Computers Environment and Urban Systems最新文献

{"title":"Multi-frequency street-level urban noise modeling and mapping through street view and remote sensing image fusion","authors":"Yan Zhang ,&nbsp;Entong Ke ,&nbsp;Mei-Po Kwan ,&nbsp;Libo Fang ,&nbsp;Mingxiao Li","doi":"10.1016/j.compenvurbsys.2026.102401","DOIUrl":"10.1016/j.compenvurbsys.2026.102401","url":null,"abstract":"<div><div>Urban noise pollution has become the third most significant environmental health threat following air and water pollution, while traditional noise modeling methods suffer from limitations including high costs, limited coverage, and an exclusive focus on total decibel values while neglecting frequency characteristics. This study proposes a method that combines street view imagery (SVI) and remote sensing imagery (RSI) to achieve precise modeling and mapping of multi-frequency noise exposure at the urban street scale. Using Xiangzhou District, Zhuhai City as a case study, we utilized approximately 6000 street view images and corresponding remote sensing images, and recorded 35,276 street noise audios containing 23 frequency bands (100 Hz-16,000 Hz) through volunteer cycling surveys. A multi-source fusion model was constructed based on a pre-trained vision transformer architecture, with 923 valid street noise-image paired samples used for training and validation. The sensitivity results demonstrate that: (1) the proposed multimodal fusion model achieves high predictive accuracy, with R² values for dBA prediction ranging from 0.417 to 0.649, with particularly higher accuracy observed for mid-frequency noise prediction; (2) 50-m resolution street-scale multi-frequency soundscape maps were successfully generated, providing scientific evidence for refined urban noise management; (3) explainable machine learning models revealed that buildings, roads, sidewalks, and terrain visual elements are the four most important factors affecting noise prediction, with road width showing a positive association with street noise levels. This study not only fills the gap in urban noise frequency characteristics research but also provides new methodological support for precise street-level noise pollution modeling and health-oriented urban planning. The source code is available at <span><span>https://github.com/giserzy/NoisePrediction</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"126 ","pages":"Article 102401"},"PeriodicalIF":8.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Urban morphology as a proxy for housing and infrastructure inequality: A machine learning approach using open building footprint data","authors":"Cassiano Bastos Moroz,&nbsp;Annegret H. Thieken","doi":"10.1016/j.compenvurbsys.2026.102402","DOIUrl":"10.1016/j.compenvurbsys.2026.102402","url":null,"abstract":"<div><div>Mapping spatial inequalities remains a major challenge, particularly in rapidly urbanizing regions. Although urban morphology offers valuable insights into the built environment, the extent to which it can serve as a proxy for urban inequalities remains underexplored. This study evaluates the potential of morphological indicators to reflect dimensions of urban precarity, which in this study refer exclusively to housing and infrastructure conditions. Using São Sebastião, Brazil, as a case study, we trained a random forest model on officially delineated slum locations, using indicators derived from Google Open Buildings, an open-access building footprint dataset, as predictors. The model achieved high accuracy in distinguishing slums from non-slums (AUC of 0.89), with over 90% of slum cells classified as either highly or very highly precarious. Validation with field observations and census data confirmed that the mapped precarity classes consistently correspond to observed conditions. Urban cells classified as more precarious are associated with smaller buildings, narrower and unpaved streets, less durable roof materials, and reduced access to basic infrastructure such as piped water, sewage, and garbage collection. These consistent gradients across precarity levels suggest that urban form is, to a significant extent, associated with these housing and infrastructure conditions. However, despite the scalability and reproducibility of the proposed approach, limitations persist, particularly in morphologically complex urban environments, where local knowledge and more advanced datasets may be necessary. Overall, this study provides evidence that urban morphological indicators can approximate key dimensions of urban precarity, especially those related to housing and infrastructure, even if they do not directly measure them.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"126 ","pages":"Article 102402"},"PeriodicalIF":8.