Machine Learning-Based Hourly Frost-Prediction System Optimized for Orchards Using Automatic Weather Station and Digital Camera Image Data

Spring frosts damage crops that have weakened freezing resistance after germination. We developed a machine learning (ML)-based frost-classification model and optimized it for orchard farming environments. First, logistic regression, decision tree, random forest, support vector models were trained using balanced Korea Meteorological Administration (KMA) Automated Synoptic Observing System (ASOS) frost observation data March from the last 10 years (2008–2017). Random forest showed good classification performance selected as main techniques, which fields based on initial occurrence times. The training period was then extended to March–April 20 (2000–2019). Finally, applied KMA ASOS April 2020, not used in previous steps, RGB extracted by digital cameras installed an Gyeonggi-do. successfully classified 117 of 139 cases domestic 35 37 camera observations. assumption time helped most improving model. These results clearly indicate ML has applicable accuracy farming.