【摘要】 提出了一种基于支持向量机的半脆弱数字音频水印算法。该算法不但可以容忍一定程度的常规信号处理操作,而且可以检测出对多媒体数据的恶意篡改并定位篡改区域。具有以下特点:充分利用多级置乱技术及支持向量机技术准确定位篡改区域,在篡改定位过程中,无需原始水印参与;利用水印的归一化相关系数,可进行完整性认证并给出篡改程度的度量。实验结果表明,该算法计算简单,有很好的抗裁剪性能,同时对篡改的定位也很精确。更多还原

【Abstract】 A novel semi-fragile digital audio watermarking scheme based on support vector machine is presented. The proposed algorithm can tolerate general signal processing, detect any spiteful tamper on marked multimedia data and locate spatial regions. The features of the proposed algorithm are as follows: It take full advantage of multilevel scrambling and support vector machine to locate tampered region. The algorithm can locate tampered regions without the help from the original watermark. By using the Normalized Cross-Correlation, the proposed scheme can not only authenticate the integrity of the audio signal but also evaluate the temper degree. The experimental results show that the proposed algorithm has the advantages such as simple computation complexity, good robustness against shearing attack, and accurate location for tamper.更多还原