International Journal of Signal Processing, Image Processing and Pattern Recognition Vol.10, No. 1 (2017), pp.35-48 http://dx.doi.org/10.14257/ijsip.2017.10.1.05 ISSN: 2005-4254 IJSIP Copyright ⓒ 2017 SERSC Variant of LSB Steganography Algorithm for Hiding Information in RGB Images Ashis Kumar Mandal l and M N M Kahar 2 1 Department of Computer Science and Engineering, Hajee Mohammad Danesh Science and Technology University, Dinajpur, Bangladesh 2 Faculty of Computer Systems & Software Engineering, University Malaysia Pahang, Pahang, Malaysia 1* ashis@hstu.ac.bd, 1 ashis.62@gmail.com, 2 mnizam@ump.edu.my Abstract Steganography has been emerged as a prominent technique for communicating secret information in a proper multimedia career. In image steganography, least significant bit (LSB) substitution method is popular but vulnerable to numerous security attacks. Hence, it is always imperative to make the steganographic technique more robust and secure. This paper presents a new approach to hiding text messages within red-green-blue (RGB) images based on LSB technique. Unlike the standard LSB technique, where message bits are directly embedded in the LSB positions, our proposed algorithm encodes these message bits before embedding in the LSB positions. This embedding and encoding process are done on the basis of most significant bit (MSB) values of RGB and the concept of odd and even parities of those pixels respectively. The main goal of the presented method is to improve the robustness and the image quality of the stego-image. The performance of stego-image quality is measured by mean square of error (MSE), peak signal-to-noise ratio (PSNR), structural similarity measure (SSIM), bits ratio (BR), and histogram analysis. Experimental results and comparative studies have revealed that the proposed technique is able to conceal message effectively, with there being little difference between the cover image and the stego-image. Results also show that our proposed approach can embed reasonable amount of information. Keywords: Image steganography, LSB, MSB, odd and even parity, RGB image 1. Introduction Information hiding research domain has attracted many researchers in producing the reliable technique to ensure the protection of digital contents from numerous security attacks, especially in the online domain. Such techniques seen in the literature include cryptography, watermarking and steganography [1-2]. One of the popular approaches is the steganography, which is the art and science of hiding information by embedding messages within cover media such as text, image, sound or video files [3]. For example, in image steganography, the cover medium is the image whereby information is embedded by altering some nonessential components of that image. Usually, in image steganography, there are two ways of steganographic algorithms, transform domain and spatial domain. Steganography in image transform domain includes performing a series of changes to the cover image before hiding the information. Discrete cosine transformation (DCT), fourier transformation (FT) and discrete wavelet transformation (DWT) are some techniques used for this purpose [4]. On the other hand, steganography in the image spatial domain generally uses a direct least significant bit (LSB) replacement technique — encoding at the level of the LSBs [5].  Corresponding Author