High capacity secure dynamic multi-bit data hiding using Fibonacci Energetic pixels

Steganography and Steganalysis are becoming increasingly relevant in information forensics and hiding data in the higher bitplanes without keeping any perceptible signature into the image is a challenging problem in this area. In this paper, we propose a unique solution to this problem using Fibonacci numbers as base. The pixels are selected from the busy part of the image where noticeable changes in pixel intensities occur. The business of the pixels is determined by their Fibonacci energy. The pixels values are converted into Fibonacci base and their corresponding Fibonacci energies are estimated by the Fibonacci expansion of pixel intensities. The set of energetic pixels are considered according to the descending order of their energy values. The binary data are concealed into higher bitplanes (up to 5) of the Fibonacci base of the pixel intensities. We theoretically derive some nice combinatorial properties related to distortion of pixel intensities and also experimentally show that our algorithm withstands against visual, structural and statistical attacks. The average embedding capacity is 3.98 bpp and average PSNR is 39.59 dB. We also demonstrate that our method is capable of resisting from the series of benchmark tests provided by StirMark 4.0 .