Page 773 - Cloud computing: From paradigm to operation
P. 773
Video processing and storage 4
Step 2: After receiving the human recognition request, the IVS system authenticates the user information,
requests the cloud computing platform to create the intelligent human recognition function, and then
responds to the CU with the information from the intelligent human recognition function.
Step 3: After receiving the response of the IVS system, the CU sends the request to the cloud computing
platform.
Step 4: The cloud computing platform downloads the relevant video files from the media storage unit (MSU),
processes the video files in parallel and then sends the human recognition results back to the CU.
Step 5: When the intelligent human recognition task finishes, the cloud computing platform destroys the
intelligent human recognition function to release the corresponding resources.
6.4.2 Case 2: Video synopsis
The IVS system deployed by a public security department produces massive surveillance video continuously
and the cost of data storage is high. However, the public security department is usually interested in moving
objects in the surveillance video. Therefore, to save storage space, surveillance video is usually processed to
generate synopsis video before long-term storage. The objective of video synopsis is to shorten the video
time by reorganizing moving video objects in temporal and spatial dimensions. For example, by using the
video synopsis operation, a 24 h surveillance video file is transformed into a 30 min video file that contains
all moving objects in the original video file. The intelligent video synopsis function can be virtualized as
functional components in the cloud computing platform, and can be dynamically created according to user
demand.
Step 1: Users log into the IVS system through the CU, and set several functional parameters of the video
synopsis, such as the camera channel numbers, the time period of interest in the historical video and the
density of the objects. Then users send an intelligent video synopsis request to the IVS system.
Step 2: After receiving the video synopsis request, the IVS system authenticates the user information,
requests the cloud computing platform to create the intelligent video synopsis function and then responds
to the CU with the information from the intelligent video synopsis function.
Step 3: After receiving the response of the IVS system, the CU sends the request to the cloud computing
platform.
Step 4: The cloud computing platform downloads the relevant video files from the MSU, processes the video
files in parallel and then uploads the video synopsis results back to the MSU.
Step 5: When the intelligent video synopsis task finishes, the cloud computing platform destroys the
intelligent video synopsis function to release the corresponding resources.
7 Requirements for cloud computing platform supporting visual surveillance
7.1 User requirements
There are two types of user: cloud computing service consumers and cloud computing service providers.
7.1.1 Cloud computing service consumer requirements
– USR-01: A visual surveillance cloud computing (VSCC) platform is required to support registration
and de-registration of the end user through the interface provided by the system and the end user
can view and modify personal information.
– USR-02: A VSCC platform is required to support convenient end-user login to and logout from the
system. A username and password are required when an end user logs into the system.
– USR-03: A VSCC platform is recommended to support end-user view of the user access logs or other
system logs.
– USR-04: A VSCC platform is recommended to support real-time surveillance video distribution.
765