一种基于深度学习的自校准雷达测速方法及系统
By combining deep learning methods with radar speed detectors and camera-captured images, the radar speed detector is automatically calibrated, solving the accuracy problem of radar speed detectors during their service life, reducing verification costs and risks, and improving detection accuracy and efficiency.
CN117148335BActive Publication Date: 2026-07-17HUNAN INST OF METROLOGY & TEST
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUNAN INST OF METROLOGY & TEST
- Filing Date
- 2023-09-01
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Radar speed detectors are difficult to maintain accuracy during their service life, and their calibration costs are high and they pose safety risks.
Method used
A deep learning-based self-calibration radar speed measurement method is adopted. By combining images captured by the radar speedometer and camera, the target vehicle recognition neural network and speed calculation neural network are used for automatic calibration to adjust the radar speedometer to maintain accuracy.
Benefits of technology
It enables automatic calibration of radar speedometers within the calibration cycle, reducing calibration costs and risks, and improving the accuracy and computational efficiency of speed detection.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117148335B_ABST
Abstract
本发明公开了一种基于深度学习的自校准雷达测速方法及系统，方法包括：学习阶段：采集车辆经过时雷达测速仪测得的车速和相机抓拍的图像，输入速度计算神经网络进行训练；使用阶段：车辆经过时，通过雷达测速仪得到检测车速的同时，获取相机抓拍的图像，通过训练后的目标车辆识别神经网络识别固定时间间隔起、止点抓拍图像中的目标车辆，再输入训练后的速度计算神经网络，输出计算速度；根据计算速度进行检测速度的误差判断，若存在误差，则根据误差，对所述雷达测速仪进行自动调整。本发明解决雷达测速仪难以在检定周期内始终保持准确性，以及雷达测速仪的检定成本高、有安全风险的问题。
Need to check novelty before this filing date? Find Prior Art