论文部分内容阅读
目前的数字水印算法,主要集中在针对图像和视频,对音频、语音数据产品的数字水印算法研究仍然较少,需要提出更多的有效的针对音频、语音数据的数字水印算法。该文以二值图像作为数字水印信息,以音频为嵌入对象,在离散傅立叶变换域的系数上嵌入水印。在离散余弦变换域,采用扩频技术通过对变换域系数进行量化来嵌入水印信息,在水印的提取过程中不需要原始音频信号的参与。
At present, digital watermarking algorithms are mainly focused on image and video, and there is still little research on digital watermarking algorithms for audio and voice data products. Therefore, more effective digital watermarking algorithms for audio and voice data need to be proposed. This paper takes binary image as digital watermark information and audio as embedded object, and embeds watermark in the coefficients of discrete Fourier transform domain. In the discrete cosine transform domain, the watermark information is embedded by quantizing the transform domain coefficients by using the spread spectrum technique. The original audio signal is not required to be involved in the watermark extraction process.