High Secure Image Steganography based on Hopfield Chaotic Neural Network and Wavelet Transforms

Steganography is an art of hiding the information without any change in the external appearance of the cover object. Cryptography is a technique to make the information unreadable for unauthorized users. Making the data unreadable and hiding it, will make the data highly secure. Image steganography allows the user to hide a large amount of data inside an image. On transmission side Steganography is performed by choosing a Cover-Image then hiding the text within the image. On receiving side the secret text is extracted from the stego image. In this paper, High Secure steganography algorithm is proposed. This process contains three stages. In the first stage, the text is encrypted by using a traditional encryption method i.e Caeser method. In the second stage the cipher text is again encrypted by using the chaotic neural network and in the third stage the resulting encrypted text is embedded inside the image using DWT. High security can be achieved by encrypting the text using Chaotic Neural Network. The binary sequence of the encrypted text created by Chaotic neural network is unpredictable making it highly secure. The Proposed algorithm is tested against different gray scale images considering PSNR, MSE and SSIM for evaluation. It is observed that the security is increased with acceptable PSNR compared to other methods.