隱寫術(Steganography)是將大量資料隱藏於如影像、視訊等載體之技術，而隱寫分析(Steganalysis)為隱寫術的對抗技術，用於判斷載體中是否隱藏額外資訊。本研究藉由深度學習(Deep Learning)，訓練一個由密集連結的卷積神經網路(Convolutional Neural Networks，CNN)模塊為主體，設計一個針對JPEG影像隱寫術的全新隱寫分析架構。本架構結合Inception Net、ResNet與DenseNet的設計巧思與優勢，包括Inception Net中結合多種不同尺度卷積核以增加對尺度的適應性，ResNet對於特徵擷取的高重複使用率且冗餘程度低，而DenseNet可創造新特徵但冗餘程度高等特性，設計出效果更好的隱寫分析架構。本研究的模型訓練不需要人工預處理，可自動學習有效特徵。在特徵提取的部分，本機制不使用池化操作以避免抑制隱寫訊號，偵測則以三種隱寫術為對象。此外，我們設計所謂「多模型決策方法」，結合多個單一模型一起檢測，並以原先的個別單一訓練模型做為比較對象。實驗結果顯示我們的模型和方法，幾乎都超越了比較的目標，甚至是目前最先進的隱寫分析模型-SRNet。

Steganography is a technique for hiding large amounts of data in carriers such as video and video, and Steganalysis is a technique to determine whether additional information is hidden in the carrier. In our study, Deep Learning is used to train a densely connected Convolutional Neural Networks (CNN) module to design a new steganalysis architecture for JPEG image steganography. This architecture combines the design ingenuity and advantages of Inception Net, ResNet, and DenseNet.Including the inclusion of multiple scale convolution kernels in Inception Net to increase scalability, and ResNet's high reuse and low redundancy for feature extraction. DenseNet can create new features but high levels of redundancy to design a better steganalysis architecture. The model training of our study does not require manual preprocessing and can automatically learn effective features. In the feature extraction part, the module does not use the pooling operation to avoid suppressing the steganographic signal, and the detection is based on three steganography. In addition, we design a so-called "multi-model decision method" that combines multiple single models to detect together and compares with the original individual training models. The experimental results show that our models and methods almost surpass the goal of comparison, even the most advanced steganalysis model - SRNet.

[1]Kim Y , Duric Z , Richards D . Modified Matrix Encoding Technique for Minimal Distortion Steganography[J]. 2006.
[2]Fridrich J , Goljan M , Lisonek P , et al. Writing on wet paper[J]. IEEE Transactions on Signal Processing, 2005, 53(10):3923-3935.
[3]Zhang W , Wang X . Generalization of the ZZW Embedding Construction for Steganography[J]. IEEE Transactions on Information Forensics and Security, 2009, 4(3):564-569.
[4]Tomas Filler, Jan Judas, Jessica Fridrich "Minimizing Embedding Impact in Steganography using Trellis-Coded Quantization", Proc. SPIE, Electronic Imaging, Media Forensics and Security XII, San Jose, CA, January 18-20, 2010.
[5]Tomáš Pevný、Tomáš Filler、Patrick Bas . Using High-Dimensional Image Models to Perform Highly Undetectable Steganography[J]. Lecture Notes in Computer Science, 2010, 6387:161-177.
[6]Holub V , Fridrich J J . Designing steganographic distortion using directional filters.[C]// IEEE International Workshop on Information Forensics & Security. IEEE, 2012.
[7]V. Holub, J. Fridrich, T. Denemark, Universal Distortion Function for Steganography in an Arbitrary Domain, EURASIP Journal on Information Security, (Section:SI: Revised Selected Papers of ACM IH and MMS 2013), 2014(1).
[8] Guo L , Ni J , Shi Y Q . An efficient JPEG steganographic scheme using uniform embedding[C]// Information Forensics and Security (WIFS), 2012 IEEE International Workshop on. IEEE, 2012.
[9]Holub, Vojtěch, Fridrich J . [ACM Press the first ACM workshop - Montpellier, France (2013.06.17-2013.06.19)] Proceedings of the first ACM workshop on Information hiding and multimedia security - IH&MMSec "13 - Digital image steganography using universal distortion[J]. 2013:59.
