April 2017 Digital Edition
March 2017 Digital Edition
Feb. 2017 Digital Edition
Nov/Dec 2016 Digital Edition
Oct 2016 Digital Edition
New AI technology that mimics the human brain can help law enforcement and intelligence organizations rapidly identify patterns, objects and faces in large amounts of archived and live streaming video
Video is a critical element in crime prevention and investigation, yet current law enforcement systems are increasingly unable to cope. The sheer volume of surveillance material captured and stored every day is staggering, and set to rise dramatically. Adding more cameras to gather more information will only ever be useful if processes to search and analyze the mountain of data keep pace. As it stands vital information may be missed because the vast majority of video is simply never viewed.
Information technology firm Cisco estimates than in 2021 it would take more than 5 million years to watch the amount of video traffic across the globe – each month. Market researcher IHS forecasts that 127 million surveillance cameras and 400,000 body-worn cameras will ship this year, in addition to the estimated 300 million cameras already deployed. By 2020 it is predicted there will be more than 1 billion cameras operated by smart cities worldwide, providing 30 billion frames of video per day. Internet video surveillance traffic alone increased 71 per cent in 2016 according to Cisco, and is set to increase sevenfold by 2021. Globally, 3.4 per cent of all video traffic crossing the internet will be video surveillance.
Give that a major problem for surveillance operators is directed attention fatigue, where the brain naturally alternates between periods of attention and distraction, it would require a superhuman effort to identify and classify all these images. What is required is a system that is never distracted and can work in conjunction with people to reduce errors, which is what artificial intelligence-driven video systems promise.
AI in video surveillance can potentially deliver four times the performance of conventional video search – in contrast to human vigilance, which studies have shown can degrade by 95 per cent after about 20 minutes.
The cost of deep learning
Since 2012, when AI video analytics took off, the systems trained to recognize objects and facial IDs from different types of image have proved expensive to run and slow to compute, and require large datasets to generate results. These systems, which are based on convolutional neural networks (CNNs), employ an AI technique known as ‘deep learning’. They excel at churning through data but lack the ability to refine and react to streams of information gathered from the surrounding environment – which the human brain is extremely good at.
What’s more, CNNs exhibit limitations including poor noise immunity, particularly when random pixels appear in an image due to noisy sensors or lens contamination. They can serve false classifications if the network becomes confused – for example by someone wearing glasses, or if it cannot find a new face in a crowd without a large set of labelled images relating to that face being added to the database. The network parameters of CNNs need careful adjustment, and even then the accuracy rate for correct image classification may not be sufficient for video surveillance applications.
Spiking neural networks
A relatively new approach is the spiking neural network (SNN), which simulates and models the different aspects of the human brain’s operation much more closely than a CNN.
For instance, a police department that is looking for a suspect in live video streams does not have thousands of images of that suspect; nor does it have weeks to train a CNN system. In an SNN-based system, it can find patterns and people in videos in milliseconds and from a single image – which, importantly, can be as small as 24 x 24 pixels: it doesn’t need to be high definition. The system excels in recognition in low-light, low-resolution, noisy environments, making it ideal for the large amount of previously installed video surveillance systems.
Unlike current CNN technologies that require extensive pre-labelled datasets and expensive cloud-based training and acceleration, an SNN system can be implemented in software with traditional computer processors (CPUs) and trained on-premises. The one-shot technology learns in real time and requires only modest processing power – typically a Windows- or Linux-based x86 desktop computer or server – as well as consuming little energy.
This enables a greater number of law enforcement organizations to capitalize on the opportunities offered by AI. It means AI algorithms can be used with legacy systems without requiring expensive hardware or infrastructure upgrades, and it can be deployed in the field in highly secure environments that may not have cloud connectivity.
Tasks that seemed impossible for machines just a few years ago are becoming almost routine, and SNN technology has perhaps the greatest potential to bring valuable new capabilities into mainstream automated video surveillance today.
