Effectiveness of built environment interventions by season using web cameras

Introduction: Fewer than 50% of adults meet CDC guidelines for physical activity (PA). The built environment (BE) is a culprit for limited PA. The study objective was to analyze existing, online public data feeds to quantify effectiveness of BE interventions and examine the impact of seasonality on these interventions. The Archive of Many Outdoor Scenes (AMOS) has collected 330 million images of outdoor environments from 23,000 public, outdoor webcams since 2006. AMOS captures one image per camera each half hour. Many of the environments captured by AMOS are street intersections with BE improvements during the past seven years. Methods: Using AMOS, webcam photographs were captured for two weeks before and after a BE change (bike lanes, crosswalks) in Washington, DC; one week each from winter and summer. We used the Amazon Mechanical Turk (MTurk) website to crowd-source the image annotation. Results: The number of cyclists per summer photograph increased four-fold between 2009 and 2010 (F=36.72, p=0.002). There was no significant increase in pedestrians between the two summers. Analysis of winter differences is underway. Conclusion/Discussion: The use of webcam data allowed for surveillance and analysis of BE changes, as well as analysis of the impact of seasonality. The addition of the bike lane is associated with a significant four-fold increase in number of cyclist per summer scene. Seasonal moderation is currently being explored. Captured public webcam scenes and MTurks offer an inexpensive (US$12.00), discreet means to evaluate effectiveness of BE policy change and interventions.