Abstract
Least significant bit (LSB) steganography is the most widely employed technique, as it promises imperceptibility at low computational complexity while hiding the message in LSBs of cover image pixels. Recent research is focused on its capacity improvement by extending the embedding into second and third binary layers of cover image. This paper is focused to improve both capacity and imperceptibility. Considering the cover image, an advance technique for embedding is introduced based on representing cover image pixels using a new 13-bit prime series representation which increases the embedding capacity 3-times as compared to conventional LSB embedding. Moreover, an adaptive algorithm has also been proposed which automatically adjusts the chaotic key used to select the locations of embedding pixels based on the dimensions of cover and secret image. It ensures that the entire secret image is randomly spread and covered in the cover image. Furthermore, redundancy of the secret image has been reduced significantly by applying two dimensional DCT and thresholding for the coefficients. The 2-bit Reed Solomon error correction code applied to the secret information enhances security and reliability against attacks. Consequently, the simulation results illustrate minimum visual distortion effects in the proposed model along with correct recovery of secret data.