Accurate and robust 3D reconstruction of wind turbine blade leading edges from high-resolution images

Leading edge erosion of wind turbine blades reduces energy production and blade lifetime, a growing issue with larger blades. Effective monitoring is crucial to tracking erosion and controlling maintenance costs. This paper presents an image-based 3D reconstruction method for the leading edges, targeting challenges like textureless surfaces, background motion, and limited image overlap that cause existing methods to fail. Leveraging monocular depth estimation with dense image matching within a sparse reconstruction framework enables superior accuracy and robustness compared to commercial and open-source sparse 3D reconstruction software, achieving a reduction of at least 38% in reprojection error in the conducted experiments. Furthermore, this paper proposes a loss term for Gaussian Splatting-a recent dense reconstruction paradigm-which makes it possible to obtain dense reconstructions without missing patches, in contrast to traditional methods, while maintaining fine surface detail.