Computational Urban Science

Recent advancements in artificial intelligence (AI) have seen the emergence of smart video surveillance (SVS) in many practical applications, particularly for building safer and more secure communities in our urban environments. Cognitive tasks, such as identifying objects, recognizing actions, and detecting anomalous behaviors, can produce data capable of providing valuable insights to the community through statistical and analytical tools. However, artificially intelligent surveillance systems design requires special considerations for ethical challenges and concerns. The use and storage of personally identifiable information (PII) commonly pose an increased risk to personal privacy. To address these issues, this paper identifies the privacy concerns and requirements needed to address when designing AI-enabled smart video surveillance. Further, we propose the first end-to-end AI-enabled privacy-preserving smart video surveillance system that holistically combines computer vision analytics, statistical data analytics, cloud-native services, and end-user applications. Finally, we propose quantitative and qualitative metrics to evaluate intelligent video surveillance systems. The system shows the 17.8 frame-per-second (FPS) processing in extreme video scenes. However, considering privacy in designing such a system results in preferring the pose-based algorithm to the pixel-based one. This choice resulted in dropping accuracy in both action and anomaly detection tasks. The results drop from 97.48% to 73.72% in anomaly detection and 96% to 83.07% in the action detection task. On average, the latency of the end-to-end system is 36.1 seconds.

Babak Rahimi Ardabili

Armin Danesh Pazho

Ghazal Alinezhad Noghre

Christopher Neff

Sai Datta Bhaskararayuni

Arun Ravindran

Shannon Reid

Hamed Tabkhi

Comput.Urban Sci.

Understanding Policy and Technical Aspects of AI-enabled Smart Video Surveillance to Address Public Safety

In this commentary, we describe the current state of the art of points of interest (POIs) as digital, spatial datasets, both in terms of their quality and affordings, and how they are used across research domains. We argue that good spatial coverage and high-quality POI features — especially POI category and temporality information — are key for creating reliable data. We list challenges in POI geolocation and spatial representation, data fidelity, and POI attributes, and address how these challenges may affect the results of geospatial analyses of the built environment for applications in public health, urban planning, sustainable development, mobility, community studies, and sociology. This commentary is intended to shed more light on the importance of POIs both as standalone spatial datasets and as input to geospatial analyses.

Achilleas Psyllidis

Song Gao

Yingjie Hu

Eun-Kyeong Kim

Grant McKenzie

Ross Purves

May Yuan

Clio Andris

Points of Interest (POI): a commentary on the state of the art, challenges, and prospects for the future

Urban vitality is an essential indicator of an area’s capacity to promote lively social and economic activities. Urban functional areas can play different roles throughout the day, and urban vitality may exhibit significant intraday temporal dynamics. However, few studies have evaluated the dynamic vitality throughout the day among various urban functional areas or explored how the built environment influences this attribute. To bridge this gap, we assessed the vitality dynamics in intensity, variability, and night ratio. We then examined the influencing factors of urban vitality in Central Shanghai using heatmap and point of interest (POI) data. We found significant differences in the intensity, variability, and night ratio of urban vitality among different urban functional areas. The difference in vitality intensity was more significant than the variability and night ratio between weekdays and weekends. The built environment significantly affected urban vitality, but its role differed among the various urban functional areas. Overall, describing urban vitality from a dynamic perspective could improve our understanding of the differences in attracting and maintaining human activities among different urban functional areas.

