A Framework for Modi-Lipi Script Based On CAPTCHA
Abstract
This paper presents a novel framework for generating CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) utilising the Modi Lipi script. Modi Lipi, a historical script primarily used for writing Marathi, presents a unique challenge for automated recognition due to its cursive and often complex conjunct characters. This inherent complexity is leveraged to create robust security measures. The proposed framework encompasses three main stages: Modi Lipi Character Generation, which dynamically creates diverse and warped script images; CAPTCHA Design and Integration, focusing on usability and security parameters for web applications; and Evaluation and Validation, where the generated CAPTCHAs are tested against both human users and state-of-the-art optical character recognition (OCR) and machine learning attacks. Our preliminary results indicate that Modi Lipi-based CAPTCHAs achieve a significantly lower success rate against automated attacks than standard Latin- or Devanagari-script CAPTCHAs, while maintaining a high degree of solvability for human users familiar with the script. This work contributes a culturally significant and effective method for enhancing digital security while simultaneously aiding in the preservation and popularisation of the Modi Lipi script.
Fields of Study & Subjects
User Authentication and Security Systems
Spam and Phishing Detection
Adversarial Robustness in Machine Learning
Publication details
- Journal / source
- International Journal on Advanced Electrical and Computer Engineering
- Volume
- 15
- Issue
- 1S
- Pages
- 388–395
- Publication date
- 19 January 2026
- Language
- EN