Vol.37 No.10

Journal of Xi'an Jiaotong University

Oct..2003

Super Resolution Image Reconstruction Based on Projections onto Convex Sets
and Line Process Modeling
Huang Hua,Kong Lingli,Qi Chun,Zhu Shihua
(School of Electronics and Information Engineering,Xi'an Jiaotong University, Xi'an 710049, China)
Abstract:In order to improve the super resolution reconstruction of image, a method of projections onto convex sets (POCS) based on the line process modeling is proposed. According to the important role of the prior information playing in the image reconstruction, the line process modeling describing the edge information is considered as the prior information and used in the smoothness constraint projection. The experiment results demonstrate that the proposed method gives better supper resolution results in the visual effect and higher signal to noise ratio than the original method. Especially under high decimation rate, the ringing artifact is reduced obviously.
Keywords:super resolution;image reconstruction;projections onto convex sets;line process modeling