Shuwei Tang

Na Ta

How the built environment affects the spatiotemporal pattern of urban vitality: A comparison among different urban functional areas

Urban perception is a hot topic in current urban study and plays a positive role in urban planning and design. At present, there are two methods to calculate urban perception. 1) Using a model to learn image features directly automatically; 2) Coupling machine learning and feature extraction based on expert knowledge (e.g. object proportion) method. With two typical streets in Wuhan as the study area, video data were recorded and used as the model input. In this study, two representative methods are selected: 1) End to end convolution neural network (CNN-based model); 2) Based on full convolution neural network and random forest (FCN + RF-based model). By comparing the accuracy of two models, we analyze the adaptability of the model in different urban scenes. We also analyze the relationship between CNN-based model and urban function based on POI data and OSM data, and verify its interpretability. The results show that the CNN-based model is more accurate than FCN + RF-based model. Because the CNN-based model considers the topological characteristics of the ground objects, its perception results have a stronger nonlinear correlation with urban functions. In addition, we also find that the CNN-based model is more suitable for scenes with weak spatial heterogeneity (such as small and medium-sized urban environments), while the FCN + RF-based model is applicable to scenes with strong spatial heterogeneity (such as the downtown areas of China’s megacities). The results of this study can be used as a reference to provide decision support for urban perception model selection in urban planning.

Ruifan Wang

Shuliang Ren

Jiaqi Zhang

Yao Yao

Yu Wang

Qingfeng Guan

A comparison of two deep-learning-based urban perception models: which one is better?

The Local Climate Zone (LCZ) classification is already widely used in urban heat island and other climate studies. The current classification method does not incorporate crucial urban auxiliary GIS data on building height and imperviousness that could significantly improve urban-type LCZ classification utility as well as accuracy. This study utilized a hybrid GIS- and remote sensing imagery-based framework to systematically compare and evaluate different machine and deep learning methods. The Convolution Neural Network (CNN) classifier outperforms in terms of accuracy, but it requires multi-pixel input, which reduces the output’s spatial resolution and creates a tradeoff between accuracy and spatial resolution. The Random Forest (RF) classifier performs best among the single-pixel classifiers. This study also shows that incorporating building height dataset improves the accuracy of the high- and mid-rise classes in the RF classifiers, whereas an imperviousness dataset improves the low-rise classes. The single-pass forward permutation test reveals that both auxiliary datasets dominate the classification accuracy in the RF classifier, while near-infrared and thermal infrared are the dominating features in the CNN classifier. These findings show that the conventional LCZ classification framework used in the World Urban Database and Access Portal Tools (WUDAPT) can be improved by adopting building height and imperviousness information. This framework can be easily applied to different cities to generate LCZ maps for urban models.

Kwun Yip Fung

Zong-Liang Yang

Dev Niyogi

Improving the local climate zone classification with building height, imperviousness, and machine learning for urban models

Digital twins are considered to be a new starting point for today’s smart city construction. This paper defines the concepts of digital twins and digital twin cities, discusses the relationship between digital twins and smart cities, analyzes the characteristics of smart cities based on digital twins, and focuses on the five main applications of smart cities based on digital twins. Finally, we discuss the future development of smart cities based on digital twins.

Li Deren

Yu Wenbo

Shao Zhenfeng

Smart city based on digital twins

Accurate and prompt traffic data are necessary for the successful management of major events. Computer vision techniques, such as convolutional neural network (CNN) applied on video monitoring data, can provide a cost-efficient and timely alternative to traditional data collection and analysis methods. This paper presents a framework designed to take videos as input and output traffic volume counts and intersection turning patterns. This framework comprises a CNN model and an object tracking algorithm to detect and track vehicles in the camera’s pixel view first. Homographic projection then maps vehicle spatial-temporal information (including unique ID, location, and timestamp) onto an orthogonal real-scale map, from which the traffic counts and turns are computed. Several video data are manually labeled and compared with the framework output. The following results show a robust traffic volume count accuracy up to 96.91%. Moreover, this work investigates the performance influencing factors including lighting condition (over a 24-h-period), pixel size, and camera angle. Based on the analysis, it is suggested to place cameras such that detection pixel size is above 2343 and the view angle is below 22°, for more accurate counts. Next, previous and current traffic reports after Texas A&M home football games are compared with the framework output. Results suggest that the proposed framework is able to reproduce traffic volume change trends for different traffic directions. Lastly, this work also contributes a new intersection turning pattern, i.e., counts for each ingress-egress edge pair, with its optimization technique which result in an accuracy between 43% and 72%.

