For over 10 years, IntelliVision® has provided advanced facial recognition analytics with an exceptionally high accuracy rate on public standard data sets. Our facial recognition software detects faces in real-time and provides an alert when a match is found with a face in the database. Around the world, IntelliVision AI-based face recognition is used in thousands of cameras for access control, VIP greeting, shoplifter and unwanted person applications.
As we continue to grow and adapt to the needs of our customers, we have introduced new analytics to help address global concerns during COVID-19.
New Technology to Keep You Safe
Recently, we announced the addition of our face mask detection to provide business owners a solution to returning to work in addition to boosting the public’s confidence in safety during the times of COVID-19. According to Vaidhi Nathan, SVP Cameras and Analytics at Nortek Security & Control (NSC), face mask detection is the first of a number of products to be released by IntelliVision as part of a broader COVID-19 protection strategy.
“The IntelliVision face mask detection analytic ensures that employees, shoppers and customers are complying with safety guidelines, and [the technology] is designed to send an alert if a mask is not detected,” said Nathan.
We are committed to helping our customers and the market return to work safely through the IntelliVision family of AI video and audio analytics solutions including:
Facial recognition
License plate recognition (LPR/ANPR)
Object left/removed
Object classification (humans, vehicles, pets, and airplanes)
Intelligent object detection
Intrusion/perimeter watch
Stay tuned for new developments and the next release of solutions coming soon!
Automatic license plate or number plate recognition (ALPR/ANPR) provides immediate benefits for law enforcement, parking agencies, security teams, government buildings and more. While LPR is used in many different cases, it’s proven to be effective at deterring crime, analyzing and collecting data, and providing agencies with real-time solutions.
In addition to the many LPR user, our License Plate Recognizer products also support Smart Parking applications. Our AI-based license plate detection, recognition and search solution has an accuracy as high as 98 percent and is used to detect and recognize license plates on moving and stationary vehicles. The technology can detect and recognize number plate from over 25 countries and can also detect different license plate styles, including regular and stacked license plates.
Added Benefits of ALPR/ANPR
Vehicle monitoring is a large part of outdoor security and surveillance, and LPR is an integral part of this task. The product works with live cameras as well as archived video. See how our advanced technology can work for you:
Supports processing of single frames and video streams
Compares detected plates against a watch list database and provides real-time alerts on plate matches
Provides detailed reports
Handles different plate styles, e.g. regular/stacked
Can be used for electronic toll collection (ETC)
Can be connected to gate systems for automated parking access control
Our LPR technology has advanced features and is easy to deploy with a wide variety of cameras.
Other important features include:
Detect, recognize and search for license plates in real-time or archived footage
Enables automated matching against a watchlist with real-time alerting
Allows vehicles to be tracked across multiple cameras or locations
Records and logs all license plates at a scene for later forensic investigation
The traffic light was invented in 1868 and the first one was installed in London, outside the Houses of Parliament. It was operated by gas and, sadly, blew up, killing a policeman. According to the United States Access Board, there are more than 300,000 traffic lights in the US today and a good estimate worldwide would be in the millions.
Most traffic signals today are controlled by counting the number of cars that go over a wire induction loop buried in the street. The metal in the car disturbs a magnetic field and a counter is incremented, unless you’re riding a motorcycle that is. The induction loops often aren’t sensitive enough to detect your bike, and you’re left standing by the light, keeping the engine ticking over. Induction loops are also difficult and expensive to maintain or repair. You have dig up the road to replace them when they go wrong.
Step forward the video camera. By mounting video cameras on light poles above the street, equipped with video analytics software to count the vehicles passing by, traffic signals can be easily and efficiently controlled. Video cameras are much cheaper than the induction loops and easy to adjust or repair. They can transmit their video stream to a small server in the roadside cabinet and the analytics can then be sent to the controller in the same cabinet. It’s easy to program for multiple lanes, turns and rights-of-way, and by adding analytics which can recognize public transit or emergency vehicles, these can be given priority too.
Although video camera-based systems can be expensive, they can relieve the need for other even more expensive measures like new roads. One of the other advantages of video cameras is that they can also be used for incident monitoring.
Face recognition is increasingly being used in access control for primary or secondary authentication, giving users an automated, frictionless experience. Access cards are easy to duplicate and to lose, leaving offices and secure facilities at greater risk of unwanted intrusion. But you always carry your face with you, and anti-spoofing technology makes sure it’s yours. Today’s multi-core camera chips allow you to embed face recognition software in the camera itself, so you don’t even need a server.
