River network automated selection method based on heterogeneous graph convolutional networks

doi:10.11947/j.AGCS.2025.20240337

Abstract

Abstract:

River network selection is a map generalization process in which important rivers are selected and other rivers are discarded, due to space limitations when scaling down from large-scale to small-scale maps. Traditional deep learning methods typically focus on homogeneous graphs with a single type of relationship between river segments, which limits their ability to fully utilize the complex connectivity information between segments. This often results in lower selection accuracy and compromised topological connectivity. To address these issues, this paper introduces an automated river network selection method based on heterogeneous graph convolutional networks. In this method, river segments are represented as nodes, and their connections as edges. These edges are categorized into three types based on different relationship characteristics, creating a heterogeneous graph of the river network. The river network data and corresponding selection labels are input into the relational graph convolutional networks (RGCN) model, which aggregates information and classifies the nodes. River segments are selected based on the classification results, achieving automation in the selection process. Experiments using river network data at scales of 1∶24 000 and 1∶250 000 show that the proposed method significantly improves selection accuracy. Key performance metrics, including precision, recall, F₁ score and AUC, all exceed 92%. Additionally, the method reduces river network discontinuities and better preserves the topological connectivity and shape similarity of the river network.

Key words: map generalization, river network selection, deep learning, heterogeneous graph of the river network, RGCN

CLC Number:

P208

Yaqing WANG, Zhonghui WANG. River network automated selection method based on heterogeneous graph convolutional networks[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(7): 1332-1345.

