Photovoltaic panels identification in aerial images

The solar power generation is increasing rapidly around the world. These systems are often installed on roofstops and other structures that are difficult to access. Despite this difficulty, regular inspection of the system is a requirement to ensure efficiency. As a result, a challenge arise to develop efficient and low-cost inspection methods to monitor structural failures, dirt and other characteristics of the system. Thus, in this work an Unmanned Aerial Vehicle was used to carry out the inspection, obtaining critical images that are necessary to the solar panels maintenance and a machine learning algorithm was applied to identify them.