Race and gender bias in face recognition can be an issue if the recognition models have been built with insufficiently diverse datasets. Often, the datasets are biased towards the gender and race of the developers themselves, who tend to be white males. But there are datasets available with millions of faces of all races and both genders. At IntelliVision we have ensured that our facial recognition models have been built and tested with multiple ethnicities and genders, ensuring that everyone can be detected.
Adding face recognition to turnstiles and gates can improve both security and speed, leading to a frictionless experience for staff or for people passing through checkpoints. Image databases can be created from existing employee records or passport data.
But what about “spoofing”, where someone holds up a photograph or even a cellphone video of the person’s face? Won’t that allow the wrong person through? That’s where “anti-spoofing” comes in. Normally this requires a 3D or stereo camera to detect depth and movement. But at IntelliVision we have implemented anti-spoofing in regular 2D cameras, by detecting “liveness” over a number of successive frames. This ensures that only real faces are detected and recognized.
The IntelliVision booth was swamped at IoT World 2019 in Santa Clara! Facial recognition and video analytics at the edge are perfect for embedding smart cameras in IoT devices and smart machines. We also had a lot of interest in our ADAS and driver monitoring solutions.
The central monitoring station – the heart of many security systems.
At the heart, albeit remote, of many home and commercial security systems is the central monitoring station. Banks of computer screens are watched day and night by human operators looking for break-ins, loiterers and other suspicious events. While they provide a valuable service there are problems associated with the use of human operators – there are only so many screens one operator can monitor, and the human brain is wired to start ignoring stimuli of the same kind over time. In other words, the more events and alerts there are and the longer an operator has been watching, the less likely it is that an important event will be noticed. And because most CCTV surveillance cameras in home or commercial use are just dumb cameras, they can transmit a lot of alerts requiring human intervention or verification.
Not only is this a problem for home and business owners, it also limits the number of customers that a central station can effectively monitor. Traditional CCTV surveillance systems that just stream video or use PIR (Passive Infrared) motion detection, are extremely prone to false alarms, requiring either video verification or a loss of accuracy and business.
Human/vehicle detection
The way to decrease false alarms, improve event detection accuracy and thereby allow a monitoring station business to grow its customer base, is to use AI in the form of video analytics. Video analytics can analyze a stream of video, frame by frame, in real-time, and provide alerts only when real events occur, detecting the presence of humans or vehicles where they shouldn’t be at that hour. Or detecting when a person spends time in a restricted area – loitering – or a vehicle is detected at a business after hours, when no-one should be there.
The way to decrease false alarms, improve event detection accuracy and thereby allow a monitoring station business to grow its customer base, is to use AI in the form of video analytics. Video analytics can analyze a stream of video, frame by frame, in real-time, and provide alerts only when real events occur, detecting the presence of humans or vehicles where they shouldn’t be at that hour. Or detecting when a person spends time in a restricted area – loitering – or a vehicle is detected at a business after hours, when no-one should be there.
But the real benefit of AI-based video analytics is not what they detect, but what they ignore. An animal crosses the camera at night; it’s fall and leaves are coming off the trees; snow is falling; headlights are detected from the highway a few hundred yards away. Video analytics discards all of these potential alerts only sending the ones that actually matter.
The advantages of video analytics are these:
Reduced false alarms
Improved monitoring accuracy
Less strain on operators
Less need for video verification
Ability to add more customers without more operators
Detecting human shapes
How do AI-based video analytics work? In the past, in the early years of video analytics, programmers would create reams and reams of algorithms to determine what was a valid motion in a video frame and what wasn’t. Video frames were analyzed by these algorithms and the only way they could be improved was by the programmer writing more or better algorithms. The accuracy of the solution depended on the skill of the programmer. But a few short years ago, the mathematical concept of the neural network was born, closely mimicking the workings of the human brain in how it learns. One of the applications of the neural network is in machine learning. The computer can teach itself what is a human shape by scanning thousands of shapes of humans in different poses, at different angles, and in different lighting.
Neural networks and machine learning – this is what we mean by artificial intelligence or AI in the realm of video analytics. The analytics get better and more accurate the more data is passed through them. Note that this doesn’t happen with live, real-time data, but at the development stage. It takes a lot of processing power and usually high-powered graphical acceleration (GPUs) to process all that video data. The result is a compact analytics engine that can run on a regular server, or even inside the camera itself.
