Traffic sign data processing method, electronic device and storage medium

Abstract

The invention discloses an electronic device, a traffic sign data processing method and a storage medium. The method includes: acquiring collected traffic sign data, the traffic sign data including each traffic sign has a corresponding vertex set data, the traffic sign The board comprises a zebra crossing; Calculate the center point of each traffic sign in the plurality of traffic signs; Calculate the first distance from the center point of each traffic sign to a preset vertex of the traffic sign where it is located, and obtain each The first distance corresponding to the traffic sign board; filter out the zebra crossing data from the traffic sign data, and determine the zebra crossing set, wherein each piece of zebra crossing data in the zebra crossing set includes the vertex set data of the zebra crossing, the central point of the zebra crossing and the corresponding zebra crossing The first distance; based on the set of zebra crossings, determine the processed zebra crossings. The embodiment of the present application can automatically detect overlapping, redundant or wrong position of traffic signs to improve the accuracy of map making.

Description

technical field [0001] The invention relates to the field of automatic driving, in particular to a traffic sign data processing method, an electronic device and a storage medium. Background technique [0002] In the field of autonomous driving, the drawing of crowdsourced high-precision maps usually requires the construction of semantic geometry layers first. The drawing cost of crowdsourcing collection vehicles is low, so crowdsourcing collection vehicles are usually used to collect map-related data, and they will be extensively arranged to collect high-quality data and increase the update frequency of high-precision maps. At the same time, due to the low accuracy of crowdsourcing collection data , the distribution of traffic signs is complex, and it is inevitable that misjudgment will occur when the program processes traffic sign data when constructing a semantic geometry layer, resulting in overlapping, redundant or wrong position of traffic signs. These errors require m...

AI Technical Summary

Problems solved by technology

The drawing cost of crowdsourcing collection vehicles is low, so crowdsourcing collection vehicles are usually used to collect map-related data, and they will be extensively arranged to collect high-quality data and increase the update frequency of high-precision maps. At the same time, due to the low accuracy of crowdsourcing collection data , the distribution of traffic signs is complex, and it is inevitable that misjudgment will occur when the program processes traffic sign data when constructing a semantic geometry layer, resulting in overlapping, redundant or wrong position of traffic signs

These errors require manual visual inspection of the results of traffic sign fusion processing. The overlapping, redundant or wrong position of traffic signs is very time-consuming and easy to miss.

Images