A Comparison of Data Hiding in Digital Colour Images Using the Improved Least Significant Bit Algorithm Based on the Three Color Channels with (2:2:4) and (2:3:3) Arrangements

Abstract

The study aimed to introduce the concept of steganography, which is the science of concealing secret information within other media such as text, video, audio, or images. In image steganography, the digital  image serves as a cover image in which the secret message is hidden, and the resulting image after embedding is known as the stego image. The work relied on the Improved Least Significant Bit (Improved LSB) technique by embedding data inside an RGB colour image in a way that preserves image quality while providing larger storage capacity for hidden information. The hiding process was implemented using the same algorithm in two modes: (2:2:4) and (2:3:3). In the first mode, 2 bits of secret data were hidden in the least significant part of the red channel (R), 2 bits in the green channel (G), and 4 bits in the blue channel (B). In the second mode, 2 bits were hidden in the red channel, while 3 bits were hidden in each of the green and blue channels. This algorithm is based on color theory, which states that the human eye is more sensitive to red and green than to blue. The results confirmed successful data hiding in both cases without noticeable distortion or visible artifacts, and the hidden data was retrieved with no loss, without requiring the original image or a location map. The implementation was carried out in MATLAB, where a program was designed for data embedding and extraction. Results of both methods were compared by evaluating image quality using Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), and Bit Error Rate (BER). Additionally, histograms were generated to illustrate color distribution differences between the original and stego images.