While Google Street View (GSV) has been increasingly available for large-scale examinations of urban landscapes, little is known about how to use this promising data source more cautiously and effectively. Using data for Santa Ana, California, as an example, this study provides an empirical assessment of the sensitivity of GSV-based streetscape measures and their variation patterns. The results show that the measurement outcomes can vary substantially with changes in GSV acquisition parameter settings, specifically spacing and direction. The sensitivity is found to be particularly high for some measurement targets, including humans, objects, and sidewalks. Some of these elements, such as buildings and sidewalks, also show highly correlated patterns of variation indicating their covariance in the mosaic of urban space.

The presence of employment centers provides the potential for reducing commuting distance. However, employment centers have distinct attributes, which may lead to varied impacts on commuting outcomes. We examine how proximity to employment centers can influence commuting distance with consideration of the heterogeneity of employment centers and workers. Specifically, we consider various attributes of employment centers related to location, persistency, job density, industry diversity, and size and analyze their impacts on the commuting patterns of low- and high-income workers using panel (2002-2019) data. Our analysis of the Los Angeles region shows that increasing proximity to the nearest employment center decreases commuting distance even after controlling for the job attributes located in the neighborhood of workers. The results further suggest that employment centers are not equal in terms of their impact on commute distance and that their impact is different for commuters from different income groups. Our findings contribute to the literature by deciphering the location and attributes of employment centers that may exert a greater impact on commuting patterns.

Objectives: Despite theoretical interest in how dimensions of the built environment can help explain the location of crime in micro−geographic units, measuring this is difficult. Methods: This study adopts a strategy that first scrapes images from Google Street View every 20 meters in every street segment in the city of Santa Ana, CA, and then uses machine learning to detect features of the environment. We capture eleven different features across four main dimensions, and demonstrate that their relative presence across street segments considerably increases the explanatory power of models of five different Part 1 crimes. Results: The presence of more persons in the environment is associated with higher levels of crime. The auto−oriented measures—vehicles and pavement—were positively associated with crime rates. For the defensible space measures, the presence of walls has a slowing negative relationship with most crime types, whereas fences did not. And for our two greenspace measures, although terrain was positively associated with crime rates, vegetation exhibited an inverted−U relationship with two crime types. Conclusions: The results demonstrate the efficacy of this approach for measuring the built environment.

There is interest in understanding which characteristics of metropolitan areas impact the length of time or distance residents spend commuting. We utilize two measures recently introduced to the urban literature capturing distinct dimensions of employment decentralization –the level of employment deconcentration and employment spatial dispersion in metropolitan areas – to assess how they are related to commuting patterns across metropolitan areas. These two measures of urban/metropolitan spatial structure avoid challenges in identifying “job centers” and allow for a more systematic investigation of how employment decentralization affects commuting patterns. Furthermore, we detect key differences for the implications of these measures for commuting across 329 US metropolitan regions based on their population size. We find that greater employment deconcentration in very small MSAs is associated with longer commute times and distances, whereas greater employment deconcentration in large or very large MSAs is associated with shorter commutes. And whereas spatial dispersion is not related to commute times in very small MSAs, greater spatial dispersion is associated with longer commutes in very large MSAs. This study also shows that the spatial pattern of employment in regions, captured by these new measures, is associated with the proportion of very short and very long duration commutes.

Despite the substantial number of studies on the relationships between crime patterns and built environments, the impacts of street-level built environments on crime patterns have not been definitively determined due to the limitations of obtaining detailed streetscape data and conventional analysis models. To fill these gaps, this study focuses on the non-linear relationships and threshold effects between built environments and local crime patterns at the level of a street segment in the City of Santa Ana, California. Using Google Street View (GSV) and semantic segmentation techniques, we quantify the features of the built environment in GSV images. Then, we examine the non-linear relationships and threshold effects between built environment factors and crime by applying interpretable machine learning (IML) methods. While the machine learning models, especially Deep Neural Network (DNN), outperformed negative binomial regression in predicting future crime events, particularly advantageous was that they allowed us to obtain a deeper understanding of the complex relationship between crime patterns and environmental factors. The results of interpreting the DNN model through IML indicate that most streetscape elements showed non-linear relationships and threshold effects with crime patterns that cannot be easily captured by conventional regression model specifications. The non-linearities and threshold effects revealed in this study can shed light on the factors associated with crime patterns and contribute to policy development for public safety from crime.

Criminological theories have posited that the built environment impacts where crime occurs; however, measuring the built environment is difficult. Furthermore, it is uncertain whether the built environment differentially impacts crime in high-disadvantage neighborhoods. This study extracts features of the built environment from Google Street View images with a machine learning semantic segmentation strategy to create measures of fences, walls, buildings, and greenspace for over 66,000 street segments in Los Angeles. Results indicate that the presence of more buildings on a segment was associated with higher crime rates and had a particularly strong positive relationship with robbery and motor vehicle theft in low-disadvantage neighborhoods. Notably, fences and walls exhibited different relationships with crime. Walls, which do not allow visibility, were strongly negatively related to crime, particularly for robbery and burglary in high-disadvantage neighborhoods. Fences, which allow visibility, were associated with fewer robberies and larcenies, but more burglaries and aggravated assaults. Fences only exhibited a negative relationship with violent crime when they were located in low-disadvantage neighborhoods. The results highlight the importance of accounting for the built environment and the surrounding level of disadvantage when exploring the micro-location of crime.

While street view imagery has accumulated over the years, its use to date has been largely limited to cross-sectional studies. This study explores ways to utilize historical Google Street View (GSV) images for the investigation of neighborhood change. Using data for Santa Ana, California, an experiment is conducted to assess to what extent deep learning-based semantic segmentation, processing historical images much more efficiently than visual inspection, enables one to capture changes in the built environment. More specifically, semantic segmentation results are compared for (1) 248 sites with construction or demolition of buildings and (2) two sets of the same number of randomly selected control cases without such activity. It is found that the deep learning-based semantic segmentation can detect nearly 75% of the construction or demolition sites examined, while screening out over 60% of the control cases. The results suggest that it is particularly effective in detecting changes in the built environment with historical GSV images in areas with more buildings, less pavement, and larger-scale construction (or demolition) projects. False-positive outcomes, however, can emerge due to the imperfection of the deep learning model and the misalignment of GSV image points over years, showing some methodological challenges to be addressed in future research.