We use proprietary and third party's cookies to improve your experience and our services, identifying your Internet Browsing preferences on our website; develop analytic activities and display advertising based on your preferences. If you keep browsing, you accept its use. You can get more information on our Cookie Policy
Cookies Policy
SAFECITY.Story.DATA/CONTEXT MANAGEMENT.Data.MultimediaAnalysis.VideoProcessing&AnalysisEngine - FIWARE Forge Wiki

SAFECITY.Story.DATA/CONTEXT MANAGEMENT.Data.MultimediaAnalysis.VideoProcessing&AnalysisEngine

From FIWARE Forge Wiki

Jump to: navigation, search
Name VideoProcessing&AnalysisEngine
Goal To provide the required video processing and analysis algorithms in order to visually perceive potential threat cases in camera monitored areas
Description Component that includes a rich set of automated or semi-automated (where required) image/video processing algorithms such as:
  • Region/Object segmentation (car, person, animal, vehicle, etc.) by color, shape, texture, edges, etc.
  • Motion Estimation
  • Moving Object Segmentation
  • Moving Object Tracking (person, vehicle, animal, etc.)
  • Foreground/Background Detection
  • Face detection
  • Face identification
  • Orphan object detection
  • Car detection and tracking
  • Licence Plate recognition
  • Line drawing
  • Change detection
  • Activity Detection
  • Event Classification and Detection (use of the respective ontology and pattern recognition algorithms)
  • Suspicious Behaviour Classification and Detection (use of the respective ontology and pattern recognition algorithms) such as detection of loitering, fights, races, falls, movement against the flow of traffic, etc.

A sub-component of the engine further performs video pre-processing for video content filtering and selective video data streaming at the ARATOS gateway, which ideally is performed in real-time at low processing and memory capabilities hardware. The tasks performed include the following:

  • Noise filtering
  • Illumination changes compensation
  • Activity detection
  • Change detection

In many cases, the type and sequence of the video processing and analysis steps is driven by the predefined safety and security related knowledge domain representations to lead to the desired situation insights generation.

Version 1.0 Source Athens Information Technology - AIT, Sofia Tsekeridou
Scope Platform Epic {{{Epic}}} Theme
Status Pending MoSCoW priority MST Relative priority
Chapter Data/Context Management, Enabler Multimedia Analysis
Stakeholder Owner Athens Information Technology - AIT, Sofia Tsekeridou
Complexity Rationale The efficiency of the video analytics engine and its real-time operation are crucial factors in automated camera-enabled surveillance systems

Personal tools
Create a book