Abstract
The aims of improving steganographic method are divided into two groups: the first is to make the hiding capacity as high as possible; the second is to make the visible distortion as low as possible. The higher the visual quality of the stego-image, the less suspicious it becomes, which can increase security. However, the distortion caused by embedding data into images is not predictable and typically image dependent. If the user has a database of possible cover images, finding a suitable cover image that can sustain high visual quality after embedding is challenging. Thus, an automatic cover selection method is needed. In this paper, the problem of visual quality of the stego-image is tackled as a classification problem, where a CNN-based classifier is employed to select images that can have high imperceptibility after the process of embedding. To achieve that, a CNN was trained to classify images into “High Quality” and “Low Quality”. The CNN was based on SqueezeNet architecture, and was trained in two scenarios; transfer learning and learning from scratch. The two classifiers were able to achieve very high classification accuracies of F1 = 0.926 and 0.904.
Similar content being viewed by others
References
Al-Qershi OM, Khoo BE (2014) Controlling hiding capacity using image characteristics with a 2D-DE data hiding scheme. AEU Int J Electron Commun 68(4):346–350
Bas P, Filler T, Pevný T (2011) Break Our Steganographic System the ins and outs of organizing BOSS BT—information hiding. In Proceedings of the 13th international conference on information hiding, pp 59–70
Cruz N, Lobos-Tsunekawa K, Ruiz-del-Solar J (2018) Using convolutional neural networks in robots with limited computational resources: detecting NAO robots while playing Soccer BT—RoboCup 2017: Robot World Cup XXI, pp 19–30
Cychnerski J, Brzeski A, Boguszewski A, Marmolowski M, Trojanowicz M (2017) Clothes detection and classification using convolutional neural networks. In 2017 22nd IEEE international conference on emerging technologies and factory automation (ETFA), pp 1–8
El Abbadi NK (2013) Cover optimization for image in image steganography. Int J Comput Sci Issues 10(1):556–564
Evsutin O, Kokurina A, Meshcheryakov R (2018) Approach to the selection of the best cover image for information embedding in JPEG images based on the principles of the optimality. J Decis Syst 27(sup1):256–264
Grībermans D, Jeršovs A, Rusakovs P (2016) Development of requirements specification for steganographic systems. Appl Comput Syst 20(1):40–48
Han X, Zhong Y, Cao L, Zhang L (2017) Pre-trained AlexNet architecture with pyramid pooling and supervision for high spatial resolution remote sensing image scene classification. Remote Sens 9(8):848
Hajduk V, Levický D (2016) Cover selection steganography. In 2016 International symposium ELMAR, pp 205–208
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Hetherington J, Lessoway V, Gunka V, Abolmaesumi P, Rohling R (2017) SLIDE: automatic spine level identification system using a deep convolutional neural network. Int J Comput Assist Radiol Surg 12(7):1189–1198
Holub V, Fridrich J, Denemark T (2014) Universal distortion function for steganography in an arbitrary domain. EURASIP J Inf Secur 2014:1–13
Holub V, Fridrich J (2012) Designing steganographic distortion using directional filters. In IEEE international workshop on information forensics and security (WIFS), pp 234–239
Hradiš M, Kotera J, Zemcık P, Šroubek F (2015) Convolutional neural networks for direct text deblurring. In Proceedings of BMVC, vol 10, p 2
Huang W, Zhao X (2016) Novel cover selection criterion for spatial steganography using linear pixel prediction error. Sci China Inf Sci 59:1–3
Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K (2016) SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size, pp 1–13. arXiv preprint arXiv:1602.07360
ImageNet. http://www.image-net.org
Johnson NF, Jajodia S (1998) Exploring steganography: seeing the unseen. Computer (Long Beach Calif) 31(2):26–34
Kouider S, Chaumont M, Puech W (2012) Technical points about adaptive steganography by oracle (ASO). In Proceedings of the 20th European signal processing conference (EUSIPCO), pp 1703–1707
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems, pp 1097–1105
Lin G, Wu Q, Qiu L, Huang X (2018) Image super-resolution using a dilated convolutional neural network. Neurocomputing 275:1219–1230
Ng H-W, Nguyen VD, Vonikakis V, Winkler S (2015) Deep learning for emotion recognition on small datasets using transfer learning. In Proceedings of the 2015 ACM on international conference on multimodal interaction, pp 443–449
Ouyang W et al (2015) Deepid-net: deformable deep convolutional neural networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2403–2412
Peng M, Wang C, Chen T, Liu G (2016) Nirfacenet: a convolutional neural network for near-infrared face identification. Information 7(4):61
Quattoni A, Torralba A (2009) Recognizing indoor scenes. In: 2009 IEEE conference on computer vision and pattern recognition, pp 413–420. https://ieeexplore.ieee.org/document/5206537. Accessed 24 Aug 2021
Sajedi H, Jamzad M (2010) Using contourlet transform and cover selection for secure steganography. Int J Inf Secur 9(5):337–352
Sajedi H, Jamzad M (2010) Selecting a reliable steganography method. In Proceedings of the international conference on multimedia computing and information technology (MCIT), pp 69–72
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In Proceedings of international conference on learning representations, ICLR,
Subhedar MS, Mankar VH (2013) Performance evaluation of image steganography based on cover selection and contourlet transform. In 2013 International conference on cloud and ubiquitous computing and emerging technologies, pp 172–177
Sun Y, Liu F (2010) Selecting cover for image steganography by correlation coefficient. In Proceedings of the second international workshop on education technology and computer science, vol 2, pp 159–162
Szegedy C et al (2015) Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1–9
Wu S, Liu Y, Zhong S, Liu Y (2015) What makes the stego image undetectable? In Proceedings of the 7th international conference on Internet multimedia computing and service, pp 47–52
Xie M, Jean N, Burke M, Lobell D, Ermon S (2016) Transfer learning from deep features for remote sensing and poverty mapping. In Proceedings of the 13th AAAI conference on artificial intelligence,
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Hamid, N., Sumait, B.S., Bakri, B.I. et al. Enhancing visual quality of spatial image steganography using SqueezeNet deep learning network. Multimed Tools Appl 80, 36093–36109 (2021). https://doi.org/10.1007/s11042-021-11315-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-021-11315-y