A method and device for generating a digital orthophoto map

Abstract

This specification discloses a method and device for generating a digital orthophoto map, which can first obtain each frame of environment images collected in a target area and the corresponding laser point cloud. Afterwards, each frame of laser point cloud is projected into the corresponding environment image, and the three-dimensional coordinates of the space points corresponding to the environment pixels in each environment image are determined. The first weight of each spatial point is determined according to the three-dimensional coordinates of each spatial point, the camera position when collecting each frame of laser point cloud, and the collection time. Then, each spatial point is projected onto the road orthophoto map, and the target pixel value of each image pixel is determined according to the pixel value of the environmental pixel corresponding to each spatial point and the first weight, so as to generate the road orthophoto map . By combining the laser point cloud and the environment image, the generated road orthophoto map contains relatively rich road texture information, which saves the generation cost of the road orthophoto map and improves the accuracy of subsequent road pavement element information extraction.

Description

technical field [0001] The present application relates to the technical field of high-precision maps, and in particular to a method and device for generating a digital orthophoto map. Background technique [0002] Road pavement element information (such as lane lines, curbs, etc.) is an important basic element in high-precision maps, which can be used to locate vehicle poses and plan driving paths. Therefore, when making a high-precision map, it is also necessary to extract and label road surface element information from the road image. It can be seen that the accuracy and information richness of road image maps affect the effect of high-precision maps. [0003] At present, the method of collecting road image maps generally needs to be equipped with multiple lidar devices on the collection vehicle to collect relatively dense laser point clouds. By traversing the target area, the laser point cloud on the road of the target area is collected. According to the collected lase...

AI Technical Summary

Problems solved by technology

[0004] However, the road reflectance intensity map generated based on laser point cloud cannot clearly express the texture information of the road surface, which may easily lead to errors in the extraction of road surface element information, and the road reflection is generated based on the dense laser point cloud collected by multiple laser radar devices. Higher cost for rate-intensity plots

Images