Two-Phase Method for Image Authentication and Enhanced Decoding

A two-phase method for image authentication using side information from the communication channel, followed by improved error correction using reliable data, is presented. The functionality of the proposed method is divided into two phases. In the first phase, important fragments of an image, called the region of interest, such as an eye or face, are authenticated using side information from the channel. If the authentication succeeds, the second phase follows. In the second phase, the authenticated fragments are provided as a feedback to the channel decoder. This feedback is used to improve the results of channel decoding. By using a varying length of authenticated fragments at the receiver side, varying levels of coding gains are achieved. The first phase achieves limited image correction, together with the authentication of the region of interest. The second phase improves the decoding results using the image data, which are already declared correct and authentic by the first phase. The image authentication algorithm and error correcting codes mutually benefit from each other in the proposed method: the error correcting codes help in image authentication and successful image authentication helps in improving the results of error correcting codes. The applicability of the proposed scheme is supported by simulation results for images, where bit error rate curves are shown in addition to the visual image enhancements. A security analysis of the proposed method is given in the end.