A New Hiding Method based on DCT and Arnold Transform

Abstract

In this paper a new steganography method is suggested to be applied using Discrete Cosine Transform (DCT), Arnold Transform and block shift technique. DCT is calculated on blocks of independent pixels of cover image. Arnold Transform and block shift are used as a pre-treatment for secret image before hiding process. While block shift alone is used as a pre-treatment for cover image pre-treatment for cover image before hiding process. Some of the well-known fidelity measures like Peak Signal to Noise Ratio, SSIM, and Accuracy Rate have been used to assess the quality of the modified image and extracted image. The results show that the stego-image is closed related to the original cover image, when the Peak Signal to Noise Ratio is up to 55 dB, SSIM is close to one. The recovered secret image is extracted with a good quality if the stego-image has no attack. Attacks have been applied to the proposed hiding method in order to measure the robustness like (add noise and cropping).This system overcomes the demerits of using single level of hiding. That is either using cryptography or steganography. Moreover, one more thing to add is it requires only the computation time of single level hiding, because visual cryptography requires no computation to decrypt the information.