Street-level Data Research Articles

SolarSAM: Building-scale Photovoltaic Potential Assessment Based on Segment Anything Model (SAM) and Remote Sensing for Emerging City

Driven by advancements in photovoltaic (PV) technology, solar energy has emerged as a promising renewable energy source due to its ease of integration onto building rooftops, facades, and windows. For emerging cities, the lack of detailed street-level data presents a challenge for effectively assessing the potential of building-integrated photovoltaic (BIPV). To address this, this study introduces SolarSAM, a novel BIPV evaluation method that leverages satellite imagery and deep learning techniques, and an emerging city in northern China is utilized to validate the model performance. SolarSAM segmented various building rooftops using text prompt-guided semantic segmentation during the process. Separate PV models were then developed for Rooftop PV, Facade-integrated PV, and PV windows, using this segmented data and local climate information. The potential for BIPV installation, solar power generation, and city-wide power self-sufficiency were assessed, revealing that the annual BIPV power generation potential surpassed the city’s total electricity consumption by a factor of 2.5. Economic and environmental analysis were also conducted for the BIPVs on different buildings; the levelized cost of electricity is 0.18-0.41 CNY/kWh, and the annual total carbon reduction is 7.08×107 T CO2. These findings demonstrated the model’s performance and revealed the potential for BIPV power generation.

Deep-Learning-Based Real-Time Visual Pollution Detection in Urban and Textile Environments

The environmental physiognomy of an area can significantly diminish its aesthetic appeal, rendering it susceptible to visual pollution, the unbeaten scourge of modern urbanization. In this study, we propose using a deep learning network and a robotic vision system integrated with Google Street View to identify streets and textile-based visual pollution in Dhaka, the megacity of Bangladesh. The issue of visual pollution extends to the global apparel and textile industry, as well as to various common urban elements such as billboards, bricks, construction materials, street litter, communication towers, and entangled electric wires. Our data collection encompasses a wide array of visual pollution elements, including images of towers, cables, construction materials, street litter, cloth dumps, dyeing materials, and bricks. We employ two open-source tools to prepare and label our dataset: LabelImg and Roboflow. We develop multiple neural network models to swiftly and accurately identify and classify visual pollutants in this work, including Faster SegFormer, YOLOv5, YOLOv7, and EfficientDet. The tuna swarm optimization technique has been used to select the applied models’ final layers and corresponding hyperparameters. In terms of hardware, our proposed system comprises a Xiaomi-CMSXJ22A web camera, a 3.5-inch touchscreen display, and a Raspberry Pi 4B microcontroller. Subsequently, we program the microcontroller with the YOLOv5 model. Rigorous testing and trials are conducted on these deep learning models to evaluate their performance against various metrics, including accuracy, recall, regularization and classification losses, mAP, precision, and more. The proposed system for detecting and categorizing visual pollution within the textile industry and urban environments has achieved notable results. Notably, the YOLOv5 and YOLOv7 models achieved 98% and 92% detection accuracies, respectively. Finally, the YOLOv5 technique has been deployed into the Raspberry Pi edge device for instantaneous visual pollution detection. The proposed visual pollutants detection device can be easily mounted on various platforms (like vehicles or drones) and deployed in different urban environments for on-site, real-time monitoring. This mobility is crucial for comprehensive street-level data collection, potentially engaging local communities, schools, and universities in understanding and participating in environmental monitoring efforts. The comprehensive dataset on visual pollution will be published in the journal following the acceptance of our manuscript.

Residential turnover and crime—Evidence from administrative data for England and Wales

Abstract This article combines administrative data on housing transactions with monthly crime data for the years 2014–18 to investigate the link between residential turnover and crime. Using a panel of 34,753 small neighbourhoods in England and Wales as well as street-level data for London, estimates from models with multiple fixed effects suggest that higher turnover is associated with higher overall crime rates. These effects are small and driven by increases in property and violent crime. Using information on price growth, I document differential effects for high-turnover, high price growth (‘attractive’) and high turnover, low price growth (‘declining’) areas.

A PILOT STUDY OF URBAN POI MAPPING USING CROWDSOURCED STREET-LEVEL IMAGERY AND DEEP LEARNING

Abstract. Point-of-interest (POI) data contains rich semantic and spatial information, having a wide range of applications including land use, transport planning and driving navigation. However, urban POI mapping traditionally requires a lot of manpower and material resources, which only few institutions or enterprises can afford to. With the increasing amount of street-level imagery, it is possible to directly extract POI-related information from such data and automatically map the distribution of urban POIs. In the pilot study, we mainly focused on extracting POIs from billboards in street-level imagery. Firstly, the you only look once (YOLO) algorithm was considered to locate billboards in the imagery, then an optical character recognition (OCR) model was adopted to extract POI-related semantic information from the detected billboard, and finally the extracted semantic text was further processed to obtain POI results. The preliminary study shows that it is a promising way of mapping urban POIs from crowdsourced street-level data using deep learning techniques.

