A new project from the University of Oxford (UK), the University of Portsmouth (UK) and Johns Hopkins University (US) aims to harness the power of the human brain to identify and classify galaxies and stars. On the Galaxy Zoo website, users are asked to identify the objects in photographs as spiral or elliptical galaxies, the direction of rotation, or if the photo depicts a star or merger of galaxies. The site launched yesterday and says they have already had an “amazing response.”
“The human brain is actually better than a computer at pattern recognition tasks like this. Whether you spend five minutes, fifteen minutes or five hours using the site your contribution will be invaluable,” said Kevin Schawinski of Oxford University of the project.
Researchers hope that the Galaxy Zoo project will yield a greater understanding of how galaxies are distributed across the universe and perhaps even improve upon the current model of the universe itself. “It will be great to have all the galaxies classified; itÄôs as fundamental as knowing if a human is male or female,” said Professor Bob Nichol from the University of Portsmouth. Along the way, researchers hope that participants will have a little fun. The web site will keep statistics about who are the most accurate and voracious galaxy catalogers, as well as let users print out posters of galaxies they have identified.
The Galaxy Zoo team was inspired by the ‘Stardust@home’ project, in which NASA employed the use of distributed brain power to look for interstellar dust. So what other projects can you donate some spare brain cycles to it the interest of science?
- Stardust@home – I mentioned this one above. Users are invited to sort through over a million “focus movies” — magnified images of blocks of aerogel that were exposed to dust grains on the Comet Wild-2. Sounds like fun.
- Clickworkers – Another NASA study, this time to identify and classify craters on Mars.
- Distributed Proofreaders – In order to digitize old books, they must be scanned, page by page, and converted to text using optical character recognition software. Unfortunately, old type faces often foul up the OCR software, so images of the scanned pages need to be compared to the OCR text. I did this type of work as part of a research project in college scanning old maritime manuals and logs, and it can be tedious.
- reCAPTCHA – This is one of the more clever implementations of a distributed brain project. reCAPTCHA uses spam-fighting CAPTCHA images to help identify words that OCR software can’t figure out when scanning old books. See our full review.
- Google Image Labeler – Improve Google’s image search technology by tagging images. Earn points based on the complexity of your tag (i.e., ‘bird’ will get you fewer points than ‘yellow crested cockatoo’) and compete against other image labelers.
Do you know of any other distributed brain projects? Would you participate in any of these? Let us know in the comments below.
