Learning Shared and Cluster-Specific Dictionaries for Single Image Super-Resolution
Tingting Yao,Zhiyong Wang,Yu Luo,Yue Liang,Qing Hu,David Dagan Feng
DOI: https://doi.org/10.1109/access.2019.2929773
IF: 3.9
2019-01-01
IEEE Access
Abstract:Recently, many multi-dictionary-based sparse representation methods have been proposed for single-image super-resolution (SISR) by learning a sub-dictionary for each specific type of visual content (i.e., a cluster). Although promising reconstruction results have been achieved for certain scenarios, due to the similarity shared among different types of visual content, each sub-dictionary is not learned as discriminatively as expected, which could compromise the visual quality of reconstructed high-resolution (HR) images. In this paper, we propose a novel shared and cluster-specific dictionaries learning method for SISR, where cluster-specific dictionaries can be learned as separate as possible and a dictionary shared among clusters is learned explicitly to explore similarity among clusters. First, low-resolution (LR) and HR image patches extracted from training images are jointly grouped into visual clusters. Then, a shared sub-dictionary and a set of cluster-specific sub-dictionaries are learned simultaneously under the sparse representation framework. In addition, the group sparsity constraint, the locality constraint, and the incoherence penalty term are incorporated into a unified framework to preserve the relationship and structure among the training data in dictionary learning. Finally, an anchored neighborhood regression method is devised to pre-calculate a projection matrix of each learned dictionary atom from LR to its HR space so that an HR image can be more efficiently recovered in the reconstruction stage. Comprehensive experimental results on two widely used benchmark datasets demonstrate the effectiveness and robustness of our method against a number of state-of-the-art methods. The reconstruction results of real-world maritime surveillance images further indicate that the proposed method is well suitable for practical applications.