Wind Turbine Blade Surface Defect Detection Method Based on Improved YOLOv8

As a critical component of wind turbines, wind turbine blades often have surface defects that pose a threat to their operational safety. Due to the complexity of these surface defects, the backbone feature network of the widely used YOLOv8 defect detection model exhibits limited capability in extracting features from such intricate defects. To address this issue, this paper proposes an improved YOLOv8-based method for wind turbine blade surface defect detection. First, the SPDConv module is embedded into the backbone network to enhance the capture of critical features and improve detection accuracy. Simultaneously, P2 detection head is added to strengthen the model’s multi-scale detection capability, further boosting detection accuracy. Additionally, WIoU is adopted as the loss function, which incorporates more geometric information and constraints to accelerate bounding box regression speed and enhance target localization accuracy.The improved model was trained and tested using the wind turbine blade dataset, and the experimental results show that it achieves a mean Average Precision of 85.3%. Compared with the original YOLOv8 model, the enhanced model exhibits a 2.1% improvement in mAP, along with 2.7% and 2.5% increases in precision and recall rates, respectively. These findings demonstrate that the improved model achieves higher accuracy and better detection performance for surface defect detection on wind turbine blades.