Data Innovators Lisa Dolev

Published on October 24th, 2016 | by Joshua New

0

5 Q’s for Lisa Dolev, Founder and CEO of Qylur Intelligent Solutions

The Center for Data Innovation spoke with Lisa Dolev, founder and chief executive officer of Qylur Intelligent Solutions, a security screen tehnology company based in Palo Alto, California. Dolev discussed how their technology can make security screening less invasive by using machine learning and the Internet of Things.

This interview has been lightly edited.

Joshua New: Qylur’s flagship product is the Qylatron, a security screening system that uses machine learning to identify dangerous items. Can you explain how Qylatron works?

Lisa Dolev: The Qylatron Entry Experience transforms venue entry points by integrating and connecting multiple operations: greeting and welcoming guests, validating tickets and passes, performing security screening, and engaging guests in the venue brand and experience, all in one step. Qylatron leverages patented technologies including machine learning, the Industrial Internet of Things, and fused sensors.

The solution is comprised of a self-service automated kiosk that can engage up to five guest parties at a time; operator workstations, mobile devices and managers apps; and analytical data services for real-time reporting and operations optimization. Large color panels on the kiosk cell doors give “wait” or “proceed” indications to guests. Guests arriving in front can scan their tickets to open cell doors and are greeted personally. Once bags are inserted and doors closed, the system screens the bags while guests proceed through person screening gates. If Qylatron flags a bag for review, cell doors will remain locked until a guard reviews the result, after which bags can be manually searched or released to the guest.

Qylatron’s automatic detection is based on machine learning technology that is trained on vast quantities of bags contents and threats, to achieve the best available performance, especially on explosives that are practically impossible for human screeners to detect. Auto-detection constantly learns and self-improves based from its own scan information and performance that it gathers on the ground.

Qylatron also works in collaboration with human screeners as second-tier reviewers and to detect additional site-specific banned items.

New: What kind of things can the Qylatron detect?  How does it differ from, say, a standard metal detector and X-ray scanner at an airport?

Dolev: Qylatron detects guns, explosives, and explosive devices within guest bags. In other words, the most dangerous and top-concern threats at large public venues. Qylatron screening is embedded in an entry process that is designed to engage guests in an enjoyable activity that is linked to the brand and experience they expect to find at the venue inside. To this end, Qylatron greets customers, personally if ID scanning is used, presents branded content on video screens, and offers a hassle-free, self-service experience.

Qylatron takes security screening to another level of accuracy and sophistication by using multi-view X-ray images and applying machine learning-based detection that evolves and gets optimized over time. Qylatron can use the guest ID to adjust the level of detection to the guest’s risk level. In addition, Qylatron lets screeners mark the items they detect on touch screen icons, thus informing other staff members and creating system records for reporting and learning.

New: One of the Qylatron’s biggest selling points is that it’s faster than traditional security screening technology. How much faster? Is there any tradeoff between speed and reliability?

Dolev: The Qylatron effective rate of screening is affected by many site-specific factors and will thus vary among sites. With Qylatron’s ability to deliver the highest accuracy screening results at a consistent rate, the least reliance on human judgement, and the most respectful and private process, it simply cannot be compared to any alternative method, as they are notorious for their invasiveness and deficiencies even when many personnel are involved. Any alternative that would try to match the Qylatron’s benefits would have to move people a lot slower.

New: Qylur also enables different Qylatron systems to share data with each other. What does this accomplish? How significantly can these systems improve over time?

Dolev: Qylatrons pool together the information they learn from the ongoing scans they perform, contributing to a vast and growing pool of data that enables constant improvement of the detection engine. Qylatrons at similar type sites, such as amusement parks, or domestic airports in the United States, are even more closely related, forming a “social network of machines” of mutually-relevant information that is used for accelerated learning. The timeframe to show dramatic results in alarm performance can be in mere days, depending on the pace of screening.

New: Qylur helped provide security screening for the Olympic village at the Rio Olympics. How did this go?

Dolev: The Qylatron was in use throughout the Olympic games, including the Olympic Village main entry site, which was identified as a very-high-security point. The Qylatron was deployed and operated through ISDS Ltd., which was an official supplier and sponsor to Rio 2016, and performed to the customer’s complete satisfaction.

Tags: , , ,


About the Author

Joshua New is a policy analyst at the Center for Data Innovation. He has a background in government affairs, policy, and communication. Prior to joining the Center for Data Innovation, Joshua graduated from American University with degrees in C.L.E.G. (Communication, Legal Institutions, Economics, and Government) and Public Communication. His research focuses on methods of promoting innovative and emerging technologies as a means of improving the economy and quality of life. Follow Joshua on Twitter @Josh_A_New.



Leave a Reply

Your email address will not be published. Required fields are marked *

Back to Top ↑

Show Buttons
Hide Buttons