文章基本信息

标题：Audio Zero-Watermarking Based on Discrete Hartley Transform and Non-Negative Matrix Factorization
本地全文：下载
作者：Pranab Kumar Dhar ; Emtiaj Hasan ; Tetsuya Shimamura 等
期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2017
卷号：17
期号：8
页码：11-18
出版社：International Journal of Computer Science and Network Security
摘要：This paper proposes a zero-watermarking method based on discrete Hartley transform (DHT) and non-negative matrix factorization (NMF) for audio signal. Initially, the DHT is applied to the original audio signal to obtain the DHT coefficients. Then, the DHT coefficients are divided into an arbitrary number of frames. The coefficients of each frame are arranged into a square matrix and NMF is applied to each matrix. The summation of the coefficients of each weight matrix is calculated and the fractional part of this summation is selected. The fractional part is converted into integer value and modulation operation is performed to generate a binary pattern. XOR operation is applied between the watermark data and generated binary pattern. This operation generates a secret key, which is used later to extract the watermark. The proposed method does not embed any watermark into original audio, indicating that it is imperceptible. Experimental results show that the proposed method is robust against various attacks such as noise addition, re-sampling, re-quantization, low-pass filtering, echo, MP3 compression and so on. Moreover, it shows superior performance than the state-of-the-art watermarking methods in terms of robustness.
关键词：Copyright protection Discrete Hartley transform Non-negative matrix factorization Zero watermarking