The invention provides a power transmission line locking pin defect detection method based on aerial images, and belongs to the technical field of power inspection, and the method comprises the steps:marking aerial images, carrying out the classification and storage of image data, and building a defect data set of a locking pin, which comprises a training set and a test set; performing neural network training on the Faster R-CNN model applying the ResNeXt-101 feature extraction network by using the training set, and improving a feature extraction process by using a GN algorithm, a DCNv2 algorithm and an FPN algorithm to generate a training model; and evaluating the training model by using the test set, and after the training model meets the precision requirement, detecting the locking pindefect of the aerial image of the routing inspection power transmission line. The unmanned aerial vehicle can automatically inspect and detect the defects of the power transmission line locking pin,and can adapt to various types and scales of power transmission line aerial image locking pin defects, so that the inspection result is not influenced by objective factors such as the skill level of inspectors and weather and terrain, and the unmanned aerial vehicle is more convenient, safer and more efficient than manual inspection.