Eptascape

Spec Sheet
Intelligent Video
MPEG-7 Overview
Video Demos

Intelligent Video Surveillance

The task of preventive video surveillance relies mostly on the ability of humans to detect meaningful changes in the scene being monitored to detect a possible breach of security or a dangerous situation. Detecting critical situations as early as they arise can have a major impact in taking effective countermeasures and in the effectiveness of a video surveillance system.

Typical functions in video surveillance systems using the Eptascape software may involve identifying objects as they appear or disappear in a scene or being able to identify and count how many people are currently in the scene, as well as calculate how quickly they are moving and in which directions.

At the foundation of Eptascape intelligent video surveillance is the task of automatic analysis of video content the way humans perceive visual information. Making scene understanding possible requires a paradigm shift from pixel and frame-based imaging to an object-based one. Segmentation of an image into meaningful objects and their subsequent tracking in real-time are the challenges that need to be addressed to extract the semantic of video objects acquired by cameras in the field.

Eptascape approach takes in consideration advances in Computer Vision, together with the availability of inexpensive, powerful hardware, to design products that use automatic scene analysis for the task of increasing the efficiency of surveillance personnel in promptly detecting potentially dangerous situations at an affordable cost.

Examples (Click on links to see video demos)

• Baggage detection
  • detecting unattended luggage in airports and other transportation facilities
• Intrusion Detection
  • detecting people or vehicles outside allowed areas
  • implementing a virtual fence (wire trap) like detection of crossing of a door or climbing over a wall/fence
  • advanced motion detection applications in general
• Tail-gating
  • detecting two or more people going through an access gate at the same time with the same badge
• Direction flow
  • detecting people going against the allowed flow like in an airport exit
    

Intelligent Transport Systems

• Detecting pedestrians
  • waiting in a queue in potentially hazardous areas
• Detecting vehicles
  • blocking for a breakdown or an accident
  • outside allowed areas
  • going the wrong direction
  • running through red lights
  • waiting in a queue