Alcatraz AI’s Facial Authentication Enables Frictionless Access Control

Download PDF version

Effective access control can be achieved without the use of cards using a new generation of secure facial authentication enabled by artificial intelligence and machine learning. Alcatraz AI is introducing a system that deploys a sensing device, about the size of a badge reader, with multiple color and infrared cameras that can detect facial features and confirm an identity.

Real-time 3-D facial mapping avoids anyone using a photograph, video or mask to spoof the system and confirms there is a real person that matches the stored facial image.

System Helps In Tailgating Mitigation

Deep neural networks, powered by NVIDIA, enable the system to achieve new levels of frictionless access control, says Vince Gaydarzhiev, CEO of Alcatraz AI. Computer processing is achieved at the edge to ensure speedy and secure access control.

We saw an opportunity to create a system that solves issues of tailgating and addresses the need for security without increasing friction"“We saw an opportunity to create a system that solves issues of tailgating and addresses the need for security without increasing friction,” says Gaydarzhiev. The accuracy of the system lessens the need for security guards, he says.

The Silicon Valley startup, currently with 20 employees, was founded in early 2016 by a team from Apple, NVIDIA and Lily Robotics with a goal of targeting mid- to large-sized corporations that currently have deployed badging systems. The company has raised close to $6M from venture capital firms and individuals, and Johnson Controls/Tyco has invested in the startup.

Alcatraz AI’s sensor device, mounted near a door, confirms a user’s identity and communicates the user’s badge number to the existing access control infrastructure. “The system improves the facial profile every time, using the neural network to be even more accurate in the future,” says Gaydarzhiev. He says it is the industry’s first “instant one-factor authentication for multi-person in-the-flow sensing.”

In the future, the frictionless system simply recognises the user and opens the door

The system is less expensive than previous facial authentication systems and does not require users to be very close to the reader

Easy Enrollment And Deployment

Enrollment in the system is easy. Companies can deploy a separate enrollment station, or any reader can be used for enrollment. After badging in a couple of times, the face matching system “enrols” the face with the associated badge number, thus allowing the user to dispense with the badge altogether.

In the future, the frictionless system simply recognises the user and opens the door. A user company can quickly deploy the system at locations where thousands of employees have access, without requiring employees to go to HR for enrollment.

Gaydarzhiev says accuracy of the system is no less than that of iris scanning, and the accuracy is configurable for specific needs. He says the system is less expensive than previous facial authentication systems and does not require users to be very close to the reader. Facial authentication is also more flexible than iris scanning or fingerprinting.

Detecting Intent From Positioning Of Eyes

The system detects intent from the positioning of the eyes and body to avoid opening a door unintentionallyIn contrast to near field communication (NFC) or Bluetooth systems, the technology does not require a compatible smart phone or have issues of communication range. There is no need for users to stop and perform an action or gesture to signal intent. The system detects intent from the positioning of the eyes and body to avoid opening a door unintentionally, says Gaydarzhiev.

Alcatraz AI is targeting high-tech enterprises, including healthcare, government and eventually banks. Currently they have three pilot installations among large global software companies and are undergoing trials with some government agencies. Today, they sell direct to end users, but the intent is to develop a dealer channel that will account for most of the sales.

Download PDF version

Author profile

Larry Anderson Editor, SecurityInformed.com & SourceSecurity.com

An experienced journalist and long-time presence in the US security industry, Larry is SecurityInformed.com's eyes and ears in the fast-changing security marketplace, attending industry and corporate events, interviewing security leaders and contributing original editorial content to the site. He leads SecurityInformed's team of dedicated editorial and content professionals, guiding the "editorial roadmap" to ensure the site provides the most relevant content for security professionals.

