Recently, the solar power generation has attracted much attention and market is growing. Although it is more common than in the past, there is not enough specialist for maintenance of the solar panel. Because solar panels are exposed to the outdoor environment, frequent failures can occur. So, we propose a solar panel detection system using the drones, the thermal cameras, and the RGB cameras. We used the characteristics of the fault of solar panel in the thermal camera, and detected the position of the photovoltaic panel and the array using the RGB camera. DETECTING FAULTY SOLAR PANELS BASED ON THERMAL IMAGE PROCESSING