Deep Shearlet Residual Learning Network for Single Image Super-Resolution

Recently, the residual learning strategy has been integrated into the convolutional neural network (CNN) for single image super-resolution (SISR), where the CNN is trained to estimate the residual images. Recognizing that a residual image usually consists of high-frequency details and exhibits cartoon-like characteristics, in this paper, we propose a deep shearlet residual learning network (DSRLN) to estimate the residual images based on the shearlet transform. The proposed network is trained in the shearlet transform-domain which provides an optimal sparse approximation of the cartoon-like image. Specifically, to address the large statistical variation among the shearlet coefficients, a dual-path training strategy and a data weighting technique are proposed. Extensive evaluations on general natural image datasets as well as remote sensing image datasets show that the proposed DSRLN scheme achieves close results in PSNR to the state-of-the-art deep learning methods, using much less network parameters.

Automated summary

This paper proposes a deep shearlet residual learning network (DSRLN) based on shearlet transform for estimating residual images, trained in the shearlet transform domain to provide optimal sparse approximation of cartoon-like images. To address the large statistical variation among shearlet coefficients, a dual-path training strategy and data weighting technique are introduced. Extensive evaluations show that DSRLN achieves close results in PSNR to state-of-the-art deep learning methods with fewer network parameters.