Yalong Pi

Nick Duffield

Amir H. Behzadan

Tim Lomax

Visual recognition for urban traffic data retrieval and analysis in major events using convolutional neural networks

Abstract Streets are an important component of urban public spaces and also a high-incidence area for urban crime. However, current research mainly involves adult crime, or fails to distinguish between adult and juvenile crime, which poses a severe challenge to the prevention of juvenile delinquency. Juveniles have lower self-control abilities and are more likely to be influenced by external environmental factors to trigger criminal behavior compared to adults. Therefore, this study uses New York’s Manhattan district as an example, based on CPTED and social disorganization theories, and utilizes street view data and deep learning techniques to extract street environment indicators. The GWR model is used to explore the influence mechanism of urban street environment on juvenile crime. The results of this study, considering spatial heterogeneity, demonstrate the impact of various physical environmental indicators of urban streets on juvenile delinquency, and reveal that some street indicators have differentiated effects on crime in different areas of the city. Overall, our research helps to uncover the relationship between juvenile delinquency and the built environment of streets in complex urban settings, providing important references for future urban street design and juvenile delinquency prevention.

Bingcheng Li

Gang Li

Li Lan

Annan Jin

Zhe Lin

Yatong Wang

Xiliang Chen

The influence mechanism of urban street environment on juvenile delinquency based on multi-source data fusion: a case study of Manhattan, New York

Abstract The middle and lower reaches of the Yangtze River are frequently affected by the Western Pacific Subtropical High (WPSH) in summer. This leads to phenomena including air subsidence, high temperatures, low rainfall, and weak winds, all of which affect the urban heat island (UHI) effect. Currently, there are few studies on the influence of WPSH on the UHI effect. In this study, we analysed the temporal and spatial distributions of the influence of WPSH on the UHI effect by establishing two scenarios: with and without WPSH. We calculated the UHI intensity and the urban heat island proportion index (UHPI) to analyse the temporal and spatial distributions of the UHI effect. The geographical detector method was then used to analyse the factors influencing UHI. The results indicate the strong heat island effect during the day in provincial capitals and some developed cities. The area of high UHI intensity was larger under the influence of WPSH than in the years without WPSH. WPSH affected UHPI at both day and night, although the effect was more pronounced at night. The factors affecting daytime UHI intensity are mainly POP and NTL, O3 plays a large role in the years with WPSH control. The main factors affecting the UHI intensity at night are AOD, POP and NTL were mainly factors in the years without WPSH control, POP and WPSH were mainly factors in the years with WPSH control. The interactions of the factors are mainly POP and multi-factors during the daytime, and DEM and multi-factors during the nighttime. It was found that the UHI intensity was enhanced under the control of the WPSH, and the influencing factors of the diurnal UHI differed with and without the WPSH control, which ultimately provides realistic suggestions for mitigating the intensity of the UHI in areas affected by the WPSH.

Jie Deng

Geying Lai

Ao Fan

Effects of the Western Pacific Subtropical High on the urban heat island characteristics in the middle and lower reaches of the Yangtze River, China

