International Journal of Scientific Research and Engineering Development

Phishing attacks continue to rank among the most pervasive threats in modern cybersecurity, wherein adversaries construct visually deceptive websites to harvest user credentials and sensitive data. Conventional defenses—including URL blacklisting, SSL certificate verification, and machine-learning-based classifiers—are predominantly reactive and consistently fail against zero-day phishing domains and cloned branding assets. This paper introduces a proactive anti-phishing framework based on Dynamic Session-Based Visual Cryptography (DS-VC) integrated with Least Significant Bit (LSB) Steganography. A secret authentication image is split into two shares using a (2,2) Visual Cryptography scheme: Share-1 resides on the client device, while Share-2 is generated dynamically per session by binding a SHA-256 hash of the session token, server domain, and a timewindow counter, then invisibly embedded within the website banner using LSB steganography. On each login visit, the client extracts Share-2 from the banner, overlays it with Share-1, and verifies reconstruction of the secret. A phishing site that copies the banner cannot produce a valid session-specific Share-2; reconstruction yields noise and the user is immediately alerted. A proof-of-concept prototype implemented with Python, Flask, NumPy, and Pillow demonstrates end-to-end verification in under 150 ms, confirming the feasibility of the proposed approach.

Omkar Abaso Bawdekar,"An Anti-Phishing Framework Using Dynamic Session-Based Visual Cryptography and Steganography" International Journal of Scientific Research and Engineering Development, V9(2): Page(900-904) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.