期刊
IEEE TRANSACTIONS ON CYBERNETICS
卷 51, 期 1, 页码 115-125出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCYB.2019.2952710
关键词
Feature extraction; Image reconstruction; Spatial resolution; Convolutional neural networks; Training; Graphics; Convolutional neural network; multiscale learning; residual learning; single-image super-resolution (SISR)
类别
资金
- National Natural Science Foundation of China [61702129, 61772149, 61562013, 61866009, 61572099, U1701267]
- China Post-Doctoral Science Foundation [2018M633047]
- Guangxi Science and Technology Project [2019GXNSFFA245014, AD18281079, 2017GXNSFDA198025, AD18216004, AB17195057, AA18118039]
This article addresses the challenges of high parameter and complex structure in CNN models for SISR by introducing novel local wider residual blocks (LWRBs) and various strategies. The proposed cascading residual network (CRN) and enhanced residual network (ERN) achieve superior results compared to advanced methods, while effectively utilizing low-level features.
Deep convolutional neural networks (CNNs) have contributed to the significant progress of the single-image super-resolution (SISR) field. However, the majority of existing CNN-based models maintain high performance with massive parameters and exceedingly deeper structures. Moreover, several algorithms essentially have underused the low-level features, thus causing relatively low performance. In this article, we address these problems by exploring two strategies based on novel local wider residual blocks (LWRBs) to effectively extract the image features for SISR. We propose a cascading residual network (CRN) that contains several locally sharing groups (LSGs), in which the cascading mechanism not only promotes the propagation of features and the gradient but also eases the model training. Besides, we present another enhanced residual network (ERN) for image resolution enhancement. ERN employs a dual global pathway structure that incorporates nonlocal operations to catch long-distance spatial features from the the original low-resolution (LR) input. To obtain the feature representation of the input at different scales, we further introduce a multiscale block (MSB) to directly detect low-level features from the LR image. The experimental results on four benchmark datasets have demonstrated that our models outperform most of the advanced methods while still retaining a reasonable number of parameters.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据