Image hiding using neighborhood similarity
Ali Mohammad Nickfarjam, Hossein Ebrahimpour-Komleh, A. Pourshabanan Najafabadi · 2014
In this paper, a new method for image hiding is presented which takes advantages of Particle Swarm Optimization (PSO) and neighborhood similarity features in order to embed pixels of secret image in best positions of host image. Most Significant Bits (MSBs) of secret image pixels are utilized to hide in Least Significant Bits (LSBs) of host image pixels. Three feature functions and three corresponding coefficients are defined to find appropriate pixels for hiding. The proposed technique employs the ability of three features for neighborhood similarity to improve embedding performance as well as making better visually proprieties of the cover image. The presented method defines a special secret key for each host image based on PSO. The novelty of using neighborhood similarity features with LSB-replacement causes embedding performance improvement. The experimental results show the superiority of this approach over the LSB-based and evolutionary-based methods.