Journal article
430 Training neural networks to identify built environment features for pedestrian safety
Injury prevention, v 28(Suppl 2), pp A65-A65
20 Nov 2022
Abstract
BackgroundWe used panoramic images and neural networks to measure street-level built environment features with relevance to pedestrian safety.MethodsStreet-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.ResultsThirty 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.ConclusionNeural 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 Outcomes1) Describe how neural networks can be used for road safety research; 2) Describe challenges of using neural networks.
Metrics
12 Record Views
Details
- Title
- 430 Training neural networks to identify built environment features for pedestrian safety
- Creators
- Alex Quistberg - Drexel UniversityCristina Isabel Gonzalez - Department of Biomedical Engineering, Universidad de los Andes, Bogota, ColombiaPablo Arbeláez - Department of Biomedical Engineering, Universidad de los Andes, Bogota, ColombiaOlga Lucia Sarmiento - Department of Public Health, School of Medicine, Universidad de los Andes, Bogota, ColombiaLaura Baldovino-Chiquillo - Department of Public Health, School of Medicine, Universidad de los Andes, Bogota, ColombiaQuynh Nguyen - School of Public Health, University of Maryland, College Park, USATolga Tasdizen - Department of Electrical and Computer Engineering, University of Utah, Salt Lake City, USALuis Angel Guzman Garcia - Department of Civil and Environmental Engineering, Universidad de los Andes, Bogota, ColombiaDario Hidalgo - Department of Logistics and Transportation, Pontifica Universidad Javeriana, Bogota, ColombiaStephen J Mooney - University of WashingtonAna V Diez Roux - Drexel UniversityGina Lovasi - Drexel University
- Publication Details
- Injury prevention, v 28(Suppl 2), pp A65-A65
- Publisher
- BMJ Publishing Group Ltd
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Urban Health Collaborative
- Web of Science ID
- WOS:000898210800195
- Other Identifier
- 991019306092004721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Public, Environmental & Occupational Health