Intelligent video in the physical security industry has historically focused on applications such as perimeter detection, wrong-way detection and loitering detection. More recently, business intelligence algorithms have been used by retailers to monitor queue lengths, count the number of customers entering a store and provide heat maps and people flow analysis.
Intelligent video in the physical security industry has historically focused on applications such as perimeter detection, wrong-way detection and loitering detection. More recently, business intelligence algorithms have been used by retailers to monitor queue lengths, count the number of customers entering a store and provide heat maps and people flow analysis. However, with the explosion of interest in intelligent video and the success of products such as Microsoft's Kinect, will 2013 be the year of intelligent video?
Forums discussing this concept are not new. In 2011, IMS Conferences ran an intelligent video conference in Los Angeles and the Embedded Vision Alliance has regularly been bringing together engineers from leading companies to discuss the industry. These discussions have covered markets as diverse as consumer and military.
In non-security applications, video analytics can also provide real value. For example, the medical industry has been looking at using analytics to identify that a patient has rolled out of bed or that an employee has not washed their hands following an examination. In the case of the patient falling out of bed, the solution alerts someone to go to their aid. In the other case, a simple reminder is likely to be enough to ensure that hygiene protocols are met. Another example is in the automotive industry where applications include driver drowsiness and lane departure alerts as well as, potentially, vehicles that can assess the dangers on the road and drive themselves.
One imminent application of intelligent video is likely to be in digital signage. The application is based around a small camera which is able to recognize the age and gender of the person viewing the advert. Similar to tailored adverts on social media websites, an automatic decision can be made on which advert to present.
At a simplistic level, this could mean advertising perfume to a woman and aftershave to a man. However, the potential exists to expand this offering. For example, if the person is standing beside a pram then baby and child related products could be advertised. If the person is identified as being within a specific age range, then products tailored to that age demographic could be advertised. Provided that the customer is not offended by the advertisement decisions, this is one use of video analytics where the return on investment is clear.