People mentioned in this article

Vince Gaydarzhiev Alcatraz AI

Related categories
Access control software

In case you missed it

Entrance Control Vs Access Control: Similarities And Differences

Entrance control and access control - of the physical kind - are common terms in the security industry which are often used interchangeably, but should they be? Having worked both sides of the fence, with previous roles at TDSi and HID and now the Major Accounts and Marketing Manager at Integrated Design Limited, Tony Smith highlights the subtle but important differences between these two terms and the systems they refer to, outlining how they should work together to achieve optimal security. Access control is a system which provides discriminating authentication Access control provides a discriminating authentication process and comprises the software or hardware that defines the criteria for acceptance or denial Used to describe a system which performs identification of users and authentication of their credentials (deciding whether or not the bearer of those credentials is permitted admission) access control is an incredibly broad term. Access control provides a discriminating authentication process and comprises the software or hardware that defines the criteria for acceptance or denial of an individual to a restricted area. Entrance control – such as security turnstiles - takes the output of that validation and has the capability to see whether that criteria is being adhered to, either granting or denying access as appropriate. Entrance control is the hardware responsible for keeping people honest If access control verifies authorized personnel using their credentials – their face, fingerprints, PIN number, fob, key card etc – and decides whether or not they are permitted access, entrance control is the hardware which enforces that decision by making users present their credentials in the correct way, either opening to allow pedestrian access or remaining closed to bar entry and potentially raising an alarm. For example, a card reader acts as an access control device, recognizing the card holder as having the correct permissions and saying ‘yes, this person can pass’. But, it’s the entrance control system – a turnstile, for example – which actually physically allows or denies access. Physical access and video surveillance Some entrance control systems don’t feature a physical barrier, however. Fastlane Optical turnstiles will not physically stop an unauthorized person from passing through, and instead alarm when someone fails to present valid credentials, alerting security staff that a breach has occurred. These kinds of turnstiles are suited to environments which just need to delineate between the public and secure side of an entrance, with less need to physically prevent unauthorized users from entering. State of the art access control integrations have been installed for award-winning complex, The Bower It’s also possible to capture video footage of any incidents, allowing security personnel to identify users failing to abide by the access control system’s rules, using It’s also possible to capture video footage of incidents, allowing security personnel to identify users failing to abide by access control system rules the footage to decide on the level of response required. The breach could have been the result of a member of staff being in a hurry and failing to show their card before passing through, in which case they can be reminded about the security protocol. Or, it could be an unidentified person who needs to be escorted from the premises. Entrance control and access control working together For optimum security, access control and entrance control should work together, with the entrance control system enhancing the use of the access control system, making it more efficient and better value for money. The two can’t effectively operate without each other. Security turnstiles, for example, require something to tell them that someone is about to enter – the access control system does this – and, the access control system needs a method of stopping people when they don’t badge in correctly. The two systems are complementary.

ATMs Provide Convenience For Bank Customers, But They Have Vulnerabilities

Most customers interface with their financial institutions using automated teller machines (ATMs), which have security issues. However, there are solutions available to combat all current security threats, and the cost of protection is coming down. The ATM industry is therefore in a position to minimize losses, while ensuring consumers continue to get the vital cash they need to lead their daily lives. It is important for the ATM industry to constantly innovate to meet new security challenges. So what innovations are we going to see in the next five years? Contactless technology Contactless technology will be a great help against ATM skimming, in which criminals steal personal information at ATM machines. Contactless is already being used in some European countries, and the number is increasing. Not having to insert a card into the ATM removes the opportunity to trap cards and also gets around the problem of “foreign” devices installed to read cards. So contactless technology, which some saw as the end of cash, can help make ATMs and cash more secure. Data capture form to appear here! Not having to insert a card into the ATM removes the opportunity to trap cards Biometrics are certain to be used increasingly to bolster ATM security. Finger, palm, vein, iris and facial recognition all have potential in this respect. Any of these may in the future be used with or without cards, PINs and one-time codes. Speed of operation in relation to biometrics could ultimately govern their use at ATMs. There may also be privacy issues that need to be addressed. The ATM vestibule environment must add security with proper security and surveillance equipment. ATM vestibules, or lobbies, are installed for many good reasons. For one, more convenient, 24/7 locations equals better customer retention for a bank, offering comfort and convenience. 24/7 access to ATMs, night drops, coin counters, online banking kiosks, and other self-service solutions are very much in demand. Second, ATM vestibules protect customers from inclement weather and provide a more comfortable banking environment (however, vagrancy can be an issue; therefore ATM vestibules should require card access). Security and surveillance solutions can’t just be for show. ATMs and crime A new crime wave is hitting automated teller machines (ATMs); the common banking appliances are being rigged to spit out their entire cash supplies into a criminal’s waiting hands. The common banking appliances are being rigged to spit out their entire cash supplies into a criminal’s waiting hands The crime is called “ATM jackpotting” and has targeted banking machines located in grocery shops, pharmacies and other locations in Taiwan, Europe, Latin America and the United States. Rough estimates place the total amount of global losses at up to $60 million. The protection of ATMs ATMs in supermarkets and pharmacies tend to be targeted because they may not be as well-protected, and store personnel likely would not know who is authorized to work on the ATM. In contrast, anyone approaching an ATM at a bank location would be more likely to be challenged. ATM jackpotting originated back in 2010 when Barnaby Jack, a New Zealand hacker and computer expert, demonstrated how he could exploit two ATMs and make them dispense cash on the stage at the Black Hat computer security conference in Las Vegas. Since then, malware has been created and made available on the “Dark Web” that can instruct an ATM to dispense all its cash on demand. ATM jackpotting ATM jackpotting is a combination of a physical crime and a cyberattack ATM jackpotting is a combination of a physical crime and a cyberattack. Typically, a criminal with a fake ID enters a grocery shop or pharmacy posing as an ATM technician, then uses a crowbar to open the top of the ATM – the “top hat” – to gain access to the personal computer that operates the machine. Once he or she has access to the PC, they remove the hard drive, disable any anti-virus software, install a malware program, replace the hard drive and then reboot the computer. The whole operation takes about 30 seconds. The malware then enables the thief to remotely control the ATM and direct it to dispense all its cash on command. If a legitimate customer approaches the machine in the meantime, it can operate as usual until activated otherwise by the malware. Catch up on part one and part two of our banking security mini series.

