Image Authentication Technique Based on Digital Watermarking using Clustering

Abstract

The digital images are being widely used in computer applications. These can be modified without any limit by using different image processing software. As a result the concerned person can suffer from financial and public damage. To overcome this, a reliable digital image authentication scheme is required to be developed. Digital watermarking can be used as a solution to this problem. It is an efficient way to guarantee authenticity and integrity of digital images. A watermark is embedded in to the digital image which uniquely identifies the ownership and provides the security to it and can be easily extracted. In this paper, an image authentication scheme based on digital watermarking using clustering is proposed. We have used a combination of DCT and clustering together with image scrambling technique. DCT is used for watermark embedding process and Fuzzy C-Means is used for clustering. In order to improve the security of the proposed Digital Watermarking scheme, Arnold Transform is used that scrambles watermark before embedding it in to the cover image. Performance of the proposed work is evaluated based on Peak Signal-to-Noise Ratio (PSNR) and Normalized Correlation (NC).