Ownership Authentication and Tamper Detection in Digital Images via Zero-Watermarking

This paper presents a new zero-watermarking algorithm for digital images. The main challenge in every zero-watermarking algorithm is to create a highly capable, reliable, and distortion-free scheme. Unlike other zero-watermarking methods, our method not only focusing on ownership authentication, in the present proposal, we extend the capabilities of the zero-watermarking algorithm to detect tampering in digital images. Using a matching-pixel criterion in spatial domain in conjunction with a data hiding via bit replacing, compression and encryption, the algorithm obtains a zero-watermarking code used to authenticate the ownership and detect a tamper in the image. Experimental results show the effectiveness of the proposed method to detect tampering such as copy-move, copy-paste, text addition, constant average and cropping manipulations, as well as its robustness to prove the ownership authentication against these attacks.