Does Stop and Search Reduce Crime? Evidence from Street-Level Data and a Surge in Operations Following a High-Profile Crime

Abstract This paper uses a surge in stop and search operations following a high-profile murder to look at their effect on recorded crime. Difference-in-difference estimates using detailed geocoded data at the street-level suggest a doubling to trebling of the number of searches in streets close to the place of the murder. IV estimates on the effects of stops and searches on crime suggest little effect on property crime, weapons offences and violent crime or drug offences. Some specifications find reductions in anti-social behaviour, suggesting that any effects on crime are due to an increased police presence on the streets.

Microscale Walkability Modelling. The Case of Athens City Centre

To date, several macro-level walkability measures have been proposed, but microscale and audit-based walkability approaches prove to be highly effective to support realistic, quick and cheap mechanisms for pedestrian-friendly environments. Yet, walkability audits are time- and cost-intensive solutions, because they require several streetscape observations. This study aims to investigate whether a multiple linear regression model of urban form- and function-related variables can effectively predict an audit-based average walkability indicator. For this purpose, we use a virtual, brief and reliable audit tool (MAPS-Mini) in Athens city centre in order to collect street-level data and in turn to construct a microscale walkability indicator (dependent variable). Moreover, our approach suggests a flexible statistical model of open-source data, with six exploratory variables of the macro-level built environment: angular integration, population density, transit stop density, pedestrian street density, retail and entertainment activity density, and building height. The results indicate that audit-based average walkability scores can be effectively estimated, as the regression model can explain about 82% of the variation. Furthermore, the density of retail and entertainment activities was indicated as the strongest correlate of more walking-friendly streetscapes, while some urban policy implications include the promotion of footpath repairs and better-engineered crossings.

A modified two-process Knox test for investigating the relationship between law enforcement opioid seizures and overdoses

Recent research has shown an association between monthly law enforcement drug seizure events and accidental drug overdose deaths using cross-sectional data in a single state, whereby increased seizures correlated with more deaths. In this study, we conduct statistical analysis of street-level data on law enforcement drug seizures, along with street-level data on fatal and non-fatal overdose events, to determine possible micro-level causal associations between opioid-related drug seizures and overdoses. For this purpose, we introduce a novel, modified two-process Knox test that controls for self-excitation to measure clustering of overdoses nearby in space and time following law enforcement seizures. We observe a small, but statistically significant ( p < 0.001), effect of 17.7 excess non-fatal overdoses per 1000 law enforcement seizures within three weeks and 250 m of a seizure. We discuss the potential causal mechanism for this association along with policy implications.

Robust LiDAR Feature Localization for Autonomous Vehicles Using Geometric Fingerprinting on Open Datasets

Localization is a key task for autonomous vehicles. It is often solved with GNSS but due to multipath the performance is often not sufficient. Feature localization systems using LiDAR can deliver an accurate localization but the creation of the necessary feature maps is an effortful task. With digitization of urban planning processes a lot of street level data is being generated and increasingly becomes openly available. We propose a novel feature localization system which utilizes geometric fingerprinting to robustly associate features to a feature map generated from this open data from the city of Berlin. With this association, we perform a precise localization of a vehicle in areas spanning over several square kilometers using an optional IMU, the vehicle's CAN-odometry and an initial pose estimate. We evaluated our system with our autonomous car in real world scenarios and achieved a centimeter precision localization accuracy outperforming a high-cost GNSS. The source code will be published at https://github.com/dcmlr/fingerprint-localization .

Predicting bicycling and walking traffic using street view imagery and destination data

Few studies predict spatial patterns of bicycling and walking across multiple cities using street-level data. This study aims to model bicycle and pedestrian traffic at 4145 count locations across 20 U.S. cities using new micro-scale variables: (1) destinations from Google Point of Interest data (e.g., restaurants, schools) and (2) pixel classification from Google Street View imagery (e.g., sidewalks, trees, streetlights). We applied machine learning algorithms to assess how well street-level variables predict bicycling and walking rates. Adding street-level variables improved out-of-sample prediction accuracy of bicycling and walking activities. We also found that street-level variables (10-fold CV R2: 0.82–0.88) may be a useful alternative to Census data (0.85–0.88). Macro-scale factors (e.g., zoning) captured by Census data and micro-scale factors (e.g., streetscapes) captured in our street-level data are both useful for predicting active travel. Our models provide a new tool for estimating and understanding the spatial patterns of active travel.

