Reconstruction of Superresolution Image Using Generalized Gaussian Markov Random Fields Model

An image super-resolution reconstruction method based on generalized Gauss-Markov random fields (GGM- RF)model is presented in this paper. The process of searching solution and experimental results are presented and ana- lyzed. Compared with Compound Markov and Huber-Markov random models, GGMRF model has the merits of easier solving and reduced computational expense, because it does not need to discriminate edge or line process. The experi- mental results show that for the case of lightly noised image, this method has a better visual effect on the reconstructed image.