The invention discloses a traffic anomaly detection method and device and a model training method and device, and relates to the technical field of artificial intelligence, in particular to the field of computer vision and the technical field of deep learning. The specific implementation scheme of the method is characterized by determining objects in traffic data, wherein the objects comprise traffic facilities and traffic participants; determining correlation parameters betwen the objects in the traffic data; and according to the correlation parameters between the objects, using a pre-trained traffic anomaly detection model to obtain a detection result of traffic anomaly. Abnormal conditions can be reported and broadcasted in time; and the cause of the abnormal conditions are analyzed and judged.
本申请公开了交通异常的检测方法、模型的训练方法、装置,涉及人工智能技术领域,尤其涉及计算机视觉领域和深度学习技术领域。具体实现方案为:确定交通数据中的对象,对象包括交通设施以及交通参与者;确定交通数据中的对象之间的关联参数;根据对象之间的关联参数,利用预先训练的交通异常检测模型,得到交通异常的检测结果。可以对异常情况及时上报和广播;对异常情况发生的原因进行解析判断。
Traffic anomaly detection method and device and model training method and device
交通异常的检测方法、模型的训练方法及装置
2021-06-04
Patent
Electronic Resource
Chinese
Traffic violation model training process anomaly detection method
European Patent Office | 2024
|European Patent Office | 2023
|Construction method of urban traffic anomaly detection model
European Patent Office | 2020
|