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

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

使用空间矢量数据对建筑模式进行分类的图卷积神经网络

DOI:
10.1016/j.isprsjprs.2019.02.010
复制
发表时间:
2019-04
影响因子:
12.7
通讯作者:
Yin Hongmei
Yin Hongmei
中科院分区:
工程技术1区
文献类型:
--
作者:
Yan Xiongfeng;Ai Tinghua;Yang Min;Yin Hongmei

文献摘要

参考文献

被引文献

相似文献

机器学习方法,特别是卷积神经网络(CNN),已成为许多学科科学研究不可或缺的一部分。然而,这些强大的方法通常无法对空间矢量数据进行模式分析和知识挖掘,因为在大多数情况下,此类数据并不是底层的网格或数组结构,而只能建模为图结构。本研究引入了一种新颖的图卷积,通过使用图傅里叶变换和卷积定理将其从顶点域转换为傅里叶域中的逐点乘积。此外,还提出了图卷积神经网络(GCNN)架构来分析图结构的空间矢量数据。本研究的重点是构建模式分类的经典任务,该任务仍然受到设计规则的使用和特定模式的手动提取特征的限制。将表示分组建筑物的空间矢量数据建模为图形,并研究各个建筑物的特征指数以收集输入变量。这些图的模式特征是通过训练标记数据直接提取的。实验证实,GCNN 在识别规则和不规则模式方面产生了令人满意的结果,从而比现有方法取得了显着的改进。综上所述,GCNN 在图结构空间矢量数据的分析方面具有巨大的潜力以及进一步改进的空间。
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.
通过多级图形分区和建筑分组提取建筑模式
DOI: 10.1016/j.isprsjprs.2016.10.001
发表时间: 2016-12
影响因子: 12.7
作者:
Du Shihong;Shu Mi;Luo Liqun;Cao Kai;Du SH
通讯作者: Du SH
DOI: 10.1016/j.compenvurbsys.2009.07.005
发表时间: 2009-09
期刊: Comput. Environ. Urban Syst.
影响因子: --
作者:
Patrick Lüscher;R. Weibel;D. Burghardt
通讯作者: Patrick Lüscher;R. Weibel;D. Burghardt
基于城市形态和格式塔理论的自动化建筑概括
DOI: 10.1080/13658810410001702021
发表时间: 2004-07-01
影响因子: 5.7
作者:
Li, Z;Yan, H;Chen, J
通讯作者: Chen, J
用于表征当地城市模式的扩展最小生成树方法
DOI: 10.1080/13658816.2017.1384830
发表时间: 2018-01-01
影响因子: 5.7
作者:
Wu, Bin;Yu, Bailang;Wu, Jianping
通讯作者: Wu, Jianping
DOI: --
发表时间: 2007
期刊: --
影响因子: --
作者:
Guo Ren-zhong
通讯作者: Guo Ren-zhong