Determining the optimal generalization operators for building footprints using an improved graph neural network model

References

• Abbaspour RA, Chehreghan A, Chamani M. 2021. Multi-scale polygons matching using a new geographic context descriptor. Appl Geomat. 13(4):885–899. doi:10.1007/s12518-021-00396-x.
   Web of Science ®Google Scholar
• Abhilash B, Busari E, Syranidou C, Linssen J, Stolten D. 2022. Classification of building types in Germany: a data-driven modeling approach. Data. 7(4):45. doi:10.3390/data7040045.
   Web of Science ®Google Scholar
• Ai T, Cheng X, Liu P, Yang M. 2013. A shape analysis and template matching of building features by the Fourier Transform Method. Comp Environ Urban Syst. 41:219–233. doi:10.1016/j.compenvurbsys.2013.07.002.
   Web of Science ®Google Scholar
• Ai T. 2021. Some thoughts on deep learning enabling cartography. Acta Geodaetica et Cartograhica Sinica. 50:1170–1182. doi:10.11947/j.AGCS.2021.20210091.
   Google Scholar
• Altmann A, Toloşi L, Sander O, Lengauer T. 2010. Permutation importance: a corrected feature importance measure. Bioinformatics. 26(10):1340–1347. doi:10.1093/bioinformatics/btq134.
   PubMed Web of Science ®Google Scholar
• An X, Zhu Y, Yan X. 2023. Building selection method supported by convolutional neural network. Acta Geodaetica et Cartographica Sinica. 52:1574–1583. doi:10.11947/j.AGCS.2023.20220216.
   Google Scholar
• Balboa JL, Ariza-López FJ. 2008. Generalization-oriented road line classification by means of an artificial neural network. GeoInformatica. 12:289–312. doi:10.1007/s10707-007-0026-z.
   Web of Science ®Google Scholar
• Basaraner M, Cetinkaya S. 2017. Performance of shape indices and classification schemes for characterising perceptual shape complexity of building footprints in GIS. International Journal of Geographical Information Science. 31(10):1952–1977. doi:10.1080/13658816.2017.1346257.
   Web of Science ®Google Scholar
• Bouvrie JV. 2006. Notes on convolutional neural networks. Cambridge, MA: Massachusetts Institute of Technology.
   Google Scholar
• Chen J, Chen H. 2021. Edge-featured graph attention network. arXiv. preprint arXiv210107671 doi:10.48550/arXiv.2101.07671.
   Google Scholar
• Cheng B, Liu Q, Li X. 2015. Local perception-based intelligent building outline aggregation approach with back propagation neural network. Neural Process Lett. 41(2):273–292. doi:10.1007/s11063-014-9345-x.
   Web of Science ®Google Scholar
• Courtial A, Ayedi AE, Touya G, Zhang X. 2020. Exploring the Potential of deep learning segmentation for mountain roads generalisation. IJGI. 9(5):338. doi:10.3390/ijgi9050338.
   Google Scholar
• Courtial A, Touya G, Zhang X. 2021a. Can graph convolution networks learn spatial relations? Abstr Int Cartogr Assoc. 3:1–2. doi:10.5194/ica-abs-3-60-2021.
   Google Scholar
• Courtial A, Touya G, Zhang X. 2021b. Generative adversative networks to generalize urban areas in topographic maps. In: XXIV ISPRS Congress (2021 Edition). p. 15–22. nice (en ligne), france. doi:10.5194/isprs-archives-XLIII-B4-2021-15-2021. Hal-03306595
   Google Scholar
• Courtial A, Touya G, Zhang X. 2022. Representing vector geographic information as a tensor for deep learning based map generalisation. AGILE GIScience Ser. 3:1–8. doi:10.5194/agile-giss-3-32-2022.
   Google Scholar
• Deng M, Tang J, Liu Q, Wu F. 2018. Recognizing building groups for generalization: a comparative study. Cartogr Geogr Inform Sci. 45(3):187–204. doi:10.1080/15230406.2017.1302821.
   Web of Science ®Google Scholar
• Duchêne C, Touya G, Taillandier P, Gaffuri J, Ruas A, Renard J. 2018. Multi-agents systems for cartographic generalization: feedback from past and on-going research. PhD dissertation, Institut National de l’Information Géographique et Forestière doi:10.13140/RG.2.2.35489.92006.
   Google Scholar
• Feng Y, Thiemann F, Sester M. 2019. Learning cartographic building generalization with deep convolutional neural networks. IJGI. 8(6):258. doi:10.3390/ijgi8060258.
   Google Scholar
• Fiedukowicz A. 2020. The role of spatial context information in the generalization of geographic information: using reducts to indicate relevant attributes. IJGI. 9(1):37. doi:10.3390/ijgi9010037.
