Data and cognition dual-driven building group generalization results and scale consistency assessment

doi:10.11947/j.AGCS.2025.20240490

Abstract

Abstract:

The consistency between cartographic generalization results and their target scales constitutes a critical aspect of generalization quality evaluation. This process involves multidimensional factors including quantitative features, structural characteristics, and cognitive attributes. Traditional methods face difficulties in determining quantitative metrics when combining multiple evaluation indicators and encounter challenges in integrating domain knowledge such as spatial cognition. To address these limitations, this paper proposes a dilated graph convolutional network (DGCNN)-based recognition model for assessing scale consistency between building groups and their generalization results. The model adopts a dual data-driven and cognition-driven strategy to measure feature changes before and after generalization across three spatial-cognitive levels: global structure, local structure, and individual characteristics. It leverages multi-scale generalized datasets for training. Experimental results demonstrate that the proposed model effectively recognizes consistency between cartographic generalization outcomes and target scales.

Key words: building group, cartographic generalization, deep learning, quality assessment, spatial structure

CLC Number:

P208

Nina MENG, Fengmei LI, Xiaodong ZHOU. Data and cognition dual-driven building group generalization results and scale consistency assessment[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(7): 1318-1331.

Figures/Tables 20

Fig. 1

Tab. 1

Fig. 2

Fig. 3

Fig. 4

Tab. 2

Fig. 5

Fig. 6

Fig. 7

Tab. 3

Fig. 8

Fig. 9

Tab. 4

Fig. 10

Fig. 11

Fig. 12

Tab. 5

Tab. 6

Fig. 13

Fig. 14

References 36

[1]	李志林, 王继成, 谭诗腾, 等. 地理信息科学中尺度问题的30年研究现状[J]. 武汉大学学报(信息科学版), 2018, 43(12): 2233-2242.
	LI Zhilin, WANG Jicheng, TAN Shiteng, et al. Scale in geo-information science: an overview of thirty-year development[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 2233-2242.
[2]	艾廷华. 深度学习赋能地图制图的若干思考[J]. 测绘学报, 2021, 50(9): 1170-1182. DOI: . doi: 10.11947/j.AGCS.2021.20210091
	AI Tinghua. Some thoughts on deep learning enabling cartography[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(9): 1170-1182. DOI: . doi: 10.11947/j.AGCS.2021.20210091
[3]	KNURA M. Learning from vector data: enhancing vector-based shape encoding and shape classification for map generalization purposes[J]. Cartography and Geographic Information Science, 2024, 51(1): 146-167.
[4]	GAO Aji, AI Tinghua, YU Huafei, et al. A vector-based coastline shape classification approach using sequential deep learning model[J]. International Journal of Applied Earth Observation and Geoinformation, 2024, 129: 103810.
[5]	YAN Xiongfeng, AI Tinghua, YANG Min, et al. Graph convolutional autoencoder model for the shape coding and cognition of buildings in maps[J]. International Journal of Geographical Information Science, 2021, 35(3): 490-512.
[6]	于洋洋, 贺康杰, 武芳, 等. 面状居民地形状分类的图卷积神经网络方法[J]. 测绘学报, 2022, 51(11): 2390-2402. DOI: . doi: 10.11947/j.AGCS.2022.20210134
	YU Yangyang, HE Kangjie, WU Fang, et al. Graph convolution neural network method for shape classification of areal settlements[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(11): 2390-2402. DOI: . doi: 10.11947/j.AGCS.2022.20210134
[7]	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.
[8]	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.
[9]	DONADIO C, BRESCIA M, RICCARDO A, et al. A novel approach to the classification of terrestrial drainage networks based on deep learning and preliminary results on solar system bodies[J]. Scientific Reports, 2021, 11(1): 5875.
[10]	YAN Xiongfeng, AI Tinghua, YANG Min, et al. A graph convolutional neural network for classification of building patterns using spatial vector data[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 150: 259-273.
[11]	孟妮娜, 王安东, 周校东. 建筑物线型排列模式识别的图卷积神经网络方法[J]. 测绘科学技术学报, 2019, 36(6): 627-631.
	MENG Nina, WANG Andong, ZHOU Xiaodong. A graph convolutional neural network method for pattern recognition of linear building alignment[J]. Journal of Geomatics Science and Technology, 2019, 36(6): 627-631.
[12]	王米琪, 艾廷华, 晏雄锋, 等. 图卷积网络模型识别道路正交网格模式[J]. 武汉大学学报(信息科学版), 2020, 45(12): 1960-1969.
	WANG Miqi, AI Tinghua, YAN Xiongfeng, et al. Grid pattern recognition in road networks based on graph convolution network model[J]. Geomatics and Information Science of Wuhan University, 2020, 45(12): 1960-1969.
