Abstract: Wind turbine blades are prone to defects such as cracks, sand holes, and edge damage in harsh environments. To address the issues of low efficiency, poor anti-interference capability, and insufficient accuracy inherent in traditional manual and conventional inspection methods, a systematic review of machine vision-based surface defect detection technologies for wind turbine blades is provided. Three mainstream approaches are systematically classified, including traditional image processing, deep learning-based vision methods, and UAV-vision fusion.Through a comparative analysis of their advantages, limitations, and applicable scenarios, this review demonstrates that traditional image processing offers a simple workflow and low cost, making it suitable for low-complexity scenarios. Deep learning-based vision methods can automatically extract high-level features, significantly improving the accuracy and robustness of defect recognition in complex backgrounds. The integration of UAVs with vision systems enables rapid, full-coverage inspection without blind spots while reducing the risks associated with high-altitude operations. However, current technologies still face challenges including environmental interference, insufficient labeled data, and deployment difficulties. Future research will focus on multi-sensor fusion, lightweight models, and intelligent collaborative detection. The review aims to provide technical support for intelligent wind power operation and maintenance and holds significant importance for innovation in renewable energy technology.