AI-based video analysis for traffic monitoring

Video data from these surveillance cameras is ex-tremely huge. When it requires extracting information about an object or event from the surveillance camera without knowing the exact time, it is very challenging to immediately identify it from billions of video frames. Thus, how to effectively extract information from video surveillance has a strategically important role in practical applications. In this paper, we introduce a system to manage and retrieve surveillance videos based on the indexing of moving objects. The system contains object detection and features extraction, video indexing, retrieval processing mechanism, and fast previewing. Specifically, this system uses information from moving objects to index video content. As a result, in the retrieval phase, the system allows both text-based query and image-based query mechanisms. The system performance are evaluated on traffic surveillance videos and got promising results.