Road data generation method and device, electronic equipment and storage medium

Abstract

The invention provides a road data generation method and device, electronic equipment and a storage medium. The method comprises the steps of constructing a generative adversarial network comprising ageneration model and a discrimination model; obtaining first custom data, a plurality of real road images in multiple types of scenes, segmentation label images corresponding to the real road imagesand scene control parameters corresponding to the multiple types of scenes as training samples, and training the generative adversarial network to obtain an optimal generation model based on the generation model, wherein the scene control parameters represent scene types; generating a to-be-generated road image and a segmentation label image corresponding to the to-be-generated road image based onthe optimal generation model, the scene control parameter corresponding to the to-be-generated road image and the second custom data, wherein the first custom data and the second custom data have thesame distribution rule. According to the invention, the problem that the time cost and the labor cost are high when road images of a specific scene and corresponding traffic line labels are obtainedcan be solved, and the road data obtaining efficiency is improved.

Description

technical field [0001] The present invention relates to the field of intelligent transportation, in particular to a road data generation method, device, electronic equipment and storage medium. Background technique [0002] With the continuous development of science and technology, intelligent transportation technology is also constantly innovating. Improving the road traffic perception system will help to promote more intelligent and informatized urban traffic, and can accelerate the comprehensive transportation industry in the direction of safety, efficiency and intelligence. develop. [0003] Existing road perception models generally use semantic segmentation methods based on deep learning. In order to achieve better segmentation results in different scenarios, such as strong light scenes, night scenes, and rainy weather scenes, this method requires a large number of scene images. data for training. However, it is difficult to obtain a large number of road scene picture...

AI Technical Summary

Problems solved by technology

[0004]In view of this, the object of the present invention is to provide a road data generation method, device, electronic equipment and storage medium, which are used to solve the problem of obtaining road images of specific scenes and It corresponds to the problem of high time cost and labor cost when labeling traffic lines, and improves the efficiency of obtaining road data

Images