About this document
Urbansci 09 00151 by nguemtomarco is a document available to read on EtoBox.
This article presents a data-driven machine learning approach to predict urban heat islands by utilizing building volume data derived from 3D urban models. The study employs Gaussian blurring to enhance the correlation between air temperature and building volume, and evaluates model accuracy using advanced metrics like SSIM and LPIPS instead of traditional metrics like MSE. The findings aim to assist urban planners in creating more sustainable urban environments by integrating environmental parameters into
- Author
- nguemtomarco
- Language
- EN