Radial Distortion Correction of UAV Images Based on Geometric Properties of PV Modules

Photovoltaic (PV) installation capacities increase both domestically and internationally. Unmanned aerial vehicle (UAV) technology shows excellent potential in PV power plant inspection, especially in combination with infrared thermography, which provides an effective early fault detection method. However, images captured by UAVs may be affected by radial distortion, which causes straight lines in the images to become curved, affecting the accuracy of PV module identification and fault diagnosis. Since PV module have the geometrical property that the lines are parallel or perpendicular to each other, this paper proposes an image correction method based on this property, which evaluates the correction effect by calculating the number of parallel and perpendicular pairs of lines in the image. We use an iterative approach to adjust the distortion correction parameters. The proposed method effectively corrects the image distortion caused by camera lens while maintaining the integrity and accuracy of critical structures in the image. For experimental validation, we collected infrared thermal images using a UAV at a PV power plant in Jiangsu Province, China. Compared with existing methods, the method proposed in this paper has significant advantages in avoiding overcorrection and maintaining the structural integrity of PV modules, which can help in subsequent PV module identification and fault diagnosis.