This summer I am off in the far away land known as Connecticut at the Research Experience for Undergraduates at Fairfield University. Thankfully, the wonders of the series of tubes allow me to blog from this desolate land.
When I tell people that I am doing research in the field of Computer Security they think of hackers and viruses and guns and Global Thermonuclear War. Sadly, my research group is only focusing on the topic of CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart). Though it may sound boring in comparison, I actually have gotten pretty excited about it and will attempt to share some of that excitement.
For those of you who don’t know, a CAPTCHA is the name assigned to those annoying tests that you need to take when signing up for an account (an example can be found in the upper left corner of this post). The basic idea is that only humans should be able to pass these tests so that spammers and other bad people cannot go around and write a script to create 7 million email accounts in 3 seconds for the purposes of spamming us.
A real challenge is that computers are getting better at solving text based CAPTCHAs because of improvements in the field of Optical Character Recognition. Just last month hackers almost completely broke Google’s system; thankfully Google fixed the issue rapidly. Those kind of events force the tests to become harder in order to remain relevant, meaning that humans will have a harder time reading them. This is why, as far as I can tell, everybody hates CAPTCHAs.
The specifics of our research is in a CAPTCHA that uses handwriting and/or cursive text as opposed to the standard printed text. Handwriting and cursive text are harder for computers to decode because it is difficult for them to even isolate the letters from each other. You can see a short presentation here that I gave which contains some more information and references.
What do you think of CAPTCHA?
I’ll leave you with a cool video explaining how ReCAPTCHA works.