追踪车辆的轨迹识别方法、装置、设备及存储介质

By acquiring vehicle tracking images and road construction images, and combining frame rate segmentation processing and a preset detection model, accurate vehicle trajectory recognition was achieved, solving the accuracy and precision problems of vehicle trajectory recognition in existing technologies and improving the effect of vehicle detection and tracking.

CN115588181BActive Publication Date: 2026-07-17DONGFENG LIUZHOU MOTOR

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
DONGFENG LIUZHOU MOTOR
Filing Date
2022-09-15
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies cannot effectively guarantee the accuracy and precision of vehicle trajectory recognition, especially in methods using deep learning and two-stage networks.

Method used

By acquiring vehicle tracking images and road construction images of the target tracking road, the position and vehicle identification code of each tracked vehicle are determined, and the trajectory is drawn by combining the road construction image. Frame rate segmentation processing and preset detection models are used to match vehicles and connect trajectories to achieve accurate vehicle trajectory recognition.

Benefits of technology

It improves the accuracy and precision of vehicle trajectory recognition, meets the needs of special scenarios, and enhances the effect of vehicle detection and tracking.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115588181B_ABST
    Figure CN115588181B_ABST
Patent Text Reader

Abstract

本发明属于车辆追踪技术领域,公开了一种追踪车辆的轨迹识别方法、装置、设备及存储介质。该方法包括:获取目标追踪道路的车辆追踪图像和道路构建图像;根据所述车辆追踪图像确定各追踪车辆的追踪位置和各追踪车辆的车辆标识码;根据所述道路构建图像、所述追踪位置以及所述车辆标识码进行轨迹绘制,得到各追踪车辆的车辆行驶轨迹;根据各追踪车辆的车辆行驶轨迹完成对各追踪车辆的轨迹识别。通过上述方式,实现对追踪车辆的检测和跟踪,并将追踪车辆在目标追踪道路上进行演示,能够满足特殊场景的需求,并提升了追踪车辆识别的准确性和精度。
Need to check novelty before this filing date? Find Prior Art