Search and rescue operations in emergency situations are often stymied by the inability to find victims hidden amid rubble or trapped in tunnels. But what if you could somehow see beyond solid objects? That is no longer the realm of science fiction or comic book superheroes like Superman and his X-ray vision.
A new form of photography, called femto-photography, is emerging. It exploits the finite speed of light, high-speed lasers, and ultra-fast imaging to overcome the seemingly impossible task of recording what is beyond the line of sight. Developed by the Media Lab’s Camera Culture group at the Massachusetts Institute of Technology in collaboration with Bawendi Lab in the department of chemistry at MIT, femto-photography emits short laser pulses and analyzes multibounce reflections—known as “echoes of light”—to estimate hidden geometry behind solid objects. Basically, “We take the same picture over and over again 10 million times,” Ramesh Raskar, project director and an associate professor at MIT’s Media Lab, said. “It takes about an hour to do this,” he added. “It’s the world’s fastest slowest camera.”
It’s also expensive: The lab prototype costs an estimated $500,000. But Raskar, who spoke at TEDGlobal, believes it could be done for a few thousand dollars: Many of the components—like streak tubes, used for high-speed image analysis—are already available, which would help make the device a viable commercial product beyond the small market for purely scientific purposes.
One potential market is of course search and rescue equipment for use in hazardous conditions like post-earthquake relief missions. In this case, rescuers could shoot beams of light into an area; then, by observing changes in the reflection, receive information if something is moving around out of view. Another application would be medical imaging. Instead of building a smaller and smaller endoscope to go deeper into the body, Raskar explained, you could place the scope at a convenient location and look more easily into cavities in the lungs, for example, or around obstacles.
“It’s the same principle as looking for a car coming around the bend, out of the line of sight,” Raskar said, noting that collision avoidance could also be a useful application. The idea is always the same: “You send a beam of light to what is visible, and it reflects what is hidden, and this comes back to the camera. With multiple bounces of light you take pictures”—to see through obtrusive buildings and walls. Who needs Superman?