Border Checkpoints Go High Tech

Science fiction is quickly becoming science fact

• By John Merlino
• Mar 01, 2015

Imagine this: You’re looking around the room and people are getting retinal scans. Others are presenting RFID cards. Some are undergoing fingerprint imaging. Did you accidentally walk onto the set of some futuristic blockbuster movie? Guess again. You’re actually standing at a U.S. border checkpoint, waiting to be processed.

It’s not unusual nowadays to find domestic and global entry points leveraging technologies once considered to be science fiction. Prime examples are the U.S. Customs GOES program (Global Online Enrollment System for U.S. Customs and Border Protection Trusted Traveler Program) and its Canadian counterpart NEXUS, which are applying a host of emerging technologies to pre-approve citizens and expedite passage at ports of entry, including everything from photo IDs with embedded RFID chips to iris scans and hand imaging.

Employing these new screening technologies allows United States and Canadian border agencies to concentrate the bulk of their efforts on potentially higherrisk travelers and goods, in order to secure the integrity of our borders. It’s a dual authentication procedure that links passport documents with biometric data. Of course agents still conduct random screening interviews, but the use of technology brings a whole new level of efficiency to checkpoint processing.

The United States is currently discussing how to leverage advanced technology to pre-screen travelers seeking admittance across its southern borders. Going hand-in-hand with the discussion on immigration reform is the debate on how to work with our southern neighbors to create a secure worker identification credential that can authenticate a person with biometric data such as hand, retinal or facial geometry.

While GOES and NEXUS are in their infancy, the proliferation of these kinds of technologies holds tremendous promise for helping border agencies more tightly control legitimate traffic and block nefarious passage. At “working” labs on the U.S. and Canadian borders the efficacy of these programs are still being tested. But as beta sites they can be great proving grounds for the wider use of multiple authentication credential programs at our borders. They, along with other Trusted Traveler programs such as the airline passenger screening program called “TSA-Pre” seem to be gaining acceptance, clearly demonstrating that the traveling public is willing to work with the security agencies protecting our country.

Out of the Lab and Into the Field

Many of the advanced technologies border checkpoints are using today have been niche solutions in high security environments for decades. Crossover for tools like fingerprint scanning and hand geometry imaging was slow because initially these applications required huge amounts of data to be stored locally. But with advances in processing power, bandwidth capacity, data compression and storage efficiencies, now imaging databases can be cost-effectively archived in remote server farms and accessed quickly over a secure network for speedy validation.

Another development contributing to successful crossover has been the introduction of higher-sensitivity visual processors which enable biometric programs to efficiently capture the images of a person’s iris and retina in a single scan. With the addition of software specially designed to manage metadata, border agencies can now efficiently link any biometric data with other information stored on travelers’ credentials to help further expedite border processing.

While streamlining traffic flow is proving valuable at border checkpoints, it doesn’t take much of a leap of imagination to see that the technology might eventually play a key role in other high throughput environments such as airports, seaports and other Customs verification check points.

Bringing Facial Recognition into Play

As facial recognition software continues to improve in accuracy and reliability, it, too, will inevitably wind its way into the border checkpoint toolkit. At some point, it might even replace the current retinal/iris scanning technologies used by GOES and NEXUS as the preferred screening solution.

Like other imaging technologies, limitations in processing power, bandwidth capacity, data compression and storage efficiencies had been hampering the widespread adoption of facial recognition software— but no more. With far fewer constraints, many facial recognition solutions are now able to integrate multiple image sensors (cameras) with portals that direct traffic flow. A subject’s image can be captured multiple times, collated and digitally stored using an algorithm matrix that analyzes specific facial features.

While the procedure requires high processor overhead from multiple software and hardware platforms, new open standards-based architecture, high-megapixel network cameras and sophisticated analytics are helping to accelerate the accuracy, reliability, usability and timely retrieval of these enhanced images.

Augmenting the Human Factor

In the aftermath of 9/11, we’ve forever changed our views on vigilance against internal and external threats. Nowadays the average person has come to expect closer scrutiny and screening in nearly all aspects of their daily lives—from traffic cams and toll booth recorders to TSA advanced imaging machines at airports; omnipresent surveillance cameras in office buildings, schools and shopping centers; and now even wearable cameras on police officers.

Without technology to augment human screening, law enforcement and security staff would be quickly overwhelmed by the volume of tasks needing to be processed.

Numerous studies conducted by the military and private enterprises have repeatedly demonstrated the inherent limitations on how well humans manage multiple tasks simultaneously. Considering the complexities of a border crossing—or the similar challenges of airport screening—which often includes country clearances, checking for contraband and identifying individuals on a watch list—technology, of necessity, becomes a critical force multiplier. Technologies that can match an image with a known watch list can greatly reduce human error. Knowing that they have the support of these increasingly reliable and sophisticated systems lowers security agents’ stress and enables them to function for longer periods of time with increased situational awareness.

Acknowledging Possible Downsides

There are definitely measurable upsides to technologies like worker and visitor credentials that make their future use imminent. While the goal of these types of systems is to allow greater “authorized” access, some may argue that the resultant higher throughput at our borders will present the opportunity for more unlawful entry. The possibility certainly exists. But using these credential systems to pre-screen travelers will give agents more time to discern the subtle clues that can only be detected through human interaction.

Deploying additional technologies such as facial recognition against known watch lists also has a great upside in helping to detect persons of interest. As these solutions continue to mature the essential question becomes: How can we get smarter, faster and better at securing our borders?

What the Next Innovations Might Be

Due to a number of new developments in enhanced image quality, advanced compression and rapid storage and retrieval of relevant video, IP video has begun to outpace analog technology, especially in high security environments. Increased processing power at the edge provides the opportunity for manufacturers to embed greater intelligence in the camera itself—such as motion detection, tampering alarms, cross-line detection and more—which allows border agents to anticipate and respond more quickly to threats.

Now viewed as smarter edge appliances, network cameras remove much of the legacy server overhead of the past because the software is managed directly on the device rather than from the remote server. As processing power continues to grow, this new system configuration will ultimately pave the way for facial recognition programs operating in-camera that are capable of obtaining a “target/suspect” and reporting that to a security agent far removed from the location.

But for these sophisticated analytic programs, often created by third-party developers, to succeed they need to follow a few basic tenets:

• Use an open platform with a standards based approach, such as ONVIF/ Open Network Video Interface which allows them to work with other open applications.
• Build on native applications that provide integration capabilities through APIs/Application programming interfaces, or use a servicesbased approach with a requisite software development kit/SDK.
• Have the capability to leverage the enhanced processing capabilities of a robust processor within the device itself.
• Conform to the inherent security protocols of a mature and secure operating system, a necessary prerequisite to obtaining authorization to operate on a Federal or DoD/ Military network.

Balancing the Need for Homeland Security against Personal Privacy

It’s important to recognize that hightech screening and surveillance come at a cost. The ongoing great debate on privacy versus ready access to information is definitely heating up.

There is no doubt that the interests of national security and the privacy of our society will often be at odds. However, polls seem to indicate that a growing percentage of people are okay with surrendering a bit of privacy if the end result is a safer, more secure world.

This article originally appeared in the March 2015 issue of Security Today.