Data hiding in multiple images has been a significant research direction in information security. How to reasonably design the embedding strategy to spread the payload among multiple images is still an open issue. In this paper, we propose an embedding strategy on fusing multiple features. We utilize the typical characteristic parameters of gray level co-occurrence matrix, the image entropy and the shape parameter to describe image complexity. Furthermore, we combine with the number of cover images, the number of cover images assigned to steganographer and the size of cover image to estimate the steganographic capacity of each image. The strategy is implemented together with some state-of-the-art single image steganographic algorithms. Experimental results demonstrate that the security performance of the proposed strategy is higher than that of the state-of-the-art embedding strategy against the blind universal pooled steganalysis.

隐藏在多个图像中的数据一直是信息安全的重要研究方向。如何合理设计嵌入策略以在多个图像之间分布有效载荷仍然是一个悬而未决的问题。在本文中，我们提出了一种融合多种特征的嵌入策略。我们利用灰度共生矩阵的典型特征参数，图像熵和形状参数来描述图像复杂度。此外，我们结合封面图像的数量，分配给隐写术者的封面图像的数量以及封面图像的大小，以估计每个图像的隐写能力。该策略与一些最新的单图像隐写算法一起实施。