ARCHIVED - Rubbing out red eye, and other adventures in digital imaging

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

iSYS Corporation

April 26, 2007— Ottawa, Ontario

First photo: original with red-eye; second photo: after iSYS Automated Red-Eye Correction has been applied

First photo: original with red-eye; second photo: after iSYS Automated Red-Eye Correction has been applied

Millions of digital camera users now print their pictures at photo kiosks located in stores across the country. The process is downright simple, so much so that it can be easy to overlook some truly sophisticated innovations that improve the final product.

Among the most remarkable of these innovations is a system for removing the glowing red colour that often appears in photographs when a camera flash reflects off the blood vessels in a subject's eyes. A small St. John's company has made it possible for someone at a photo kiosk to push just one button and make this annoying problem go away — regardless of where the red eyes might appear in the image.

That button activates PixFix™ Red-Eye, a patent pending system introduced by iSYS Corporation in 2002. Although the firm only has three employees, they worked in collaboration with the National Research Council Industrial Research Assistance Program (NRC-IRAP), which provides a range of both technical and business advisory services along with potential financial support to growth-oriented and innovative small and medium-sized Canadian enterprises. Delivered by an extensive integrated network of Industrial Technology Advisors (ITAs) — a group of some 260 professionals in 100 communities across the country — NRC-IRAP supports innovative research, development, and commercialization of new products and services.

In this case, NRC-IRAP helped iSYS develop software that merely replaces the red spot with a black one to match the subject's pupil. However, placing that spot correctly in the image was a technical challenge that had daunted photo printing firms.

"We just kind of swept in there," says Vice-President of Business Development John Guzzwell, who recalls the flurry of interest from industry giants such as Sony when iSYS first showcased its product at a major European trade show.

In fact, PixFix™ Red-Eye has since been licensed to12 companies in nine countries in North America, Europe and Asia. The system now checks for red-eye problems in billions of photos every year, whether it be at large commercial printing operations, familiar consumer kiosks, or desktop printers designed especially for handling photographs.

And in 2006 such success also won the prestigious Manning Innovation Award for Guzzwell, Patricia LeFeuvre, and Rodney Hale, the trio who incorporated iSYS in 1999. They had been working together at C-CORE, a research venture based at Memorial University of Newfoundland, and wanted to explore the commercial prospects of new imaging technologies.

Their earliest work took place in fish-processing plants, where they developed algorithms that enabled cameras to determine if fillets travelling on a conveyor belt had any visible defects.

With the backing of NRC-IRAP, the company subsequently used their image analysis know-how to develop PixFix™ Red-Eye, a product that had to be ready to tackle a rapidly growing marketplace. Guzzwell points out that even with the revenue that the product now provides to the company, iSYS continues to maintain a dynamic relationship with NRC-IRAP to refine and expand its product offerings.

"You can't do development work that's unfunded and still survive," says Guzzwell. "Just the fact that we can undertake these projects and develop these software products is fabulous."

Enquiries: Media relations
613-991-1431
media@nrc-cnrc.gc.ca

NRC-IRAP
1-877-994-4727
publicinquiries.irap-pari@nrc-cnrc.gc.ca

Stay connected

Subscribe

Date modified: