Heat exposure assessment and comfort path recommendations for leisure jogging based on street view imagery and GPS trajectories

Abstract

Heat exposure significantly affects outdoor leisure activities and health. However, limited research has investigated the dynamic heat exposure assessments and comfort-based route recommendations for leisure jogging. This study addresses this gap by developing a framework that assesses real-time heat exposure and recommends comfortable routes for joggers based on street view images and GPS tracks. Specifically, we utilize Baidu street view images to dynamically estimate the mean radiant temperature (MRT). Individual joggers’ heat exposure is then assessed using real-time MRT measurements and jogging trajectories. To enhance jogger comfort, we developed a dynamic route recommendation method that combines the jogging comfort index with user preferences using a genetic algorithm, offering personalized jogging routes. Empirical analysis using actual data in Chengdu reveals that (1) MRT and heat risk areas exhibit significant spatiotemporal heterogeneity and clustering. (2) About 33.8 % of jogging tracks overlap with thermal risk regions, with 11 % intersecting high-risk areas, indicating substantial heat exposure for joggers in July. (3) The recommended jogging routes reduce heat exposure risk by 10 % compared to actual routes and a baseline model, demonstrating the approach's effectiveness. (4) The recommended routes, tailored to various needs, outperform the benchmark model, particularly in avoiding dynamic heat risk and providing personalized experiences. These insights can help guide leisure joggers to routes that offer a more comfortable and individualized environment.