文献类型：期刊文献

英文题名：A novel multi-image visually meaningful encryption algorithm based on compressive sensing and Schur decomposition

作者：Ye, Guodong[1];Pan, Chen[1];Dong, Youxia[1];Jiao, Kaixin[1];Huang, Xiaoling[1]

机构：[1]Guangdong Ocean Univ, Fac Math & Comp Sci, Zhanjiang 524088, Peoples R China

年份：2021

卷号：32

期号：2

外文期刊名：TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES

收录：SCI-EXPANDED(收录号：WOS:000557716500001)、、EI(收录号：20203309033943)、Scopus(收录号：2-s2.0-85089184031)、WOS

基金：The authors would like to thank the anonymous reviewers for their valuable comments and the editor's helpful suggestions. This work was supported in part by the National Natural Science Foundations of China (No. 61702116, No. 61972103), the Natural Science Foundation of Guangdong Province of China (No. 2019A1515011361), the Postgraduate Education Innovation Project of Guangdong Ocean University (No. 201925), and the Project of Enhancing School with Innovation of Guangdong Ocean University of China (No. Q18306).

语种：英文

外文关键词：Cryptography - Image compression - Discrete cosine transforms - Image coding - Matrix algebra - Discrete wavelet transforms

外文摘要：Image encryption algorithm is an important telecommunication technology for protecting personal security. In this article, we propose a novel multi-image visual encryption algorithm based on compressive sensing (CS) and Schur decomposition. First, multiple images are sparsed by a discrete wavelet transform, and the compressed images are obtained by zigzag confusion and CS. Second, these compressed images are combined according to random sequences and scrambled again to obtain the secret image. Third, the carrier image is subjected to a two-level lifting wavelet transform where the low-frequency and high-frequency subbands are selected to perform a discrete cosine transform (DCT). Fourth, Schur decomposition is employed on the secret image to calculate an upper triangular matrix and a unitary matrix (orthogonal matrix) to be embedded into the DCT coefficients of the previous low-frequency and high-frequency subbands, respectively. With an inverse transformation, the final visually meaningful encrypted image is achieved. Here, the initial position of the zigzag confusion is related to the plain image, which can effectively resist known-plaintext and chosen-plaintext attacks. The simulation results demonstrate that the proposed algorithm has high security and good imperceptibility.