Multi-Channel Convolutional Neural Networks for Image Super-Resolution

Shinya OHTANI  Yu KATO  Nobutaka KUROKI  Tetsuya HIROSE  Masahiro NUMA  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E100-A   No.2   pp.572-580
Publication Date: 2017/02/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E100.A.572
Type of Manuscript: Special Section PAPER (Special Section on Image Media Quality)
super-resolution,  resolution enhancement,  convolutional neural networks,  CNN,  deep learning,  

Full Text: PDF>>
Buy this Article

This paper proposes image super-resolution techniques with multi-channel convolutional neural networks. In the proposed method, output pixels are classified into K×K groups depending on their coordinates. Those groups are generated from separate channels of a convolutional neural network (CNN). Finally, they are synthesized into a K×K magnified image. This architecture can enlarge images directly without bicubic interpolation. Experimental results of 2×2, 3×3, and 4×4 magnifications have shown that the average PSNR for the proposed method is about 0.2dB higher than that for the conventional SRCNN.