Abstract: Because of the large area and long distance of distribution network line corridors, the recognition accuracy of tree barriers around distribution network lines is low. Therefore, an online recognition of distribution network line corridor tree barriers based on airborne laser point cloud semantic segmentation is studied. The laser radar equipment mounted on the helicopter is used to collect the data of the tree obstacles, and the collected data is pre-processed. Semantic segmentation algorithm is used to extract the feature information of point cloud in LinkNet network and assign corresponding category labels. After cluster analysis of category labels, a recognition function is constructed for recognition. The experimental results show that the clustering recognition effect is better and can meet the requirement of high precision recognition. The laser radar mounted on the helicopter was used to conduct a comprehensive scan of the line channel, and the distance between the tree barrier and the wire was successfully measured to be 14.15 m, which was more consistent with the actual situation, and significantly improved the accuracy of tree barrier recognition.