The invention belongs to the field of intelligent traffic monitoring and provides a traffic abnormity detection method and system. According to the method and system, a normal traffic video image sequence is divided into video block sequences, the number of lenses in the video block sequences is detected, a Gaussian model of the number of the lenses in the video block sequences is established, and abnormity detection is conducted on tested traffic video images through the Gaussian model. Compared with the existing manual analysis mode, monitoring intelligentization is realized, the real-time performance is high, and the labor cost is lowered; meanwhile, the block and lens technology enables the detection speed to be higher, the method and system are particularly suitable for the field of traffic video monitoring with a large processing data amount, a complex scene and fast changes, and an accident site can be positioned quickly and accurately.
Traffic abnormity detection method and system
2015-04-22
Patent
Elektronische Ressource
Englisch
Proactive Traffic Abnormity Warning System
Springer Verlag | 2023
|Europäisches Patentamt | 2021
|Automobile tire health abnormity detection system and detection method
Europäisches Patentamt | 2023
|Beidou-based traffic abnormity scene pre-avoiding device
Europäisches Patentamt | 2023
|