PV power plant inspection by image mosaicing techniques for IR real-time images

Operation and maintenance play a crucial role for the performance of PV power plants. In this experimental paper a novel approach is proposed included planning and mapping of PV power plants by digital image processing techniques. For this purpose, aerial thermovisual photography is carried out in a PV plant by a light Unmanned Aerial System (UAS) with a thermograph camera onboard. The main objective of current paper is to develop a digital image processing algorithm in order to provide an Image Mosaicing (IM) for richer description of the plant by merging many infrared (IR) images. The algorithm is able to analyze the image mosaicing (IM) in order to count total number of PV modules, defect and healthy modules in PV strings. The results have demonstrated that proposed digital image processing (DIP) algorithm is very efficient and reliable for analysis of infrared (IR) images to provide a depth perspective of PV power plants during procedure inspection.