Related Experiment Video
Updated: Aug 28, 2025

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
Assessing Park Quality with a Wearable Video Device and an Unmanned Aerial System
Richard R Suminski1, Gregory M Dominick1, Eric Plautz1
1Department of Behavioral Health and Nutrition, University of Delaware, Newark, DE 19726, USA.
Abstract:
Parks are ideal places for promoting physical activity, which is vital for achieving and sustaining good health. Thus, it is important to develop and provide the best methods for assessing aspects of parks that could influence physical activity. This study examined the use of high-tech video capture for describing park quality. Videos were obtained with a wearable video device (WVD) and an unmanned aerial system (UAS) at 28 and 17 parks, respectively. In-person audits of park attributes were performed using the Physical Activity Readiness Assessment (PARA) instrument while video was simultaneously captured. The PARA provides quality ratings of park attributes that range from poor to good. Kappa statistics were calculated to compare in-person PARA outcomes with PARA outcomes obtained by reviewing the WVD and UAS videos. Substantial and almost-perfect agreements were found between WVD and in-person PARAs on the quality of features and amenities and the severity of incivilities. Agreements between UAS and in-person PARAs on feature and amenity quality and incivility severity were unacceptable (mostly fair and moderate). In conclusion, being able to reliably assess park quality using video provides advantages over in-person assessments (e.g., retrospective analysis). In addition, it sets up the possibility of utilizing computer vision to automate the video analysis process.

