Aiming at the low automation and efficiency of unmanned aerial vehicle (UAV) power inspection mode in image fast detection, a transmission line component detection method combining single-stage multi-frame predictive detector (SSD) with feature pyramid network (FPN) is proposed, and the insulator fault is detected. On the basis of SSD target detection, the FPN feature pyramid structure is added to locally integrate the feature information between layers. The experimental results show the superiority of the proposed method. The experimental results show that the method has good detection effect for large, medium and small size targets in component detection, and the detection accuracy is about 90%, and the detection accuracy in insulator fault detection is 87.4%. It provides a reference for the development of transmission line component detection technology.