|
|
 |
| |
|
|
|
Secure Image Forensic Marking Algorithm using 2D Barcode and Off-axis Hologram in DWT-DFRNT Domain |
|
|
|
PP: 513S-520S |
|
|
|
Author(s) |
|
|
|
De Li,
JongWeon Kim,
|
|
|
|
Abstract |
|
|
| In this study, a robust, secure forensic marking algorithm was implemented using processes for generating holograms of 2D barcodes containing copyright information and embedding the holograms in the discrete wavelet transform- (DWT-) discrete fractional random transform (DFRNT) domain. In the proposed algorithm, the off-axis holograms of the 2D barcodes in which information is hidden are robust against attacks and are embedded in the transformed domains of the original images. The DFRNT enhances security by randomly mixing information so that it is embedded in unpredictable positions in certain frequency spaces. Therefore, DFRNT was combined with DWT and then used as a dual domain to ensure simultaneous robustness and security. Different holograms were produced with little signal interference by changing the coordinate values and separation angles, thereby ensuring security as well as enabling the embedding of multiple holograms in one image. The algorithm was also designed so that bit errors occurring during detection could be corrected by the self-error correction function of 2D barcodes. Its robustness was examined in several experiments, including compression, noise addition, and rotation experiments. The detection bit errors for all 2D barcodes used in the experiments were less than 3%. This confirms that the algorithm can accurately extract hidden information from detected 2D barcodes. |
|
|
|
|
|
|
|
