Enhancing the Embedding Capacity in Line-based Cubism-Like Image

Anita Davamani K, Amudha S

Article (PDF)


Line-based Cubism- like Image, Cover Image, Stego Image, Discrete Cosine Transform (DCT), Gamma Correction.


Data hiding has wide applications in military areas, especially in covert communications and in secret keeping. In the existing system data is directly hidden into cover image and sent to receiver side. When this method is used, it is easy for the hackers to extract the hidden data. Hence a new type of computer art, called Line-based Cubism-Like Image, which keeps lines and regions from multiple viewpoints are proposed. Cubism artists transform a natural scene into geometric form in paintings by breaking up, analyzing and reassembling objects in the scenes from multiple viewpoints. In this method, the prominent line segments which are present in the source image are detected and rearranged to form regions with cubism flavor. Data is hidden during the process of recoloring the region in the generated art image. By using this technique data hiding is skillfully done with minimal distortion. This data embedding process is proved to be reversible, that is, the cover image along with the hidden data can be recovered successfully. This way of combining art image creation and data hiding, which may be called aesthetic data hiding, is a new idea of information hiding. Attracted by the art exhibited by the image, people hopefully will pay no attention to the hidden data in the art image, and via this camouflage effect, the embedded data can be kept securely or transmitted securely. This method also suffers from a drawback, that is, the files which are one third of the size of the cover image can only be embedded. Larger files cannot be embedded. As an enhancement measure, DCT compression technique is used to compress those files which are larger in size. Thus by this compression technique, larger files can also be embedded in the cover image. Gamma correction is also done to enhance the quality of the cover image.