About the author:
Bob Beachler is Senior Vice President of Marketing and Business Development at BrainChip. He can be reached at: [email protected]
TSA announced on December 4 that it will charge a fee of $85 per individual to participate for five years in the Pre[check] trusted traveler program, which will enable cleared passengers to move through airport security on an expedited basis.
“Successful applicants will be eligible to receive expedited screening at participating U.S. airport security checkpoints, including use of a dedicated screening lane and more limited physical screening,” says a TSA notice published in the Federal Register.
Prospective participants in the program must provide biometric and biographic information to TSA.
“TSA has taken a number of actions to implement its intelligence-driven, risk-based approach to aviation security,” the agency explains in its published notice. “These actions include expedited screening for passengers 12-years old or younger or 75-years old or older and for U.S. military personnel. These steps enhance aviation security by permitting TSA to focus its limited security resources on passengers who are more likely to pose a threat to civil aviation, while also facilitating and improving the commercial aviation travel experience for the public.”
Dr. Wallace Loh,
The DHS Homeland Security Academic Advisory Council will hold a public meeting in the Ronald Reagan Building in Washington, DC, on Oct. 24 at which it will explore a wide range of academic questions, including how DHS can attract student interns, student veterans and recent graduates to jobs at the department.
According to a DHS notice in the Federal Register on Oct. 9, the academic advisory panel will meet at the Ronald Reagan International Trade Center at 1300 Pennsylvania Ave., NW, Floor B, Room B1.5-10, Washington, DC 20004 from 10 AM to 4 PM.
The group will hear updates from five of its subcommittees -- student and recent graduate recruitment, homeland security academic programs, academic research and faculty exchange, international students, and campus resilience -- and explore a number of pressing questions, including:
- How Black, Hispanic and Native American graduates can take advantage of DHS internship and job opportunities;
- How to enhance existing relationships between FEMA’s Emergency Management Institute and the higher education community;
- How universities can effectively communicate to DHS their emerging scientific findings and technologies.
Last March, DHS Secretary Janet Napolitano appointed Dr. Wallace Loh, the president of the University of Maryland in College Park, MD, as the advisory panel’s first chairman. Here are the names and affiliations of the panel’s members, as of March 2012:
- Dr. Joseph E. Aoun, President of Northeastern University in Boston, Massachusetts
- Dr. Lezli Baskerville, President and CEO of the National Association for Equal Opportunity in Higher Education in Washington, D.C.1
- Ms. Carrie L. Billy, President of the American Indian Higher Education Consortium in Alexandria, Virginia
- Dr. Walter G. Bumphus, President and CEO of the American Association of Community Colleges in Washington, D.C.
- Dr. David M. Dooley, President of the University of Rhode Island in Kingston, Rhode Island
- Dr. Royce C. Engstrom, President of the University of Montana in Missoula, Montana
- Dr. Antonio R. Flores, President and CEO of the Hispanic Association of Colleges and Universities in San Antonio, Texas
- Dr. Rufus Glasper, Chancellor of the Maricopa Community Colleges in Tempe, Arizona
- Dr. Jay Gogue, President of Auburn University in Auburn, Alabama
- Ms. Marlene M. Johnson, Executive Director and CEO of the Association of International Educators (NAFSA) in Washington, D.C.
- Dr. Eric W. Kaler, President of the University of Minnesota in the Twin Cities, Minnesota
- Dr. R. Bowen Loftin, President of Texas A&M University in College Station, Texas
- Dr. Wallace Loh, President of the University of Maryland in College Park, Maryland
- Dr. Gail O. Mellow, President of LaGuardia Community College in Long Island City, New York
- Hon. Ruby G. Moy, President and CEO of the Asian Pacific Islander American Association of Colleges and Universities in Washington, D.C.
- Dr. Hunter R. Rawlings, III, President of the Association of American Universities in Washington, D.C.
- Dr. John Sexton, President of New York University in New York City, New York
- Rear Admiral Sandra Stosz, Superintendent of the U.S. Coast Guard Academy in New London, Connecticut
- Dr. Dianne Boardley Suber, President of Saint Augustine’s College in Raleigh, North Carolina
- Dr. Holden Thorp, Chancellor of the University of North Carolina at Chapel Hill in Chapel Hill, North Carolina
Further information is available from Lindsay Burton, of the DHS office of academic engagement, at 202-447-4686 or [email protected].
