News

Student Using Human Brain Mapping to Power Image Search Engine

There are some exciting new innovations in search coming up on the horizon, and one of those is a project currently being worked on by a Canadian computer science grad student, who is mapping the way the human brain works and applying it to technology that will eventually power a search engine dedicated to visual images. Kris Woodbeck, a University of Ottawa Master’s degree student, has been working with the Canadian government’s Technology Transfer and Business Enterprise (TTBE) office to secure a patent for his project. The patent will form the intellectual property for a startup devoted to image search, which is expected to be launched some time next year.

Already Woodbeck has created a prototype search engine based on the pending patent. The search engine mimics the way the brain processes visual information, and takes advantage of the currently available graphics processing capabilities of computers. Woodbeck sees a lot of potential for the image search area of search engines, and says that currently available technology just isn’t up to par:

“All the major (search) players don’t really look at visual content,” he argued. “Their results aren’t very good. You’ll see a search engine, for instance, that uses the color of the picture to classify it. It’s very new field. It’s in its infancy because everyone is using metadata. Nobody’s looking at the content, because it’s so slow to process images.”

Woodbeck hasn’t fully hashed out all of the company’s business plans, but says that he sees potential for his patent beyond just generic object recognition. He expects that there will be opportunities to license the intellectual property to biometrics sectors (such as facial recognition), and even the medical and military markets.

He has been testing his prototype search engine on academic data sets containing somewhere between 60,000 to 100,000 images, and has been trying to classify the images according to certain labels. Woodbeck said that he will soon start using web crawlers to begin indexing images so that the search engine will have at least some content ready to go for its launch in 2008.

For his efforts, Kris Woodbeck won his university’s “Innovator of the Year” award, which was awarded last week.

You Might Also Like

Comments are closed.

6 thoughts on “Student Using Human Brain Mapping to Power Image Search Engine

  1. If you are interested in such technologies – welcome. Our prototype works in Russia and now contains about 2 000 000 indexed photos. It’s open for public testing, English version coming soon :)