Enhanced Facade Parsing for Street-Level Images Using Convolutional Neural Networks

Façade parsing is an essential process before the 3-D modeling of digital or virtual 3-D city models. The existing grammar-based approaches for façade parsing rely on strong prior knowledge but can obtain façade parts with better structure. Pixelwise-segmentation-based approaches achieve façade parsing with much less knowledge but the resulting structure of façade parts is normally incomplete. Both these approaches are restricted by their high reliance on the data set. Therefore, they cannot be applied for façade parsing with complex scenes. To address this issue, we built a large street-level data set by taking Mapillary images as the training data for more general scenes. At the same time, we propose a new pipeline based on convolutional neural network (CNN) that combines pixelwise segmentation and global object detection to achieve better results for facade parsing. Our pipeline can be applied to façade images after rectification and street-level façade images with complex scenes. The result of the ablation study demonstrates that the design of our pipeline is effective. We test our pipeline on the classic ECP2011 data set and our new large street-level data set. Our pipeline achieves state-of-the-art results for both the data sets: an accuracy of 98.2% and the mean average precision (mAP) of 98.8% on the ECP2011 data set as well as the mAP of 81.1% for façade parts parsing on our street-level data set.

FULLY AUTOMATIC FEATURE-BASED REGISTRATION OF MOBILE MAPPING AND AERIAL NADIR IMAGES FOR ENABLING THE ADJUSTMENT OF MOBILE PLATFORM LOCATIONS IN GNSS-DENIED URBAN ENVIRONMENTS

Abstract. Mobile Mapping (MM) has gained significant importance in the realm of high-resolution data acquisition techniques. MM is able to record georeferenced street-level data in a continuous (laser scanners) and/or discrete (cameras) fashion. MM’s georeferencing relies on a conjunction of Global Navigation Satellite Systems (GNSS), Inertial Measurement Units (IMU) and optionally on odometry sensors. While this technique does not pose a problem for absolute positioning in open areas, its reliability and accuracy may be diminished in urban areas where high-rise buildings and other tall objects can obstruct the direct line-of-sight between the satellite and the receiver unit. Consequently, multipath measurements or complete signal outages impede the MM platform’s localisation and may affect the accurate georeferencing of collected data. This paper presents a technique to recover correct orientation parameters for MM imaging platforms by utilising aerial images as an external georeferencing source. This is achieved by a fully automatic registration strategy which takes into account the overall differences between aerial and MM data, such as scale, illumination, perspective and content. Based on these correspondences, MM data can be verified and/or corrected by using an adjustment solution. The registration strategy is discussed and results in a success rate of about 95 %.

The link between crime risk and property prices in England and Wales: Evidence from street-level data

This paper uses street-level data on property sales and crime rates for England and Wales to investigate compensating differentials for crime risk. My identification strategy relies on the use of non-parametric regional time trends on various levels of spatial aggregation as well as various fixed effects for streets and wider areas to control for unobserved amenities and regional economic conditions. The data comes from transaction data collected by the land registry and recently published crime maps for the whole of the UK. My estimates, which are robust to a range of sensible specification changes, suggest that each case of anti-social behaviour per ten population in the same street leads to an approximately 0.6–0.8% drop in property prices, while a corresponding increase in violent crime decreases prices by roughly 0.6–1.6% and a corresponding increase in non-violent crime by about 0.2–0.4%. The majority of estimates are at the upper end of these intervals. Estimates for robbery, burglary and vehicle crime are either zero or positive, but are possibly biased because of reverse causality. Crime outside of the respective street does not appear to matter, which is consistent with earlier findings. Expressed in monetary terms each case of anti-social behaviour costs society between £5000 and £6700 and each violent crime between £5000 and £13,300. The results confirm estimates based on prior willingness-to-pay studies and other studies using smaller areas such as single cities.

Exploratory Cumulative Risk Assessment (CRA) Approaches Using Secondary Data

ABSTRACT Cumulative risk assessments (CRAs) include the examination of risks posed by multiple stressors and include population-specific vulnerabilities and susceptibilities. In this case study, we assess potential hearing impairment hazard due to joint exposure from noise and volatile organic compounds (VOCs) in order to examine the strengths and limitations of using secondary data on exposure and health effects for a CRA. Block group-level noise categories were estimated using modeled street-level data. A quantile regression model of sociodemographic and personal predictors from the 1999–2000 U.S. National Health and Nutrition Examination Survey VOC dataset was used along with block group-level sociodemographic and personal variables to estimate VOC exposures. Hazard indices (HIs) for potential hearing impairment due to joint noise and VOC exposures were calculated. County-averaged HIs for hearing impairment ranged from 0.8 (10th total VOCs percentile and 45–60 dB) to 1.7 (90th total VOCs percentile and 71–75 dB). Limitations of the exposure and health effects data included issues combining heterogeneous data and a lack of established threshold levels for combined low-level exposures; yet, this case study illustrates that screening-level CRAs, including non-chemical stressors, can be accomplished with publicly available data and existing methods.

Developing and testing a street audit tool using Google Street View to measure environmental supportiveness for physical activity

BackgroundWalking for physical activity is associated with substantial health benefits for adults. Increasingly research has focused on associations between walking behaviours and neighbourhood environments including street characteristics such as pavement availability and aesthetics. Nevertheless, objective assessment of street-level data is challenging. This research investigates the reliability of a new street characteristic audit tool designed for use with Google Street View, and assesses levels of agreement between computer-based and on-site auditing.MethodsThe Forty Area STudy street VIEW (FASTVIEW) tool, a Google Street View based audit tool, was developed incorporating nine categories of street characteristics. Using the tool, desk-based audits were conducted by trained researchers across one large UK town during 2011. Both inter and intra-rater reliability were assessed. On-site street audits were also completed to test the criterion validity of the method. All reliability scores were assessed by percentage agreement and the kappa statistic.ResultsWithin-rater agreement was high for each category of street characteristic (range: 66.7%-90.0%) and good to high between raters (range: 51.3%-89.1%). A high level of agreement was found between the Google Street View audits and those conducted in-person across the nine categories examined (range: 75.0%-96.7%).ConclusionThe audit tool was found to provide a reliable and valid measure of street characteristics. The use of Google Street View to capture street characteristic data is recommended as an efficient method that could substantially increase potential for large-scale objective data collection.

WALKYOURPLACE – EVALUATING NEIGHBOURHOOD ACCESSIBILITY AT STREET LEVEL

Abstract. The popularity of a neighbourhood is often explained by its perceived "higher" quality of life. Good access to shops, restaurants, parks, etc., is seen as an indicator that reflects improved quality of life. We present a web-based tool for assessment of accessibility to such services. The system evaluates in real time an area that is accessible using pedestrian, transit, and cycling infrastructure. The accessible area is evaluated using "quality of life" indicators, such as the number of grocery stores, shopping and recreation facilities, and local crime within that area. This tool sets itself apart from pre-computed and neighbourhood-level walkability indices, because it makes use of detailed street-level data, rather than block-level generalizations. It uses real network travel time, and, when transit data are provided, permits the creation and evaluation of accessibility areas for a combination of travel modes such as walking with transit use.

Neighborhood conditions are associated with maternal health behaviors and pregnancy outcomes

Women residing in neighborhoods of low socioeconomic status are more likely to experience adverse reproductive outcomes; however, few studies explore which specific neighborhood features are associated with poor maternal health behaviors and pregnancy outcomes. Based upon our conceptual model, directly observed street-level data from four North Carolina US counties were used to create five neighborhood indices: physical incivilities (neighborhood degradation), social spaces (public space for socializing), walkability (walkable neighborhoods), borders (property boundaries), and arterial features (traffic safety). Singleton birth records (2001–2005) were obtained from the North Carolina State Center for Vital Statistics and maternal health behavior information (smoking, inadequate or excessive weight gain) and pregnancy outcomes (pregnancy-induced hypertension/pre-eclampsia, low birthweight, preterm birth) were abstracted. Race-stratified random effect models were used to estimate associations between neighborhood indices and women’s reproductive behaviors and outcomes. In adjusted models, higher amounts of physical incivilities were positively associated with maternal smoking and inadequate weight gain, while walkability was associated with lower odds of these maternal health behaviors. Social spaces were also associated with inadequate weight gain during pregnancy. Among pregnancy outcomes, high levels of physical incivilities were consistently associated with all adverse pregnancy outcomes, and high levels of walkability were inversely associated with pregnancy-induced hypertension and preterm birth for Non-Hispanic white women only. None of the indices were associated with adverse birth outcomes for Non-Hispanic black women. In conclusion, certain neighborhood conditions were associated with maternal health behaviors and pregnancy outcomes.