期刊
NEURAL NETWORKS
卷 139, 期 -, 页码 77-85出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.neunet.2021.02.005
关键词
Global dense block; Fast dense residual network; Down-sampling block; Global dense residual learning; Text image representation and recognition
资金
- National Natural Science Foundation of China [NSFC 62072151, 61822701, 62036010, 61806035, U1936217]
- Anhui Provincial Natural Science Fund, China for Distinguished Young Scholars [2008085J30]
- Fundamental Research Funds for Central Universities of China [JZ2019HGPA0102]
This paper investigates how to enhance the local and global feature learning abilities of DenseNet by fully exploiting the hierarchical features from all convolutional layers. The proposed Dense Residual Network (DRN) utilizes refined residual dense blocks (r-RDB) and global dense blocks (GDB) to adaptively learn global dense residual features in a holistic way, leading to enhanced results compared with other related deep models.
Deep Convolutional Neural Networks (CNNs), such as Dense Convolutional Network (DenseNet), have achieved great success for image representation learning by capturing deep hierarchical features. However, most existing network architectures of simply stacking the convolutional layers fail to enable them to fully discover local and global feature information between layers. In this paper, we mainly investigate how to enhance the local and global feature learning abilities of DenseNet by fully exploiting the hierarchical features from all convolutional layers. Technically, we propose an effective convolutional deep model termed Dense Residual Network (DRN) for the task of optical character recognition. To define DRN, we propose a refined residual dense block (r-RDB) to retain the ability of local feature fusion and local residual learning of original RDB, which can reduce the computing efforts of inner layers at the same time. After fully capturing local residual dense features, we utilize the sum operation and several r-RDBs to construct a new block termed global dense block (GDB) by imitating the construction of dense blocks to adaptively learn global dense residual features in a holistic way. Finally, we use two convolutional layers to design a down-sampling block to reduce the global feature size and extract more informative deeper features. Extensive results show that our DRN can deliver enhanced results, compared with other related deep models. (C) 2021 Elsevier Ltd. All rights reserved.
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