   Google Scholar
• Galkin F, Aliper A, Kuznetsov I, Gladyshev VN, Zhavoronkov A. 2018. Human microbiome aging clocks based on deep learning and tandem of permutation feature importance and accumulated local effects. bioRxiv. doi:10.1016/j.isci.2020.101199.
   Google Scholar
• Gong L, Cheng Q. 2019. Exploiting edge features for graph neural networks. arXiv. Preprint. arXiv1809.02709: 9203–9211 doi:10.48550/arXiv.1809.02709.
   Google Scholar
• Hu Y, Liu C, Li Z, Xu J, Han Z, Guo J. 2022. Few-shot building footprint shape classification with relation network. IJGI. 11(5):311. doi:10.3390/ijgi11050311.
   Google Scholar
• Karsznia I, Sielicka K. 2020. When traditional selection fails: how to improve settlement selection for small-scale maps using machine learning. IJGI. 9(4):230. doi:10.3390/ijgi9040230.
   Google Scholar
• Karsznia I, Wereszczyńska K, Weibel R. 2022. Make it simple: effective road selection for small-scale map design using decision-tree-based models. IJGI. 11(8):457. doi:10.3390/ijgi11080457.
   Google Scholar
• Kipf T, Welling M. 2017. Semi-supervised classification with graph convolutional networks. arXiv. preprint. arXiv1609.02907. doi:10.48550/arXiv.1609.02907.
   Google Scholar
• Lamy S, Ruas A, Demazeu Y, Jackson M, Mackaness WA, Weibel R. 1999. The application of agents in automated map generalization. In: Proceedings of the 19th Ica Meeting. Ottawa, Canada 14, p. 160–169.
   Google Scholar
• Lee J, Jang H, Yang J, Yu K. 2017. Machine learning classification of buildings for map generalization. IJGI. 6(10):309. doi:10.3390/ijgi6100309.
   Google Scholar
• Li C, Wu W, Yin Y, Wu P, Wu Z. 2022. A multi‐scale partitioning and aggregation method for large volumes of buildings considering road networks association constraints. Transat GIS. 26(2):779–798. doi:10.1111/tgis.12885.
   Web of Science ®Google Scholar
• Li Y, Lu X, Yan H, Wang W, Li P. 2022. A skeleton-line-based graph convolutional neural network for areal settlements: shape classification. Appl Sci. 12(19):10001. doi:10.3390/app121910001.
   Google Scholar
• Li Z, Yan HW, Ai T, Chen J. 2004. Automated building generalization based on urban morphology and gestalt theory. Inter J Geogr Inform Sci. 18(5):513–534. doi:10.1080/13658810410001702021.
   Web of Science ®Google Scholar
• Li Z. 2007. Digital map generalization at The Age of Enlightenment: a review of the first forty years. The Cartographic Journal. 44(1):80–93. doi:10.1179/000870407X173913.
   Web of Science ®Google Scholar
• Lin T, Goyal P, Girshick R, He K, Dollár P, Facebook AI, Research FAIR. 2017. Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell. 42(2):318–327. doi:10.1109/TPAMI.2018.2858826.
   Google Scholar
• Liqiang Z, Hao D, Dong C, Zhen W. 2013. A spatial cognition-based urban building clustering approach and its applications. Inter J Geogr Inform Sci. 27(4):721–740. doi:10.1080/13658816.2012.700518.
   Web of Science ®Google Scholar
• Liu C, Qian H, Wang X, He H, Xie L, Wang C. 2016. Linkage matching method of road and residential land using urban skeleton line network. c. 45:1485–1494. doi:10.11947/j.AGCS.2016.20160221.
   Google Scholar
• Liu C, Qian H, He H, Wang X, Xie L. 2016. A Construction Method of Road and Residence Correlation Based on Urban Skeleton Network. In: Yuan H, Geng J, Bian F, editors. geo-spatial knowledge and intelligence. GRMSE 2016. Communications in computer and information science, 699. Singapore: Springer.
   Google Scholar
• Liu C, Zhai R, Qian H, Gong X, Wang A, Wu F. 2023. Identification of drainage patterns using a graph convolutional neural network. Transactions in GIS. 27(3):752–776. doi:10.1111/tgis.13041.
   Web of Science ®Google Scholar
• Lyu Z, Sun Q, Ma J, Xu Q, Li Y, Zhang F. 2022. Road network generalization method constrained by residential areas. IJGI. 11(3):159. doi:10.3390/ijgi11030159.
   Google Scholar
• MacEachren AM. 1985. Compactness of geographic shape: comparison and Evaluation of measures. Geografiska Annaler: Series B, Human Geography. 67(1):53–67. doi:10.2307/490799.
