Keyless steganography in spatial domain using Energetic pixels

Abstract

Steganography is the field of hiding messages in apparently innocuous media (e.g. images). Hiding messages in the pixel intensities of images is a popular approach in spatial domain steganography. However, since most schemes do not consider the image content when choosing the message carrying pixels, they are readily defeated by visual and statistical attacks. We show how to distribute the message in selective parts of an image, particularly in the places where a large change in the color/intensity occurs, using a variety of embedding schemes. Our definition of energetic pixels captures this notion of 'busy' area in an image and our embedding techniques keep the energy function invariant between the cover and the stego images for lossless data recovery, without the necessity of sharing a key or a pass-phrase between the sender and the receiver. We show that our approach provides minimum visual disturbance and can overcome popular first order statistical tests. © Springer-Verlag 2012.