[13]	YANG Min, JIANG Chenjun, YAN Xiongfeng, et al. Detecting interchanges in road networks using a graph convolutional network approach[J]. International Journal of Geographical Information Science, 2022, 36(6): 1119-1139.
[14]	LI Chengming, ZHANG Honggang, WU Pengda, et al. A complex junction recognition method based on GoogLeNet model[J]. Transactions in GIS, 2020, 24(6): 1756-1778.
[15]	肖天元, 艾廷华, 余华飞, 等. 地图综合图卷积神经网络点群简化方法[J]. 测绘学报, 2024, 53(1): 158-172. DOI: . doi: 10.11947/j.AGCS.2024.20220584
	XIAO Tianyuan, AI Tinghua, YU Huafei, et al. A point cluster simplification approach of graph convolutional neural network for map generalization[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(1): 158-172. DOI: . doi: 10.11947/j.AGCS.2024.20220584
[16]	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.
[17]	COURTIAL A, TOUYA G, ZHANG Xiang. Deriving map images of generalised mountain roads with generative adversarial networks[J]. International Journal of Geographical Information Science, 2023, 37(3): 499-528.
[18]	COURTIAL A, EL AYEDI A, TOUYA G, et al. Exploring the potential of deep learning segmentation for mountain roads generalisation[J]. ISPRS International Journal of Geo-Information, 2020, 9(5): 338.
[19]	KARSZNIA I, ADOLF A, LEYK S, et al. Using machine learning and data enrichment in the selection of roads for small-scale maps[J]. Cartography and Geographic Information Science, 2024, 51(1): 60-78.
[20]	LEE J, JANG H, YANG J, et al. Machine learning classification of buildings for map generalization[J]. ISPRS International Journal of Geo-Information, 2017, 6(10): 309.
[21]	COURTIAL A, TOUYA G, ZHANG X. Generative adversarial networks to generalise urban areas in topographic maps[J]. ISPRS-International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2021, 43B4: 15-22.
[22]	武芳, 巩现勇, 杜佳威. 地图制图综合回顾与前望[J]. 测绘学报, 2017, 46(10): 1645-1664. DOI: . doi: 10.11947/j.AGCS.2017.20170287
	WU Fang, GONG Xianyong, DU Jiawei. Overview of the research progress in automated map generalization[J]. Acta Geodaetica et Cartographica Sinica, 2017, 46(10): 1645-1664. DOI: . doi: 10.11947/j.AGCS.2017.20170287
[23]	TOUYA G. Social welfare to assess the global legibility of a generalized map[C]//Proceedings of 2012 Geographic Information Science. Berlin: Springer, 2012: 198-211.
[24]	邓红艳, 武芳, 翟仁健, 等. 基于多维约束空间的自动制图综合质量评估模型[J]. 中国矿业大学学报, 2006, 35(5): 667-672.
	DENG Hongyan, WU Fang, ZHAI Renjian, et al. Quality evaluation model of automated cartographical generalization based on a multi-dimensional constraint space[J]. Journal of China University of Mining & Technology, 2006, 35(5): 667-672.
[25]	STOTER J, BURGHARDT D, DUCHÊNE C, et al. Methodology for evaluating automated map generalization in commercial software[J]. Computers, Environment and Urban Systems, 2009, 33(5): 311-324.
[26]	COURTIAL A, TOUYA G, ZHANG X. Constraint-based evaluation of map images generalized by deep learning[J]. Journal of Geovisualization and Spatial Analysis, 2022, 6(1): 13.
[27]	FAN Hongchao, MENG Liqiu, CHENG Tao. Special issue on the recent trends of GeoAI techniques for SDGs[J]. Journal of Geodesy and Geoinformation Science, 2024, 7(1): 2.
[28]	YANG Min, YUAN Tuo, YAN Xiongfeng, et al. A hybrid approach to building simplification with an evaluator from a backpropagation neural network[J]. International Journal of Geographical Information Science, 2022, 36(2): 280-309.
[29]	RUAS A, HOLZAPFEL F. Automatic characterisation of building alignments by means of expert knowledge[C]//Proceedings of the 21st International Cartographic Conference (ICC). Durban: IEEE, 2003: 1604-1616.
[30]	MACEACHREN A M. Compactness of geographic shape: comparison and evaluation of measures[J]. Geografiska Annaler: Series B, Human Geography, 1985, 67(1): 53-67.
[31]	YAN Haowen, WEIBEL R. An algorithm for point cluster generalization based on the Voronoi diagram[J]. Computers & Geosciences, 2008, 34(8): 939-954.
[32]	ZHANG Xiang, AI Tinghua, STOTER J, et al. Building pattern recognition in topographic data: examples on collinear and curvilinear alignments[J]. GeoInformatica, 2013, 17(1): 1-33.
[33]	BORING E G. A history of experimental psychology[M]. New York: Appleton-Century-Crofts, 1950, 1: 284-290.
[34]	GESCHEIDER G A. Psychophysics: the fundamentals[M]. Mahwah: Lawrence Erlbaum Associates, 1997, 10: 53-58.
[35]	MONTELLO D R. Cognitive map-design research in the twentieth century[J]. Cartography and Geographic Information Science, 2002, 29(3), 283-304.
[36]	ZHANG Muhan, CUI Zhicheng, NEUMANN M, et al. An end-to-end deep learning architecture for graph classification[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2018, 544: 4438-4445.

