Review and comparison of captcha approaches and a new captcha model

Abstract

Captcha is an abbreviation of Completely Automated Public Turing test to Tell Computers and Humans Apart. It is a technique used for verification purposes in many applications. Software developers often use this method to get rid of the malicious requests that automatic tools will make. The proliferation of bots and spam tools in particular has led to the emergence and rapid spread of such technologies. Captcha has been developed to prevent such malicious activities, but today's simple image-based Captcha applications can be solved by OCR methods. In this study, an image based CAPTCHA was developed. In this captcha method, which is created with high noise value, it is very difficult to solve with OCR because the noise does not decrease even when the picture is converted to binary. In addition, the existing captcha approaches were examined and a security comparison was given.