2011 | OriginalPaper | Buchkapitel
A De-noising Method Based on Sparse Representation under the Framework of MCA
verfasst von : Lei Li, Yuemei Ren
Erschienen in: Advances in Computer Science, Intelligent System and Environment
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Due to the good ability of Contourlet transform to represent image edges, and the effectiveness of redundant dictionary for capturing various geometrical structural features of images, it is possible to weaken pseudo-Gibbs phenomena in the process of image de-noising. According to Meyer’s image cartoon- texture model, under the framework of morphological component analysis, a method for image de-noising based on Contourlet transform and learned dictionary is proposed. The method used Contourlet transform to represent the cartoon component, and constructed a redundant dictionary by learning algorithm to represent the texture component sparsely. Experimental results show that, in comparison with wavelet-based de-noising methods and some algorithms based on learned-dictionary lonely, our method has better de-noising ability, preserves more edge, contour and detail image information and improves the peak signal-to-noise ratio.