A New Cost Function for Spatial Image Steganography Based on 2D-SSA and WMF

As an essential tool for secure communications, adaptive steganography aims to communicate secret information with the least security cost. Inspired by Ranking Priority Profile (RPP), we propose a novel two-step cost function in this paper. The RPP mainly includes three rules, i.e. Complexity-First rule, Clustering rule and Spreading design function. We use two-dimensional Singular Spectrum Analysis (2D-SSA) Weighted Median Filter (WMF) designing 2D-SSA is employed selecting key components clustering embedding positions, which follows rule. Also, deploy smooth resulting image produced WMF. Extensive experiments have shown efficacy of proposed method, has improved performance over four benchmarking approaches against non-shared selection channel attack. It also provides comparable selection-channel-aware scenarios, where best results are observed when relative payload 0.3 bpp or larger. Besides, approach much faster than other model-based methods.