Digital Image Watermarking: Novel Algorithms

Watermarking is widely acknowledged as a trustworthy mechanism to provide security to the multimedia applications. In this book, five different novel watermarking algorithms are described to address the problem of digital multimedia (images) security. Different approaches are implemented for gray scale and multichannel color image watermarking such as Discrete Wavelet Transform with subband threshold computing, Complex Number Principle, Discrete Cosine Transforms and Lifting Wavelet Transform with Singular Value Decomposition. These algorithms exhibit different characteristics such as adaptive basis, spread spectrum, spatial domain, frequency domain, blind, semi-blind and multiple watermarks embedding in color images. The algorithms are evaluated using performance measuring parameters such as Peak Signal to Noise Ratio between original and watermarked image, Correlation Coefficient and Bit Error Rate between original and retrieved watermark. To evaluate performance, benchmark standard 'StirMark4.0' attacks are applied on watermarked images. The watermarking algorithms presented in this book are promising with better precision, good fidelity and higher robustness.