High definition map is an essential infrastructure to realize automatic driving technology, and lane line is an important part of lane level road network of high definition. In the past, lane detection of high definition map was mostly based on the data of vehicle camera, which had efficiency problem caused by limited imaging range, need for perspective transformation and multiple stitching. In this paper, based on UAV aerial images, U-Net network is used to identify road areas and filter noise in non-road areas. HSL color transform and Sobel operator are used to calculate lane color and edge gradient features respectively. Otsu algorithm is used to automatically determine feature segmentation threshold to obtain binary Lane feature map. Local maximum algorithm is used to determine the initial position of sliding window. Sliding window algorithm and polynomial detection are used to fit lane lines. The experimental result shows that in this paper, on the premise of ensuring certain detection accuracy, the detection length of a single lane line exceeds 100 meters, and the road detection efficiency reaches 25.2m/s. Compared with the lane line detection algorithms based on vehicle-mounted camera data, our method has obvious efficiency advantages.