A roadside target detection method and device for travel map danger warning

By constructing a roadside target detection framework based on Transformers and using a ground plane encoder and decoder for correction, the problem of low accuracy in roadside visual target detection is solved, and higher accuracy in traffic hazard identification and safety warning is achieved.

CN117523172BActive Publication Date: 2026-06-26TSINGHUA UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TSINGHUA UNIVERSITY
Filing Date
2023-09-26
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

The lack of high-precision roadside visual target detection methods in existing technologies makes it impossible for travel map platforms to effectively identify potential traffic hazards, affecting the accuracy of user safety warning systems.

Method used

A Transformers-based roadside target detection framework, including a visual encoder, a ground plane encoder, and a ground plane guided decoder, is adopted. Scene-level ground plane cues are captured through a self-attention mechanism and corrected by combining the ground plane equation graph, thereby improving the accuracy of 3D target detection.

Benefits of technology

It improves the accuracy of target detection in roadside monocular images, enabling more accurate identification of potential hazards on the road and providing high-precision safety warning data for travel map platforms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117523172B_ABST
    Figure CN117523172B_ABST
Patent Text Reader

Abstract

The application provides a roadside target detection method and device for travel map danger warning, the method comprises the following steps: processing a roadside image by using a feature extraction model to obtain visual features and ground plane features; processing the visual features by using a visual encoder to obtain visual embedding features; processing the ground plane features by using a ground plane encoder to obtain ground plane embedding features; processing the visual embedding features and the ground plane embedding features by using a ground plane guided decoder to obtain a ground plane perception object query; processing the ground plane perception object query by using a detection head to obtain a target detection result; processing the ground plane features by using a ground plane predictor to obtain a ground plane equation graph; and calculating a target depth value by using two-dimensional pixel coordinates of a target bottom surface center, the ground plane equation graph and camera parameters. The application effectively improves the target detection accuracy of the roadside monocular image and provides a key data basis for the danger warning system of the travel map.
Need to check novelty before this filing date? Find Prior Art