By Anthony Hildebrand
http://www.info4security.com/story.asp?sectioncode=9&storycode=4120956&c=1
Video analytics is the most important breakthrough in the application of CCTV. This was the message from Patrick Meaney at last week’s Next Generation CCTV conference.
Meaney, founder of website www.videoanalytics.info, gave his presentation at the event in London, and said that CCTV had been initially installed as a crime prevention measure, but hadn’t delivered the goods in this respect – which is where video analytics comes in.
He said video analytics had the potential to become “the third forensic science, in addition to fingerprinting and DNA”.
Operator assistance
He quoted figures on control room operator fatigue and concentration loss, and said the review of recorded video could be equally as impaired.
Meaney said the advantage of video analytics for monitoring live video is in being able to automatically draw the operator’s attention to relevant events, while for recorded video it can automate the analysis of the footage, extracting only events or incidents of interest.
But, he said, the analytics industry had been damaged by the performance of the earliest systems to hit the market. These were oversold, and failed to deliver on exaggerated claims, Meaney said. He said he had seen one major control room where staff routinely switched off the analytics system due to the constant alarms it triggered.
“The technology is still being oversold,” Meaney said. “The customer has got to be absolutely clear about what they want, and not take things at face value.” He said the technology is till being oversold, and that customers need to be educated about “the art of the possible”.
For instance, he said: “To make it work to its ideal level you do really need real-time video.” He said that for the first time, “scene depth” can be considered when using real-time, high definition video.
Changing conditions
Meaney went into the differences between simple motion detection – which is still sometimes described as ‘analytics’ or ‘intelligent video’ – and the more discriminatory video analytics.
To cope with changing environmental conditions, he said you need a video analytics system with adaptive algorithms specifically designed to take these factors into account.
He said there were a number of factors to be considered when implementing a system: Does it do what you need it do? Is it configured for the application and environment?
Training criticism
He also had some criticism for the levels of education and training being provided to operators, both from suppliers and the end-user management.
“It staggers me,” he said. “Usually the operator training is pretty lamentable in terms of operators understanding and knowing what they’re working with.”
Meaney also stressed the cost benefits of using video analytics systems.
“In the current economic environment, technology is perceived to be adding real value into the mix,” he said. “For the first time we can talk about security in terms of a return on investment, and not a grudge purchase.”
This was able to be demonstrated through time savings in reviewing footage and in following up incidents.