AI-based Security Software Is The Answer To Vandalism And Theft Prevention

On November 2019 in Stockton, California, surveillance footage found that vandals shot out glass windows and doors in many places in a small business complex (FOX40). The intruders broke in only to leave with nothing, proving their intent was solely to vandalize the property. Meanwhile, it was reported that a trio of ATM thieves struck around 9 times across many different locations inside Brooklyn and Queens within just over a month in fall 2019 (ATM Marketplace). On average, the cost of vandalism to SMB is around $3,370 per incident (US Small Business Administration), including a staggering 692 vehicle vandalism claims per day. Likewise, the average cost of theft to SMB is about $300 per shoplifting incident and $1,500 per employee theft incident, which accounts for 38% and 34.5% of all theft instances, respectively (National Retail Security Survey). High-performance artificial intelligent systems can automate the monitoring tasks Vandalism and theft have proven time and time again to be inconvenient and deconstructively harmful towards SMB. However, these financial burdens can be prevented with the use of the right security system. AI-based security systems with Deep Learning contain many features that many SMB owners find advantageous in their pursuit to stop unwarranted and unwanted money loss. Intrusion and loitering detection The first of many features that can help with vandalism and theft prevention is Intrusion Detection. High-performance artificial intelligent systems can automate the monitoring tasks for high-risk sites to provide a high level of security and security personnel monitoring efficiency. Traditional intrusion detection systems detect objects based on size and location, but they do not recognize the type of objects. Now, Intrusion Detection (Perimeter Protection) systems with cutting-edge, built-in AI algorithms to recognize a plethora of different object types, can distinguish objects of interest, thus significantly decreases the false-positive intrusion rate. The more advanced AI-based systems, like those we offered at IronYun, enable the users to draw ROIs based on break-in points, areas of high-valuables, and any other preference to where alerts may be beneficial. Similarly, AI Loitering Detection can be used to receive alerts on suspicious activity outside any given store. The loitering time and region of interest are customizable in particular systems, which allows for a range of detection options. Advanced loitering detection software as such can detect and trigger real-time alerts for both people loitering and/or vehicles that are illegally parked in certain areas of interest. A benefit, which only certain advanced systems contain, is the ability to send trigger actions to 3rd-party systems in reaction to receiving an alert of loitering and/or intrusion detection. These trigger actions can be set to contact authorities immediately and/or trigger a scare tactic alarm or announcement to intruder/loiterer. Certain Face Recognition and License Plate Recognition software can record individual people/vehicles Face and license plate recognition In addition to the activity detection solutions, certain Face Recognition and License Plate Recognition software can record individual people/vehicles and use pre-configured lists to identify particular faces or plates that may be of interest, such as those in watchlists. These systems can also enable the users to upload images of faces not in the lists and search for them in the camera recording. For instance, if a person is detected several times loitering outside a store, one may save one of the detection photos into the watchlist, and set up an alert when said face is recognized again outside the building in the future. The alerts will help to deter and prevent vandalism or theft, and notify the authorities to the scene before the troublemaker completes the act. The main attributes of high-performance Face Recognition systems which maximize assistance with vandalism and theft management include: Face match rate > 90% with good camera angles and lighting. Processing multiple streams and multiple faces per image. Live face extraction and matching to databases of thousands of faces within 3 seconds. State-of-the-art AI security software with Deep Learning allows the user to no longer need to install special LPR cameras If the watchlist individual is wearing a mask or their face is not in view of the camera, their license plate may be a good indicator. If a particular car is detected several times loitering in the parking lot or street outside a store, the user can set the alerts for such car to get notified in the future. With an AI solution like this, common street cameras should be equipped with LPR capabilities. So, state-of-the-art AI security software with Deep Learning allows the user to no longer need to install special LPR cameras. high-performance alert mechanisms A high-performance AI solution, in addition to having high accuracy, should be able to: Easily integrate with 3rd-party systems Work well with all ONVIF IP cameras including infrared and thermal ones (for Intrusion detection) Analyzes video streams in real time and trigger alerts within a few seconds Send alerts to multiple VMSs, connect with signaling devices such as loud speakers or flashing lights Send email notifications to security staff and police departments Send notification on mobile device using AI NVR mobile app Maintains a record of all alerts to provide evidence of intrusion and loitering instances for police and insurance agencies. To assist in theft and vandalism prevention, AI-based security systems with deep learning will do all of the tedious work for you. Their low cost and high performance also make them the most accessible security solutions in the market with large return on investment. Stopping crimes is a difficult, ongoing challenge, but with the right AI software, business vendors and police departments can do so with more ease.