An entity-level conformal spatial alignment model for multi-source building matching optimization

doi:10.11947/j.AGCS.2025.20250162

Abstract

Abstract:

Accurate matching and integration of multi-source building vector data are crucial for urban spatial analysis and applications. However, the prevalent spatial inconsistency significantly hinders matching accuracy and data fusion quality. Existing spatial alignment methods either adopt a “matching-then-alignment” pipeline that suffers from a circular dependency between alignment quality and matching accuracy and thus fails to fundamentally improve matching conditions, or utilize global/local transformation models that struggle to accurately correct entity-level nonlinear distortions and often introduce geometric deformations that interfere with subsequent morphology-based matching criteria. To address these limitations, this paper proposes a novel entity-level conformal spatial alignment model for multi-source building matching optimization, employing an innovative “vector-raster-vector” collaborative workflow. This approach independently executes prior to the matching process: first, it extracts building centroids to construct a Delaunay triangulation representing spatial structures and rasterizes it; second, it applies a global-local progressive feature matching strategy in the raster domain to efficiently identify high-confidence corresponding points; subsequently, it constructs a continuous displacement field based on reliable correspondences. Crucially, this displacement field drives each source building polygon to undergo an overall rigid translation, accurately correcting positional deviations while strictly preserving their inherent geometric shapes, ultimately ensuring spatial validity through topological conflict detection and resolution mechanisms. Experimental results demonstrate that this method significantly enhances spatial consistency among multi-source data, with an average Hausdorff distance reduction of 18.23% and substantial improvements in F₁ scores ranging from 1.09 to 6.65 percentage points across various downstream matching algorithms. The experiments confirm the effectiveness and application potential of this method as a preprocessing strategy for enhancing the accuracy and quality of multi-source building vector data matching and integration.

Key words: building matching, spatial alignment, vector data, displacement field, vector-raster-vector, spatial correction

CLC Number:

P208

Yue QIU, Fang WU, Renjian ZHAI, Haizhong QIAN, Zhekun HUANG, Bo LI. An entity-level conformal spatial alignment model for multi-source building matching optimization[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(12): 2262-2275.

