For a considerable amount of time, text-based CAPTCHA (Completely Automated Public Turing Test to Tell Computers and Humans Apart) systems have been the main line of security against web platform automated attacks. The effectiveness and usability of conventional CAPTCHA designs have encountered difficulties despite their widespread use, such as accessibility problems and susceptibility to sophisticated automated tactics. This study examines new developments in text-CAPTCHA methods to improve usability and security. The paper begins by surveying the landscape of text-CAPTCHA methods and examining their underlying principles and vulnerabilities. Subsequently, it delves into innovative approaches such as Transparent Decoy CAPTCHA, named TDC, which utilises contextual understanding to present challenges that are more resistant to automated solving algorithms. This study discusses the accessibility in CAPTCHA systems, presenting innovative remedies that guarantee inclusivity for users with impairments while yet upholding strong security against automated attacks. The text explores the significance of user experience in CAPTCHA design and reduces user frustration while maintaining security. The usefulness and feasibility of TDC techniques are evaluated through empirical assessments and case studies, considering criteria such as security aspects, usability and accessibility. An alphabetical dataset consisting only of capital letters has been developed for the TDC character recognition challenge, and it has obtained an accuracy of 96.1%. The findings offer valuable insights into CAPTCHA technology and provide direction for the creation of more secure and user-friendly systems to address the increasing challenges faced by automated threats on the web.