Facial recognition software failed to identify Boston Marathon bombing suspects, despite photos available in databasesApril 23, 2013 11 Email article | Print article
Even though the whole ordeal is over, and Boston is slowly picking up the pieces after the Boston Marathon bombing, lots of people want to know why this happened. Another relevant question is how the two suspects evaded law enforcement for so long before finally being discovered. One reason is failure of highly-expensive facial recognition software.
The Boston Police Commissioner Edward Davis, told the Washington Post that facial recognition software used by the department was not able to identify either of the two suspects.
According to the Post:
“The technology came up empty even though both Tsarnaevs’ images exist in official databases: Dzhokhar had a Massachusetts driver’s license; the brothers had legally immigrated; and Tamerlan had been the subject of some FBI investigation.”
It seems silly that the software failed, especially considering how advanced modern technology is these days. Then again, I’m willing to bet the software used at the Boston PD is probably outdated. It’s no secret that it costs a lot of money to stay on the bleeding edge, and that’s money the state probably isn’t willing to invest. Keep in mind I’m just speculating here, I actually have no clue how updated the Boston PD’s facial recognition software is.
Regardless, facial recognition systems don’t work too well when scanning low resolution images like those taken from a surveillance video and low-quality cell phone cameras. (I’m not even going to bother to generalize that statement about all cell phone cameras because I know better; a lot of smartphones are just as powerful as professional point and shoot cameras now.)
Even so, surveillance equipment was partly responsible for authorities being able to identify the suspects, as the Post notes:
“The work was painstaking and mind-numbing: One agent watched the same segment of video 400 times. The goal was to construct a timeline of images, following possible suspects as they moved along the sidewalks, building a narrative out of a random jumble of pictures from thousands of different phones and cameras. It took a couple of days, but analysts began to focus on two men in baseball caps who had brought heavy black bags into the crowd near the marathon’s finish line but left without those bags.”
I don’t know what to think of all this information, but I will say in this day and age it’s unacceptable. What good is facial recognition software that probably cost millions of dollars if law enforcement must manually identify persons of interest? I’m not suggesting everyone should have pulled out their phones when the bombs went off to get the high-resolution photos or videos required for facial recognition to work; I’m saying you’d think the expensive technology would at least be able to flag the suspects involved regardless of the quality of photos/videos given.
What do you think? Let us know in the comments below!