Improving the security of image watermarking based on the combination of discrete wavelet transform, singular value decomposition and discrete cosine conversion methods
Subject Areas : Multimedia Processing, Communications Systems, Intelligent Systemshossein nematizadeh 1 , Mohammad Tahghighi Sharabian 2
1 - Faculty of Computer Engineering, Raja University of Qazvin, Qazvin, Iran
2 - Faculty of Electrical and Computer Engineering, Zanjan Islamic Azad University of Zanjan , Zanjan Iran
Keywords: SVD, Watermarks, Combination, DWT, DCT,
Abstract :
Introduction: One of the methods of ensuring information security is the use of encryption methods, but these methods are not able to hide the existence of information. In order to hide information, algorithms called Steganography have been created. Steganography is the method of hiding important data in a file or ordinary message, in order to prevent detection by others. This secret information is extracted to the original state at the destination. The use of Steganography can be combined with encryption as an additional step to hide or protect data. Steganography can be used to hide almost any type of digital content, including text, images, video, or audio content. Often the content to be hidden is encrypted before the Steganography process to provide an extra layer of protection. Steganography is more focused on keeping information hidden, while cryptography is more involved with the issue of ensuring access to information. According to the type of cover signal and also the insertion algorithm, different steganography methods have been presented, which in terms of hiding capacity and security are different. This issue has also led to reproduction, redistribution, and illegal digital media. Since copying and changing digital data has become easy and undetectable, its investigation is also of great importance.Method: In this article, a new method for non-blind image hiding that is resistant to affine transformation and normal image manipulation is used. The proposed Steganography method is based on additive discrete wavelet transform and singular value decomposition. After using RWDT for overlay and hidden images, we use SVD for their LL subbands. Then we modify the singular values of the overlay image using the singular values of the visual masker.Result: Analytical studies on the extracted watermarked image show that the Steganography method is capable and resistant against Salt & Pepper attacks with a scale of 0.1 out of 1.0 and Gaussian attack with a rate of 0.01 and the watermarked image is well recovered. Discussion: One of the most common cryptography techniques in the field of transformation is the modification of the coefficients obtained from the Singular Value Decomposition (SVD) of the image mask. The proposed algorithm has a good performance against rotation and cutting attacks, also the Steganography based on multiple SVD has performed poorly against these two attacks.
_||_