Tech Columnist Parmy Olson published an article on those inventors who turned billions of Phones into cameras
Taking a selfie is one of the easiest and quickest things you can do on your smartphone. But as with any landmark invention, it took decades and plenty of graft to develop the camera technology that lives in your pocket.
A TRIO OF ENGINEERS BEHIND THE INVENTION OF THE IMAGE SENSING TECHNOLOGY
A trio of engineers behind the invention of the image-sensing technology found in billions of smartphones, camera phones, PCs and hospital scanning technology, won the £1 million ($1.3 million) Queen Elizabeth Prize for engineering on Wednesday, and spoke about where the image-sensor technology they developed should go in the future.
“I feel gobsmacked and very thankful to the Queen Elizabeth prize for this honor,” said one of the engineers, Eric Fossum. “I’m still continuing to do research into very-high sensitivity sensors, so we can work under very dark conditions. I’m working on trying to count every single proton and particle of light as it comes into the sensor.”
IT TOOK THREE GENERATIONS OF WORK IN ACROSS THE WORLD
It took three generations of work in across the world to eventually come up with the technology to produce cameras small and cheap enough to fit in a smartphone.
The first stage of the digital camera revolution came in the 1970s with the charge-coupled device (CCD) developed by British engineer Michael Tompsett. In 1972 he used a shoebox-sized camera to take the first digital colour photo of his wife, Margaret. It was blurrier than what you’d get on film back then, but it made the cover of Electronics Magazine and marked the beginning of a new wave of digital imaging.
NEXT CAME THE RAPID COMMERCIALIZATION OF IMAGE SENSORS IN JAPAN
Next came the rapid commercialization of image-sensors in Japan in the 1980’s, sparked by Nobukazu Teranishi’s invention of the pinned photodiode, a semiconductor device that converts light into an electric current. Teranishi, who worked for NEC Corporation at the time, figured out how to produce higher-resolution images with smaller pixels.
Finally in 1993, more than two decades after that first blurry photo of Margaret Tompsett, engineer Eric Fossum figured out how to miniaturize the technology. Working out of NASA’s Jet Propulsion Laboratory in Pasadena, Calif., he developed the CMOS sensor to try and protect camera images in space from being ruined by high-energy cosmic particles. The sensor turned out to be a cheaper and more energy-efficient alternative for digital cameras everywhere.
Years later and sitting in a conference room ahead of their award presentation from Princess Anne, daughter of Queen Elizabeth II, the engineers were circumspect on Wednesday about what their inventions had led to in the decades since.
“Selfies are something I didn’t think about,” says Fossum. “Or silly cat videos.”
The proliferation of millions of cameras in cities for constant surveillance was on their minds too. “It boggles my imagination,” Fossum added. “Putting cameras on drones that you can fly into your neighbor’s yard. Who would have thought this would have been an issue?”
Teranishi said cameras needed to become smarter to tackle the privacy problem. “Future cameras will not provide images,” he told FORBES, but recognize objects like fire when they need to. “It doesn’t have to conflict with privacy.”
Fossum disagreed somewhat here. “I don’t know if you can get rid of the visual information part,” he later said on the sidelines. “To reduce the false alarm rate you may want to be able to look through the monitor.”
This points to the next big next big step that image-recognition technology is now facing: incorporating artificial intelligence software.
That will be fundamental to products like self-driving cars. The fatal crash of a self-driving Tesla Model S car last year in the U.S. occurred in part because the forward facing sensors (which included a camera, a radar and ultrasonic sensors) failed to distinguish the white side of a tractor trailer against a brightly-lit sky.
Teranishi said engineers now needed to develop technology that could better distinguish between different materials like the side of a tractor trailer. Broadly speaking, the image sensor and processor within camera sensors needed to become more tightly integrated.
“We need more test driving,” Teranishi added. “We need more learning.”
The Queen Elizabeth Prize is modelled on the Nobel Prize and is said to be the most prestigious in the field of engineering. Its two previous winners are Tim Berners-Lee and the team behind the development of the World Wide Web in 2013 and Dr. Robert Langer in 2015. Lancer developed the controlled-release large molecule drugs delivery systems, which the prize’s organizers said had benefited the lives of more than 2 billion people worldwide.
The three engineers who won the prize this week, along with George Smith, an American winner of the Nobel Prize in Physics whose work in electronic memory design proved the basis for Tompsett’s original CCD sensor, will be formally honored with a ceremony at Buckingham Palace later this year.
All three of the prize winners spoke of donating their prize money to organisations that promote engineering to kids and young entrepreneurs.”I’m still an active professor at Dartmouth,” said Fossum. “We try to tell students it’s okay to not get the right answer the first time. That’s not what they’re taught at school. The process of engineering is to try, and then you do it again and again, and then it gets better.
“It’s not just the students but the teachers we’re trying to communicate this to. They have to understand there has to be room for this creative process, and trial and error that has to happen. Don’t give up if it doesn’t happen the first time. That’s my message.