Knowledge Commons of Institute of Automation,CAS
CNN-G: convolutional neural network combined with graph for image segmentation with theoretical analysis | |
Lu, Yi1; Chen, Yaran1; Zhao, Dongbin1; Liu, Bao2; Lai, Zhichao2; Chen, Jianxin3 | |
发表期刊 | IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS |
2021 | |
卷号 | 0期号:0页码:0 |
摘要 | Deep Convolutional Neural Network (CNN), although recognized to be considerably successful in its application to semantic segmentation, are inadequate for extracting the overall structure information, for its representing images with the data in Euclidean space. To improve this inadequacy, a new model in the graph domain that transforms semantic segmentation into graph node classification is proposed for semantic segmentation. In this model, the image is represented by a graph, with its nodes initialized by the feature map obtained by a CNN, and its edges reflecting the relationships of the nodes. The node relationships that are taken into consideration include distance-based ones and semantic ones, respectively calculated with Gauss kernel function and attention mechanism. Graph Neural Network is also introduced in this model for the classification of graph nodes, which can expand the receptive field without the loss of location information and combine the structure with the feature extraction. Most importantly, it is theoretically concluded that the proposed graph model takes the same role as a Laplace regularization term in image segmentation, which has been proven by multiple comparative experiments that show the effectiveness of the model in image semantic segmentation. The learned attention is visualized by the heatmap to validate the structure learning ability of our model. The results of these experiments show the importance of structure information in image segmentation. Hence an idea of deep learning combined with graph structural information is provided in theory and method. |
关键词 | Graph neural network, image segmentation, self-attention, structure pattern learning. |
DOI | DOI:10.1109/TCDS.2020.2998497. |
收录类别 | SCI |
七大方向——子方向分类 | 强化与进化学习 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/40611 |
专题 | 多模态人工智能系统全国重点实验室_深度强化学习 复杂系统管理与控制国家重点实验室 |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.Peking Union Medical College Hospital 3.Beijing University of Chinese Medicine |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Lu, Yi,Chen, Yaran,Zhao, Dongbin,et al. CNN-G: convolutional neural network combined with graph for image segmentation with theoretical analysis[J]. IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS,2021,0(0):0. |
APA | Lu, Yi,Chen, Yaran,Zhao, Dongbin,Liu, Bao,Lai, Zhichao,&Chen, Jianxin.(2021).CNN-G: convolutional neural network combined with graph for image segmentation with theoretical analysis.IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS,0(0),0. |
MLA | Lu, Yi,et al."CNN-G: convolutional neural network combined with graph for image segmentation with theoretical analysis".IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 0.0(2021):0. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
CNN-G Convolutional (5636KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论