科研之友
微信
新浪微博
Facebook
分享链接
摘要
The reversible image authentication (RIA) method is capable of detecting the tampered regions, and is able to reconstruct the original image if the marked image is un-tampered. The existing RIA methods successfully detect the integrity of digital images and have the ability to recover the original image from its marked version. However, they fail to detect the tampered regions with equal modification in brightness. Moreover, the existing methods also provide no mechanism to protect those un-embeddable blocks and thus a larger detection error rate can be introduced. In this paper, instead of embedding block-independent authentication code (AC) used in other methods, we embed hash value of block features to avoid the tampered regions that are undetectable in prior works. Besides, we also embed one bit AC into the least significant bit (LSB) of those un-embeddable blocks for the authentication purpose while achieving reversibility. The experimental results show that the proposed method offers a more accurate detection result while providing a comparable image quality.