Abstract Rapid replacement of vegetated land with impermeable land (built-up areas) is a major factor in the increase in Land Surface Temperature (LST), while increased LST worsens the temperature in cities and creates the Surface Urban Heat Island (SUHI) effect. The study aims to measure vegetation loss and Land Surface Temperature of the Gazipur district between 2000 and 2020 and explore the relationship among Normalized Difference Vegetation Index (NDVI), LST, and Urban Thermal Field Variance Index (UTFVI). The Landsat TM/OLI images with minimum cloud coverage have been used to derive different indices. The mean NDVI values are 0.21, 0.16, and 0.22 in 2000, 2010, and 2020 respectively which indicates a general improvement in the health of the vegetation. Besides, the highest LST values throughout 20 years, represent a general increasing trend. As a consequence, different land covers have experienced fluctuations in mean temperature. The result shows that the mean temperature of bare land, buildup, vegetation, and waterbody has increased by 4.77, 2.01, 2.25, and 2.23 °C respectively from 2000 to 2020. The strongest SUHI zone’s area grew by about 28% between 2000 and 2020. Additionally, the highest index value of UTFVI was 0.39 in 2000 and grew to 0.43 in 2010. It changed to 0.49 in 2020, or ten years later. Thus, the SUHI effect’s increasing intensity is visible. Also, regression analysis has been used to explore the correlation between the derived indices. Stakeholders from different sectors like urban planners and policymakers may take insights from this study to work to promote greenery for a healthy urban environment.

Mehereen Salam

Md. Kamrul Islam

Israt Jahan

Md. Arif Chowdhury

Assessing the impacts of vegetation loss and land surface temperature on Surface Urban Heat Island (SUHI) in Gazipur District, Bangladesh

Abstract Bus Rapid Transit (BRT) proves its effectiveness in alleviating traffic congestion, especially in urban areas. The implementation of Transit Signal Priority (TSP) for BRT has shown a significant reduction in delays. However, in densely populated urban areas, this priority can inadvertently cause additional delays for other modes of transportation. In this paper, we propose a control strategy for Regionally Coordinating Bus Priority Signals Control (RCBPSC) at urban intersections. The aim is not only to reduce bus delays but also to consider minimizing delays for pedestrians and other vehicles. To achieve this, we modeled two consecutive intersections along the Amman BRT. Essentially, we evaluated three different control scenarios in addition to the current base scenario. These scenarios include adaptive traffic signal control, RCBPSC with no signal timing optimization, and RCBPSC with signal optimization. Simulation results indicate that the adaptive traffic signal timing has the worst operational performance in terms of average delay and Level of Service (LOS) compared to the base scenario. Additionally, the results show that BRT delays significantly decrease at both intersections when we implement RCBPSC scenarios. When implementing RCBPSC with optimization scenarios, the results show an average reduction of more than 60% in intersection delay, a decrease in emissions of more than 50%, and an improved LOS for system users compared to the base scenario. The findings of this work can help agencies improve the current operational condition of BRT when implementing RCBPSC.

Esra’a Al-Hyasat

Taqwa I. Alhadidi

Evaluating the performance of implementing regionally coordinating bus priority signals under different control schemes

Abstract Infectious diseases usually originate from a specific location within a city. Due to the heterogenous distribution of population and public facilities, and the structural heterogeneity of human mobility network embedded in space, infectious diseases break out at different locations would cause different transmission risk and control difficulty. This study aims to investigate the impact of initial outbreak locations on the risk of spatiotemporal transmission and reveal the driving force behind high-risk outbreak locations. First, we built a SLIR (susceptible-latent-infectious-removed)-based age-stratified meta-population model, integrating mobile phone location data, to simulate the spreading process of an infectious disease across fine-grained intra-urban regions (i.e., 649 communities of Shenzhen City, China). Based on the simulation model, we evaluated the transmission risk caused by different initial outbreak locations by proposing three indexes including the number of infected cases (CaseNum), the number of affected regions (RegionNum), and the spatial diffusion range (SpatialRange). Finally, we investigated the contribution of different influential factors to the transmission risk via machine learning models. Results indicate that different initial outbreak locations would cause similar CaseNum but different RegionNum and SpatialRange. To avoid the epidemic spread quickly to more regions, it is necessary to prevent epidemic breaking out in locations with high population-mobility flow density. While to avoid epidemic spread to larger spatial range, remote regions with long daily trip distance of residents need attention. Those findings can help understand the transmission risk and driving force of initial outbreak locations within cities and make precise prevention and control strategies in advance.