Figures/Tables 20

Fig. 1

Tab. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Fig. 11

Fig. 12

Tab. 2

Tab. 3

Fig. 13

Fig. 14

Tab. 4

Tab. 5

Fig. 15

References 30

[1]	王家耀, 王光霞, 江南, 等. 地图学原理与方法[M]. 3版. 北京: 科学出版社, 2023.
	WANG Jiayao, WANG Guangxia, JIANG Nan, et al. Principles and methods of cartography[M]. 3rd ed. Beijing: Science Press, 2023.
[2]	LIU Jingnan, ZHAN Jiao, GUO Chi, et al. Data logic structure and key technologies on intelligent high-precision map[J]. Journal of Geodesy and Geoinformation Science, 2020, 3(3): 1-17.
[3]	STOTER J, POST M, VANALTENA V, et al. Fully automated generalization of a 1∶50k map from 1∶10k data[J]. Cartography and Geographic Information Science, 2014, 41(1): 1-13.
[4]	何宗宜. 地图数据处理模型的原理与方法[M]. 武汉: 武汉大学出版社, 2004.
	HE Zongyi. Elements and methods of model for cartographical data processing[M]. Wuhan: Wuhan University Press, 2004.
[5]	张青年. 顾及密度差异的河系简化[J]. 测绘学报, 2006, 35(2): 191-196.
	ZHANG Qingnian. Generalization of drainage network with density differences[J]. Acta Geodaetica et Cartographica Sinica, 2006, 35(2): 191-196.
[6]	张园玉, 李霖, 金玉平, 等. 基于图论的树状河系结构化绘制模型研究[J]. 武汉大学学报(信息科学版), 2004, 29(6): 537-539, 543.
	ZHANG Yuanyu, LI Lin, JIN Yuping, et al. Structured design of dendritic river networks based on graph[J]. Geomatics and Information Science of Wuhan University, 2004, 29(6): 537-539, 543.
[7]	武芳, 谭笑, 翟仁健, 等. 基于知识表达和推理的河网自动选取[J]. 辽宁工程技术大学学报, 2007, 26(2): 183-186.
	WU Fang, TAN Xiao, ZHAI Renjian, et al. Automated river selection based on cartographic knowledge representation and reasoning[J]. Journal of Liaoning Technical University, 2007, 26(2): 183-186.
[8]	谭笑. 基于知识的线状水系要素自动综合研究[D]. 郑州: 信息工程大学, 2005.
	TAN Xiao. Research of linear drainage features automated generalizaton based on knowledge[D]. Zhengzhou: Information Engineering University, 2005.
[9]	邵黎霞, 何宗宜, 艾自兴, 等. 基于BP神经网络的河系自动综合研究[J]. 武汉大学学报(信息科学版), 2004, 29(6): 555-557.
	SHAO Lixia, HE Zongyi, AI Zixing, et al. Automatic generalization of river network based on BP neural network techniques[J]. Geomatics and Information Science of Wuhan University, 2004, 29(6): 555-557.
[10]	翟仁健, 武芳, 邓红艳, 等. 基于遗传多目标优化的人工河网自动选取模型[J]. 测绘学报, 2008, 37(1): 108-113.
	ZHAI Renjian, WU Fang, DENG Hongyan, et al. An automated selection model of ditch based on multi-objective optimization by genetic algorithm[J]. Acta Geodaetica et Cartographica Sinica, 2008, 37(1): 108-113.
[11]	SEN A, GOKGOZ T. An experimental approach for selection/elimination in stream network generalization using support vector machines[J]. Geocarto International, 2015, 30(3): 311-329.
[12]	WANG Di, QIAN Haizhong. Graph neural network method for the intelligent selection of river system[J]. Geocarto International, 2023, 38(1): 2252762.
[13]	YU Huafei, AI Tinghua, YANG Min, et al. Integrating domain knowledge and graph convolutional neural networks to support river network selection[J]. Transactions in GIS, 2023, 27(7): 1898-1927.
[14]	KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[EB/OL]. [2024-03-04]. https://arxiv.org/abs/1609.02907v4.
[15]	HAMILTON W L, YING R, LESKOVEC J. Inductive representation learning on large graphs[EB/OL]. [2024-03-04]. https://arxiv.org/abs/1706.02216v4.
[16]	张瑞. 面向异构图节点分类问题的图卷积网络研究[D]. 上海: 华东师范大学, 2023.
	ZHANG Rui. Graph convolutional networks for node classification tasks on heterophilic graphs[D]. Shanghai: East China Normal University, 2023.
[17]	SCHLICHTKRULL M, KIPF T N, BLOEM P, et al. Modeling relational data with graph convolutional networks[C]//Proceedings of the 15th European Semantic Web Conference. Berlin: Springer, 2018: 593-607.
[18]	贾香恩, 董一鸿, 朱锋, 等. 异构图卷积网络研究进展[J]. 计算机工程与应用, 2021, 57(9): 36-49.
	JIA Xiang'en, DONG Yihong, ZHU Feng, et al. Research progress of heterogeneous graph convolutional networks[J]. Computer Engineering and Applications, 2021, 57(9): 36-49.
[19]	刘一漩, 张晶, 赵进勇, 等. 赤水河弯曲河段地貌特征参数统计分析[J]. 水生态学杂志, 2022, 43(3): 44-50.
	LIU Yixuan, ZHANG Jing, ZHAO Jinyong, et al. Statistical analysis of geomorphological parameters in meandering sections of Chishui River[J]. Journal of Hydroecology, 2022, 43(3): 44-50.
[20]	WANG Wenning, YAN Haowen, LU Xiaomin, et al. Drainage pattern recognition method considering local basin shape based on graph neural network[J]. International Journal of Digital Earth, 2023, 16(1): 593-619.
[21]	王文宁. 多形态河系的流域基本单元特征分析、识别及相似度计算[D]. 兰州: 兰州交通大学, 2023.
	WANG Wenning. Characteristics analysis, recognition, similarity calculation of the catchment basic units for multiple drainage patterns[D]. Lanzhou: Lanzhou Jiatong University, 2023.
[22]	艾廷华, 刘耀林, 黄亚锋. 河网汇水区域的层次化剖分与地图综合[J]. 测绘学报, 2007, 36(2): 231-236, 243.
	AI Tinghua, LIU Yaolin, HUANG Yafeng. The hierarchical watershed partitioning and generalization of river network[J]. Acta Geodaetica et Cartographica Sinica, 2007, 36(2): 231-236, 243.
[23]	赵春燕. 水系河网的Horton编码与图形综合研究[D]. 武汉: 武汉大学, 2004.
	ZHAO Chunyan. Research on horton code and graphic generalization of catchment[D]. Wuhan: Wuhan University, 2004.
[24]	黄子叶, 王易初, 倪晋仁. 汉江流域河网分级特征研究[J]. 北京大学学报(自然科学版), 2021, 57(2): 351-360.
	HUANG Ziye, WANG Yichu, NI Jinren. Hierarchical characteristics of river network in Hanjiang basin[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2021, 57(2): 351-360.
[25]	高翔. 基于关系图卷积神经网络的多标签事件预测[J]. 电讯技术, 2023, 63(4): 457-465.
	GAO Xiang. Multi-label event prediction based on relational graph convolutional neural network[J]. Telecommunication Engineering, 2023, 63(4): 457-465.
[26]	徐冰冰, 岑科廷, 黄俊杰, 等. 图卷积神经网络综述[J]. 计算机学报, 2020, 43(5): 755-780.
	XU Bingbing, CEN Keting, HUANG Junjie, et al. A survey on graph convolutional neural network[J]. Chinese Journal of Computers, 2020, 43(5): 755-780.
[27]	THANAPALASINGAM T, VAN BERKEL L, BLOEM P, et al. Relational graph convolutional networks: a closer look[J]. PeerJ Computer Science, 2022, 8: e1073.
[28]	YU Huafei, AI Tinghua, YANG Min, et al. A recognition method for drainage patterns using a graph convolutional network[J]. International Journal of Applied Earth Observation and Geoinformation, 2022, 107: 102696.
[29]	刘怀湘, 王兆印. 典型河网形态特征与分布[J]. 水利学报, 2007, 38(11): 1354-1357.
	LIU Huaixiang, WANG Zhaoyin. Morphological feature and distribution of typical river networks[J]. Journal of Hydraulic Engineering, 2007, 38(11): 1354-1357.
[30]	杨飞, 王中辉. 河系几何相似性的层次度量方法[J]. 地球信息科学学报, 2021, 23(12): 2139-2150.
	YANG Fei, WANG Zhonghui. River systems geometric similarity measurement method under hierarchical analysis[J]. Journal of Geo-information Science, 2021, 23(12): 2139-2150.

类别	分类标准	特征指标
河段的几何特征	描述河段的几何形态和物理属性	河段长度、河段曲率、汇入角、伸长比、圆度比
河段的空间关系	描述河段之间的关系及其在河网中的结构性知识	河流间距、目标河段支流数、左右支流数量对称性、Strahler编码、Horton编码

模型	精确率/（%）	召回率/（%）	F₁值/（%）	AUC
RGCN	92.76	92.48	92.62	0.925 9
GraphSAGE	89.37	90.01	89.69	0.901 4
GCN	88.13	88.63	88.37	0.889 1
GAT	87.24	87.39	87.31	0.875 6

指标	目标结果	RGCN	GraphSAGE	GCN	GAT
Sim＿RN	0.547 5	0.546 4	0.528 1	0.521 7	0.483 1
R_C	0.261 8	0.259 1	0.191 9	0.200 7	0.185 1

图结构	精确率/（%）	召回率/（%）	F₁值/（%）	AUC
异构图	92.76	92.48	92.62	0.925 9
同构图	90.49	90.11	90.30	0.904 3