Figures/Tables 15

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Tab. 1

Tab. 2

Tab. 3

Tab. 4

Fig. 5

Tab. 5

Tab. 6

Fig. 6

Fig. 7

Fig. 8

Fig. 9

References 42

[1]	武芳, 王泽根, 蔡忠亮, 等. 空间数据库原理[M]. 武汉: 武汉大学出版社, 2017: 230-231.
	WU Fang, WANG Zegen, CAI Zhongliang, et al. Spatial database principle[M]. Wuhan: Wuhan University Press, 2017: 230-231.
[2]	杨飞, 华一新, 李响, 等. 基于多粒度时空对象数据模型的城市基础设施建模与管理[J]. 地球信息科学学报, 2021, 23(11): 1984-1997.
	YANG Fei, HUA Yixin, LI Xiang, et al. An urban facilities modeling and management method based on the multi-granularity spatiotemporal object data model[J]. Journal of Geo-information Science, 2021, 23(11): 1984-1997.
[3]	张新长, 华淑贞, 齐霁, 等. 新型智慧城市建设与展望：基于AI的大数据、大模型与大算力[J]. 地球信息科学学报, 2024, 26(4): 779-789.
	ZHANG Xinchang, HUA Shuzhen, QI Ji, et al. Progress and prospects of new smart city construction: AI-based big data, big models and big computing power[J]. Journal of Geo-information Science, 2024, 26(4): 779-789.
[4]	张永军, 张祖勋, 龚健雅. 天空地多源遥感数据的广义摄影测量学[J]. 测绘学报, 2021, 50(1): 1-11. DOI: . doi: 10.11947/j.AGCS.2021.20200245
	ZHANG Yongjun, ZHANG Zuxun, GONG Jianya. Generalized photogrammetry of spaceborne, airborne and terrestrial multi-source remote sensing datasets[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(1): 1-11. DOI: . doi: 10.11947/j.AGCS.2021.20200245
[5]	李德仁. 新基建时代地理信息产业的机遇与挑战[J]. 中国工业和信息化, 2020(12): 52-57.
	LI Deren. Opportunities and challenges of geographic information industry in the new infrastructure era[J]. China Industry & Information Technology, 2020(12): 52-57.
[6]	杨明旺, 赵丽科, 叶林峰, 等. 基于卷积神经网络的遥感影像建筑物提取方法综述[J]. 地球信息科学学报, 2024, 26(6): 1500-1516.
	YANG Mingwang, ZHAO Like, YE Linfeng, et al. A review of convolutional neural networks related methods for building extraction from remote sensing images[J]. Journal of Geo-information Science, 2024, 26(6): 1500-1516.
[7]	DOYTSHER Y. A rubber sheeting algorithm for non-rectangular maps[J]. Computers & Geosciences, 2000, 26(9/10): 1001-1010.
[8]	WEI R, MURRAY A T. An integrated approach for addressing geographic uncertainty in spatial optimization[J]. International Journal of Geographical Information Science, 2012, 26(7): 1231-1249.
[9]	赵东保, 盛业华, 张卡. 利用几何矩和叠置分析进行多尺度面要素自动匹配[J]. 武汉大学学报(信息科学版), 2011, 36(11): 1371-1375.
	ZHAO Dongbao, SHENG Yehua, ZHANG Ka. An algorithm for muti-scale one-to-many areal feature matching based on geometry moments and overly analysis[J]. Geomatics and Information Science of Wuhan University, 2011, 36(11): 1371-1375.
[10]	刘凌佳. 多尺度面实体匹配方法及其融合应用研究[D]. 武汉: 武汉大学, 2018.
	LIU Lingjia. Research on methods and integration applications of polygonal object matching on multi-scale datasets[D]. Wuhan: Wuhan University, 2018.
[11]	CHEN S, SHI W, ZHOU M, et al. CDasXORNet: change detection of buildings from bi-temporal remote sensing images as an XOR problem[J]. International Journal of Applied Earth Observation and Geoinformation, 2024, 130(1): 103836.
[12]	TONG Xiaohua, LIANG Dan, JIN Yanmin. A linear road object matching method for conflation based on optimization and logistic regression[J]. International Journal of Geographical Information Science, 2014, 28(4): 824-846.
[13]	WANG Yanxia, CHEN Deng, ZHAO Zhiyuan, et al. A back-propagation neural network-based approach for multi-represented feature matching in update propagation[J]. Transactions in GIS, 2015, 19(6): 964-993.
[14]	李德仁, 龚健雅, 张桥平. 论地图数据库合并技术[J]. 测绘科学, 2004, 29(1): 1-4.
	LI Deren, GONG Jianya, ZHANG Qiaoping. On the conflation of geographic databases[J]. Science of Surveying and Mapping, 2004, 29(1): 1-4.
[15]	张新长, 何显锦, 孙颖, 等. 多尺度空间数据联动更新技术研究现状及展望[J]. 测绘学报, 2022, 51(7): 1520-1535. DOI: . doi: 10.11947/j.AGCS.2022.20220095
	ZHANG Xinchang, HE Xianjin, SUN Ying, et al. Advance and future development of the multi-scale spatial data linkage updating[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(7): 1520-1535. DOI: . doi: 10.11947/j.AGCS.2022.20220095
[16]	SCHORCHT M, HECHT R, MEINEL G. Comparative study on matching methods for the distinction of building modifications and replacements based on multi-temporal building footprint data[J]. ISPRS International Journal of Geo-Information, 2022, 11(2): 91-107.
[17]	DUAN Weiwei, CHIANG Yaoyi, LEYK S, et al. Automatic alignment of contemporary vector data and georeferenced historical maps using reinforcement learning[J]. International Journal of Geographical Information Science, 2020, 34(4): 824-849.
[18]	翟仁健. 基于全局一致性评价的多尺度矢量空间数据匹配方法研究[D]. 郑州: 信息工程大学, 2011.
	ZHAI Renjian. Research on automated matching methods for multi-scale vector spatial data based on global consistency evaluation[D]. Zhengzhou: Information Engineering University, 2011.
[19]	蓝振家. 顾及邻域相似性的面实体匹配最优化方法[D]. 武汉: 武汉大学, 2019.
	LAN Zhenjia. Optimization method for areal feature matching considering context-dependent similarity[D]. Wuhan: Wuhan University, 2019.
[20]	朱欣焰, 刘凌佳. 一种用于空间数据整合的建筑物面实体对齐方法[J]. 武汉大学学报(信息科学版), 2018, 43(12): 2269-2277.
	ZHU Xinyan, LIU Lingjia. A building polygonal object alignment approach for spatial data integration[J]. Geomatics and Information Science of Wuhan University, 2018, 43(12): 2269-2277.
