Abstract:
The least significant bit (LSB) insertion method is a simple steganographic algorithm
that takes the least significant bit in some bytes of the cover medium and swaps them
with a sequence of bytes containing the secret data in order to conceal the information in
the cover medium. However its imperceptibility and hiding capacity are relatively low.
This is as revealed by the statistical characteristics of its resultant stego images
compared to the original cover images.
To increase the level of imperceptibility and the hiding capacity in the LSB insertion
method, this research proposes an enhanced LSB method that employs a selective and
randomized approach in picking specific number of target image bits to swap with the
secret data bits during the embedding process. To facilitate the selective picking of the
target image bits, a variation of the standard minimal linear congruential pseudo random
number generator (LCG) is used. The message digest (digital signature) of a user
supplied password is used to seed the LCG and to extract the message from the cover
medium.
In measuring the effectiveness of the proposed method, the study adopted an
experimental research design where the statistical characteristics of the proposed method
stego images were compared with those of the traditional LSB method in a comparative
experiment designed to establish the levels of image distortion (noise) introduced in the
original cover image when either of the methods is used under the same payload and
image.
The experiment results indicated improved levels of imperceptibility and hiding capacity
in the proposed method.
