Abstract
The emergence of digital multimedia and the proliferation of its use have raised major concerns about the protection of intellectual property. In response to these concerns, digital watermarks have emerged as a possible solution for protecting the intellectual property of digital content. In this paper, we derive estimates of data-hiding capacity of balanced multiwavelet transforms. This class of transforms, relatively new, has useful properties for image processing applications as shown in this paper. Furthermore, we will investigate the relevance of two closely related statistical models, developed for scalar wavelets, for modeling the statistics of balanced multiwavelet transform coefficients. Finally, we will present performance results of a spread spectrum watermarking system that is based on this new transform.