组合NMF和PCA的图像哈希方法

Image Hashing method via combination of NMF and PCA

图像哈希函数在内容认证、数据库搜索和数字水印中有广泛的应用。本文提出了一种新颖的基于非负矩阵分解(NMF)和主成分分析(PCA)的感知哈希方法,通过对哈希生成两阶段框架的详细分析,NMF被设计用来捕获图像的局部信息,而PCA用来捕获全局信息和信息压缩,图像之间的相似程度通过哈希的归一化相关值来确定。该方法在两阶段中都采用了伪随机处理,增强了算法安全性。实验表明建议的方法对JPEG压缩、图像滤波等内容保持操作具有较好的稳健性,同时有较好的分类能力。

Image hash function has extensive applications in content authentication, database search, and water- marking. This paper presents a novel perceptual hashing method based on non-negative matrix factorizations （NMF） and principal component analysis （PCA）. Through the detailed analysis of two-stage framework of generating Hash, NMF is used to capture the local features of image, and PCA is used to capture whole information and information compression. The similarity of images is determined by Hash normalization correlation. Two stages of the method use pseudorandom processing that enhances security of the algorithm. Test results indicate that the proposed method is robust against content-preserving modifications such as JPEG compression, and at the same time, is capable of excellent classification.