Blind image watermark analysis using higher order statistics

Over the past two decades, great efforts have been made to develop digital watermarking techniques for copyright protection or content verification of digital products. Currently, most of watermark detection methods are designed based on the corresponding specific watermark embedding procedures, which is called symmetrical digital watermarking. In this paper, we propose a general steganalysis scheme to detect the existence of watermark in an image no matter what kind of watermark embedding scheme is used. In the proposed method, discrete wavelet transform (DWT) and higher order local autocorrelations are used to differentiate the intrinsic statistical characteristics between non-watermarked images and watermarked images. Support vector machines (SVMs) are then used to classify these characteristics. Numerical experimental results show that the proposed scheme describes the intrinsic statistical characteristics and the proposed steganalysis method is effective.