U prof helps NASA get clearer Mars images

Jenna Ross

Professor Guillermo Sapiro might have developed an imaging technology used on the Mars rovers, but he has little use for high-tech equipment at home.

“I have one of those standard Olympus digital cameras, and it’s more technology than I need,” Sapiro said. “I actually reduce the quality of the photos before I print them.”

Sapiro’s work, however, has the opposite aim.

While employed by Hewlett-Packard Development Company in the mid-1990s, Sapiro and his colleagues conceived a method to compress images without a loss in quality. This image-compression algorithm now plays an integral part in processing photos taken by NASA’s roving spacecraft, including the two currently on Mars.

“My algorithm is walking Mars,” said Sapiro, who came to the University in 1997. “It’s pretty cool.”

Mostafa Kaveh, head of the electrical and computer engineering department, agreed.

“(Sapiro’s) work is so fundamentally strong,” Kaveh said. “This is only one example of his tremendously creative contributions to science.”

The Spirit and Opportunity rovers, part of NASA’s Mars Exploration Rover project, landed on Mars Jan. 3 and 24, respectively. Within hours of landing, Spirit sent black-and-white and color pictures of the planet’s surface.

Sapiro said the rovers use two techniques for transmitting and compressing these photos. The first is a loss technology, which looks complete but is not scientifically accurate.

“This is used for the photos you see on CNN,” he said.

Sapiro helped create the second technology, the international standard for lossless compression, which preserves the true images the rover takes.

“It is exactly what the robot sees,” Sapiro said. “When the scientists need accurate data, they will use the lossless photos.”

Sapiro said it is difficult to tell by looking at the photos which technique was used, but he heard the first landing pictures taken by the Spirit rover used his algorithm.

“The quality of the data is incredible,” he said. “In such adversarial conditions, it’s taking gorgeous pictures.”

Although seeing the photos his technology transmitted is good, Sapiro said, the project is most impressive to him.

“Mine is a very small contribution to this huge operation,” he said. “The rovers are incredible. It’s a tremendous scientific accomplishment.”

When Sapiro learned that NASA lost contact with the Spirit rover, he was not fazed.

“You think it’s shocking to hear that there’s a bug in this highly important system, but these guys don’t have three lines of code to deal with – they have millions of lines of code,” Sapiro said. “Even if they both died today, it would remain an amazing achievement.”

Sapiro’s interest in image processing began in Israel, where he studied at the Technion Israel Institute of Technology as an undergraduate in 1985.

He attended seminars given by professors who explained their fields of work, one of whom spoke about image processing.

“Those 50 minutes with that professor was the reason I do what I do,” Sapiro said.

Now Sapiro is the professor.

He said he wanted to go back to academia after working for Hewlett-Packard in California for three years. Although he planned to return to Israel, he was tempted by an interview at the University.

Kaveh helped recruit Sapiro to the University.

“We had heard a great deal about his mathematical talents and engineering prowess,” Kaveh said. “The first chance we had, we worked to bring him here.”

Sapiro has taught courses in a number of areas, including brain imaging, image processing and signals and systems.