[10]Wang C , Ni J . An efficient JPEG steganographic scheme based on the block entropy of DCT coefficients[C]// IEEE International Conference on Acoustics. IEEE, 2012.
[11]https://medium.com/@chih.sheng.huang821/%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92-%E6%94%AF%E6%92%90%E5%90%91%E9%87%8F%E6%A9%9F-support-vector-machine-svm-%E8%A9%B3%E7%B4%B0%E6%8E%A8%E5%B0%8E-c320098a3d2e
[12]Kodovsky, J., Fridrich, J. and Holub, V. (2012) Ensemble Classifiers for Steganalysis of Digital Media. IEEE Transactions on Information Forensics and Security, 7, 432-444.
[13]Fridrich, J., et al. (2011) Breaking HUGO—The Process Discovery. Information Hiding-international Conference, 6958, 85-101. https://doi.org/10.1007/978-3-642-24178-9_7
[14]Jan, K. and Fridrich, J. (2011) Steganalysis in High Dimensions: Fusing Classifiers Built on Random Subspaces. Proceedings of SPIE—The International Society for Optical Engineering, 7880, 181-197.
[15]Fridrich, J. and Kodovsky, J. (2012) Rich Models for Steganalysis of Digital Images. IEEE Transactions on Information Forensics and Security, 7, 868-882.
[16]Holub, V. and Fridrich, J. (2013) Random Projections of Residuals for Digital Image Steganalysis. IEEE Transactions on Information Forensics and Security, 8, 1996-2006.
[17]Pevny T, Bas P, Fridrich J. Steganalysis by subtractive pixel adjacency matrix[J]. IEEE Transactions on information Forensics and Security, 2010, 5(2): 215-224.
[18]Fridrich J, Kodovsky J. Rich models for steganalysis of digital images[J]. IEEE Transactions on Information Forensics and Security, 2012, 7(3): 868-882.
[19]Holub V, Fridrich J. Random projections of residuals for digital image steganalysis[J]. IEEE Transactions on Information Forensics and Security, 2013, 8(12): 1996-2006.
[20]Li B, Li Z, Zhou S, et al. New steganalytic features for spatial image steganography based on derivative filters and threshold LBP operator[J]. IEEE Transactions on Information Forensics and Security, 2018, 13(5): 1242-1257.
[21]Pevny T, Fridrich J. Merging Markov and DCT features for multi-class JPEG steganalysis[C]//Security, Steganography, and Watermarking of Multimedia Contents IX. International Society for Optics and Photonics, 2007, 6505: 650503.
[22]Kodovsky J, Fridrich J. Steganalysis of JPEG images using rich models[C]//Media Watermarking, Security, and Forensics 2012. International Society for Optics and Photonics, 2012, 8303: 83030A.
[23]Holub V, Fridrich J. Low-complexity features for JPEG steganalysis using undecimated DCT[J]. IEEE Transactions on Information Forensics and Security, 2015, 10(2): 219-228.
[24]Holub V, Fridrich J. Phase-aware projection model for steganalysis of JPEG images[C]//Media Watermarking, Security, and Forensics 2015. International Society for Optics and Photonics, 2015, 9409: 94090.
[25]Song X, Liu F, Yang C, et al. Steganalysis of adaptive JPEG steganography using 2D Gabor filters[C]//Proceedings of the 3rd ACM workshop on information hiding and multimedia security. ACM, 2015: 15-23.
[26]Denemark, T., Sedighi, V., Holub, V., Cogranne, R. and Fridrich, J. (2014) Selection-Channel-Aware Rich Model for Steganalysis of Digital Images. IEEE International Workshop on Information Forensics and Security, 2015, 48-53.
[27]Tang, W., Li, H., Luo, W. and Huang, J. (2014) Adaptive Steganalysis against WOW Embedding Algorithm. In: Uhl, A. Katzenbeisser, S. Kwitt, R. and Piva, A., Eds., 2nd ACM IH&MMSec, ACM Workshop on Information Hiding & Multimedia Security, Salzburg, 91-96.