In a two-year project, the Port of Portland, OR, recently completed a security upgrade that implemented the Transportation Worker Identification Credential (TWIC) standard for smart cards and improved access control for pedestrian, vehicle and rail routes leading to two multi-purpose, multi-modal terminals.
The program was funded, in part, by a grant from the Department of Homeland Security.
According to Forrest Gist, who was principal project manager for CH2M Hill [www.ch2m.com], the consulting engineers for the port upgrade (and now with Critigen) [www.critigen.com], “The Port was a great partner in the project. It was a nice blend of security hardening and advanced technology and communications.”
Gist worked with Dan Pippenger, the port security director; George Seaman, the project manager; and Jay Ostlund, the assistant project manager.
Products used in the project included an access control system from AMAG [www.amag.com], a video management system from Verint [www.verint.com], guard houses from BIG Enterprises [www.bigbooth.com] and thermal imaging cameras from FLIR [www.flir.com].
The Port of Portland services more than 800 vessel calls each year, accounting for more than 14 million tons of cargo, including bulk, break-bulk, containers and automobiles. Terminal operations continue night and day, year round, so the security infrastructure needs to be effective, regardless of weather or lighting conditions.
The port’s Terminal 4 has seven berths that can handle vehicle, bulk and liquid bulk cargo. Terminal 6 is a deep draft container terminal that spreads across 300 acres. It also services vehicles and break-bulk cargo. Each terminal can process 1,000 trucks per day, in addition to all of their normal rail traffic.
Some of the new security systems at the port focused on the cargo and its containers. The solutions included an optical character recognition (OCR) system, which scans shipping container markings and matches these markings to their truck’s license plate. In addition, radiation portal monitors were installed to scan containers for abnormal levels of radiation, which could indicate the presence of a dirty bomb.
Other parts of the security upgrade involved the terminals’ physical security, including the installation of improved guardhouses, reinforced fencing and an upgraded access control system, based on the implementation of the nationwide Transportation Worker Identification Credential (TWIC) system.
One of the largest security problems, common to many ports, was that certain areas were nearly impossible to secure physically. For example, it is impractical, expensive or dangerous to use fencing in railway areas because trains enter the terminal access areas at all hours of the day and night. This creates is an obvious vulnerability, because anyone can walk down the tracks and enter the port’s property before security personnel knew about it.
Similarly, waterfront areas are impractical to secure physically. Fences aren’t a viable solution because of the round-the-clock access required by ships, equipment and longshoremen.
The port’s solution to these challenges was to install FLIR thermal cameras, which enabled port security personnel to monitor rail access points 24 hours a day, and receive alarms whenever anyone crossed onto port property. The thermal cameras could effectively monitor large waterfront areas, such as Terminal 6, which is almost three miles long, and send alarms to the security operations guardhouse for evaluation and response.
Since thermal cameras make video images from heat, rather than light, and cannot be fooled by poor weather, darkness or camouflage, they work 24/7, and can see intruders from farther away than comparable CCTV cameras.
The Port of Portland’s thermal cameras, operating alongside daylight and lowlight video cameras, provide an overlapping mix of video coverage. Fixed thermal cameras of 19, 35, 50 and 100mm focal links monitor stationary areas and choke points. Port security also uses PTZ-50 and 35 x 140 multi-sensor cameras in central locations to scan larger areas, while maintaining a zoom capability for more detailed threat analyses.
By coupling the FLIR cameras with a video analytics package, the port created a virtual perimeter that helps the security teams detect movement in areas that are otherwise difficult or impossible to secure physically.
Editor’s Note: Critigen [www.critigen.com] is a global technology consultancy. Divested from CH2M Hill’s $6.4 billion engineering business, Critigen offers its expertise to worldwide clients to develop technology-based approaches. The Port of Portland project was completed under CH2M Hill.