An Enhanced Captcha System Based on Variable-Length Accented Character

Completely Automated Public Turing Test to Tell Humans and Computers Apart (CAPTCHA) is aimed at distinguishing humans from computers using text, images, audio, or video. Text-based CAPTCHAs are the most common but face vulnerabilities due to their limited use of Latin characters. NaijaCAPTCHA system was developed to generate fixed-length five-character codes by integrating accented characters into text-based CAPTCHA codes. It was enhanced to yield variable-length accented character-based CAPTCHA codes, utilizing Latin and accented characters to bolster the security of online transactions. The design was implemented using JavaScript, PHP, HTML, and CSS. The Enhanced NaijaCAPTCHA (E-NaijaCAPTCHA) comprises four modules: CAPTCHA Generator, Obfuscator, Display Unit, and Database. The generator employed the Gimpy algorithm to create codes of varying lengths (4 to 7 characters) with two accented characters. The Obfuscator manipulated the code's appearance through colour, text distortion, background noise, and skewing. User presentation and authentication occurred in the display unit, verifying code correctness against the obfuscated value. The evaluation of the E-NaijaCAPTCHA included usability testing with 51 diverse participants, each with 10 CAPTCHA challenges, yielding 510 CAPTCHA codes attempted, which measured user-friendliness and solving times, with security testing using the Tesseract OCR engine to assess the system's resistance to automated attacks. The bot recorded a 0.0% success rate and accuracy in decoding the CAPTCHA. The developed system, equipped with various hardware and software tools, proved user-friendly in usability tests involving diverse participants. Security evaluations confirmed its robustness against automated attacks, as the Tesseract OCR engine struggled to accurately decode the CAPTCHA codes.