Article Text

Download PDFPDF
430 Training neural networks to identify built environment features for pedestrian safety
  1. Alex Quistberg1,2,
  2. Cristina Isabel Gonzalez3,4,
  3. Pablo Arbeláez3,4,
  4. Olga Lucia Sarmiento5,
  5. Laura Baldovino-Chiquillo5,
  6. Quynh Nguyen6,
  7. Tolga Tasdizen7,
  8. Luis Angel Guzman Garcia8,
  9. Dario Hidalgo9,
  10. Stephen J Mooney10,12,
  11. Ana V Diez Roux1,2,
  12. Gina Lovasi1,2
  1. 1Urban Health Collaborative, Dornsife School of Health, Drexel University, Philadelphia, United States
  2. 2Department of Environmental & Occupational Health, Dornsife School of Public Health, Drexel University, Philadelphia, 19104
  3. 3Center for Research and Formation in Artificial Intelligence, Universidad de los Andes, Bogota, Colombia
  4. 4Department of Biomedical Engineering, Universidad de los Andes, Bogota, Colombia
  5. 5Department of Public Health, School of Medicine, Universidad de los Andes, Bogota, Colombia
  6. 6School of Public Health, University of Maryland, College Park, USA
  7. 7Department of Electrical and Computer Engineering, University of Utah, Salt Lake City, USA
  8. 8Department of Civil and Environmental Engineering, Universidad de los Andes, Bogota, Colombia
  9. 9Department of Logistics and Transportation, Pontifica Universidad Javeriana, Bogota, Colombia
  10. 10Department of Epidemiology, University of Washington, Seattle, USA
  11. 11Department of Epidemiology and Biostatistics, Dornsife School of Public Health, Drexel University, Philadelphia, USA
  12. 12Harborview Injury Prevention & Research Center, University of Washington, Seattle, USA


Background We used panoramic images and neural networks to measure street-level built environment features with relevance to pedestrian safety.

Methods Street-level features were identified from systematic literature search and local experience in Bogota, Colombia (study location). Google Street View© panoramic images were sampled from 10,810 intersection and street segment locations, including 2,642 where pedestrian collisions occurred 2015–2019; the most recent, nearest (<25 meters) available image was selected for each sampled intersection or segment. Human raters annotated image features which were used to train neural networks. Neural networks and human raters were compared across all features using mean Average Recall (mAR) and mean Average Precision (mAP) estimated performance. Feature prevalence was compared by pedestrian vs non-pedestrian collision locations.

Results Thirty features were identified related to roadway (e.g., medians), crossing areas (e.g., crosswalk), traffic control (e.g., pedestrian signal), and roadside (e.g., trees) with streetlights the most frequently detected object (N=10,687 images). Neural networks achieved mAR=15.4 versus 25.4 for humans, and a mAP=16.0. Bus lanes, pedestrian signals, and pedestrian bridges were significantly more prevalent at pedestrian collision locations, whereas speed bumps, school zones, sidewalks, trees, potholes and streetlights were significantly more prevalent at non-pedestrian collision locations.

Conclusion Neural networks have substantial potential to obtain timely, accurate built environment data crucial to improve road safety. Training images need to be well-annotated to ensure accurate object detection and completeness.

Learning Outcomes 1) Describe how neural networks can be used for road safety research; 2) Describe challenges of using neural networks.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.