分类	特征名称	度量方法	形式化表示或计算公式	说明
个体特征	形状	紧密度	Tig（m）=4πArea（m）/P²（m）	Area（m）和P（m）分别为建筑物m的面积和周长
	尺寸	建筑物面积	Area（m）	Area（m）和P（m）分别为建筑物m的面积和周长
	位置	建筑物几何中心		P（x_i，y_i）为建筑物顶点坐标
局部特征	方向	建筑物最小外接矩形的最长边与x轴正轴之间的夹角	Direct（m）=Angle（SMBR（m），x）	SMBR（m）为建筑物最小外接矩形的最长边
	方向关系	相邻建筑物的Voronoi方向关系	Dir＿voronoi（m_i，m_j）	m_i、m_j表示相邻的两个建筑物
	距离关系	相邻建筑物的邻近距离	Distance（m_i，m_j）
	中线长	邻近建筑物形心连线中点与建筑物群中心的距离	L（m_i，m_j）
	中线夹角	两个相邻中线构成的夹角	α（L_p，L_q）	L_p、L_q为相邻的中线
	群轮廓	建筑物群凸壳的面积	Area（group）	V（x_i，y_i）为建筑物群凸壳的顶点
	群中心	建筑物群凸壳的几何中心		V（x_i，y_i）为建筑物群凸壳的顶点
整体特征	群平均面积	建筑物群的平均面积		m_i表示群内的每个建筑物，n表示建筑物数量
	群平均周长	建筑物群的平均周长
	群密度	群内建筑物数量与群凸壳面积的比值	Density（group）=n/Area（group）

分类	特征名称	表示	计算公式	说明
个体变化特征	形状变化	C_s	C_s=｜Tig（V）－Tig（O）｜	V表示综合后的建筑物，O表示综合前的建筑物
	尺寸变化	C_a	C_a=｜Area（V）－Area（O）｜
	位置变化	C_p	C_p=｜Loacat（V）－Loacat（O）｜
	方向变化	C_d	C_d=｜Dierct（V）－Direct（O）｜，C_d∈[0°，180°]
局部变化特征	方向关系变化	L_v	L_v=｜Dir＿voronoi（V_i，V_j）－Dir＿voronoi（O_i，O_j）｜，L_v∈[0°，180°]	V_i、V_j表示综合后相邻的建筑物，O_i、O_j表示综合前相邻的建筑物
	距离关系变化	L_d	L_d=｜Distance（V_i，V_j）－Distance（O_i，O_j）｜	V_i、V_j表示综合后相邻的建筑物，O_i、O_j表示综合前相邻的建筑物
	初始距离	L_o	Distance（O_i，O_j）	VL_p、VL_q表示综合后相邻的中线，OL_p、OL_q表示综合前相邻的中线
	中线长变化	L_l	L_l=｜L（V_i，V_j）－L（O_i，O_j）｜
	相邻中线夹角变化	L_α	L_α=｜α（VL_p，VL_q）－α（OL_p，OL_q）｜
整体变化特征	群轮廓变化	G_l	G_l=｜Area（group_v）－Area（group_O）｜	group_v表示综合后的建筑物群组，group_O表示综合前的建筑物群组
	群中心变化	G_c	G_c=｜Center（group_v）－Center（group_O）｜
	群平均面积变化	G_a	G_a=｜Area＿mean（group_v）－Area＿mean（group_O）｜
	群平均周长变化	G_p	G_p=｜Perim＿mean（group_v）－Perim＿mean（group_O）｜
	群分布密度变化	G_d	G_d=｜Density（group_v）－Density（group_O）｜

名称	配置信息
操作系统	Ubuntu 20.04子系统、Windows 11系统
开发语言	Python 3.8.10
框架	Py Torch 1.13.1
GPU	NVIDIA GeForce GTX 1050Ti
CPU	AMD Ryzen 5 2600

建筑物群匹配对的类型		One-hot编码
综合前	综合后	One-hot编码
1∶10 000	1∶20 000	[1，0，0]
1∶10 000	1∶25 000	[0，1，0]
1∶10 000	1∶30 000	[0，0，1]

测试集样本	预测：[1，0，0]	预测：[0，1，0]	预测：[0，0，1]
标签：[1，0，0]	65	3	4
标签：[0，1，0]	4	65	2
标签：[0，0，1]	5	4	57