首页  >  科学研究  >  科研成果  >  正文
科研成果
博士生晏雄锋的论文在ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING 刊出
发布时间:2019-04-29 10:47:44     发布者:易真     浏览次数:

标题: A graph convolutional neural network for classification of building patterns using spatial vector data

作者: Yan, XF (Yan, Xiongfeng); Ai, TH (Ai, Tinghua); Yang, M (Yang, Min); Yin, HM (Yin, Hongmei)

来源出版物: ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING  : 150  : 259-273  DOI: 10.1016/j.isprsjprs.2019.02.010  出版年: APR 2019  

摘要: Machine learning methods, specifically, convolutional neural networks (CNNs), have emerged as an integral part of scientific research in many disciplines. However, these powerful methods often fail to perform pattern analysis and knowledge mining with spatial vector data because in most cases, such data are not underlying grid-like or array structures but can only be modeled as graph structures. The present study introduces a novel graph convolution by converting it from the vertex domain into a point-wise product in the Fourier domain using the graph Fourier transform and convolution theorem. In addition, the graph convolutional neural network (GCNN) architecture is proposed to analyze graph-structured spatial vector data. The focus of this study is the classical task of building pattern classification, which remains limited by the use of design rules and manually extracted features for specific patterns. The spatial vector data representing grouped buildings are modeled as graphs, and indices for the characteristics of individual buildings are investigated to collect the input variables. The pattern features of these graphs are directly extracted by training labeled data. Experiments confirmed that the GCNN produces satisfactory results in terms of identifying regular and irregular patterns, and thus achieves a significant improvement over existing methods. In summary, the GCNN has considerable potential for the analysis of graph structured spatial vector data as well as scope for further improvement.

入藏号: WOS:000464088400018

语言: English

文献类型: Article

作者关键词: Building pattern classification; Graph convolutional neural network; Machine learning; Spatial vector data; Graph Fourier transform; Deep learning

地址: [Yan, Xiongfeng; Ai, Tinghua; Yang, Min] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Hubei, Peoples R China.

[Yin, Hongmei] Chinese Acad Surveying & Mapping, Beijing 100830, Peoples R China.

通讯作者地址: Ai, TH (通讯作者)Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Hubei, Peoples R China.

电子邮件地址: xiongfeng.yan@whu.edu.cn; tinghuaai@whu.edu.cn; yangmin2003@whu.edu.cn; mayyin@whu.edu.cn

影响因子:5.994


信息服务
学院网站教师登录 学院办公电话 学校信息门户登录

版权所有 © 武汉大学资源与环境科学学院
地址:湖北省武汉市珞喻路129号 邮编:430079 
电话:027-68778381,68778284,68778296 传真:027-68778893    邮箱:sres@whu.edu.cn