DCT based Fuzzy Image Watermarking

Durgansh Sharma; Manish Prateek; Tanushyam Chattopadhyay

Durgansh Sharma 1Jaipuria Institute of Management, Noida, India;
Manish Prateek 2Centre for Information Technology, College of Engineering Studies, UPES, Dehradun, India;
Tanushyam Chattopadhyay 3R&D, Innovation Lab, Tata Consultancy Services, Kolkata, India;

Keywords: Fuzzy Logic, Fuzzy Inference System, Human Visual System (HVS), Robust Image Watermarking,

Abstract

Image watermarking implants an ownership parameter in a digital image or video. This paper works on authentication of the image
file. Robust image watermarking technique is used to model the Human Visual System (HVS) using Fuzzy Logic. The proposed Fuzzy
Logic System is trained using inference rules considering three sensitivities of HVS as Brightness, Contrast and Edge Sensitivity. The
Fuzzy network uses the image captured in realtime and computes block wise for producing a single output weighting factor used to
embed unique identification numbers generated from the confidential data as watermark, which is for authorization and verification
of the original image. The robustness of the watermark embedded is checked by Stirmark image processing attacks. Recovered
watermark’s computed value using SIM(X, X*) parameter for the image verified it as good watermark recovery process.