Abstract: Nowadays, social network multimedia websites are great convenient for the content distribution while it brings some security problems, such as privacy, piracy, disclosure of sensitive contents and so on. Aiming at copyright protection, the copy detection technology of multimedia contents becomes a hot topic. Previous block based copy detecting methods are sensitive to geometry attacks, an image copy detection method based on eigen value feature and the transfer matrix feature of the entropy matrix are used for copy matching. SVM technology is used to verify the fusion result. The experiments show that the detection rate is improved while it can achieve some robustness to noise, compression, rotation and scale etc.

Key words: image entropy, feature fusion, copy detection

摘要： 目前流行的社会网络为媒体分发带来极大便利，但同时带来一系列的安全隐患，例如隐私、盗版、敏感内容泄露等等。为追踪非法拷贝，针对版权保护的多媒体内容拷贝检测技术是一个研究热点。传统的拷贝检测方法难以做到分块结构和几何鲁棒性的统一，提出了一种基于熵矩阵的特征值特征和变换矩阵特征的融合方法，用于拷贝的特征匹配，利用SVM技术对融合效果进行了验证。实验证明，除改善了检测率外，对噪声、压缩、旋转、缩放等处理有一定的鲁棒性。

关键词: 图像熵, 特征融合, 拷贝检测