Chaotic image encryption algorithm based on frequency domain scrambling
Musheer Ahmad, Omar Farooq, Jonathan Blackledge · 2010
This letter proposes an image encryption algorithm using scrambling by exploiting the features of chaotic maps suited for cryptography. The proposed algorithm performs scrambling and masking of image pixels using states of chaotic maps in a secure manner. A novel multi-level blocks scrambling scheme is employed in frequency domain to overcome the drawbacks of spatial domain scrambling. At each level of scrambling, the non-overlapping blocks of frequency coe cients are shu ed using random control parameters. To improve the statistical characteristics from cryptographic viewpoint, mixing operation is done using keystream extracted from one-dimensional chaotic map and the plain-image. As a result, the algorithm resists the chosen-plaintext/chosen-ciphertext/known-plaintext attacks, as mixing depends on plain-image. Moreover, the experimental results of keyspace, sensitivity to secret keys, entropies, gray-level distribution and maximum deviation confirm that the proposed algorithm provides high security and can be applied to protect digital images over communication channels.