, based on coupled spatial and temporal anisotropic diffusions. This approach consists in smoothing a sequence while keeping all discontinuities caused by edges and moving objects. Our approach is based on a low level detection of moving areas in a noisy sequence, in order to associate temporal diffusion to a spatial diffusion process, thus denoising the sequence more efficiently.

Menu