Wind power is crucial for the energy transition. Wind turbine blades, which capture wind energy, require effective defect detection to ensure reliable operation. The integration of drone aerial photography and machine vision can efficiently detect surface defects on these blades. This paper reviews recent developments in drone-based wind turbine blade defect detection. It begins with an overview of blade characteristics and defect types. Four detection methods are compared, highlighting the advantages and technical processes of drone-visual inspection. Traditional image processing and machine learning methods for image stitching, defect segmentation, and feature extraction are summarized, alongside deep learning approaches for defect classification, recognition, and segmentation. Relevant datasets and performance metrics are organized, and the paper concludes by identifying challenges and discussing potential solutions K . eywords:UAV aeri