Work place: Department of Mathematics, Shantou University, Shantou, Guangdong, 515063, P. R. China
E-mail:
Website:
Research Interests:
Biography
Junqin Zhao, female, is a master degree candidate at department of mathematics in Shantou University.
By Weichuang Guo Junqin Zhao Ruisong Ye
DOI: https://doi.org/10.5815/ijigsp.2014.11.07, Pub. Date: 8 Oct. 2014
A great many chaos-based image encryption schemes have been proposed in the past decades. Most of them use the permutation-diffusion architecture in pixel level, which has been proved insecure enough as they are not dependent on plain-images and so cannot resist chosen/known plain-image attack usually. In this paper, we propose a novel image encryption scheme comprising of one permutation process and one diffusion process. In the permutation process, the image sized is expanded to one sized by dividing the plain-image into two parts: one consisting of the higher 4bits and one consisting of the lower 4bits. The permutation operations are done row-by-row and column-by-column to increase the speed of permutation process. The chaotic cat map is utilized to generate chaotic sequences, which are quantized to shuffle the expanded image. The chaotic sequence for permutation process is dependent on plain-image and cipher keys, resulting in good key sensitivity and plain-image sensitivity. To achieve more avalanche effect and larger key space, a chaotic Bernoulli shift map based bilateral (i.e., horizontal and vertical) diffusion function is applied as well. The security and performance of the proposed image encryption have been analyzed, including histograms, correlation coefficients, information entropy, key sensitivity analysis, key space analysis, differential analysis, encryption rate analysis etc. All the experimental results suggest that the proposed image encryption scheme is robust and secure and can be used for secure image and video communication applications.
[...] Read more.Subscribe to receive issue release notifications and newsletters from MECS Press journals