High security image steganography using IWT and graph theory

Steganography conceals the secret information inside the cover medium. There are two types of steganography techniques available practically. They are spatial domain steganography and Transform domain steganography. The objectives to be considered in the steganography methods are high capacity, imperceptibility and robustness. In this paper, a Color image steganography in transform domain is proposed. Reversible Integer Haar wavelet transform is applied to the R, G and B planes separately and the data is embedded in a random manner. Random selection of wavelet coefficients is based on the graph theory. This proposed system uses three different keys for embedding and extraction of the secret data, where key1(Subband Selection - SB) is used to select the Wavelet subband for embedding, key2(Selection of Co-effecients-SC) is used to select the co-efficients randomly and key3 (Selection of Bit length-SB) is used to select the number of bits to be embedded in the selected co-efficients. This method shows good imperceptibility, High capacity and Robustness.