A new approach in Steganography of digital images using saliency map algorithm and Pixel Neighbors

• Published in: International journal of computer science and information security Vol. 14; no. 7; p. 566
• Main Authors: Ghasemi, Meysam, Emadi, Mehran
• Format: Journal Article
• Language: English
• Published: Pittsburgh L J S Publishing 01.07.2016
• Subjects: Accuracy; Algorithms; Channels; Computer information security; Greek civilization; Media; Methods; Pixels; Policies; Steganography
• ISSN: 1947-5500

In current years, called the age of communication, communication can be different from a simple interaction. Sometimes due to security issues, the information that is going to be communicated between a sender and a receiver should not have its real appearance. This is where the discussion of steganography arises. Steganography includes a large set of communication methods that hide the presence of the message. Some of these methods include micro points, digital signature, and steganography channels. Despite great progress in methods available, the carrier file content has not received much attention. In fact, all the existing algorithms work based on its designer's policy, not the file content. Thus, the purpose of this study is to focus on the content of the media file. In the proposed procedure, after selecting the media image, the image prominences will be selected by AIM algorithm and accurately separated from the surrounding area using active contour. Next, the input chosen for steganography is embedded in the new image without the prominent part, applying pixel neighbors method and LSB. In this study, color and black and white images will be used for steganography employing stated method. Based on the results of the tests, the proposed method has an average of 75 for PSNR peak signal noise rate that in some approaches has more than 15 percent improvement over other methods.