07-004: Image Based Turing Test


This technology is an image based CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) that can authenticate that the user of a web service is a human being.

The use of CAPTCHAs in web interfaces, it is hoped will keep such Bots from misusing the web service. Human users will be easily able to authenticate themselves as being human by passing the tests while machines will typically fail. Some practical examples of web services where CAPTCHAs are required are: online polls, preventing spammers from getting free e-mail ids, preventing chat bots from irritating people in chat rooms, preventing automated online dictionary attacks in password systems, preventing unruly search engine bots from indexing private web pages, preventing web bots from adding advertisements to comment fields in Blogs etc. As the web replaces human to human physical interaction such examples are bound to increase.

A common type of CAPTCHA requires that the user type the letters of a distorted image, sometimes with the addition of an obscured sequence of letters or digits that appears on the screen. A person’s ability to pass this CAPTCHA may be related to the person’s familiarity with the language that the letters or digits originate.

To get past the cultural bias of present CAPTCHA’s, this technology requires that the user to match groups of related images. To make the test more difficult for a computer o pass, many of the images are distorted. Additionally, the images may be pictures of faces taken from different angles and under different lighting conditions.