A Framework of Component Temperature Monitoring System for Row Crops Based on High Spatial Resolution Data

Component temperature holds great importance in vegetation evapotranspiration estimation, drought monitoring, and smart agricultural management. In this study, a new framework of component temperature monitoring system for row crops is proposed based on high spatial resolution data gained by ground measurement. Data acquisition of the system is designed as a thermal radiometer that is placed on a platform and moved horizontally along a straight line with a fixed viewing angle. Component temperature separation can be achieved by identifying characteristic points from the brightness temperature curve observed by the thermal radiometer. To suppress the effect of noise and get the best estimate of unknown parameters, a Bayesian inversion algorithm based on Bayesian inference is applied in component temperature retrieval. The validation result shows high accuracy of the method in the low noise scenario, with RMSEs lower than 0.75K for the three components, which highlights its potential for component temperature inversion using ground-measured data.