Entertainment venues, whether it is a stadium, casino or even cinemas, are designed to ensure that as many people as possible are able to come and go freely without feeling like they are under constant surveillance. This makes it difficult for security operations and the venue’s ground staff to identify which individuals are on site at any particular time.
From a security perspective, known criminals or troublemakers may be on a blacklist, but that list is only useful in keeping the peace and ensuring everyone has a good time if there is someone looking at the faces of people as they enter the venue. And when considering a casino, for example, Franz Kersten, Panasonic SA’s, channel manager says the human resources required to keep an eye on people coming in at any one time is enormous, even if the job is left to operators in a control room – who also have other responsibilities to take care of at the same time.
The solution, Kersten says, is to make use of the latest facial recognition solutions to automatically identify those people and then advise your staff that they are on site. Of course, he admits that facial recognition is not a simple function and there are many solutions on the market making promises they can’t deliver.
Panasonic recently launched its new facial recognition solution in conjunction with its new i-Pro Extreme camera range. The solution is made up of i-Pro cameras with built-in face matching intelligence, along with Panasonic’s FacePro server-based software which has been enhanced with deep learning algorithms to take facial recognition to a new level.
Edge face matching
Kersten explains that the system works as follows: the FacePro software is able to control the i-Pro Extreme cameras settings via intelligent Auto (iA) mode to enable them to get the best shot of people’s faces in various conditions. This allows the system to identify faces that are difficult to recognise with conventional technologies, including faces at an angle of up to 45-degrees to the left or right, or 30-degrees up or down, and those partially hidden by sunglasses. The company’s ‘Best Shot’ function ensures that the best image of a face is then sent to the server where it is compared to a database of faces to identify people of interest.
The on-camera face matching can deal with up to eight faces per second. The best image is then sent to the server, saving bandwidth as it does not transmit a video clip, only an image. The server is then able to perform its recognition function on a database of up to five million faces in only three seconds. Furthermore, each server is able to accommodate up to ten cameras.
The accuracy of the system is such that it has been installed at an International Airport’s immigration control where it is used to compare the image in people’s digital passports with the image taken at the arrival gate. Kersten adds that FacePro can accept images from third-party providers as well as the i-Pro Extreme cameras, ensuring customers are still able to make use of their existing surveillance infrastructure.
More than a blacklist
The possibilities for the new solution from Panasonic goes further than identifying people on a blacklist, however. As an example, Kersten says casinos can also use it to identify high net worth individuals as they arrive so that the venue’s staff can meet them as they walk in and ensure they receive VIP treatment.
Similarly, Panasonic also provides automated number plate recognition (ANPR) using the same infrastructure, which can be used to identify VIP visitors as they drive into the location, or any other previously identified people the company wants to know about when they come on site.
© Technews Publishing (Pty) Ltd | All Rights Reserved