The embodiment of the invention relates to the field, and discloses a traffic signal lamp detection method. The method comprises the steps of firstly obtaining an image collected by a camera, then obtaining a bounding box, a label and a confidence coefficient of a traffic signal lamp in the image based on a deep learning algorithm, finally determining the position of the traffic signal lamp according to the bounding box, and meanwhile, determining the state of the traffic signal lamp according to the label and the confidence coefficient. The detection method provided by the embodiment of the invention can accurately detect and identify the state and the position of the traffic signal lamp.
本发明实施例涉及领域,公开了一种交通信号灯的检测方法,该方法首先获取摄像头所采集的图像,然后基于深度学习算法,获取所述图像中交通信号灯的边界框、标签和置信度,最后根据所述边界框,确定所述交通信号灯的位置,同时,根据所述标签和置信度,确定所述交通信号灯的状态,本发明实施例提供的检测方法能够准确检测和识别交通信号灯的状态和位置。
Traffic signal lamp detection method and device, and automatic driving automobile
一种交通信号灯的检测方法、装置及自动驾驶汽车
2021-01-29
Patent
Elektronische Ressource
Chinesisch
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Automatic correction device of traffic signal lamp and traffic signal lamp system
Europäisches Patentamt | 2022
|Traffic signal lamp recognition device for unmanned driving
Europäisches Patentamt | 2021
|Signal lamp detection method, device and system and automatic driving vehicle
Europäisches Patentamt | 2024
|Automatic adjusting device for traffic signal lamp panel
Europäisches Patentamt | 2021
|