[21]	张付兵, 温伯威, 郭丽萍, 等. 基于机器学习的多源矢量同名面实体几何不一致性识别和处理方法[J]. 测绘科学技术学报, 2024, 40(4): 411-417.
	ZHANG Fubing, WEN Bowei, GUO Liping, et al. A method for recognizing and processing geometric inconsistencies in multi-source identical polygonal vector entities based on machine learning[J]. Journal of Geomatics Science and Technology, 2024, 40(4): 411-417.
[22]	KOVALERCHUK B, DOUCETTE P, SEEDAHMED G, et al. Automated vector-to-raster image registration[C]//Proceedings of 2008 Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XIV. Orlando: SPIE, 2008: 69660W.
[23]	杜永葛, 霍亮, 程宏宇, 等. 矢量数据校正工具集的设计与实现[J]. 矿山测量, 2017, 45(1): 9-13.
	DU Yongge, HUO Liang, CHENG Hongyu, et al. Design and implementation of vector data correction toolkit[J]. Mine Survey, 2017, 45(1): 9-13.
[24]	王艳东, 邵鑫, 刘波, 等. 一种利用Mask R-CNN的遥感影像与矢量数据配准方法[J]. 武汉大学学报(信息科学版), 2022, 47(4): 623-631.
	WANG Yandong, SHAO Xin, LIU Bo, et al. A registration method of remote sensing image and vector data using mask R-CNN[J]. Geomatics and Information Science of Wuhan University, 2022, 47(4): 623-631.
[25]	RUIZ-LENDÍNEZ J J, MAĆKIEWICZ B, MOTEK P, et al. Method for an automatic alignment of imagery and vector data applied to cadastral information in Poland[J]. Survey Review, 2019, 51(365): 123-134.
[26]	YUAN Lang, LI Yuxia, YANG Chao, et al. Road vectorization based on image pixel tracking and attribute matching method[C]//Proceedings of 2020 IEEE International Geoscience and Remote Sensing Symposium. Waikoloa: IEEE, 2020: 3199-3202.
[27]	FAN Hongchao, ZIPF A, FU Qing, et al. Quality assessment for building footprints data on OpenStreetMap[J]. International Journal of Geographical Information Science, 2014, 28(4): 700-719.
[28]	郝燕玲, 唐文静, 赵玉新, 等. 基于空间相似性的面实体匹配算法研究[J]. 测绘学报, 2008, 37(4): 501-506.
	HAO Yanling, TANG Wenjing, ZHAO Yuxin, et al. Areal feature matching algorithm based on spatial similarity[J]. Acta Geodaetica et Cartographica Sinica, 2008, 37(4): 501-506.
[29]	汪汇兵, 唐新明, 邱博, 等. 运用多算子加权的面要素几何匹配方法[J]. 武汉大学学报(信息科学版), 2013, 38(10): 1243-1247.
	WANG Huibing, TANG Xinming, QIU Bo, et al. Geometric matching method of area feature based on multi-weighted operators[J]. Geomatics and Information Science of Wuhan University, 2013, 38(10): 1243-1247.
[30]	LIU Lingjia, DING Xiaohui, ZHU Xinyan, et al. An iterative approach based on contextual information for matching multi-scale polygonal object datasets[J]. Transactions in GIS, 2020, 24(4): 1047-1072.
[31]	郭黎, 崔铁军, 郑海鹰, 等. 基于空间方向相似性的面状矢量空间数据匹配算法[J]. 测绘科学技术学报, 2008(5): 380-382.
	GUO Li, CUI Tiejun, ZHENG Haiying, et al. Arithmetic for area vector spatial data matching on spatial direction similarity[J]. Journal of Geomatics Science and Technology, 2008(5): 380-382.
[32]	邵世维. 基于几何特征的多尺度矢量面状实体匹配方法研究与应用[D]. 武汉: 武汉大学, 2012.
	SHAO Shiwei. Researches and applications on polygon entity matching for multi-scale vector data based on geometric features[D]. Wuhan: Wuhan University, 2012.
[33]	POTJE G, CADAR F, ARAUJO A, et al. XFeat: accelerated features for lightweight image matching[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024: 2682-2691.
[34]	LOWE D G. Object recognition from local scale-invariant features[C]//Proceedings of the 7th IEEE International Conference on Computer Vision. Kerkyra: IEEE, 1999: 1150-1157.
[35]	BAY H, TUYTELAARS T, VAN GOOL L. SURF: speeded up robust features[M]//Computer Vision-ECCV 2006. Berlin: Springer, 2006: 404-417.
[36]	RUBLEE E, RABAUD V, KONOLIGE K, et al. ORB: an efficient alternative to SIFT or SURF[C]//Proceedings of 2011 International Conference on Computer Vision. Barcelona: IEEE, 2011: 2564-2571.
[37]	XU Shibiao, CHEN Shunpeng, XU Rongtao, et al. Local feature matching using deep learning: a survey[J]. Information Fusion, 2024, 107: 102344.
[38]	BALAKRISHNAN G, ZHAO A, SABUNCU M R, et al. VoxelMorph: a learning framework for deformable medical image registration[J]. IEEE Transactions on Medical Imaging, 2019, 38(8): 1788-1800.
[39]	MENG Mingyuan, BI Lei, FENG Dagan, et al. Non-iterative coarse-to-fine registration based on single-pass deep cumulative learning[M]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2022. Cham: Springer Nature Switzerland, 2022: 88-97.
[40]	MENG Mingyuan, BI Lei, FULHAM M, et al. Non-iterative coarse-to-fine Transformer networks for joint affine and deformable image registration[M]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2023. Cham: Springer Nature Switzerland, 2023: 750-760.
[41]	AI Tinghua, ZHANG Xiang, ZHOU Qi, et al. A vector field model to handle the displacement of multiple conflicts in building generalization[J]. International Journal of Geographical Information Science, 2015, 29(8): 1310-1331.
[42]	XU Yongyang, LI Jun, XIE Xuejing, et al. Matching the building footprints of different vector spatial datasets at a similar scale based on one-class support vector machines[J]. International Journal of Geographical Information Science, 2024, 38(8): 1555-1582.

