Improved and Secure Differential LSB Embedding Steganography

With the growth of information and communication technologies and applications, requirement for the information assurance and security enhancement in data is rising exceptionally. Image steganography is one of the potential solutions for it. Least Significant Bit (LSB) based steganography is categorized as a popular and widely utilized technique in spatial domain. Conventional methods applied in LSB based steganography mostly focus on increasing the capacity of embedded information and imperceptibility while the problem of security still needs to be considered because the LSB embedding is vulnerable to several common attacks on the data like Additive White Gaussian Noise (AWGN) attack, geometric attacks and much more. The proposed work provides an innovative approach to enhance the covert transmission aspect of the system. The secret information is pre-processed by using latest devised right translated gray substitution box which maps the information to a new format. This process enhances the security of the information as data retrieval is impossible without the knowledge of mapping rule and secret key. To further emphasis on correct retrieval Bose–Chaudhuri– Hocquenghem (BCH) error correcting codes are used. Differential embedding to the LSBs of chaotically selected pixels is another step to make detection difficult and avoid error to propagate. The chaotic selection of pixels is optimized by using Genetic Algorithm (GA) which is a way forward to increase Peak Signal to Noise Ratio (PSNR) of the cover image. Simulation results reveal authenticity of the proposed scheme.