Improved Performance by Parametrizing Wavelet Filters for Digital Image Watermarking

In recent years, watermarking has become an attractive field in areas like copyright protection, image authentication and biomedical engineering. Many literatures have reported about discrete wavelet transform (DWT) watermarking techniques for data security. However, DWT based watermarking schemes are found to be less robust against image processing attacks. In this paper, an attempt is made to develop a scheme based on Redundant Discrete Wavelet Transform (RDWT). The robustness and security of the proposed RDWT scheme is improved by parametrizing the scaling and wavelet filters by Pollen’s parametrizing technique. An Independent Component Analysis (ICA) based detector is also applied, which extracts watermark directly in spatial domain rather than in transform domain. The unique feature of ICA is that it does not require any transformation during extraction. The results reveal that the proposed scheme produces high peak signal to noise ratio (PSNR) values and similarity measure under various image processing attacks.