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146006779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digitally mediated accessibility: A metric combining human perception and generative AI","authors":"Mingzhi Zhou ,&nbsp;Yuling Yang","doi":"10.1016/j.compenvurbsys.2025.102391","DOIUrl":"10.1016/j.compenvurbsys.2025.102391","url":null,"abstract":"<div><div>Accessibility metrics often fail to align with actual human behavior due to incomplete spatial knowledge and perceptual biases. The digital era has intensified this gap. Platforms like real-time navigation and social media fundamentally reshape how people acquire information and perceive their spatial options. However, conventional accessibility metrics overlook this digital mediation and struggle to capture large-scale human perception. This study bridges this gap by proposing a novel framework to analyze accessibility through the lens of digital information acquisition and perception. Focusing on discretionary activities, we use restaurant access in Shenzhen as a case study. Specifically, we leverage data from Baidu Map (navigation) and Dianping (ratings) to quantify digitally acquired attributes like travel time, price, and reviews. We then employ a two-stage method to model public perception: first, a human survey identifies how people perceive these digital attributes; second, these findings are integrated with Generative AI (GenAI) in a few-shot learning approach to model city-wide perceptions. Finally, these perceptions are incorporated into the calculation of the digitally mediated accessibility metric, which integrates digital information acquisition and perception. Our findings reveal that the digitally mediated accessibility metric uncovers geographic inequalities in restaurant access that conventional metrics overlook. This research advances accessibility theory by introducing a framework for quantifying digitally mediated accessibility and demonstrates the potential of GenAI in scaling human perception modeling for spatial analysis.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102391"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning street view representations based on a spatiotemporal contrastive learning framework","authors":"Yong Li ,&nbsp;Yingjing Huang ,&nbsp;Fan Zhang","doi":"10.1016/j.compenvurbsys.2025.102393","DOIUrl":"10.1016/j.compenvurbsys.2025.102393","url":null,"abstract":"<div><div>Street view imagery has become an important data source for urban studies, supporting various urban tasks such as environmental perception and socioeconomic predictions. Classic methods predominantly rely on handcrafted features or supervised machine learning to derive information from the images. However, these methods often fail to capture the hierarchical semantics of urban environments: at the visual layer they cannot selectively represent dynamic versus static objects, while at the higher contextual layer they cannot abstract the collective ambience of a scene beyond tangible visual content, which in turn limits their effectiveness in tasks such as place recognition and socioeconomic inference. Essentially, this limitation arises because different urban tasks rely on fundamentally different invariances across space and time. To address this challenge, we propose the spatiotemporal contrastive learning framework, a novel self-supervised framework that systematically organizes representation learning for urban scenes. This framework defines distinct pre-training strategies by selectively contrasting what remains invariant versus what changes across the dimensions of space and time, enabling the model to isolate specific urban features like dynamic elements, static structures, or neighborhood ambiance. The validation experiments confirm that each contrastive strategy produces specialized representations that significantly outperform established baselines on their corresponding tasks. This study provides not only a novel representation framework but also a rigorous benchmark that enhances the applicability of visual data in urban science. The code is available at <span><span>https://github.com/yonglleee/UrbanSTCL</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102393"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-resolution urban land use change modeling via sequential classifiers","authors":"Yves M. Räth ,&nbsp;Adrienne Grêt-Regamey ,&nbsp;Maarten J. van Strien","doi":"10.1016/j.compenvurbsys.2026.102400","DOIUrl":"10.1016/j.compenvurbsys.2026.102400","url":null,"abstract":"<div><div>Urban land use change models are vital tools for anticipating spatial development and its socio-economic and environmental impacts. Yet most models treat urban areas as thematically homogeneous, overlooking variation in residential and economic intensity. We present a high-resolution model for Switzerland’s densely populated Swiss Plateau (1999 settlements, hectare resolution). Using two sequential XGBoost classifiers, our model first predicts urban growth or shrinkage, then assigns one of 27 urban land use classes based on residential density, job density, and economic sector. Trained on five-year intervals (1995–2015) and validated with 2020 data, it achieves 92.3% accuracy for urban extent and a fuzzy kappa of 0.692 for class predictions. Transitions are shaped by neighborhood effects. Projections to 2050 show core cities densify most (+300 ha high density), while peri-urban and residential municipalities expand mainly at low to medium intensities (+3.7% area). Scenario testing illustrates how strategic projects reshape land use beyond intervention sites, supporting informed planning across diverse futures.