Automatic extraction method and system of using point cloud data to perform identification and positioning of pole-like road objects characterized by reducing manpower and time of making a High-Definition map

Disclosed is an automatic extraction method of using the point cloud data to perform identification and positioning of pole-like road objects, which includes: receiving the point cloud data of a target area; segmenting the point cloud data to obtain non-ground point cloud data; cross-sectionally sampling the non-ground point cloud data at equal interval and equal thickness in the direction perpendicular to the ground to obtain N sampling layers; selecting a reference layer from the sampling layers, wherein the remaining sampling layers are the comparison layers, and comparing each point cloud in the reference layer with at least one point cloud in each comparison layer to filter out the point clouds satisfying the characteristics of pole-like objects in the reference layer; clustering the filtered point clouds to obtain pole-like object cluster point clouds; and calculating a three-dimensional coordinate for each pole-like object cluster point cloud to serve as the coordinate of a pole-like road object.