3 results about "Edge computing" patented technology

The invention discloses a computing migration method of a 5G mobile application in a mixed environment, which comprises the following steps of: firstly, acquiring a demand of the computing applicationon edge computing equipment, acquiring idle resources of the edge computing equipment around a mobile terminal, and screening out available edge computing equipment according to the demand; selectingwhether an application migration target is an edge device or a cloud device according to whether an available edge computing device exists or not; and generating a corresponding calculation strategyfor each migration target; calculating the time delay of each strategy according to each edge calculation strategy to form a calculation strategy time delay set; and selecting a calculation strategy with the lowest time delay as a final strategy to carry out calculation migration. In the process of searching the computing migration device, the pruning strategy is adopted, the device meeting the computing application requirement is searched all the time, a searching range is narrowed, and time delay consumption is reduced. The edge device and the cloud end are fully utilized. The edge device isfully used and meanwhile execution of the computing application is guaranteed with the cloud end as a guarantee.

The invention provides a bus-mounted video passenger OD analysis system, method and device and a storage medium, and belongs to the field of intelligent traffic and computer vision.The method comprises the following steps: encoding and decoding bus video streams in real time through intelligent edge computing equipment; detecting and tracking each passenger appearing in the bus video through a target tracking network; judging the passenger getting-on and getting-off state detected and tracked in the video, and storing and taking the snapshot information of the getting-on and getting-off passenger in the moving process; extracting feature vectors of snapshots of boarding and alighting passengers through a pedestrian re-recognition algorithm, and storing the feature vectors into a database; capturing data such as a bus GPS (Global Positioning System) through intelligent vehicle-mounted equipment, and storing the data into a database; and performing passenger similarity matching and data association through an OD association algorithm, so that the travel OD of the bus passengers is obtained through analysis; The problem that passenger OD analysis is not accurate in an existing bus scene is effectively solved, and the method can be conveniently deployed and transplanted to city-level large-scale application.