Remote-sensing surface temperature downscaling algorithm

Abstract

The invention discloses a remote-sensing surface temperature downscaling algorithm. The algorithm comprises the following steps: step one, implementing a nuclear driving algorithm to obtain a high-resolution surface temperature; step two, on the basis of an image fusion algorithm, acquiring a medium-resolution surface temperature; and step three, carrying out weighted combination on the high-resolution surface temperature obtained at the step one and the medium-resolution surface temperature obtained at the step two to obtain a high-resolution surface temperature. According to the remote-sensing surface temperature downscaling algorithm based on nuclear driving and image fusion, the surface temperature of a time sequence is processed by downscaling to obtain a high resolution. Because of the nuclear drive method, the spatial resolution of the auxiliary data is high; the image fusion method has the higher temporal prediction ability; and with utilization of advantages of the two kinds of methods, temperature downscaling is carried out by using the multi-source remote sensing data. Compared with the separate nuclear drive method, the remote-sensing surface temperature downscaling algorithm is suitable for the heterogeneous regions better; and compared with the separate fusion method, the remote-sensing surface temperature downscaling algorithm can obtain a multi-phase surface temperature with higher resolution.

Description

technical field [0001] The invention relates to a downscaling algorithm for remote sensing surface temperature. Background technique [0002] Land surface temperature (LST) is an important parameter in environmental research, it can be used to simulate radiative transfer and energy budget, retrieve surface carbon, monitor urban heat island, also be used for forest fire detection, and as an indicator of climate change, etc. [0003] Thermal infrared images obtained from satellite observations have fine spatial resolution or fine temporal resolution, which will be helpful for environmental research based on surface temperature. However, there is a limit between spatial and temporal resolution, i.e. sensors with high spatial resolution tend to have lower temporal resolution and vice versa. [0004] Existing approaches to improve the resolution of land surface temperature can be divided into two categories: the first is nuclear-driven methods that downscale land surface tempera...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that in the current method for improving the resolution of the surface temperature, the nuclear driving method cannot downscale the low-resolution temperature of the long-term series, and the temperature after the downscaling of the image fusion method cannot predict the low-resolution temperature. Technical issues with recorded change information

Images