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102400"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Why same datasets yield different environment–activity relationships? Hidden uncertainties in geospatial processing methods","authors":"Haochen Shi ,&nbsp;Lingzi Xu ,&nbsp;Ding Ma ,&nbsp;Feng Gao ,&nbsp;Shaoying Li","doi":"10.1016/j.compenvurbsys.2025.102398","DOIUrl":"10.1016/j.compenvurbsys.2025.102398","url":null,"abstract":"<div><div>Crowdsourced individual trajectory data have become a valuable resource for examining environment–activity relationships at the streetscape scale. Such analyses critically depend on two key geo-processing decisions: (1) the trajectory assignment method (Hidden Markov Models [HMM] vs. buffer-based approaches), and (2) the spatial delineation of built environment variables. While it is intuitively understood that methodological choices can influence results, systematic evaluations of their combined effects remain limited. This study addresses the gap through a comparative analysis of different combinations of assignment methods and spatial ranges, using walking and cycling trajectory data from the historic urban core of Guangzhou, China. The findings reveal: (1) Assignment methods significantly affect both the statistical and spatial properties of trajectory allocation. The HMM approach produces finer representations of walking and cycling activity, while buffer-based methods capture broader trends due to the lack of probabilistic decision-making. This also explains why cycling data are more sensitive to assignment choices than walking data. (2) In combination with spatial range, assignment methods jointly influence both linear and non-linear correlation patterns between the built environment and activity. These effects are amplified in non-linear models compared to linear ones. These findings carry important methodological implications, highlighting previously hidden uncertainties embedded in common analytical workflows. The study also extends the discussion of the Modifiable Areal Unit Problem (MAUP) to trajectory-based streetscape research, underscoring the need for careful spatial decision-making in studies of active mobility.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102398"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using mobile phone data for quantifying large-scale household-level disaster recovery","authors":"Tessa Swanson ,&nbsp;Seth Guikema","doi":"10.1016/j.compenvurbsys.2025.102395","DOIUrl":"10.1016/j.compenvurbsys.2025.102395","url":null,"abstract":"<div><div>Natural disasters often result in evacuations, travel disruptions, power outages, school closures, and closures of other facilities affecting the ability of individuals to maintain their typical daily patterns. Visits to home and work follow regular patterns that may be interrupted due to a natural hazard. These disruptions impact productivity and well-being. However, there does not currently exist a way to estimate how long individuals' home and work routines were interrupted at the geographic scale of a large natural hazard. Surveys provide useful information, but only for small samples of the affected population. With surveys alone, we cannot model and understand the extent of recovery time across a large set of households with diverse experiences of the disruption. This lack of a method to assess widespread household-scale recovery of normal daily patterns is the key gap we address in this paper. We develop an approach to use location-based services data from smartphones to capture patterns in visits to home and work and deviations from those patterns that may indicate disruption and recovery while maintaining anonymity. We introduce a Bayesian belief network-based anomaly detection method to identify household-level lack of recovery and demonstrate this approach for Hurricane Irma. Our results show the proportion of users experiencing an anomalous period and the average length of recovery, validated against the limited available survey results. These large-scale data-driven results on household recovery contribute to further analysis on the impacts of the hazard and social vulnerability on recovery at the scale of individual homes and workplaces.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102395"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145841462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thriving or surviving: Understanding the geography of financial precarity in Great Britain","authors":"Zi Ye,&nbsp;Alex Singleton","doi":"10.1016/j.compenvurbsys.2026.102399","DOIUrl":"10.1016/j.compenvurbsys.2026.102399","url":null,"abstract":"<div><div>Financial precarity, the state of economic insecurity characterised by unpredictable employment and declining social protection significantly impacts cognitive functioning, emotional stability and social inclusion. This condition stems from multiple interconnected factors: poor quality and unpredictable work, unmanaged debt, insecure asset wealth and insufficient financial resource. Despite extensive research on financial precarity's individual impacts, its geographical distribution and associated social-spatial inequalities remain poorly understood. This paper addresses this gap by introducing a new geodemographic classification of financial precarity across Great Britain. Our classification system uses small-area measurements encompassing employment patterns, income levels, asset holdings, debt obligations, and lifestyle characteristics at the neighbourhood level. By mapping financial precarity at a fine spatial scale, this research reveals how economic vulnerability varies across different localities, highlighting the uneven geography of financial insecurity between rural and urban areas, city centres and peripheries, coastal and inland communities, and how the classification groups are interwoven to the variegated patterns in and around major urban areas. This small-area approach provides sufficient detail to identify spatial patterns while enabling comparisons between local areas, offering new insights into the geographic dimensions of economic precarity in contemporary Britain.