   Web of Science ®Google Scholar
• Peura M, Iivarinen J. 1997. Efficiency of simple shape descriptors. In: Cordella LP, Arcelli C, Sanniti di Baja G, editors. Advances in Visual Form Analysis: proceedings of the 3rd International Workshop on Visual Form. Singapore: World Scientific. p. 443–451.
   Google Scholar
• Rieger MK, Coulson MR. 1993. Consensus or confusion: cartographers’ knowledge of generalization. Cartographica. 30(2-3):69–80. doi:10.3138/M6H4-1006-6422-H744.
   Google Scholar
• Rosin PL. 2000. Measuring shape: ellipticity, rectangularity, and triangularity. Machine Vision and Applications. 14(3):172–184. doi:10.1007/s00138-002-0118-6.
   Google Scholar
• Ruas A, Duchene C. 2007. A prototype generalisation system based on the multi-agent system paradigm. In: Mackaness WA, Ruas A, Sarjakoski LT, editors. Generalisation of Geographic Information. Amsterdam: Elsevier. p. 269–284.
   Google Scholar
• Sester M, Feng Y, Thiemann F. 2018. Building generalization using deep learning. Int Arch Photogramm Remote Sens Spatial Inf Sci. XLII-4:565–572. doi:10.5194/isprs-archives-XLII-4-565-2018.
   Google Scholar
• Shea KS, Master RB. 1989. Cartographic generalization in a digital environment: when and how to generalize. In: International Symposium on Computer-Assisted Cartography. p. 56–67.
   Google Scholar
• Shuai Y, Shuai H, Ni L. 2007. Polygon cluster pattern recognition based on new visual distance. In: Proceedings Volume 6753, Geoinformatics 2007: Geospatial Information Science, Nanjing, China; 675316. doi:10.1117/12.761778.
   Google Scholar
• Smilkov D, Thorat N, Kim B, Viégas F, Wattenberg M. 2017. SmoothGrad: removing noise by adding noise. ArXiv. Preprints. ArXiv abs/1706.03825. doi:10.48550/arXiv.1706.03825.
   Google Scholar
• Steiniger S, Taillandier P, Weibel R. 2010. Utilising urban context recognition and machine learning to improve the generalisation of buildings. Inter J Geogr Inform Sci. 24(2):253–282. doi:10.1080/13658810902798099.
   Web of Science ®Google Scholar
• Susetyo DB, Hidayat F. 2019. Specification of map generalization from large scale to small scale based on existing data. IOP Conf Ser Earth Environ Sci. 280(1):012026. doi:10.1088/1755-1315/280/1/012026.
   Google Scholar
• Touya G, Zhang X, Lokhat I. 2019. Is deep learning the new agent for map generalization? Inter J Cartogr. 5(2-3):142–157. doi:10.1080/23729333.2019.1613071.
   Web of Science ®Google Scholar
• Touya G. 2021. Multi-criteria geographic analysis for automated cartographic generalization. The Cartographic Journal. 59(1):18–34. doi:10.1080/00087041.2020.1858608.
   Web of Science ®Google Scholar
• Touya G, Lokhat I. 2020. Deep learning for enrichment of vector spatial databases. ACM Trans Spatial Algorithms Syst. 6(3):1–21. doi:10.1145/3382080.
   Web of Science ®Google Scholar
• Touya G, Duchêne C, Ruas A. 2010. Collaborative generalisation: formalisation of generalisation knowledge to orchestrate different cartographic generalisation processes. In: Geographic Information Science: 6th International Conference, GIScience 2010, Zurich, Switzerland, September 14–17, 2010. Proceedings 6, Springer, Berlin. p. 264–278.
   Google Scholar
• Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. 2018. Graph attention networks. ArXiv. preprint. ArXiv abs/1710.10903. doi:10.48550/arXiv.1710.10903.
   Google Scholar
• Wang W, Du S, Guo Z, Luo L. 2015. Polygonal clustering analysis using multilevel graph‐partition. Transactions in GIS. 19(5):716–736. doi:10.1111/tgis.12124.
   Web of Science ®Google Scholar
• Wang X, Qian H, He H, Chen J, Hu H. 2015. Matching multi-source residential land by blank area skeleton line mesh. Acta Geodaetica et Cartographica Sinica. 44:927–935. doi:10.11947/j.AGCS.2015.20140462.
   Google Scholar
• Wang Z, Chen J, Chen H. 2021. EGAT: edge-featured graph attention network. In: Farkaš I, Masulli P, Otte S, Wermter S, editors. Artificial neural networks and machine learning – ICANN 2021. ICANN 2021. Lecture Notes in Computer Science. Cham: Springer International Publishing. p. 12891.
