Using Computer Vision to Analyze Urban Images and Answer Crime & Place Questions

Session Submission Type: Complete Thematic Panel

Abstract/Description

Many crime and place theories emphasize visual concepts to explain why some places have more crime than others. From broken windows to fences to street signage, criminologists have theorized about how people change their behavior based on the things they observe in their environment. However, due to methodological limitations, scholars attempting to measure visual features of places have historically been limited to qualitative studies and systematic social observations that cost vasts amounts of time and money and all too frequently produce small, non-generalizable samples. Today, computer vision technologies can be used to analyze urban images to generate more sophisticated visual metrics at larger scales, creating never before seen opportunities to test questions about the linkage between place visuals and crime. To consider these possibilities, the current panel will begin with an overview of geoAI and computer vision opportunities for criminologists and present 4 studies that utilize computer vision technologies in different ways to enhance our knowledge about why some places experience more crime than others.

Sub Unit

• Area IV. Correlates of Crime / 28. Neighborhoods and Communities

Individual Presentations

• How GeoAI help Understanding Safety Perceptions? A Case Study of Perception Bias in Stockholm, Sweden - Yuhao Kang, University of South Carolina

• A Picture Worth A Thousand Words: The Effect of Murals on Crime - Maya Moritz, University of Pennsylvania

• Measuring Change in the Built Environment with Machine Learning Strategies: Problems and Prospects - Cheyenne Hodgen, University of California, Irvine; John R. Hipp, University of California, Irvine

• ‘Eyes’ on the Street: How Computer Vision and Cognitive Psychology Can Help Us Get the Gist of Neighborhood Environmental Design and Explain Crime - Riley Tucker, University of Chicago; Gaby Akcelik, University of Chicago Environmental Neuroscience Lab; Nakwon Rim, University of Chicago Environmental Neuroscience Lab; Alfred Chao, University of Chicago Environmental Neuroscience Lab; Elizabeth Gaillard, University of Chicago Environmental Neuroscience Lab; Hassaan Haq, University of Chicago Environmental Neuroscience Lab; Marc Berman, University of Chicago Environmental Neuroscience Lab

Chair

• Riley Tucker, University of Chicago

Discussant

• John M. MacDonald, University of Pennsylvania