Traffic sign tracking method, device, equipment and storage medium

Abstract

The present application relates to a traffic sign tracking method, device, equipment and storage medium. The method includes: when continuously collecting image frames from a vehicle driving scene, screening feature points in the current frame according to target contour parameters of traffic signs in the current frame; if the number of feature points is less than or equal to a preset number threshold, according to The target contour parameter locates the target area of ​​the traffic sign in the next frame; intercepts the first image block in the target area from the next frame, and predicts the first profile parameter of the traffic sign in the next frame based on the first-level prediction model; according to the first A contour parameter is used to generate an outer frame, and the second image block is intercepted from the next frame according to the outer frame, and the target contour parameter of the traffic sign in the next frame is predicted based on the secondary prediction model. If it is greater than the preset number threshold, predict the target contour parameters in the next frame according to the optical flow of the feature points; take the next frame as the current frame for iterative processing to continuously track the traffic signs. This scheme can improve the accuracy.

Description

technical field [0001] The present application relates to the field of intelligent driving, more to the field of image processing technology, and in particular to a traffic sign tracking method, device, equipment and storage medium. Background technique [0002] Nowadays, with the rapid development of science and technology and the continuous improvement of the requirements for vehicle driving safety, intelligent driving technology has gradually become one of the main research directions. Traffic signs are fixed and frequently appear in vehicle driving scenes. Accurate detection of traffic signs is very important for intelligent driving scenarios. [0003] In the traditional method, the detection and recognition of traffic signs can only perform rough area positioning detection through traditional target detection algorithms, so only rough location information can be detected, which obviously cannot meet the needs of complex intelligent driving scenarios. For example, only ...

AI Technical Summary

Problems solved by technology

[0003] In the traditional method, the detection and recognition of traffic signs can only perform rough area positioning detection through traditional target detection algorithms, so only rough location information can be detected, which obviously cannot meet the needs of complex intelligent driving scenarios. For example, only To detect traffic signs based on rough area positioning, it is basically difficult to achieve continuous tracking of traffic signs, resulting in low accuracy

Images