Physical environment features that predict outdoor active play can be measured using Google Street View images

Randy Boyes; William Pickett; Ian Janssen; David Swanlund; Nadine Schuurman; Louise Masse; Christina Han; Mariana Brussoni

doi:10.1186/s12942-023-00346-3

Physical environment features that predict outdoor active play can be measured using Google Street View images

Int J Health Geogr. 2023 Sep 28;22(1):26. doi: 10.1186/s12942-023-00346-3.

Authors

Randy Boyes^{1

2}, William Pickett^{3

4}, Ian Janssen^{3

5}, David Swanlund⁶, Nadine Schuurman⁶, Louise Masse⁷, Christina Han⁸, Mariana Brussoni⁸

Affiliations

¹ Department of Public Health Sciences, Queen's University, Kingston, ON, K7L 3N6, Canada. rboyes@presagegroup.com.
² Presage Group, Inc, 3365 Harvester Road, Suite 206, Burlington, ON, L7N 3N2, Canada. rboyes@presagegroup.com.
³ Department of Public Health Sciences, Queen's University, Kingston, ON, K7L 3N6, Canada.
⁴ Faculty of Applied Health Sciences, Brock University, 1812 Sir Isaac Brock Way, St. Catharines, ON, L2S 3A1, Canada.
⁵ School of Kinesiology and Health, Queen's University, Kingston, ON, K7L 3N6, Canada.
⁶ Department of Geography, Simon Fraser University, RCB 6119/7134, Burnaby, BC, V5A 1S6, Canada.
⁷ School of Population and Public Health, University of British Columbia, British Columbia Children's Hospital, Room F508, 4480 Oak Street, Vancouver, BC, V5H 3V4, Canada.
⁸ Department of Pediatrics, School of Population and Public Health, Human Early Learning Partnership, University of British Columbia, British Columbia Children's Hospital, Room F511, 4480, Oak Street, Vancouver, BC, V5H 3V4, Canada.

Abstract

Background: Childrens' outdoor active play is an important part of their development. Play behaviour can be predicted by a variety of physical and social environmental features. Some of these features are difficult to measure with traditional data sources.

Methods: This study investigated the viability of a machine learning method using Google Street View images for measurement of these environmental features. Models to measure natural features, pedestrian traffic, vehicle traffic, bicycle traffic, traffic signals, and sidewalks were developed in one city and tested in another.

Results: The models performed well for features that are time invariant, but poorly for features that change over time, especially when tested outside of the context where they were initially trained.

Conclusion: This method provides a potential automated data source for the development of prediction models for a variety of physical and social environment features using publicly accessible street view images.

Keywords: Built environment; Child; Cities; Exercise; Play; Social factors.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Child
Environment
Humans
Machine Learning
Pedestrians*
Search Engine*
Social Environment

Grants and funding

MOP-142262/CIHR/Canada