Las Vegas Sun

November 10, 2009

Currently: 57° | Complete forecast | Log in

Group says it can prove robber is innocent

Monday, June 28, 2004 | 11 a.m.

The image on the security-camera video is fuzzy. The man brandishes a gun as he demands the convenience-store clerk open the safe, but his face isn't clear.

As he leaves the store, he passes the height strip on the doorway -- but in the video, you can't even see the numbers.

Prosecutors used that video to send 25-year-old Bryon Garnett prison for at least 40 years. But now a group that works to free the wrongly convicted claims it can make novel use of NASA science to clean up the video and prove that the Las Vegas resident is too tall to be the robber.

"Actual innocence cases are extremely rare. This is one of those cases," said Don Topham, lawyer from the Rocky Mountain Innocence Center, a Salt Lake City-based nonprofit that looks at cases in Nevada, Utah and Wyoming.

"The work of an impartial scientist proves that the person who robbed this store was 5-foot eight," Topham said. "Bryon Garnett is 6-foot-1. He's innocent and should be set free."

Clark County prosecutors don't agree, of course. But on July 6, Topham hopes a judge will set Garnett free. If that happens, it would be the first time a video-enhancing method developed by NASA scientists was used to exonerate rather than convict someone.

The method, developed out of research on reading satellite and telescope images, has been used by prosecutors in high-profile cases including the 1996 Olympics bombing and the 2002 kidnapping of Salt Lake City teenager Elizabeth Smart.

Garnett was convicted in 2001 of committing the armed robbery that took place in 2000. Prosecutors said Garnett, who had recently been released from prison for car theft and lived in an apartment complex behind the Circle K store, was the blurry man in the video from the store's security cameras.

In the video, a man in a red baseball cap enters the store, approaches the clerk behind the counter and pulls out a gun. The two bend down to the safe. A woman walks in seeking change for a $100 bill. The robber takes the bill, vaults over the counter and runs out.

"It was a simple case," recalled prosecutor Mary Hart, who tried the original case and will argue against overturning Garnett's conviction before District Judge John McGroarty on July 6.

A store clerk who was not on duty at the time of the robbery recognized the robber's jacket in the video as belonging to Garnett, who was a regular customer, Hart said.

And the on-duty clerk picked Garnett out of a lineup, Hart said, although the defense points out that the clerk was very slow to choose Garnett and originally said he wasn't totally sure.

But Topham says the linchpin of the case will be the robber's height. NASA scientist David Hathaway, based in Alabama, used the Video Image Stabilization and Registration method, which he helped invent, to sharpen the surveillance video.

In the cleaned-up copy of the video, the store's height strip was legible. Hathaway, who Topham said was originally skeptical, used the height strip as well as the door frame to conclude that the robber was 5 feet 8 inches tall, with a one-inch margin of error.

"Standing against a wall in his socks, Bryon is 6-foot-1," Topham said. "You can't make yourself five inches shorter. You can make yourself taller, but there's no way to make yourself shorter."

But Hart said the scientist's analysis, however fancy its pedigree, was inexact.

"Dr. Hathaway's equipment might be good, but when you start doing all these angles, and it's (the video) two-dimensional, how can you tell me what the height is?" Hart said. "You can't."

The robber was pitched forward, running out the door, not standing up straight, Hart said. And there's no guarantee the height strip, which has since been replaced, was accurate, he added.

But Topham's group hopes the Garnett case will succeed and thereby blaze a trail for other convicts to be cleared based on the video-analysis method.

Clearing up videos can make the license plates of a getaway car visible or someone's face recognizable, Topham said.

"I think this definitely will be used more in cases like this" if Garnett is found to be innocent, he said.

The Wisconsin Innocence Project, having heard about Garnett, is interested in using the NASA method for a similar case, Topham said.

The Rocky Mountain Innocence Center began operating two years ago and is hoping to make Garnett its first exoneration.

"Prisoners write to us and say, 'I'm in jail for a crime I didn't do,' " Topham said. "As you can imagine, we get a lot of letters."

The center focuses on what it calls "actual innocence": convicts who say they had nothing to do with the crime in question -- as opposed to those who admit they were involved but say they didn't pull the trigger. The center's lawyers look for cases in which physical evidence, often DNA, can back up someone's claim of innocence.

The refinement of DNA testing methods in the mid-1990s led to a rash of exonerations, exposing the legal system's fallibility, Topham said.

"Even if the system is 99.9 percent correct, on a national scale, that still leaves thousands of people in jail who are innocent," he said.

In Nevada, exonerating convicts is made more difficult by a statute of limitations that prevents convicts from asking for evidence to be re-examined after two years, unless they are on death row. That means there is no recourse for someone who could prove by a blood sample that he was wrongly convicted of rape in the 1980s, before the advent of DNA testing, he said.

It is critical for the public to keep in mind that the justice system has none of the safeguards of other professions such as medicine, Topham said.

'The legal system is set up so that we believe it's 100 percent effective. That's just impossible," he said.

But Hart said prosecutors are careful to be sure in every case they pursue. "When you come in and you prosecute, the last thing you want to do is put an innocent person away," he said.

The Innocence Center's defense of Garnett, he said, is "Monday quarterbacking."

archive

  • Most Read
  • Discussed
  • Most E-mailed

Calendar »

  • 10 Tue
  • 11 Wed
  • 12 Thu
  • 13 Fri
  • 14 Sat