Authentication on the front lines

The iA-thenticate platform can stand on its own or work as a PC peripheral to examine travel documents for multiple kinds of embedded security features.

Boston's Logan Airport is testing a method that verifies badges, other documents

A document security company wants to interest the federal government in a faster way to authenticate passports and IDs.

Imaging Automation Inc. of Bedford, N.H., tests embedded security features in documents.
'Prior to Sept. 11, we were very much focused on the international market' because other countries had higher security standards for travel documents, chief technology officer Jeff Setrin said. Methods include special inks, laminates, printing patterns and materials that change appearance when tampered with.

More recently, however, U.S. jurisdictions are adopting tamper-resistant documents, Setrin said. A handful of states, including West Virginia and New York, have changed their driver's licenses.

Methods are undetermined

The Homeland Security Department's $400 million U.S. Visitor and Immigration Status Indication Technology System will use a so-far undefined method for authenticating travel documents.

Setrin said U.S. security spending so far has gone mostly for X-ray machines and metal detectors. Equipment to read security-embedded documents is available only in laboratories.

'We realized there were no tools in the field to detect whether documents were authentic,' he said. 'We started shifting our focus away from the issuance side and toward the authentication side' with the company's iA-thenticate platform.

'We're using it in our badging process,' said Dennis Treece, security director at Logan International Airport in Boston. 'It's a winning technology. I'd love to see it expand into the whole passenger process.'

IA-thenticate is available either as a standalone, $6,000 platform or as a PC peripheral. Its charge-coupled-device digital camera views a document under multiple light sources, including visible, infrared and ultraviolet.

'We capture five different images of the document,' Setrin said.

Optical character recognition, pattern-matching and other applications then examine the five images to verify embedded security features.

Forensic examination verifies patterns under various lights, checks for the proper ink, looks for color and pattern changes that indicate tampering, and examines other security features.

All the data captured by OCR, bar code scanning and magnetic stripe reading combines to answer these questions:
  • Has the document expired?

  • Does it meet standards for its type?

  • Do checksums in designated fields match?

  • Are facts such as birth dates consistent throughout?

  • Is the document presenter on a government watch list?

The system typically flags about 3 percent of documents for in-depth examination, and about 10 percent of those ultimately prove to be false, Setrin said.

There are no figures for how many false documents iA-thenticate misses.

'We have some access to fraudulent documents, so we train the system to recognize those,' he said. Some customers also test with their own false documents, he said, 'but they have not shared their information with us.'

Logan airport operates two workstations to check the credentials of employees and contractors who go into secure areas.

'For years Logan has been a construction site with an airport attached,' Treece said. 'We have at any moment something in excess of 13,000 security badges issued. This gives us some assurance early in the process.'

Treece said it takes less than a second to check most types of ID. 'I suspect if you brought in a Zambian passport it might take a few seconds longer.'

IA-thenticate has caught some bad IDs at Logan. 'It doesn't happen very often,' he said, but 'it replaces eyeballing.'

About the Author

William Jackson is freelance writer and the author of the CyberEye blog.

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above