The invention discloses a traffic scene abnormal target detection method based on rules and learning. The method comprises the following steps: obtaining traffic scene video data collected in real time; performing statistics and operation on the traffic scene video data to obtain suspected abnormal motion information of a target object; and according to the suspected abnormal motion information and a traffic rule, determining a target of abnormal motion of the traffic scene. According to the embodiment of the invention, the suspected abnormal motion information of the target object is obtainedby performing statistics and operation on the acquired traffic scene video data, and the target of abnormal motion of the traffic scene is determined according to the suspected abnormal motion information and the traffic rules, so that the operation amount is small, high real-time performance is realized, and the determination of the abnormal moving target in the traffic scene is more accurate.
本发明公开了一种基于规则与学习的交通场景异常目标检测方法,所述方法包括:获取实时采集的交通场景视频数据;对所述交通场景视频数据进行统计和运算,得到目标物体的疑似异常运动信息;根据所述疑似异常运动信息和交通规则,确定交通场景异常运动的目标。本发明实施例通过将采集的交通场景视频数据进行统计和运算,得到目标物体的疑似异常运动信息,再根据疑似异常运动信息和交通规则确定出交通场景异常运动的目标,运算量小,实现高实时性,对交通场景异常运动目标的确定更加精准。
Traffic scene abnormal target detection method based on rules and learning
一种基于规则与学习的交通场景异常目标检测方法
2021-03-09
Patent
Elektronische Ressource
Chinesisch
Europäisches Patentamt | 2020
|Automatic driving solution under multi-target complex traffic scene based on reinforcement learning
Europäisches Patentamt | 2022
|Scene target-oriented traffic control efficiency evaluation method
Europäisches Patentamt | 2021
|Vectorization trajectory prediction method based on adaptive target in traffic scene
Europäisches Patentamt | 2023
|Method, device and equipment for generating playback video of abnormal traffic scene event
Europäisches Patentamt | 2023
|