Distinguishing computer-generated images from photographic images using two-stream convolutional neural network

Abstract Today’s advanced multimedia tools allow us to create photorealistic computer graphic images, effortlessly. There are various fields such as the film industry, virtual reality, video games where computer-generated (CG) images are used widely. CG images can also be misused in many ways. Therefore, there is a need of distinguishing CG images from real photographic (PG) images. This paper proposes a method to distinguish CG images from PG images using a two-stream convolutional neural network (CNN). In the proposed method, the first stream takes the advantage of the knowledge learned by the pre-trained VGG-19 network, and then this knowledge is transferred to learn the distinct features of CG and PG images. Here, we propose a second stream, that preprocesses the images using three high-pass filters which aim to help the network to focus on noise-based distinct features of CG and PG images. Finally, we propose an ensemble model to merge the outcomes of the proposed two streams. Comparative and self-analysis experiments demonstrates that the proposed method outperforms the state-of-the-art methods in terms of classification accuracy. The experimental results also show that the proposed method performs satisfactorily under the additive white Gaussian noise postprocessing operation in the images.

[1]  Siwei Lyu,et al.  How realistic is photorealistic? , 2005, IEEE Transactions on Signal Processing.

[2]  Dong-Ming Yan,et al.  Distinguishing Between Natural and Computer-Generated Images Using Convolutional Neural Networks , 2018, IEEE Transactions on Information Forensics and Security.

[3]  R. Polikar,et al.  Bootstrap - Inspired Techniques in Computation Intelligence , 2007, IEEE Signal Processing Magazine.

[4]  Wang Rangding,et al.  Classifying computer generated graphics and natural image based on image contour information , 2012 .

[5]  Larry S. Davis,et al.  Learning Rich Features for Image Manipulation Detection , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[6]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[7]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[8]  Kunj Bihari Meena,et al.  A Deep Learning Based Method to Discriminate Between Photorealistic Computer Generated Images and Photographic Images , 2020, ICACDS.

[9]  Zhenzhen Zhang,et al.  Distinguishing computer graphics from photographic images using a multiresolution approach based on local binary patterns , 2014, Secur. Commun. Networks.

[10]  Belhassen Bayar,et al.  Learning Unified Deep-Features for Multiple Forensic Tasks , 2018, IH&MMSec.

[11]  Satoshi Oyama,et al.  Fine-tuning deep convolutional neural networks for distinguishing illustrations from photographs , 2016, Expert Syst. Appl..

[12]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[13]  Jessica J. Fridrich,et al.  Rich Models for Steganalysis of Digital Images , 2012, IEEE Transactions on Information Forensics and Security.

[14]  Qi Cui.  Identifying materials of photographic images and photorealistic computer generated graphics based on deep CNNs , 2018 .

[15]  Shiguo Lian,et al.  Forensics feature analysis in quaternion wavelet domain for distinguishing photographic images and computer graphics , 2017, Multimedia Tools and Applications.

[16]  Yuewei Dai,et al.  Photorealistic computer graphics forensics based on leading digit law , 2011 .

[17]  Vipin Tyagi,et al.  A copy-move image forgery detection technique based on Gaussian-Hermite moments , 2019, Multimedia Tools and Applications.

[18]  Vipin Tyagi,et al.  Understanding Digital Image Processing , 2018 .

[19]  S. P. Ghrera,et al.  Pixel-Based Image Forgery Detection: A Review , 2014 .

[20]  Vipin Tyagi,et al.  Methods to Distinguish Photorealistic Computer Generated Images from Photographic Images: A Review , 2019, ICACDS.

[21]  Fei Peng,et al.  Identifying photographic images and photorealistic computer graphics using multifractal spectrum features of PRNU , 2014, 2014 IEEE International Conference on Multimedia and Expo (ICME).

[22]  Anderson Rocha,et al.  Computer generated images vs. digital photographs: A synergetic feature and classifier combination approach , 2013, J. Vis. Commun. Image Represent..

[23]  Hongxia Wang,et al.  Detection of Computer Graphics Using Attention-Based Dual-Branch Convolutional Neural Network from Fused Color Components , 2020, Sensors.

[24]  Wei Zhang,et al.  Distinguishing Computer-Generated Graphics from Natural Images Based on Sensor Pattern Noise and Deep Learning , 2018, Sensors.

[25]  何源,et al.  A local joint fast handoff scheme in cognitive wireless mesh networks , 2014 .

[26]  Dong-Ming Yan,et al.  Learn with diversity and from harder samples: Improving the generalization of CNN-Based detection of computer-generated images , 2020, Digit. Investig..

[27]  Ira Kemelmacher-Shlizerman,et al.  Synthesizing Obama , 2017, ACM Trans. Graph..

[28]  Jianru Xue,et al.  A statistical feature based approach to distinguish PRCG from photographs , 2014, Comput. Vis. Image Underst..

[29]  Yun Q. Shi,et al.  Distinguishing Computer Graphics from Photographic Images Using Local Binary Patterns , 2012, IWDW.

[30]  Ming He,et al.  Distinguish computer generated and digital images: A CNN solution , 2018, Concurr. Comput. Pract. Exp..

[31]  Dong-Ming Yan,et al.  Distinguishing Computer-Generated Images from Natural Images Using Channel and Pixel Correlation , 2020, Journal of Computer Science and Technology.

[32]  Junichi Yamagishi,et al.  Distinguishing computer graphics from natural images using convolution neural networks , 2017, 2017 IEEE Workshop on Information Forensics and Security (WIFS).

[33]  Xinghao Jiang,et al.  Computer Graphics Identification Combining Convolutional and Recurrent Neural Networks , 2018, IEEE Signal Processing Letters.

[34]  H. P. Chen,et al.  Detecting computer generated images based on local ternary count , 2016, Pattern Recognition and Image Analysis.

[35]  Guohua Wu,et al.  An Evaluation of Deep Learning-Based Computer Generated Image Detection Approaches , 2019, IEEE Access.

[36]  Vipin Tyagi,et al.  A hybrid copy-move image forgery detection technique based on Fourier-Mellin and scale invariant feature transforms , 2020, Multimedia Tools and Applications.

[37]  Xiangyang Luo,et al.  Identifying Computer Generated Images Based on Quaternion Central Moments in Color Quaternion Wavelet Domain , 2019, IEEE Transactions on Circuits and Systems for Video Technology.

[38]  Vipin Tyagi,et al.  Image Forgery Detection: Survey and Future Directions , 2019, Data, Engineering and Applications.

[39]  Lutz Prechelt,et al.  Automatic early stopping using cross validation: quantifying the criteria , 1998, Neural Networks.

[40]  Loui Al-Shrouf.  Development and implementation of a reliable decision fusion and pattern recognition system for object detection and condition monitoring , 2014 .

[41]  Frank Piessens,et al.  Key Reinstallation Attacks: Forcing Nonce Reuse in WPA2 , 2017, CCS.