[28]Sahiner B, Chan H P, Petrick N, et al. Classification of mass and normal breast tissue: a convolution neural network classifier with spatial domain and texture images[J]. IEEE transactions on Medical Imaging, 1996, 15(5): 598-610.
[29]Qian Y, Dong J, Wang W, et al. Deep learning for steganalysis via convolutional neural networks[C]//Media Watermarking, Security, and Forensics 2015. International Society for Optics and Photonics, 2015, 9409: 94090J
[30]Xu, G., Wu, H.Z. and Shi, Y.Q. (2016) Structural Design of Convolutional Neural Networks for Steganalysis. IEEE Signal Processing Letters, 23, 708-712
[31]https://www3.cs.stonybrook.edu/~skiena/jaialai/excerpts/node16.html
[32]Li B, Tan S, Wang M, et al. Investigation on cost assignment in spatial image steganography[J]. IEEE Transactions on Information Forensics and Security, 2014, 9(8): 1264-1277.
[33]Ye J, Ni J, Yi Y. Deep learning hierarchical representations for image steganalysis[J]. IEEE Transactions on Information Forensics and Security, 2017, 12(11): 2545-2557.
[34]Xu G. Deep convolutional neural network to detect J-UNIWARD [C]//Proceedings of the 5th ACM Workshop on Information Hiding and Multimedia Security. ACM, 2017: 67-73.
[35]Deep Residual Network for Steganalysis of Digital Images, with M. Boroumand and M. Chen, IEEE TIFS, vol. 14, no. 5, pp. 1181--1193, May 2019.
[36]C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, Z. Wojna “Rethinking the Inception Architecture for Computer Vision,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
[37]K. He, X. Zhang, S. Ren, J. Sun, "Deep residual learning for image recognition", 2015.
[38]G. Huang, Z. Liu, L. van der Maaten, K. Q. Weinberger “Densely Connected Convolutional Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
[39]ImageNet classification with deep convolutional neural networks Author： A Krizhevsky， I Sutskever ， GE Hinton Year：2012
[40] “VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION Karen” Simonyan ∗ & Andrew Zisserman + Visual Geometry Group, Department of Engineering Science, University of Oxford
[41]“Highway Networks” The Swiss AI Lab IDSIA Istituto Dalle Molle di Studi sull’Intelligenza Artificiale Universita della Svizzera italiana (USI) ` Scuola universitaria professionale della Svizzera italiana (SUPSI) Galleria 2, 6928 Manno-Lugano, Switzerland
[42]“Deep Networks with Stochastic Depth”, Gao Huang*, Yu Sun*, Zhuang Liu† , Daniel Sedra, Kilian Q. Weinberger
[43]“FRACTALNET: ULTRA-DEEP NEURAL NETWORKS WITHOUT RESIDUALS”, Gustav Larsson, Michael Maire, Gregory Shakhnarovich
[44]Aggregated Residual Transformations for Deep Neural Networks, Saining Xie1 Ross Girshick2 Piotr Dollar´ 2 Zhuowen Tu1 Kaiming H
[45]Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network, Alex Sherstinsky
[46]Yunpeng Chen1 , Jianan Li1,2 , Huaxin Xiao1,3 , Xiaojie Jin1 , Shuicheng Yan4,1 , Jiashi Feng1,”Dual Path Networks”
[47]https://www.tensorflow.org/
[48]https://medium.com/@prateekvishnu/xavier-and-he-normal-he-et-al-initialization-8e3d7a087528
[49]https://codertw.com/%E7%A8%8B%E5%BC%8F%E8%AA%9E%E8%A8%80/615856/
[50]P. Bas, T. Filler, and T. Pevný. Break our steganographic system – the ins and outs of organizing BOSS. In T. Filler, T. Pevný, A. Ker, and S. Craver, editors, Information Hiding, 13th International Conference, volume 6958 of Lecture Notes in Computer Science, pages 59–70, Prague, Czech Republic, May 18–20, 2011. Springer Berlin Heidelberg.
[51]P. Bas and T. Furon. BOWS-2. http://bows2.ec-lille.fr, July 2007.
[52]https://alaska.utt.fr/#timeline