Completely Automated Public Turing Tests to Tell Computers and Humans Apart (CAPTCHAs) are the automatic filters that are widely used these days to disallow any automated script that can perform the work of a human. CAPTCHAs are built in such a way that it is very difficult for any automated script to break them. In this paper, an approach to break text based CAPTCHAs has been proposed that first preprocesses the given CAPTCHA, segments its characters, and then recognizes the characters depending on it's features. The breaking of CAPTCHA gives strength of CAPTCHA which in turn helps to develop more robust CAPTCHA The testing results shows that the algorithm successfully breaks the CAPCHA. This algorithm will be also helpful to screen reader program. The results show that it is not a trivial task to design a CAPTCHA scheme that is both usable and robust. © 2009 IEEE.