Lossless data hiding based on histogram modification

Lossless data hiding is the technique of embedding data in an image and retrieval of the data with lossless reconstruction of original image. In this paper, we present a novel lossless data hiding scheme based on histogram modification. This technique is based on differences of adjacent pixels for embedding data and has more hiding capacity compared to existing methods. The number of message bits that can be embedded into an image equals the number of pixels associated with the peak point. Here, a histogram shifting technique is applied in order to prevent overflow and underflow problems of the pixels. For color images the hiding capacity is three times larger than the grayscale images. The performance of the algorithm has been evaluated for eight grayscale images and eight color images with hiding capacity (bits) and peak signal to noise ratio (dB) of the reconstruction as the parameters.