Digital image ownership verification based on spatial correlation of colors

In this paper, a spatial domain digital image copyright protection scheme based on Visual Cryptography (VC) and Spatial Correlation of Colors (SCC) is proposed. A binary feature matrix, extracted from the spatial correlation of host image, is used to split the watermark into two noisy binary images called shares. One of them is generated during watermark embedding phase and is registered with a trusted third party. The other is extracted during watermark extraction phase. Both these shares are combined to recover hidden watermark. When compared to the related works, the proposed scheme reduces the probability of false positives; reduces the size of shares and improves the quality of extracted watermark. Experimental results prove that the scheme is also robust to wide range of attacks.