How Technology Helped FBI Narrow Field Of Bombing Suspects

In the nearly 17 years since Atlanta's Olympic Park bombing, technology has transformed how large-scale investigations can work. Federal officials in Boston reportedly sifted through more than 10 terabytes of data — much of it images and video recorded at the marathon site. As NPR's Steve Henn reports, if you were to sit down to watch it all, it would take one person more than five years.

Copyright © 2013 NPR. For personal, noncommercial use only. See Terms of Use. For other uses, prior permission required.

STEVE INSKEEP, HOST:

Now, all of this began when the FBI released photos and videos yesterday of the two suspects in the marathon attack. Federal officials reportedly sifted through terabytes of data - an unbelievable amount of data - much of it images and videos recorded near the finish line. Now, if you were to sit down and watch it all, it would take one person years to do. However, as NPR's Steve Henn reports, in past decades, technology has transformed how these large-scale investigations play out.

STEVE HENN, BYLINE: In today's society, there are sensors almost everywhere - and I'm not just talking about surveillance cameras or even things like automatic license plate readers. Most of us walk around with a cell phone in our pocket, and it comes complete with a camera and microphone of its own. So, anytime you're in a crowd, chances are someone is rolling.

(SOUNDBITE OF EXPLOSION)

HENN: So, in Boston, Timothy Alben, who heads up the Massachusetts State Police, asked for help the day after the attack.

TIMOTHY ALBEN: There have to be hundreds, if not thousands, of photographs or videos or observations that were made down at that finish line yesterday. And I would encourage you to bring forward anything. You might not think it's significant but it might have some value to this investigation.

HENN: Alben clearly saw all these recordings as an opportunity for investigators. But the sheer volume also posed a challenge.

BRIAN CUNNINGHAM: The availability of electronic data - in this case, video data, still photo data, cellular calling records, cell tower records.

HENN: Brian Cunningham has worked at the CIA, the U.S. Department of Justice and on the national security staff at the White House. Today, he says, some of this information can be enormously valuable, but the quantity can be overwhelming.

AL SHIPP: Most people don't understand that putting more cameras doesn't necessarily yield more information.

HENN: Allen Shipp is the CEO of a company in San Francisco called 3VR. One of its first investors is In-Q-Tel, the CIA's venture capital fund. 3VR specializes in computerized video analytics, like facial recognition and object tracking. Shipp says often when investigators comb through videotapes, they're facing an information glut.

SHIPP: So, what we've done is we've indexed that information. So, when we do a video, we extract things like size, shape, color, speed, direction and let you search against that and point back to the video. So, instead of watching hours and maybe days of video, you can ask questions like show me all the red cars that went east.

HENN: Or look for anyone wearing a white baseball cap, like the suspect who allegedly set down his backpack at the site of the second explosion. Diego Simkin at 3VR says if you have high enough resolution images, their software can search by age or gender. He runs a search on the company's own security cameras looking for men.

DIEGO SIMKIN: We'll do 20 to 25.

HENN: That's pretty fine-grained.

SIMKIN: Yeah.

HENN: One of the hits is him. How old are you?

SIMKIN: I'm actually 28.

HENN: Not bad, but it's not perfect. And Shipp says this technology works best in controlled environments.

SHIPP: Well, there are its limitations. And so, you know, one thing that this technology has had a history of is overpromising and underdelivering.

HENN: The tools investigators have today are much more powerful than they were just a few years ago. But some who work in this field, like Amit Gavish, say the scene in Boston before the blast posed a challenge: it was crowded, a bit chaotic.

AMIT GAVISH: Deploying analytics at that moment, or using it for that specific moment, is extremely difficult for computer vision technologies.

HENN: At best, technologies like this make the work of human analysts go faster, perhaps help them find more images of their suspects - images they might otherwise have missed. Steve Henn, NPR News, Silicon Valley.

DAVID GREENE, HOST:

And, again, it was a chaotic night in Boston. And we'll be following these fast-moving events as they unfold over the course of the morning. Let's just go through, one more time, what occurred. Police in Boston have been on a manhunt for two suspects they believe were involved in the bombing of the Boston Marathon on Monday. Police are saying that one suspect in the bombing is dead. A second suspect is at-large in the suburb of Watertown, and they are telling people to stay in their homes. Earlier in the evening - and we believe this was related - a police officer was killed on the campus of the Massachusetts Institute of Technology. We have reporters in Boston. We have reporters following the investigation, and we'll be hearing from them all day long here on NPR News.

Copyright © 2013 NPR. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to NPR. This transcript is provided for personal, noncommercial use only, pursuant to our Terms of Use. Any other use requires NPR's prior permission. Visit our permissions page for further information.

NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.

Comments

 

Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.

Support comes from: