A secure and enhanced high embeding capacity steganography based on pixel indicator algorithm

Abstract

Distributing digital information electronically is vulnerable to a variety of attacks and attempts in breaking and revealing the secret information. Security hasbecome one of the most significant problems for distributing digital information safely. It is essential to protect digital information while communicating over insecure channels. Thus, a need exists for developing technologies that will help protecting digital information and to provide an acceptable amount of secrecy and privacy for any one connecting to and communicating over insecure channels. To add multiple layers of security, it is good practice to combine both cryptography and stenography sciences since most experts suggested that. Steganography is introduced to conceal secret information into cover object as opposed of cryptography whose goal is to protect information by transforming it into an unreadable format. A prototype, namely invisible crypto is proposed which is able to perform steganography and cryptography simultaneously. The prototype’s priority is to embed large amount and different formats of information such as PDF, EXCEL,WAV, etc using cover images like PNG, BMP. Pixel indicator high embedding capacity algorithm for RGB images is adopted and implemented in this project.Advanced Encryption standard (AES) is implemented as additional security layer for the adopted pixel indicator algorithm. Moreover, the adopted pixel indicator algorithm is enhanced in such a way to offer high embedding steganography algorithm based on RGBA images. Tests are conducted to analyze the capacity and quality performance of high embedding steganography algorithm for RGBA images. The results show that for capacity measure, using RGB images (3 channels) provides less capacity than using RGBA images (4 channels) for embedding. Some images are rejected to embed due to the low capacity using RGB images (3 channels) while using RGBA images are accepted. For overall PSNR performance, embedding using RGBA images (4 channels) has achieved better PSNR performance and better quality than using RGB images (3 channels).