Abstract
Recognizing Natural Images (NI) and Computer-Generated Images (CGI) by a human is difficult due to the use of new-age computer graphics tools for designing more photorealistic CGI images. Identifying whether an image was captured naturally or if it is a computer generated image is a fundamental research problem. For this problem, we design and implement a new Convolutional Neural Network (ConvNet) architecture along with data augmentation techniques. Experimental results show that our method outperforms existing methods by 2.09 percentage for recognizing NI and CGI images.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst
Rahmouni N, Nozick V, Yamagishi J, Echizen I (2017) Distinguishing computer graphics from natural images using convolution neural networks. In: Proceedings of IEEE ınternational workshop ınformation forensics security, Dec 2017, pp 1–6
Quan W, Wang K, Yan D-M, Zhang X (2018) Distinguishing between natural and computer-generated ımages using convolutional neural networks. IEEE Trans Inf Forens Secur 13(11), Nov 2018
Qian Y, Dong J, Wang W, Tan T (2015) Deep learning for steganalysis via convolutional neural networks. Proc SPIE 9409: 94090J, Mar 2015
Pibre L, Pasquet J, Ienco D, Chaumont M (2016) Deep learning is a good steganalysis tool when embedding key is reused for different images, even if there is a cover source-mismatch. Proc IST Electron Imag, 1–23
Ng T-T, Chang S-F, Hsu J, Pepeljugoski M (2004) Columbia photographic images and photorealistic computer graphics dataset, Columbia University, New York, NY, USA, Tech. Rep. #205-2004-5
Ng T-T, Chang S-F (2013) Discrimination of computer synthesized or recaptured images from real images. In: Sencar HT, Memon N (eds) Digital image forensics. Springer, New York, NY, USA, pp 275–309
Nagi J, Ducatelle F, Di Caro GA, Ciresan D, Meier U, Giusti FN, Schmidhuber J, Gambardella LM (2011) Max-Pooling Convolutional Neural Networks for Vision-based Hand Gesture Recognition. In: IEEE ınternational conference on signal and ımage processing applications (ICSIPA2011)
Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. preprint, arXiv:1502.03167
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Machine Learn Res 15:1929–1958
LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86:2278–2324
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Rajasekhar, K., Kumar, G.I.S. (2021). Recognition of Natural and Computer-Generated Images Using Convolutional Neural Network. In: Laxminidhi, T., Singhai, J., Patri, S.R., Mani, V.V. (eds) Advances in Communications, Signal Processing, and VLSI. Lecture Notes in Electrical Engineering, vol 722. Springer, Singapore. https://doi.org/10.1007/978-981-33-4058-9_2
Download citation
DOI: https://doi.org/10.1007/978-981-33-4058-9_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-33-4057-2
Online ISBN: 978-981-33-4058-9
eBook Packages: EngineeringEngineering (R0)