Context:

With the increasing volumes of information generated in real time, novel mechanisms area needed to ensure security so as to prevent access to unauthorized people. The conventional encryption methods are not appropriate for images, because they are prone to statistical attacks due to the strong correlation between adjacent pixels and the analysis of color gamut histograms, which can help to identify them within the image; with this aim in mind, in this paper an algorithm for image encryption using chaotic attractors is proposed.

Method:

Chaotic synchronization is used for the key management, the diffusion stage is made by means of ergodic sequences generated from the numerical solution of the Lorenz’s attractor and the permutation stage is accomplished with the wave line technique.

Results:

The proposed algorithm was tested using a set of gray-scale images obtaining suitable performance in segurity and speed, the pixel correlation is almost null and the entropy is similar to that presented in recent works with the same approach.

Conclusions:

Chaotic methods are an alternative to improve the security levels in the cryptography of images due to their properties of unpredictability and sensitivity to the initial conditions. For future work the approach presented could be applied to the encryption of color images and using different chaotic attractors.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )