|Real-time search analytics addresses one of the most important control room tasks - locating a person of interest|
If you have been to any of the many security industry tradeshows this year you will undoubtedly have seen and heard the phrase ‘next generation video analytics’. Is it just a catchy marketing phrase or is there more substance behind it? Video analytics as a technology has been with us for many years, but there has always been an air of confusion and mystery around it, in large part created by Hollywood movies, where every camera is connected, an operator can search the network and locate the villain in a matter of seconds.
In many ways, I am pleased to say that in many respects fact has caught up with fiction, with the newest video analytics solutions that are now on the market focusing on search and specifically real-time search. These solutions have been tried, tested and proven to help reduce search time from hours to minutes and even seconds.
Real-time search analytics
Real-time search analytics addresses one of the most important control room tasks - locating a person of interest. Put simply, by reducing the search time you significantly improve the chances of a favourable outcome. Whether it is working in real-time to reunite a lost child with their parents, aiding in the apprehension of a terror suspect, or working with the authorities following a major incident, every second counts.
Knowing where a person of interest
However, the technology goes one-step further than pinpointing where on the CCTV network that person currently is. Yes, knowing where a person of interest is at the current moment is vital, but in certain situations you may also want to know when and where they entered the estate, the precise route they took, who they were with and what they did before arriving at their current location.
FAQ of real-time search analytics
With the introduction of any new technology there will always be some confusion around what it is, what it can do and whether you have the infrastructure in place to consider deploying it. With that in mind here are the top ten questions I have been asked many times about real-time search analytics…
- Who can I search for? Any person of interest
- What is the search based on? Full body image, textures, colours and unique characteristics
- What can I use as a reference? Upload photo, video image or human composite
- Does it require mega-pixel cameras? Suspect Search is camera type agnostic. Images need to be in colour CIF-Full HD resolution and a frame rate of QRT or higher
- How does it help me reduce my search time? It filters out 95% of irrelevant images
- How can I learn the suspect’s locations and movements? The suspect’s route is displayed on a map
- What is the recommended environment? The technology can be used indoor or outdoor
- Does it work in real-time? Yes, real-time search can be initiated in seconds
- What are the common use cases? Intruder search, lost child, unattended bag owner, locating a witness
- Where is the technology currently in use? Transport hubs, airports, city centres, hospitals, government facilities and sporting events
Next-generation video analytics is very much in the here and now
So, how can it be used in day-to-day security operations? To best explain how it works in practice here is an example…
It is a Saturday afternoon and a family are attending a sporting event. The stadium holds 60,000 people and the concourses surrounding the entry gates are getting very crowded, with 30 minutes before the match is due to start. The family queue to go through the turnstiles and upon entering the stadium they soon get separated from their six-year-old son. Panicked, they ask the nearest steward for help, who arranges for an announcement to be made over the public-address system and, also notifies the control room.
The steward gives the control room operator a description of the young boy (short blonde hair, a light blue jacket and a small yellow backpack). He quickly enters this information in to the system by creating a human composite (otherwise known as an avatar) and in less than five seconds he is looking at a series of images where a person matching the description appeared on a surveillance camera. The rapid speed of the search is made possible because the system is capturing, indexing and storing data in real-time, from every camera located in and around the stadium complex.
The rapid speed of the search
Upon confirming that it is the boy in question the operator selects the camera feed of his last known location and the real-time feed shows him clearly at concession stands looking distressed. The nearest steward is notified and he waits with the boy until he is reunited with his parents. All of the information relating to this incident and others is logged both for training and to help the stadium staff make improvements to reduce the likelihood of incidents recurring.
This is a common scenario at a sporting event or shopping complex, but the same process can be applied to estates with hundreds or thousands of cameras across single or multiple sites, such as a sea or airport transportation hub or medical facility. What is more, the sports stadium in this example did not require any expensive camera upgrades in order to deploy real-time search analytics, as it worked with the infrastructure already in place.
Next-generation video analytics at present
Next-generation video analytics is very much in the here and now - in fact it is in real-time! What is more, it isn’t just for handling rare incidents but also daily occurrences that can absorb a lot of the control room operators’ time, and I make the point again that time is the number one critical factor when handling any incident.
The movie-makers of tomorrow are going to have to up their game in a big way, if they want to impress or maybe even inspire the security industry.