Open-source time-lapse thermal imaging camera for canopy temperature monitoring

Abstract

Remote sensing in agriculture and farming has become a popular means to manage crop production, however affordability and feasibility of different equipment can be difficult to come by. To monitor heat stress upon a crop, remote thermal imaging cameras have been shown to be effective. Unfortunately, the costs for equipment can be in the thousands. This study used FLIR's low-cost Lepton 3.5 thermal imaging camera and evaluated the camera's applicability and functionality in monitoring temperature of leaves and canopy. The Lepton 3.5 camera was used in conjunction with a Raspberry Pi 4 Model B. Various tests were conducted to compare the camera to that of a commercial handheld infrared thermometer. Both were found to decrease in accuracy with distance, with about a one degree drop when measuring past 1.8 m. In addition, the camera's frame sensitivity was tested and showed that it only deviated ±0.5 °C from the center, with a max of ±1.05 °C and a min of ± 0.27 °C in certain locations. Moreover, the performance of the camera upon a canopy was conducted. The results show the thermal imaging camera was able to differentiate temperatures across upper and lower canopy. In addition, the results also showed the thermal imaging camera was able to differentiate wet from dry leaves and the progression of the water's evaporative cooling upon the canopy. In conclusion, the open-source time-lapse thermal imaging camera demonstrates a viable option for use in agricultural fields and has proven to be utilized for monitoring canopy temperature.