Geng Tianyu, Liu Xiao-Yang, Wang Xiaodong, Sun Guiling
IEEE Trans Image Process. 2021;30:4129-4142. doi: 10.1109/TIP.2021.3069317. Epub 2021 Apr 9.
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.
最近，残差学习策略已被集成到用于单图像超分辨率（SISR）的卷积神经网络（CNN）中，其中CNN被训练来估计残差图像。认识到残差图像通常由高频细节组成并呈现出类似卡通的特征，在本文中，我们提出了一种深度剪切波残差学习网络（DSRLN），用于基于剪切波变换估计残差图像。所提出的网络在剪切波变换域中进行训练，该域为类似卡通的图像提供了最优的稀疏逼近。具体而言，为了解决剪切波系数之间的大统计变化，提出了一种双路径训练策略和一种数据加权技术。在一般自然图像数据集以及遥感图像数据集上的广泛评估表明，所提出的DSRLN方案在使用少得多的网络参数的情况下，在峰值信噪比（PSNR）方面取得了与当前最先进的深度学习方法相近的结果。
