Feasibility of Lepton 3.5 using warm-up time for Measuring Leaf Temperature of Crops

Abstract

Abstract Accurately monitoring leaf temperatures is becoming more and more critical as more studies use crop leaf temperature for irrigation, disease, and pest detection. This study aims explore the potential of using a low-cost Lepton 3.5 camera to measure crop canopy temperature. This is achieved by comparing the accuracy of the Lepton 3.5 with that of a FLIR thermal camera and an infrared thermometer. Using a custom Lepton 3.5 camera, an Implexxio LT-1T thermistor, and an MLX90614 infrared thermometer, the temperature of the target leaf of the laboratory plant was automatically measured every 5 minutes, recorded on a private cloud server, and manually measured with a handheld FLIR E8-XT as well. The performance of the custom Lepton 3.5 camera, FLIR E8-XT, and MLX90614 was compared to the highly-accurate Implexxio LT-1T thermistor using mean absolute error (MAE) and root mean squared error (RMSE). As a result of the experiment, the accuracy of MLX90614 was the highest among the three sensors. However, the accuracy of the low-cost LEPTON 3.5 module was less than ± 2°C, the same as the FLIR E8-XT, and far better than the ± 5°C error value of the device specification.