Image encryption in code based compression algorithms based on multilevel image structure model (original) (raw)

2017 9th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), 2017

Abstract

Image compression is a widely adopted technique used for effective image storage and transmission over open communication channels in cyber-physical systems. Standard cryptographic algorithms are usually used to reach this goal. Therefore, in order to organize effective and secure storage of images it is required to follow two independent and sequential procedures — compression and encryption. In the scenario of interest, it is needed to do compression and encryption transformations in reverse order to restore the original image, i.e. it is necessary to have a so-called “code book” similarly as for encryption and decryption to have a secret key. An effective way of combining these procedures for digital images is proposed in this manuscript. This research is mainly focused on the compression methods that consider significance of the initial multimedia object (for example image) different parts to increase the quality of resulting (decompressed) image. One of the most effective approaches for this task is to utilize error-correcting codes (ECC) that allow to limit the number of resulting errors (distortion) as well as to ensure the value of resulting compression ratio. Application of such codes enable to distribute errors that are added during the processing procedure according to predefined significance of the initial multimedia object elements. The approach that is based on weighted Hamming metric that makes it possible to guaranty the limitation of maximum error number (distortions) that takes into consideration predefined significance of the image zones is represented as an example. The way to use subclass of Goppa codes perfect in weighted Hamming metric when Goppa polynomials are used as a secret key is presented as well. The additional effect of such encrypted compression methods is auto-watermarking of the resulting image.

Natalia Voloshina hasn't uploaded this paper.

Let Natalia know you want this paper to be uploaded.

Ask for this paper to be uploaded.