Kang Liu

Yepeng Shi

Shang Wang

Xizhi Zhao

Ling Yin

Impact of initial outbreak locations on transmission risk of infectious diseases in an intra-urban area

Abstract The COVID-19 pandemic brought unprecedented changes to various aspects of daily life, profoundly affecting human mobility. These changes in mobility patterns were not uniform, as numerous factors, including public health measures, socioeconomic status, and urban infrastructure, influenced them. This study examines human mobility changes during COVID-19 in San Diego County and New York City, employing Latent Profile Analysis (LPA) and various network measures to analyze connectivity and socioeconomic status (SES) within these regions. While many COVID-19 and mobility studies have revealed overall reductions in mobility or changes in mobility patterns, they often fail to specify ’where’ these changes occur and lack a detailed understanding of the relationship between SES and mobility changes. This creates a significant research gap in understanding the spatial and socioeconomic dimensions of mobility changes during the pandemic. This study aims to address this gap by providing a comprehensive analysis of how mobility patterns varied across different socioeconomic groups during the pandemic. By comparing mobility patterns before and during the pandemic, we aim to shed light on how this unprecedented event impacted different communities. Our research contributes to the literature by employing network science to examine COVID-19’s impact on human mobility, integrating SES variables into the analysis of mobility networks. This approach provides a detailed understanding of how social and economic factors influence movement patterns and urban connectivity, highlighting disparities in mobility and access across different socioeconomic groups. The results identify areas functioning as hubs or bridges and illustrate how these roles changed during COVID-19, revealing existing societal inequalities. Specifically, we observed that urban parks and rural areas with national parks became significant mobility hubs during the pandemic, while affluent areas with high educational attainment saw a decline in centrality measures, indicating a shift in urban mobility dynamics and exacerbating pre-existing socioeconomic disparities.

Jaehee Park

Ming-Hsiang Tsou

Atsushi Nara

Somayeh Dodge

Susan Cassels

Examining human mobility changes during COVID-19 across socioeconomic groups: a comparative analysis of San Diego County and New York City

Abstract The spatiotemporal mobility patterns and next location prediction of fake base stations (FBS) provide important technical support for the police to prevent spam messages from FBS. However, due to the difficulty in locating their real-time locations, our understanding of the mobility patterns and predictability of FBS is still limited. Based on the crowdsourced spam data, we extract the time and potential locations of FBS and propose a Tucker-MMC method that combines Tucker decomposition with a Mobility Markov Chain (MMC) model to investigate the mobility patterns and predictability of FBS sending spam messages. First, we utilize Tucker decomposition to reflect the spatial and temporal preferences during the movement of the corresponding FBS. Then the mobility regularity and the theoretical maximum predictability of the FBS trajectories with similar mobility preferences are analyzed by entropy and Fano's inequality. A Tucker-MMC is also established for the next location prediction. The results using the spam dataset in Beijing show that the accuracy of Tucker-MMC is more than double that of the MMC. The accuracy of the actual location prediction model is more likely to approach the theoretical maximum predictability when FBS send spam messages in a shorter time, shorter transfer distance, and smaller access range.

None Yufei·Shi

Haiyan Tao

Li Zhuo

Combining the spatiotemporal mobility patterns and MMC for next location prediction of fake base stations

