Key points are not available for this paper at this time.
The goal of the research is to develop a security breach for text and image-based Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) using Deep Learning. CAPTCHA is a task used in computers to differentiate humans from machines by employing a short code. If a short code can solve that challenge and the human form succeeds in executing the submitted application, then the main motive is achieved. Therefore, the main goal is to overcome the limitations of picture and text captcha and make an automatic CAPTCHA breaker that can easily handle text and picture-based captcha without any human intervention. Additionally, the research aims to break down various types of captchas, including numeric and letter captures as well as hollow, open, and closed captures. In the context of picture-based captcha, the system gives a fixed number of pictures to form a composite unit picture. The user can select the picture that he or she wants to illustrate by clicking on the image.
Puneet et al. (Tue,) studied this question.