So stop relying on human operators to determine when an alert is real or not and make a move to AI-powered video analytics. AI – one time when artificial is actually better than the real thing.
The average human can recognize faces with an accuracy of 97.5%. As of 2018, face recognition (FR) technology is achieving accuracy rates of 95-99%, with sub-second recognition times. Of course it’s easy to throw computing power at the problem from a local server or in the cloud, but with today’s multi-core camera chips such as those from Ambarella or Qualcomm, the really interesting applications are doing face recognition at the edge.
The applications of facial recognition (FR) are almost without limit. For security we can dispense with key cards, as a camera at the office door or turnstile recognizes you and lets you in without you taking your hands out of your pockets. You may forget to bring your card, but not your face. It works just as well in apartment complexes, warehouses, hospitals, care homes, data centers or secure government installations. Retailers can quickly and automatically recognize known shoplifters or serial returners from a central database, cutting down on fraud and pilfering. Schools can ensure that only authorized people are allowed in. Banks and financial centers can match your face with your account records.
There are also added-value possibilities with FR. When your best customers come into your store or restaurant or casino you can be alerted to the presence of a VIP and give them personalized service. Better still, when you come home from work your home recognizes you and sets the temperature to your liking, and plays your favorite Led Zeppelin channel on Pandora. Or Beyonce (home personalization).
Ms Dong’s image on a public screen
Of course problems can sometimes occur. In China, a country with 170 million cameras, and 400 million more on the way, a well-known businesswoman was recently “named and shamed” for jaywalking. Her face had been recognized on the side of a bus as it sped along. The error was quickly addressed in an upgrade according to the authorities.
Credit: Joy Buolamwini
Another issue with FR is bias. The training database used in the deep learning development phase of the algorithm has a major bearing on the accuracy of the recognition, and bias can easily creep in. Developers are usually male so they may unconsciously pick more male faces to train with, and a database with predominantly Caucasian faces may have difficulty accurately recognizing Asian or African faces. “Gender Shades,” a study on some of the most popular FR systems by Joy Buolamwini, a researcher at the M.I.T. Media Lab, has shown that female black faces are badly underrepresented, leading to low accuracies.
There are also the scenic characteristics of the captured image to take into account. Poor lighting makes it harder to recognize faces – even for a human – and dark-skinned faces in low light are the most difficult. Low light conditions can be alleviated somewhat by the use of infrared lighting which some cameras will provide. Another issue is that of “liveness,” whether the face in front of the camera is that of a real person or just a picture. The algorithm has to be tuned to look for the 3D aspects of a live face such as different distances, small movements etc, instead of a static picture, to prevent spoofing. (See “anti-spoofing”)
Lastly is the thorny issue of privacy, a discussion which is more in the realms of ethics and law rather than technology. In China your face is being captured everywhere you walk whether you like it or not, and likely stored for future unspecified uses. In Europe, Facebook’s auto face-tagging has been blocked. But here in the US there is no Federal privacy law like Europe’s GDPR, and only three states – Illinois, Texas and Washington – have state-level Biometric Information Privacy Acts, though Texas and Washington do not allow private lawsuits.
As a security technology face recognition is here to stay, and
as with any new technology the issues are being addressed almost as fast as
they crop up. It may not be as accurate as iris scanning or fingerprints, but
it is quicker and less intrusive and offers a convenient and secure method of access
control, visitor tracking and criminal detection.
At ISE 2019 in Amsterdam, Bill Hensley steps into the frame while demonstrating the new ELAN range of dome cameras. The new cameras are fully integrated with IntelliVision video analytics technology, with intelligent alerts that trigger only when an object moves into the camera’s preset intrusion zone.
It seems that as soon as a security technology is invented, someone finds a way to break it. You’d think that facial recognition was a pretty fool-proof way to authenticate a real person. Your face is your face and it’s unique (apart from your evil twin that is). But what if I held a high resolution color picture of your face up to the camera? Or maybe a video on a cellphone? Could I fool a facial recognition system into thinking I was you? Maybe.
The way to ensure that the face in front of the camera is your actual face is to implement anti-spoofing. At IntelliVision…
Cities are becoming smarter. IoT is being implemented everywhere, improving the safety, security and convenience of city dwellers and workers, and the most important sensor in the Smart City is the video camera. But the days of beaming gigabytes of video into the cloud to be examined by human monitors is coming to an end.
AI-based video analytics are now being used to count cars and people, and to check for events such as crowd formation, illegal parking, and using LPR/ANPR to track vehicles either for parking convenience or for law enforcement.
