Memon, Qurban A. (2015) JPEG2000 compatible neural network based cipher. In: 5th International Conference on Computing and Informatics (ICOCI) 2015, 11-13 August 2015, Istanbul, Turkey.

Preview

PDF Download (741kB) | Preview

Abstract

In this paper, an efficient encryption technique is proposed, especially for JPEG2000 compatible images.The technique uses a multilayer neural network to generate a pseudo-random sequence for transforming wavelet subbands into cipher subbands.The neural network generator takes 64 bit key as a startup seed with additional 64 bit key for initial weights and biases.At each layer, output is calculated by several iterations to increase the complexity of the pseudorandom sequence generation.In order to examine effectiveness of this approach, various tests including correlation, histogram, key space etc. are conducted on test images, and the results demonstrate the robustness of the proposed approach.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	ISBN No: 978-967-0910-02-4 Jointly organized by: Universiti Utara Malaysia & Istanbul Zaim University
Uncontrolled Keywords:	JPEG2000 compatible cipher; neural network based random sequence; block cipher; encryption
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	School of Computing
Depositing User:	Mrs. Norazmilah Yaakub
Date Deposited:	01 Oct 2015 07:56
Last Modified:	27 Apr 2016 06:46
URI:	https://repo.uum.edu.my/id/eprint/15601

Actions (login required)

View Item