The invention relates to a monitoring video detection method for continuous lane changing and illegal turning of road vehicles, and the method comprises the steps: collecting a video image of a road through a camera, and processing the video image through a computer to judge whether there is continuous lane changing and illegal turning of an urban road vehicle or not; the method is high in detection precision, , the vehicle is detected based on the deep learning method, the vehicle is matched and tracked, the lane line is subjected to nonlinear fitting in an artificial neural network mode, theapplication range is wide, the method can not only detect a straight lane, but also detect complex conditions such as a curve and the like, the real-time performance is high, and violation behaviorsin the video monitoring vehicle can be detected in real time.
本发明涉及一种道路车辆连续变道和违规掉头的监控视频检测方法,摄像头采集道路的视频图像,通过计算机处理视频图像判断城市道路车辆有否连续变道和违规掉头;本发明检测精度高,基于深度学习方法检测车辆,并对车辆匹配跟踪,采用了人工神经网络的方式对车道线进行非线性拟合,适用范围广,本方法不仅可以检测到直线车道,还可以检测弯道等复杂情况,实时性强,本方法可以实时检测视频监控车辆中的违规行为。
Monitoring video detection method for continuous lane changing and illegal turning of road vehicles
一种道路车辆连续变道和违规掉头的监控视频检测方法
2020-05-26
Patent
Elektronische Ressource
Chinesisch
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Europäisches Patentamt | 2023
|Motor vehicle blind area lane changing illegal snapshot system
Europäisches Patentamt | 2020
|Method for monitoring illegal lane change of expressway
Europäisches Patentamt | 2023
|Illegal lane changing supervision method and system based on trajectory tracking
Europäisches Patentamt | 2024
|