Making data searchable using robust algorithms - identifying the most appropriate algorithm for the situation
Prof Shaogang (Sean) Gong, Professor of Visual Computation, Queen Mary University of London
From: Managing data session of the Video Data Analysis seminar, 28 November 2011, London
28 November 2011 Communications channel
>> Play webcastThis presentation will look at:
* Giving structure to raw data for it to be automated and searched, to deliver rapid and comprehensive searches
* Ensuring searches are quick and robust
* Filtering out the ‘noise’ in CCTV data to find the needle in the haystack
* Project work: SAMURAI (Suspicious and Abnormal Behaviour Monitoring Using a Network of Cameras for Situation Awareness Enhancement) and BEWARE (Behaviour based Enhancement of Wide-Area Situational Awareness in a Distributed Network of CCTV Cameras).
Since 7/7, vast amounts of video data is captured across the civilian and defence world. The need to manipulate this data to identify patterns, detect anomalies to deliver heightened situational awareness as quickly and reliably as possible has never been so pressing.