Abstract The outbreak of the COVID-19 Omicron variant in Shanghai in 2022 elicited complex emotions among Shanghainese during the two-month quarantine period. This paper aims to identify prevailing public themes and sentiments by analyzing social media posts from Weibo. Initially, we conducted research based on a dataset of 90,000 Weibo posts during the 2022 COVID-19 outbreak in Shanghai. By examining social media data that mirrors residents' emotional shifts and areas of focus during unforeseen circumstances, we have developed an analytical framework combining hotspot analysis and public sentiment assessment. Subsequently, we employed the Latent Dirichlet Allocation (LDA) method to conduct topic modeling on the Weibo text data. The SnowNLP sentiment classification method was then utilized to quantify sentiment values. Ultimately, we performed spatial visualization of sentiment and concern data, categorizing them into distinct time periods based on Shanghai's infection curve. This approach allowed us to investigate concern focal points, sentiment trends, and their spatiotemporal evolution characteristics. Our findings indicate that variations in public sentiment primarily hinge on the severity of the epidemic's spread, emerging events, the availability of essential resources, and the government's ability to respond promptly and accurately. It is evident that, while residents' concerns shift over time, their primary objective on social media remains expressing demands and releasing emotions. This research offers an avenue for leveraging public opinion analysis to enhance governance capacity during crises, fortify urban resilience, and promote public involvement in governmental decision-making processes.

Chao Liu

Yeyoumin Tian

Yuhao Shi

Zhiyi Huang

Yuchen Shao

An analysis of public topics and sentiments based on social media during the COVID-19 Omicron Variant outbreak in Shanghai 2022

Abstract In this paper, I advocate for a radical expansion of computational urban science to encompass a multidisciplinary, human-centered approach, addressing the inadequacies of traditional methodologies in capturing the complexities of urban life. Building on insights from Jane M. Jacobs and others, I argue that integrating computational tools with disciplines like sociology, anthropology, and urban planning can significantly enhance our understanding and management of urban environments. This integration facilitates a deeper analysis of cultural phenomena, improves urban policy design, and promotes more sustainable, inclusive urban development. By embracing qualitative research methods—such as ethnography and participatory observations—alongside computational analysis, I highlight the importance of capturing the nuanced social fabrics and subjective experiences that define urban areas. I also stress the necessity of including community stakeholders in the research process to ensure that urban science not only analyzes but also improves the lived experiences of urban populations. Furthermore, I underscore the need for ethical governance and the mitigation of biases inherent in computational tools, proposing rigorous model auditing and the inclusion of diverse perspectives in model development. Overall, this work champions a holistic approach to urban science, aiming to make cities smarter, more equitable, and responsive to their residents’ needs.

Xiao Huang

Computational urban science needs to go beyond computational

Abstract This paper reviews a large number of scholarly articles in the housing field spanning the last thirty years, from 1993 to 2022 by implementing bibliometric analysis method. We examine scientific outputs, identify influential articles, journals, international collaboration and evolution of research trends. Keywords such as “Housing price,” “Housing policy,” “Affordable housing,” “Homeownership,” “Housing market,” “Urban planning,” and “Neighborhood” have been identified as the most prevalently cited terms during this period. Furthermore, the prominence of terms such as “China,” “Gentrification,” “Public housing,” “Social housing,” “Homelessness,” “Migration,” “Urbanization,” “Energy,” “Inequality,” “Land use,” “Gender,” and “Foreclosure” have grown in importance, pointing to future research trends. The analysis also reveals that articles pertaining to the COVID-19 pandemic predominantly address the comprehensive effects of the virus on aspects of mental and physical health, consumer behavior, and economic and societal challenges.

短名	Comput.Urban Sci.
Journal Impact	2.55
国际分区	REGIONAL & URBAN PLANNING(Q2)

涉及主题	computational sciencebuilt environmenturban modellingurban development
出版信息	出版商: Springer，出版周期: ，期刊类型: journal，开源期刊: 开源
基本数据	创刊年份: 2021，原创研究文献占比: ，自引率:， Gold OA占比:
平均审稿周期	10 weeks

Computational Urban Science

阅读文献时，研飞即可帮你快速了解期刊，查询影响因子与分区，智能高亮期刊预警！

期刊引文格式

只有1位作者的期刊

有2位作者的期刊

有3位作者的期刊

有5位以上作者的期刊

书籍引用格式

学位论文引用格式

网页引用格式

专利引用格式

手工熬夜修改参考文献？研飞自动匹配期刊，一键轻松成稿，支持 Word/WPS

投稿经验分享