   Google Scholar
• Ware JM, Jones CB, Thomas N. 2003. Automated map generalization with multiple operators: a simulated annealing approach. Inter J Geograph Inform Sci. 17(8):743–769. doi:10.1080/13658810310001596085.
   Web of Science ®Google Scholar
• Wu F, Du J, Qian H, Zhai R. 2022. The development and thinking of map comprehensive intelligence research. Wuhan Daxue Xuebao. 47:1675–1687.
   Google Scholar
• Xie L, Qian H, He H, Liu C, Duan P. 2017. Settlement selection method based on case reasoning. Acta Geodaetica et Cartographica Sinica. 46:1910–1918. doi:10.11947/j.AGCS.2017.20170061.
   Google Scholar
• Xu K, Hu W, Leskovec J, Jegelka S. 2018. How powerful are graph neural networks? ArXiv. preprints. ArXiv abs/1810.00826. doi:10.48550/arXiv.1810.00826.
   Google Scholar
• Yan H,Weibel R,Yang B. 2008. A multi-parameter approach to automated building grouping and generalization. Geoinformatica. 12(1):73–89. doi:10.1007/s10707-007-0020-5.
   Web of Science ®Google Scholar
• Yan X, Ai T, Yang M, Yin H. 2019. A Graph convolutional neural network for classification of building patterns using spatial vector data. ISPRS J Photogr Remote Sens. 150:259–273. doi:10.1016/j.isprsjprs.2019.02.010.
   Web of Science ®Google Scholar
• Yan X, Ai T, Yang M, Tong X, Liu Q. 2020. A graph deep learning approach for urban building grouping. Geocarto Inter. 37(10):2944–2966. doi:10.1080/10106049.2020.1856195.
   Web of Science ®Google Scholar
• Yan X, Chen H, Huang H, Liu Q, Yang M. 2021. Building typification in map generalization using affinity propagation clustering. IJGI. 10(11):732. doi:10.3390/ijgi10110732.
   Google Scholar
• Yan X, Yuan T, Yang M. 2022. Building shape characterization representation and adaptive simplification methods. Acta Geodaetica et Cartographica Sinica. 51:269–278. doi:10.11947/j.AGCS.2022.20210302.
   Google Scholar
• Yang M, Yuan T, Yan X, Ai T, Jiang C. 2022. A hybrid approach to building simplification with an evaluator from a backpropagation neural network. Inter J Geograph Inform Sci. 36(2):280–309. doi:10.1080/13658816.2021.1873998.
   Web of Science ®Google Scholar
• Yang Y, Li D. 2020. NENN: incorporate Node and Edge Features in Graph Neural Networks. In: Proceedings of the 12th Asian Conference on Machine Learning, PMLR129, p. 593–608.
   Google Scholar
• Zaremba W, Sutskever I, Vinyals O. 2014. Recurrent neural network regularization. ArXiv. Preprints. ArXiv abs/1409.2329. doi:10.48550/arXiv.1409.2329.
   Google Scholar
• Zhang X, Ai T, Stoter J. 2008. The Evaluation of Spatial Distribution Density in Map Generalization. In: The XXI Congress of the International Society for Photogrammetry and Remote Sensing (ISPRS 2008) p. 181–188.
   Google Scholar
• Zhang X, Ai T, Stoter J, Kraak M-J, Molenaar M. 2011. Building pattern recognition in topographic data: examples on collinear and curvilinear alignments. GeoInformatica. 17(1):1–33. doi:10.1007/s10707-011-0146-3.
   Web of Science ®Google Scholar
• Zhang X, Yin W, Yang M, Ai T, Stoter J. 2018. Updating authoritative spatial data from timely sources: a multiple representation approach. Int J Appl Earth Obs Geoinf. 72:42–56. doi:10.1016/j.jag.2018.05.022.
   Web of Science ®Google Scholar
• Zhang Z, Liu T, Du P, Yang G. 2022. DGCNN recognition method of spatial map convolutional model of typical building group pattern. Wuhan Daxue Xuebao. 16:11–13. 1–
   Google Scholar
• Zhao L, Akoglu L. 2019. PairNorm: tackling Oversmoothing in GNNs. ArXiv. preprints. arXiv1909.12223. doi:10.48550/arXiv.1909.12223.
   Google Scholar
• Zhou S, Regnauld N, Roensdorf C. 2009. Generalisation log for managing and utilising a multi-representation spatial database in map production. Comput Environ Urban Syst. 33(5):334–348. doi:10.1016/j.compenvurbsys.2009.06.004.
   Web of Science ®Google Scholar