指标	定义	参数	说明
互信息（MI）	MI（A，B）=H（A）+H（B）－H（A，B）	H（•）为信息熵，A、B为标准化后的栅格图像	度量信息共享程度，值越大表明分布协同性越强
均方误差（MSE）		M、N为图像尺寸	直接反映像素级差异，值越小表明对齐精度越高
峰值信噪比（PSNR）		MAX为图像最大值	对数尺度敏感度指标，值越大表明对齐质量越高
结构相似性指数（SSIM）		μ为均值，σ为方差，C₁、C₂均为稳定性常数	综合亮度、对比度、结构相似性，越接近1越好
归一化互相关（NCC）		μ_A、μ_B均为图像均值	衡量线性相关性，值越接近1表明分布模式越相似
Dice系数		为像素集合的基数	量化空间重叠度，值越接近1表明重叠区域越大

指标	参数	说明
查准率（Precision）	TP：真正例 FP：假正例	度量匹配准确性，值越大误匹配越少
查全率（Recall）	FN：假负例	度量匹配完整性，值越大漏匹配越少
F₁值（F₁score）	Precision：查准率 Recall：查全率	综合准确性与完整性，值越大匹配性能越优

指标	对齐前	对齐后	相对提升量/（%）	绝对提升量
MSE	0.149 9	0.116 7	22.15	0.033 2
NCC	0.545 9	0.647 6	18.63	0.101 7
PSNR	8.240 9	9.327 9	13.19	1.087 0
MI	0.184 6	0.261 3	41.55	0.076 7
SSIM	0.779 3	0.816 0	4.71	0.036 7
DICE	0.640 5	0.721 2	12.60	0.080 7
Hausdorff	100.121 4	88.763 0	11.34	11.358 4
平均Hausdorff	8.660 2	7.081 6	18.23	1.578 6

指标	对齐前	本文方法		人工标记1对		人工标记5对		人工标记9对
指标	对齐前	对齐后	相对提升量/（%）	对齐后	相对提升量/（%）	对齐后	相对提升量/（%）	对齐后	相对提升量/（%）
MSE	0.149 9	0.116 7	22.15	0.151 0	－0.73	0.141 3	5.74	0.136 4	9.01
NCC	0.545 9	0.647 6	18.63	0.542 9	－0.55	0.574 0	5.15	0.588 4	7.79
PSNR	8.240 9	9.327 9	13.19	8.210 8	－0.37	8.498 9	3.13	8.652 0	4.99
MI	0.184 6	0.261 3	41.55	0.182 5	－1.14	0.204 6	10.83	0.214 9	16.41
SSIM	0.779 3	0.816 0	4.71	0.778 4	－0.12	0.788 8	1.22	0.793 8	1.86
DICE	0.640 5	0.721 2	12.60	0.638 1	－0.37	0.663 3	3.56	0.674 5	5.31
Hausdorff	100.121 4	88.763 0	11.34	88.206 1	11.90	87.080 3	13.03	87.080 3	13.03
平均Hausdorff	8.660 2	7.081 6	18.23	8.415 0	2.83	8.505 4	1.79	7.990 5	7.73

匹配方法	对齐前			对齐后			绝对提升量
匹配方法	查准率	查全率	F₁值	查准率	查全率	F₁值	查准率	查全率	F₁值
位置法	60.29	76.46	67.42	66.72	81.93	73.55	6.43	5.47	6.13
重叠法	71.74	83.39	77.13	79.83	88.14	83.78	8.09	4.75	6.65
3指标法	77.44	89.60	83.08	79.23	89.78	84.17	1.79	0.18	1.09
6指标法	64.74	79.74	71.46	71.60	85.58	77.97	6.86	5.84	6.51