A lot of image steganographic techniques (also called data hiding) conceal secret data by utilizing a reference matrix (RM), and some new types of RMs, such as the turtle shell, an octagon-shaped shell, the Sudoku table, and so on, have been proposed in recent years. In this article, we present a novel type of RM called anisotropic RM. By employing a full search strategy, we have found the optimal parameters for constructing the anisotropic RM. To judge the performance of a parameter set quickly, a theoretical peak signal-to-noise ratio (PSNR) evaluation method is proposed. In addition, we extend the RM to three-dimensional (3D) space. Experimental results show that our data hiding scheme, based on the anisotropic RM, has a better quality stego image than previous methods. Moreover, the 3D RM works better than the traditional 2D RM using the same embedding capacity.

许多图像隐写技术（也称为数据隐藏）利用参考矩阵（RM）和一些新型的RM（例如，龟壳，八角形壳，数独表等）来隐藏秘密数据。 ，近年来已提出。在本文中，我们提出了一种新型的RM，称为各向异性RM。通过采用完全搜索策略，我们发现了用于构造各向异性RM的最佳参数。为了快速判断参数集的性能，提出了一种理论上的峰值信噪比（PSNR）评估方法。此外，我们将RM扩展到三维（3D）空间。实验结果表明，基于各向异性RM的数据隐藏方案比以前的方法具有更好的质量隐身图像。此外，