StegoAppDB Database Sample

This dataset is a subset of the full StegoAppDB database (Newman 2019) and contains 51,200 images from 28 mobile phones across 10 models. Student photographers captured 100 high-resolution images per phone (2,800 total). These images were used to create 2,800 input images and 38,000 stego images. The stego images were generated by six mobile apps—five Android (MobiStego, Passlok, PixelKnot, PocketStego, SteganographyM) and one iOS (Pictograph)—at five embedding rates (5%, 10%, 15%, 20%, 25%). Through reverse-engineering, 7,600 cover images were extracted representing the preprocessed state before message embedding, providing authentic cover-stego pairs for training detection algorithms. Hidden messages were randomly extracted from Shakespeare's 634 plays (Shakespeare, 1993).This dataset is publicly available and has no copyright or privacy issues associated with it.Commercial licensing enquiries should be directed to the Iowa State University Office of Innovation Commercialization at: [e-mail], or 515-294-4740ReferencesNewman, Jennifer, Li Lin, Wenhao Chen, Stephanie Reinders, Yangxiao Wang, Min Wu, and Yong Guan. "StegoAppDB: a steganography apps forensics image database." arXiv preprint arXiv: 1904.09360 (2019).Shakespeare, William. The Complete Works of William Shakespeare. http://shakespeare.mit.edu/,1993.

本数据集为完整StegoAppDB数据库（Newman等，2019）的子集，包含来自10款型号、28台移动设备的51200张图像。学生摄影师为每台设备拍摄100张高分辨率图像，总计2800张原始素材。这些素材被用于生成2800张输入图像与38000张隐写图像。

隐写图像由6款移动应用生成：5款安卓（Android）应用（MobiStego、Passlok、PixelKnot、PocketStego、SteganographyM）与1款iOS应用（Pictograph），且分别以5%、10%、15%、20%、25%五种嵌入率完成生成。

通过逆向工程，研究团队提取得到7600张载体图像，即消息嵌入前的预处理状态图像，可提供真实的载体-隐写图像配对样本，用于训练隐写检测算法。隐藏消息均随机取自莎士比亚的634部戏剧作品（Shakespeare，1993）。

本数据集已公开发布，不存在相关版权或隐私问题。商业授权咨询请联系爱荷华州立大学创新商业化办公室，联系方式：[e-mail]，或515-294-4740。

参考文献
1. Newman, Jennifer, Li Lin, Wenhao Chen, Stephanie Reinders, Yangxiao Wang, Min Wu, 及Yong Guan. 《StegoAppDB：一款面向隐写应用取证的图像数据库》，arXiv预印本arXiv:1904.09360（2019）。
2. Shakespeare, William. 《威廉·莎士比亚全集》，http://shakespeare.mit.edu/, 1993。