Title:  A Digital Image Watermarking Technique Based On Multiple Transform Methods

Meekyeong Kim

Location: YR-459

Date: May 5, 2017

Time: 12:00 pm

Abstract

Copyright protection of digital content has become one of important issues in digital content marketplace. Digital watermarking may be used as an effective method for identifying the copyright ownership of digital content such as images, audio, and video against unauthorized use and distribution. The basic idea of digital watermarking is embedding or hiding information related to the copyright and ownership of the digital content imperceptibly as a watermark into the content.  In this dissertation, our aim is to propose an efficient digital image watermarking algorithm based on the multiple transform methods which combine the benefits of the frequency decomposition ability of discrete wavelet transform (DWT) and the unpredictable random distribution property of discrete fractional random transform (DFRNT) using 2D barcode to improve the robustness, imperceptibility, security, and extraction performance of watermarking algorithm. Experimental results present that our proposed algorithm has improved the extraction performance by accurately extracting the hidden information in the 2D barcode from the detected watermark.  Also, combining the dual transform method, DWT and DFRNT, has improved the security and robustness of the watermarking algorithm against basic image signal processing and common attacks.