Blind Image Watermark Analysis Using Feature Fusion and Neural Network Classifier

Abstract

Over the past two decades, great efforts have been made to develop digital watermarking techniques for multimedia copyright protection and authentication. However, most of watermark detection methods are designed based on the corresponding specific watermark embedding procedures. In this paper, we propose a general blind watermarking analysis scheme to recognize whether images are watermarked no matter what kind of watermark embedding schemes are used. In the proposed method, multiscale feature fusion are used to construct statistical characteristics between non-watermarked images and watermarked images. Then, RBF neural networks are used to classify these characteristics. Numerical simulations show that the proposed scheme describes intrinsic statistical characteristics and the proposed blind watermark analysis method is effective.