Existing visual cryptography (VC)-based copyright protection schemes (Amiri and Mohaddam in Multimed Tools Appl 75(14):8527–8543, 2016; Liu and Wu in IET Inf Secur 5(2):121–128, 2011) for multiple images provide meaningless shares to the owners. These shares create a suspicion that some secret information is shared. Also, these schemes require the share of every owner to prove the copyright. If any of the ownership share is not available, the copyright of these owners cannot be verified. This makes the usage of schemes restricted. To address these issues, an extended visual cryptography-based copyright protection scheme is proposed for multiple images with multiple owners. This scheme provides meaningful ownership shares to the owners, and their copyright can be verified by using a qualified set of owner shares. In this scheme, three types of shares are used, i.e., master share, ownership share and key share. The proposed scheme ensures robustness against different geometrical attacks, especially the rotation attack, as LBP and SURF together represent the host image efficiently. There is no restriction on watermark size, as SURF gives a flexibility to select any number of feature points. Usage of LBP ensures no false positive cases. Each of the ownership shares is created using the master share and the watermark. The ownership share is used to create a key share which is stored with the Trusting Authority (TA). To prove the copyright of multiple images, the ownership images and key share are superimposed to retrieve the watermark. The experimental results show that the scheme clearly verifies the copyright of digital images and is robust against several image processing attacks while having high imperceptibility. Comparisons with the existing copyright protection schemes show better performance of the proposed scheme.

中文翻译：

---

使用 LBP-SURF 描述符为多个图像和所有者扩展基于视觉密码的版权保护方案

现有的基于视觉密码学 (VC) 的版权保护方案（Amiri 和 Mohaddam in Multimed Tools Appl 75(14):8527–8543, 2016; Liu and Wu in IET Inf Secur 5(2):121–128, 2011）用于多个图像为所有者提供了无意义的分享。这些共享让人怀疑某些秘密信息被共享。此外，这些计划需要每个所有者的份额来证明版权。如果任何所有权份额不可用，则无法验证这些所有者的版权。这使得方案的使用受到限制。为了解决这些问题，针对具有多个所有者的多个图像提出了一种基于扩展视觉密码学的版权保护方案。该方案为所有者提供了有意义的所有权份额，并且可以通过使用一组合格的所有者份额来验证其版权。在这个方案中，使用三种类型的股份，即主股份、所有权股份和关键股份。所提出的方案确保了对不同几何攻击的鲁棒性，尤其是旋转攻击，因为 LBP 和 SURF 一起有效地表示了宿主图像。水印大小没有限制，因为 SURF 可以灵活地选择任意数量的特征点。LBP 的使用确保没有误报情况。每个所有权份额都是使用主份额和水印创建的。所有权份额用于创建与信任机构 (TA) 一起存储的密钥份额。为了证明多张图像的版权，将所有权图像和密钥共享叠加以检索水印。实验结果表明，该方案清晰地验证了数字图像的版权，对多种图像处理攻击具有鲁棒性，同时具有较高的不可感知性。与现有版权保护方案的比较表明所提出的方案具有更好的性能。