This paper describes ground-based thermal infrared (TIR) data collection and the development of a multivariate regression model to predict brightness surface temperature from thermal images. The statistical model estimates surface temperature regressed against thermal image red, green and blue (RGB) values with high predictive strength (R2 = 0.992) for later multi-image stitching and facet-scale spatio-temporal analyses within a Geographical Information Systems (GIS) environment.
Read the full article here: https://doi.org/10.1016/j.proeng.2017.04.295
Projects:
RP2005: Urban Micro Climates