The Aspectus VI-System™ is a video intelligence system for large-scale video surveillance networks, based on Aspectus' patented IPoIP™ (Image Processing over IP) technology. The system includes platform components and specific algorithms called Detectors. Following are the main system components.
The VI-Agent™ is a software module that performs the raw processing of images as seen by each camera in the field prior to any image compression. This process is called the "feature extraction." The result is very low-bandwidth data (usually less than 15Kbps) which is transmitted to the VI-Server™ for further processing.
The VI-Agent™ is designed for easy integration inside video devices such as IP cameras, DVRs and video encoders.
Any video device that hosts a VI-Agent™ becomes compatible with the VI-System™ and can benefit from a wide range of video intelligence capabilities.
VI-Agent™ is available both inside Aspectus' and 3rd party video devices. Aspectus offers two hardware platforms containing the VI-Agent™. One is a PC-based MPEG4 encoder and the other is an embedded MPEG4 encoder called VideoOutpost™.
The VI-Server™ is the "brain" and main component of the VI-System™ platform. It receives and processes data streams from up to thousands of video devices with embedded VI-Agent™ connected to the network. Using the IPoIP™ architecture, which supports low bandwidth bi-directional communication with VI-Agent™ enabled units, the VI-Server™ performs a wide range of software pre-configured detection missions in real time.
The server hardware can be custom configured in a variety of manners to achieve a desired level of fault tolerance and redundancy for a specific installation. The VI-Server™ software is designed to support single and multi-processor hardware on a Windows™ or Linux™ based environment.
All the system data, including event detection rules and detailed event history, resides on a central database that also provides backup, retrieval, and extensive reporting capabilities.
The VI-Sentry™ is a C2 (command & control) application which enables users to monitor and handle events, set event detection criteria and control various parameters. Using VI-Sentry™, a single operator may handle, review, and analyze automatically detected suspicious activity from thousands of cameras in the field.
VI-Sentry™ is designed to minimize the amount of visual information exposed to the operator. Only suspected events or manually selected cameras are displayed.
In order to facilitate the definition of event criteria rules, VI-Sentry™ uses a friendly wizard that guides the user step-by-step how to set the very powerful rules easily and efficiently.
The VI-Sentry™ also supports definition of complex multi-camera events (e.g. a person is detected by both camera 1 and camera 2) and also sequences (e.g. a vehicle stops for more than 1 minute, and a person leaves the vehicle unattended).