Traffic sign board information acquisition method and system for high-precision map production

Abstract

The invention, which relates to the field of high-precision map information acquisition, discloses a traffic sign board information acquisition method and system for high-precision map production. Themethod comprises the steps: acquiring to-be-recognized traffic scene image information; acquiring position information of a traffic sign board in the image; acquiring position information of a traffic sign in the image; performing attribution determination by utilizing the position information of the traffic sign in the image and the position information of the traffic sign board in the image, and attributing the traffic sign information to the corresponding traffic sign board information; and acquiring position information of the traffic sign board in a high-precision map and adding the traffic sign information and the corresponding traffic sign board information into the high-precision map information. Therefore, data are convenient to store; the traffic sign board is located accurately; and when applications like high-precision map information navigation are used, the correct guidance can be provided conveniently.

Description

technical field [0001] The invention relates to the field of high-precision map information collection, in particular to a traffic sign information collection method serving high-precision map production. Background technique [0002] High-precision maps contain a large amount of auxiliary driving information, among which traffic sign information is particularly important. The collection of traffic sign information is an important module of high-precision map production. The existing technical means are mostly traffic sign detection combined with GPS positioning measurement. [0003] At present, in terms of traffic sign detection, visual detection methods are mainly used to obtain the position information of traffic signs, such as a traffic sign detection and recognition method based on the residual SSD model with the patent number CN108960198A. block; use the residual network ResNet101 as the basic network of SSD, build a residual SSD model, carry out network training, comp...

AI Technical Summary

Problems solved by technology

[0004] In the existing schemes, it only involves the detection and recognition of traffic signs in the image, or the research on the distance measurement and positioning of traffic signs, while the high-precision map information requires more detailed traffic sign information, such as Steering information, speed limit information and other information used to guide users to drive

Images