Road marker identification method and device and storage medium

Abstract

The embodiment of the invention discloses a road marker identification method and device and a storage medium, and is applied to the technical field of information processing. The road marker identification device firstly constructs a first coordinate transformation matrix based on a two-dimensional image and three-dimensional point cloud data of a scene to be recognized, and calculates the three-dimensional point cloud data by adopting the first coordinate transformation matrix to obtain a depth image of the scene to be recognized; therefore, as long as the plane coordinate information of theroad marker is acquired from the two-dimensional image of the scene to be identified, the depth information of the corresponding position in the depth image can be determined according to the plane coordinate information of the road marker; and then spatial distribution information of the road marker is determined according to the first coordinate transformation matrix and the plane coordinate information and depth information of the road marker, thereby realizing identification of the road marker. The automatic mapping process can be realized only by means of the depth image information of the to-be-identified scene, and the road marker can be efficiently positioned.

Description

technical field [0001] The present invention relates to the technical field of information processing, in particular to a method, device and storage medium for identifying road markers. Background technique [0002] Road signs, street signs and other road markers are important road elements in high-precision maps, which provide a reference for precise vehicle positioning and road condition recognition, so that autonomous vehicles can make accurate judgments about the surrounding environment and formulate appropriate driving strategies. Strategy. [0003] With the continuous improvement of traditional extraction algorithms and the continuous development of laser point cloud deep learning algorithms, the recognition and classification of road signs such as road signs and road signs on point clouds and images can have a high accuracy and recall rate, which can be very good. Improve the automation level of high-precision map production. Generally, in high-precision map data, r...

AI Technical Summary

Problems solved by technology

This method based on ray intersection can save the step of preprocessing point cloud data, but the method of ray intersection can only locate the center point of the road marker more accurately, and the overall size and orientation cannot be determined

In order to obtain accurate results, it still needs to rely on a lot of manual work to correct the results, which affects the overall recognition efficiency

Images