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102399"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How well do street view images predict crime rates in London? A comparison with social and macro-level environmental features","authors":"Sitong Guo,&nbsp;Richard Harris,&nbsp;Rui Zhu","doi":"10.1016/j.compenvurbsys.2025.102390","DOIUrl":"10.1016/j.compenvurbsys.2025.102390","url":null,"abstract":"<div><div>In research on the causes of crime, geographic context is considered important in relation to how neighbourhood features influence crime. These features include the social and physical environmental features. Historically, measuring the impact of the physical environment – especially its micro-level characteristics – on crime has been challenging due to the lack of fine-grained quantitative data. Recent advances in computer imagery have enabled researchers to extract structured data from street view imagery, creating new opportunities to quantify features of the physical environment at this scale – particularly those visible from the streetscape perspective. However, the predictive power of these features, and particularly how they compare to more traditional sources of neighbourhood data, remain underexplored. Conducting the analysis across a large urban area also presents a significant challenge. To address these gaps, this study uses a stratified random sampling technique (stratified by classes of socio-economic deprivation/affluence) to select and extract data on micro-level environmental features from Google Street View (GSV) images. These are studied alongside other social and macro-level environmental data for 1000 Lower Super Output Areas (LSOAs) in London, using Random Forest as the core predictive model, with Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) serving as supplementary tools to predict and analyse crime rates at an additional 500 randomly sampled LSOAs. While ‘social and macro-level environmental features’ – specifically renter occupancy rate, the number of POIs, and transport accessibility scores – were generally the most important predictors of crime, for certain crimes, such as criminal damage and arson, incorporating micro-level environmental features improved model accuracy. Overall, models incorporating spatial information in the relationships between environmental and social features and crime rates, outperformed other models. This underscores the importance of considering spatial heterogeneity when analysing features influencing crime.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102390"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Realized spatial accessibility vs. potential spatial accessibility in the United States: A case study based on geospatial big data","authors":"Yaxiong Shao ,&nbsp;Wei Luo","doi":"10.1016/j.compenvurbsys.2025.102382","DOIUrl":"10.1016/j.compenvurbsys.2025.102382","url":null,"abstract":"<div><div>The COVID-19 pandemic drew significant attention to disparities in spatial access to healthcare. While potential spatial accessibility has been extensively researched, realized spatial accessibility remains relatively underexplored. This study employs geospatial big data (SafeGraph Monthly Pattern) to explore the differences between these two types of spatial accessibility using the Two-step Floating Catchment Area (2SFCA) model for the entire population at the Census Tract Level across the contiguous United States. By integrating methods such as point of interest (POI) Placekey matching, Partial Placekey, and fuzzy matching, we successfully matched SafeGraph foot traffic patterns with the American Hospital Association (AHA) survey dataset. Employing total beds as a representation of healthcare facility supply and adjusted SafeGraph visit counts as a representation of the actual healthcare service utilization, the 2SFCA model was applied to compute realized spatial accessibility. A distance decay function, derived from SafeGraph foot traffic patterns, and the same supply data along with potential demand populations were incorporated to calculate potential spatial accessibility. Results show significant differences between potential and realized spatial accessibility across the contiguous US. Compared to realized accessibility measure, the potential spatial accessibility measure significantly underestimates the spatial access to healthcare. Our approach suggests that the realized accessibility based on SafeGraph data can not only help policymakers in making more informed decisions but also serve as a catalyst in improving health access equity.</div></div>","PeriodicalId":48241,"journal":{"name":"Computers Environment and Urban Systems","volume":"125 ","pages":"Article 102382"},"PeriodicalIF":8.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}