Generated nighttime street view image to inform perceived safety divergence between day and night in high density cities: A case study in Hong Kong

Xiaotong Ye, Yuankai Wang, Jiajing Dai, Waishan Qiu

Abstract: Highlights•∼1,000 pairwise day-and-night SVIs were manually collected.•CycleGAN was deployed to generate night SVI from daytime ones.•Day and night perceived safety significantly diverge new towns.•Well-maintained building facades consistently improve perceived safety.•Untended trees and grass greatly lower the perceived nighttime safety.

External IDs:doi:10.1016/j.jum.2024.11.006