从图-地地图中学习视觉特征以发现城市形态

IF 7.1 1区 地球科学 Q1 ENVIRONMENTAL STUDIES Computers Environment and Urban Systems Pub Date : 2024-02-03 DOI:10.1016/j.compenvurbsys.2024.102076
Jing Wang , Weiming Huang , Filip Biljecki
{"title":"从图-地地图中学习视觉特征以发现城市形态","authors":"Jing Wang ,&nbsp;Weiming Huang ,&nbsp;Filip Biljecki","doi":"10.1016/j.compenvurbsys.2024.102076","DOIUrl":null,"url":null,"abstract":"<div><p>Most studies of urban morphology<span><span> rely on morphometrics, such as building area and street length. However, these methods often fall short in capturing visual patterns that carry abundant information about the configuration of urban elements and how they interact spatially. In this study, we introduce a novel method for learning morphology features based on figure-ground maps, which leverages recent developments in computer vision. Our method facilitates discovering and comparing urban form types in a fully unsupervised manner. Specifically, we examine building fabrics by 1 km patches. A visual </span>representation learning<span><span> model (SimCLR) casts each patch into a latent embedding space where similar patches are clustered while dissimilar patches are dispelled, thus generating morphology representations that entail the layout of building groups. The learned morphology features are tested in urban form typology clustering and comparison tasks in four diverse cities: Singapore, San Francisco, Barcelona, and Amsterdam, with data sourced from OpenStreetMap. </span>Clustering results show effective identification of typical urban morphology types corresponding to urban functions and historical developments. Further analyses based on the representations reveal inner- and cross-city morphological homogeneity relating to socio-economic drivers. We conclude that this method is a promising alternative for effectively describing urban patterns in morphology analysis.</span></span></p></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"109 ","pages":"Article 102076"},"PeriodicalIF":7.1000,"publicationDate":"2024-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Learning visual features from figure-ground maps for urban morphology discovery\",\"authors\":\"Jing Wang ,&nbsp;Weiming Huang ,&nbsp;Filip Biljecki\",\"doi\":\"10.1016/j.compenvurbsys.2024.102076\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Most studies of urban morphology<span><span> rely on morphometrics, such as building area and street length. However, these methods often fall short in capturing visual patterns that carry abundant information about the configuration of urban elements and how they interact spatially. In this study, we introduce a novel method for learning morphology features based on figure-ground maps, which leverages recent developments in computer vision. Our method facilitates discovering and comparing urban form types in a fully unsupervised manner. Specifically, we examine building fabrics by 1 km patches. A visual </span>representation learning<span><span> model (SimCLR) casts each patch into a latent embedding space where similar patches are clustered while dissimilar patches are dispelled, thus generating morphology representations that entail the layout of building groups. The learned morphology features are tested in urban form typology clustering and comparison tasks in four diverse cities: Singapore, San Francisco, Barcelona, and Amsterdam, with data sourced from OpenStreetMap. </span>Clustering results show effective identification of typical urban morphology types corresponding to urban functions and historical developments. Further analyses based on the representations reveal inner- and cross-city morphological homogeneity relating to socio-economic drivers. We conclude that this method is a promising alternative for effectively describing urban patterns in morphology analysis.</span></span></p></div>\",\"PeriodicalId\":48241,\"journal\":{\"name\":\"Computers Environment and Urban Systems\",\"volume\":\"109 \",\"pages\":\"Article 102076\"},\"PeriodicalIF\":7.1000,\"publicationDate\":\"2024-02-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computers Environment and Urban Systems\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S019897152400005X\",\"RegionNum\":1,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENVIRONMENTAL STUDIES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers Environment and Urban Systems","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S019897152400005X","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL STUDIES","Score":null,"Total":0}
引用次数: 0

摘要

大多数城市形态研究都依赖于形态计量学,如建筑面积和街道长度。然而,这些方法往往无法捕捉到视觉模式,而视觉模式蕴含着丰富的城市元素配置信息,以及它们如何在空间上相互作用。在本研究中,我们利用计算机视觉领域的最新发展,介绍了一种基于图形-地面地图的学习形态特征的新方法。我们的方法有助于以完全无监督的方式发现和比较城市形态类型。具体来说,我们通过 1 千米的斑块来研究建筑结构。一个视觉表征学习模型(SimCLR)将每个补丁投射到一个潜在的嵌入空间,在这个空间中,相似的补丁被聚类,而不相似的补丁则被驱散,从而生成包含建筑群布局的形态表征。学习到的形态特征在四个不同城市的城市形态类型聚类和比较任务中进行了测试:新加坡、旧金山、巴塞罗那和阿姆斯特丹的数据均来自 OpenStreetMap。聚类结果表明,有效识别了与城市功能和历史发展相对应的典型城市形态类型。基于表征的进一步分析表明,城市内部和跨城市的形态同质性与社会经济驱动因素有关。我们的结论是,这种方法是在形态分析中有效描述城市形态的一种有前途的替代方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Learning visual features from figure-ground maps for urban morphology discovery

Most studies of urban morphology rely on morphometrics, such as building area and street length. However, these methods often fall short in capturing visual patterns that carry abundant information about the configuration of urban elements and how they interact spatially. In this study, we introduce a novel method for learning morphology features based on figure-ground maps, which leverages recent developments in computer vision. Our method facilitates discovering and comparing urban form types in a fully unsupervised manner. Specifically, we examine building fabrics by 1 km patches. A visual representation learning model (SimCLR) casts each patch into a latent embedding space where similar patches are clustered while dissimilar patches are dispelled, thus generating morphology representations that entail the layout of building groups. The learned morphology features are tested in urban form typology clustering and comparison tasks in four diverse cities: Singapore, San Francisco, Barcelona, and Amsterdam, with data sourced from OpenStreetMap. Clustering results show effective identification of typical urban morphology types corresponding to urban functions and historical developments. Further analyses based on the representations reveal inner- and cross-city morphological homogeneity relating to socio-economic drivers. We conclude that this method is a promising alternative for effectively describing urban patterns in morphology analysis.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
13.30
自引率
7.40%
发文量
111
审稿时长
32 days
期刊介绍: Computers, Environment and Urban Systemsis an interdisciplinary journal publishing cutting-edge and innovative computer-based research on environmental and urban systems, that privileges the geospatial perspective. The journal welcomes original high quality scholarship of a theoretical, applied or technological nature, and provides a stimulating presentation of perspectives, research developments, overviews of important new technologies and uses of major computational, information-based, and visualization innovations. Applied and theoretical contributions demonstrate the scope of computer-based analysis fostering a better understanding of environmental and urban systems, their spatial scope and their dynamics.
期刊最新文献
Estimating the density of urban trees in 1890s Leeds and Edinburgh using object detection on historical maps The role of data resolution in analyzing urban form and PM2.5 concentration Causal discovery and analysis of global city carbon emissions based on data-driven and hybrid intelligence Editorial Board Exploring the built environment impacts on Online Car-hailing waiting time: An empirical study in Beijing
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1