A color image steganography method based on ADPVD and HOG techniques

Steganography is an important secret information communication technology in which one may send secret data without others having knowledge of its existence. This chapter proposes a pixel-based adaptive directional pixel value differencing (P-ADPVD) method. The original pixel value differencing (PVD) uses only one embedding direction for all pixels of the cover image, whereas the content of the digital images has different edge direction for each pixel. Therefore it is better to exploit this feature for embedding secret data in various edge directions. To this end, the proposed P-ADPVD method depends on a pixel-of-interest (POI) to hide secret data bits in the dominant edge direction for each pixel. Histogram of oriented gradient (HOG) algorithm is employed to find the dominant edge direction for each block of the POIs using gradient magnitude and angle information calculated from the cover image and determined by a threshold value. This methodology significantly improves the quality and security of the stegoimages without sacrificing the embedding capacity. Finally, the experimental results using various digital images prove that the P-ADPVD method provides better visual quality of the stegoimage compared with other adaptive and nonadaptive PVD-based methods.