| Abstrakt: |
The growth of internet technology in the present era is creating a pressing need to develop several new methods for copyright protection, ownership and security of the digital contents. These concerns have triggered significant research to find ways to hide data into digital media. Data hiding can be done in two domains, namely spatial domain and frequency domain. The work focuses on the robustness and data hiding capacity limitations in the spatial domain. In this paper, a layered algorithm for hiding a watermark (image/ text) in RGB (color) images based on Gödel numbering, Alphabetic Coding(AC) and Auxiliary carry watermarking method proposed by Lalitha Bhaskari, Damodaram, Avadhani(2006:666-668) proved to be more secure and robust. Unlike other methods, no prior information about the original watermark, hiding locations, strengths are needed in the proposed blind watermark scheme. The security levels are gradually enhanced from the proposed auxiliary carry watermark method combined with LSB method, to a proposed modified auxiliary carry method combined with Gödelization and alphabetic coding methods. The proposed methodology follows three layered approach for embedding the watermark into the original image ensuring no degradation of the original image and providing security to the watermark. In the first layer, watermark which is to be hidden in color image is encoded into Gödel numbers, termed as Gödelization as proposed by Lalitha Bhaskari, Avadhani, Damodaram(2009: 209- 213). In the second layer, the encoded string obtained from the first layer is compressed using AC technique. In the last layer, the resultant data which is in a compressed encoded string format is hidden in the cover image using a proposed method known as modified auxiliary carry watermark method. During the watermark extraction process, depending upon the key generated, the watermark is extracted from the watermarked image and experimental results proved that the proposed algorithm employed more data payload capacity than the traditional methods and is robust to attacks in spatial domain. The computational complexity is low, yet providing more security. [ABSTRACT FROM AUTHOR] |
