TENTACLE: Multi-Camera Immersive Surveillance System

• Main Authors: Benjamin, Justin W, Post, Benjamin L, Estes, Kyle K, Milbert, Randy L
• Format: Report
• Language: English
• Subjects: ANIMATED GRAPHICAL INTERFACE; AUTOMATED IDENTIFICATION AND TAGGING; CAMERAS; DATA BASES; EVENT; FEATURE EXTRACTION; FRAMEWORK; GRAPHICAL USER INTERFACE; IDENTIFICATION; METADATA; MULTI-CAMERA MULTI-SENSOR FUSION ALGORITHMS; PE65505F; PEOPLE AND VEHICLE TRACKING; Photography; QUERY; REAL-TIME VIDEO SURVEILLANCE; SBIR REPORTS; SBIR TOPIC AF103-032 PHASE 1; SBIR(SMALL BUSINESS INNOVATION RESEARCH); SENSOR FUSION; SURVEILLANCE; TRACKING; VIDEO SIGNALS; WUAFRL3005CV09

This final technical report describes the results of collaborative efforts by Primordial, intuVision, and All Hazards Management in conducting a phase I Small Business Innovation Research (SBIR) effort, titled AF103-032: Multicamera real-time Feature Recognition, Extraction &Tagging Automation (McFRETA), and authored by the U.S. Air Force Research Laboratory (AFRL). The goal of the effort was to design, develop, and demonstrate a scalable framework and software application to enable tracking and feature extraction of uncooperative entities capture in multiple streaming sources and enabling near real-time forensic analysis of the extracted data. During the effort, Primordial worked with intuVision to identify key features to extract from live video feeds to include geographic location, entity type, and dominate colors of the entity. Primordial then built a graphical user interface to display avatars representing real world entities in a three dimensional virtual reality using the WorldWind SDK. To prove feasibility of the system, Primordial conducted 13 internal demonstrations to validate the results of the system.