Single image super resolution based on adaptive multi-dictionary learning

Zongxu Pan,Jing Yu,Chuangbai Xiao,W Sun

DOI: https://doi.org/10.3969/j.issn.0372-2112.2015.02.001

2015-01-01

Abstract:Adaptive dictionary learning uses the low resolution image itself as training samples to make the similar patches have sparse representation over the learned dictionary ,so that extra information can be exploited from structural self-similarity by dictionary learning .In this paper ,we propose a single image super resolution method based on adaptive multi-dictionary learning .To exploit extra information from both the low resolution image itself ,and the image database ,the proposed method incorporates the idea of global dictionary learning that the image database can be used to obtain extra information into the process of adaptive dictio-nary learning .In the proposed method ,all patches in the image pyramid of the low resolution image are clustered into several groups ,then each patch satisfying a certain condition in the database is classified into one of these groups with the supervision of the clustering results ,and multi-dictionary learning is used to learn corresponding dictionaries for different groups .Experimental results demonstrate that our method achieves better result compared with ScSR ,SISR ,NLIBP ,CSSS and mSSIM methods .