期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2015
卷号：80
期号：2
出版社：Journal of Theoretical and Applied
摘要：Steganography is an art of disguising the fact that communication is going on by concealing information in other information. In general, the communication carrier can be files in many formats; however, digital images are the most common due to their frequent use on the Internet. This paper introduces an improvement on the standard least significant bit (LSB)-based image steganography technique and proposes the bit inversion method that improves the stego-images quality in 24-bit colour image. A stego-image is the outcome of an image (usually called the cover image), after a secret message is hidden in it. In this technique, the LSB�s of some pixels of the cover image are inverted, when inputs of specific patterns of some bits related to the pixels are found. In this way, less number of pixels is modified in comparison to the standard LSB method. Our focus is to obtain a high value ratio of the Peak Signal-to-Noise (PSNR) of the stego-image, to make sure that both stego-image and the original image are difficult to discern by human eyes. The proposed bit inversion method starts with the last LSBs of both green and blue colour planes that will be replaced by the first and the second most significant bits (MSB) of the secret image. The proposed method introduces two additional levels of security to the standard LSB steganography. The first level is that because only the green and blue colours are used, instead of three colors red, green, and blue in the standard LSB, the red colour will act as noise data, and thus increases the complexity of an attacker, when he/she tries to retrieve the secret message. The second level exploits the new bit inversion technique that reverses the bits of the image pixels after applying the standard LSB. Experiments have been conducted using a collection of standard images to evaluate the proposed technique, which give the Peak Signal-to-Noise Ratio (PSNR) values of 72, 61, and 70 for Lena.jpg, Babbon.jpg, and Pepper.jpg respectively. From the experiment, we also observed that by using the bit inverse technique, less numbers of pixels are modified compared with the standard LSB method.
