Residents’ Seasonal Behavior Patterns and Spatial Preferences in Public Open Spaces of Severely Cold Regions: Evidence from Harbin, China

Summary

This study develops a computer vision-based method for analysing how residents use public open spaces across all four seasons in Harbin, China. Using longitudinal video observations of a residential square, the researchers extracted pedestrian trajectories, identified five distinct behavioural patterns and examined their relationships with spatial characteristics through geographically weighted regression. The automated classification achieved an overall accuracy of 87.5%. The findings show that activity patterns and spatial preferences vary seasonally, with features such as greenery, shading, facilities and sky visibility influencing how people use outdoor spaces. The methodology offers urban planners a data-driven approach to evaluating and improving public spaces for year-round activity in cold-climate cities

More from HIT

• A Seasonal Study of the Spatial Quality of Cold Streets Based on Activity Portrait Indexes (APIs) · Research · Method · 2026
• Thermal comfort range and influence factor of urban pedestrian streets in severe cold regions · Research · 2019