A threshold-LSB based information hiding scheme using digital images

Providing protection to secret information has been given high importance and has received increasing attention from the information technology community. In this paper, we propose a new approach for information hiding based on gray-scale digital images which considers the threshold of the pixel of an image while hiding the bits of a secret message in its least-significant-bits (LSBs). The proposed method works on the secret message by converting it into a stream of bits. Then we generate pseudo random numbers to provide a random distribution of the secret bits across the image. We use m-least significant bits of an 8-bit pixel to hold the secret bits. The value of m varies depending on the threshold of the pixels. The number of bits (m) to be embedded in a pixel having small threshold is more than that of a pixel having high or medium threshold. When we consider low and high thresholds we simply replace the m-least significant bits of a pixel with the m-secret bits. For medium threshold category pixels we use exclusive-or (XOR) of m-least significant bits of the pixel with the m-secret bits. The experimental results show that the proposed scheme achieves high embedding capacity and good visual quality of stego images for m=3 and m=2 for low and non-low thresholds respectively.