Authentication System Combining Optical Character Recognition and Error Level Analysis
Swapnendu Chakrabarti, H. Singh, Deepanshu Yadav, Pragati Rana · 2024
With concerns regarding privacy at an all-time high, there is a need for research into security measures, starting off with authentication measures. The escalating threat of cyberattacks and data breaches necessitates robust authentication methods across diverse sectors, from small businesses to large enterprises. These attacks not only compromise data confidentiality but also disrupt operations, leading to severe consequences and financial burdens for organizations. The interconnected nature of modern IT infrastructures exacerbates vulnerabilities, emphasizing the urgency of addressing multifaceted challenges such as security risks, data privacy, operational disruption, technology challenges, and financial impacts. Organizations not only face immediate harm but also encounter legal consequences, with regulatory bodies imposing strict compliance requirements and significant penalties for non-compliance. A thorough literature survey has been conducted to explore feasibility of the combination of Optical Character Recognition (OCR) and Error Level Analysis (ELA) techniques to improvise upon the existing methods of facial recognition. The intent of this research work arises due to the limited research into the merging of multiple authentication techniques while improving upon the same. This combination is tried and tested upon predefined identification documents with facial recognition to prove its functionality.