图卷积网络在道路网选取中的应用Application of the graph convolution network in the selection of road network
张康,郑静,沈婕,马劲松
摘要(Abstract)：
针对现有制图综合中的道路网自动选取方法不能有效地利用道路网的空间特征问题,该文把道路网抽象为图结构,提出了使用图卷积网络来进行道路网的自动选取,并比较分析了不同的图卷积网络在道路网选取中的适用性。结果表明,图卷积网络可以通过多层卷积来自动提取不同局部范围的空间特征,从而减少空间特征的人工构建,相比传统的多层感知机(MLP)等人工智能选取方法,具有更高的选取精度。对于不同的图卷积网络模型,使用最大池化聚合的GraphSAGE获得了最优的性能。
关键词(KeyWords)： 图卷积网络;道路网;自动选取;制图综合
基金项目(Foundation): 国家自然科学基金项目（41871371）
作者(Author): 张康,郑静,沈婕,马劲松
DOI: 10.16251/j.cnki.1009-2307.2021.02.024
参考文献(References)：
- [1]刘凯．核机器学习在地图自动综合中的道路网智能选取研究[D]．南京：南京大学，2017.(LIU Kai.Research on intelligent selection of road network automatic generalization based on kernel-based machine learning[D].Nanjing:Nanjing University,2017．）
- [2]刘凯，李进，沈婕，等．基于BP神经网络和拓扑参数的道路网选取研究[J]．测绘科学技术学报，2016(3):325-330.(LIU Kai,LI Jin,SHEN Jie,et al.Selection of road network using BP neural network and topological parameters[J].Journal of Geomatics Science and Technology,2016(3):325-330．）
- [3]刘佩，袁林辉，张康，等．基于RBF神经网络的OSM道路网智能选取[J]．地理信息世界，2019(3):8-13.(LIU Pei,YUAN Linhui,ZHANG Kang,et al.Intelligent selection of OSM road network based on RBF neural network[J].Geomatics World,2019(3):8-13．）
- [4]袁林辉．集成学习与多参数在OSM道路网自动选取中的应用[D]．南京：南京大学，2018.(YUAN Linhui.Study on ensemble learning and multi-parameters system for OSM road network selection.Nanjing:Nanjing University,2018．）
- [5]马超，孙群，陈换新，等．加权网页排序算法在道路网自动选取中的应用[J]．武汉大学学报（信息科学版），2018(8):1159-1165.(MA Chao,SUN Qun,CHENHuanxin,et al.Application of weighted PageRank algorithm in road network auto-selection[J].Geomatics and Information Science of Wuhan University,2018(8):1159-1165．）
- [6]BRONSTEIN M M,BRUNA J,LECUN Y,et al.Geometric deep learning:going beyond euclidean data[J].IEEE Signal Processing Magazine,2017,34(4):18-42．
- [7]KIPF T N,WELLING M.Semi-supervised classification with graph convolutional networks[C]∥International Conference on Learning Representations.[S.l.]:[s.n.],2017．
- [8]HAMILTON W,YING Z,LESKOVEC J.Inductive representation learning on large graphs[C]∥Advances in Neural Information Processing Systems.MITPress:New York,2017:1024-1034．
- [9]VELICKOVIC P,CUCURULL G,CASANOVA A,et al.Graph attention networks[C]∥International Conference on Learning Representations.[S.l.]:[s.n.],2018．
- [10]BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[J].arXiv preprint arXiv:1312.6203,2013．
- [11]BELKIN M,NIYOGI P.Laplacian eigenmaps and spectral techniques for embedding and clustering[C]∥Advances in Neural Information Processing Systems.MIT Press:New York,2002:585-591．
- [12]SHUMAN D I,NARANG S K,FROSSARD P,et al.The emerging field of signal processing on graphs:Extending high-dimensional data analysis to networks and other irregular domains[J].IEEE Signal Processing Magazine,2013,30(3):83-98．
- [13]DEFFERRARD M,BRESSON X,VANDERGHEYNSTP.Convolutional neural networks on graphs with fast localized spectral filtering[C]∥Advances in Neural Information Processing Systems.MIT Press:New York,2016:3844-3852．
- [14]HOCHREITER S,SCHMIDHUBER J.Long shortterm memory[J].Neural Computation,1997,9(8):1735-1780．
- [15]GEMAN S,BIENENSTOCK E,DOURSAT R.Neural Networks and the Bias/Variance Dilemma[J].Neural Computation,1992,4(1):1-58．
- [16]BRADLEY A P.The use of the area under the ROCcurve in the evaluation of machine learning algorithms[J].Pattern Recognition,1997,30(7):1145-1159.