A New Secure Image Transmission Technique via Mosaic Images Using Genetic Algorithm

Abstract: A new secure image transmission method is proposed in which it transforms the secret image into secret-fragment-visible mosaic image. The size of the mosaic image is same as that of secret image. Mosaic image is created by dividing the secret image and target image into fragments of equal size and fitting these secret tile blocks into target blocks. For tile image hiding, a mapping sequence is generated using Genetic Algorithm (GA). This provides better clarity in the retrieved secret image. It also reduces the computational complexity. The quality of the original target image remains preserved while embedding the secret image. Therefore better security and robustness is assured. Embed these tile fragments into the target image based on the mapping sequence by genetic algorithm and permuted the sequence again by KBRP with a key. Color transformations are performed to make the mosaic image similar to the target image. After color transformation rotation is performed. Rotating each tile block into an optimal rotation angle with minimum root mean square error value with respect to its corresponding target blocks. For the recovery of the secret image from the mosaic image embed relevant information into the created mosaic image. Overflows/underflows in the transformed color values can also be handled by using this method. By using the same key and the mapping sequence, the secret image can be recovered.

Keywords: Image hiding, Mosaic image, Genetic Algorithm, Key Based Random Permutation, Color Transformation.