BIOMETRIC WATERMARKING USING ML AND DEEP LEARNING APPROACH

Abstract

In a generation where record safety is paramount, this paper affords a complete biometric watermarking solution that enhances document authentication through the fusion of iris and fingerprint snap shots. Utilizing the advanced Encryption standard (AES) algorithm, the proposed machine guarantees the cozy encryption of biometric statistics, developing a robust virtual watermark. This watermark is then embedded in documents to offer an additional layer of safety against unauthorized get admission to and forgery. A Convolutional Neural community (CNN) is hired for authenticating the biometric watermark, correctly distinguishing among true and counterfeit biometric capabilities. Furthermore, the gadget includes actual-time notifications and an adaptive learning mechanism, making sure continuous improvement and responsiveness to rising protection threats. With the aid of integrating these superior technologies, the proposed system offers a scalable solution for securing sensitive files and improving trust in virtual transactions