Blind digital watermarking for cartoon and map images

Cartoon/map images are synthetic graphics without complicated color and texture variation, which makes the embedding of invisible and robust digital watermarks difficult. In this research, we propose a wavelet-based, threshold-adaptive watermarking scheme (TAWS) which can embed invisible robust watermarks into various kinds of graphical images. TAWS selects significant subbands and inserts watermarks in selected significant coefficients. The inserted watermarks are adaptively scaled by different threshold values to maintain the perceptual integrity of watermarked images and achieve robustness against compression and signal processing attacks. Another major contribution of this work is that the cast watermark is retrieved without the knowledge of the original image. The so-called blind watermark retrieval technique is very useful in managing a large cartoon, trademark and digital map databases. Finally, a company logo that clearly identifies the copyright information can be embedded in cartoon and map images without serious perceptual loss. Experimental results are given to demonstrate the superior performance of TAWS.