Doctor of Philosophy (PhD)


Department of Geography and Anthropology

Document Type



Environmental criminology focuses on a specific time and location, especially between specific built environments and people's behavior. This dissertation focuses on the integration of existing theories and methodologies within and between crime analysis, image acquisition technologies, and machine learning methods to advance crime research and enhance urban safety. Three different case studies are conducted with various methods: 1) Spatiotemporal Analysis of Nighttime Crimes in Vienna, Austria; 2) Crime Prediction in Los Angeles through Convolutional Neural Networks and Google Street View Images; and 3) Street Crime Analysis in Baton Rouge, LA: A Google Street View and Machine Learning Approach.

The first case study focuses on crimes occurring during the nighttime, investigating the temporal definition of nighttime crime and the correlation between nighttime lights and criminal activities in Vienna, Austria. The study concentrates on four types of nighttime crimes: Assault, Theft, Burglary, and Robbery, conducting univariate and multivariate analyses.

The second case study explores the application of machine learning in environmental criminology, particularly through the use of Google Street View (GSV) images to predict property crime hot/coldspots in Los Angeles, California. With the capabilities of Convolutional Neural Networks (CNN), this research investigates the correlation between urban features visible in street-level images and the occurrence of property crimes such as burglary, theft, and vehicle theft.

The third case study identifies which built environments significantly impacting micro-scale crimes through the GSV images in Baton Rouge, Louisiana. With semantic image segmentation algorithm, this research examines variations in the built environmental factors from GSV across different crime rates, such as burglary, robbery, theft, and vehicle theft.

To sum up, this dissertation finds a feasibility to capture and analyze environmental crime characteristics using image data with machine learning techniques and spatial statistics methods. Among different types of crime, there are strong explanation powers with various methods and datasets for burglary.



Committee Chair

Leitner, Michael

Available for download on Thursday, May 15, 2031