The invention relates to a vehicle detection and trajectory prediction method for a roadside laser radar. The method comprises the following steps: step 1, establishing a multi-task deep neural network model for vehicle detection and trajectory prediction; 2, pre-training the multi-task deep neural network model based on the public data set; 3, performing background filtering on the point cloud by adopting a background filtering method based on voxelization processing; 4, migration training of the multi-task deep neural network model is carried out based on the self-built road side end data set; and 5, deploying the multi-task deep neural network model, and carrying out vehicle detection, tracking and trajectory prediction. Compared with the prior art, the method has the advantages that the accuracy of the detection model and the accuracy of trajectory prediction are improved, and the like.
本发明涉及一种路侧激光雷达的车辆检测和轨迹预测方法,该方法包括:步骤1:建立车辆检测和轨迹预测的多任务深度神经网络模型;步骤2:基于公开数据集对多任务深度神经网络模型进行预训练;步骤3:采用基于体素化处理的背景滤波法对点云进行背景滤波;步骤4:基于自建路侧端数据集进行多任务深度神经网络模型的迁移训练;步骤5:部署多任务深度神经网络模型,进行车辆的检测、跟踪和轨迹预测,与现有技术相比,本发明具有提升检测模型的准确率和轨迹预测的准确率等优点。
Vehicle detection and trajectory prediction method for roadside laser radar
一种路侧激光雷达的车辆检测和轨迹预测方法
2022-03-08
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06T Bilddatenverarbeitung oder Bilddatenerzeugung allgemein , IMAGE DATA PROCESSING OR GENERATION, IN GENERAL |
METHOD FOR REPAIRING VEHICLE TRAJECTORY AT SIGNALIZED INTERSECTION ON BASIS OF ROADSIDE LASER RADAR
European Patent Office | 2025
|Vehicle queuing length detection method based on roadside laser radar
European Patent Office | 2023
|Pedestrian crossing trajectory tracking and predicting method based on roadside laser radar
European Patent Office | 2022
|European Patent Office | 2021
|