PMID- 29690629
OWN - NLM
STAT- PubMed-not-MEDLINE
DCOM- 20180503
LR  - 20180518
IS  - 1424-8220 (Electronic)
IS  - 1424-8220 (Linking)
VI  - 18
IP  - 4
DP  - 2018 Apr 23
TI  - Distinguishing Computer-Generated Graphics from Natural Images Based on Sensor 
      Pattern Noise and Deep Learning.
LID - 10.3390/s18041296 [doi]
LID - 1296
AB  - Computer-generated graphics (CGs) are images generated by computer software. The 
      rapid development of computer graphics technologies has made it easier to 
      generate photorealistic computer graphics, and these graphics are quite difficult 
      to distinguish from natural images (NIs) with the naked eye. In this paper, we 
      propose a method based on sensor pattern noise (SPN) and deep learning to 
      distinguish CGs from NIs. Before being fed into our convolutional neural network 
      (CNN)-based model, these images—CGs and NIs—are clipped into image 
      patches. Furthermore, three high-pass filters (HPFs) are used to remove 
      low-frequency signals, which represent the image content. These filters are also 
      used to reveal the residual signal as well as SPN introduced by the digital 
      camera device. Different from the traditional methods of distinguishing CGs from 
      NIs, the proposed method utilizes a five-layer CNN to classify the input image 
      patches. Based on the classification results of the image patches, we deploy a 
      majority vote scheme to obtain the classification results for the full-size 
      images. The experiments have demonstrated that (1) the proposed method with three 
      HPFs can achieve better results than that with only one HPF or no HPF and that 
      (2) the proposed method with three HPFs achieves 100% accuracy, although the NIs 
      undergo a JPEG compression with a quality factor of 75.
FAU - Yao, Ye
AU  - Yao Y
AUID- ORCID: 0000-0002-7012-9307
AD  - School of CyberSpace, Hangzhou Dianzi University, Hangzhou 310018, China. 
      yyaoprivate@gmail.com.
AD  - Shenzhen Key Laboratory of Media Security, Shenzhen University, Shenzhen 518060, 
      China. yyaoprivate@gmail.com.
FAU - Hu, Weitong
AU  - Hu W
AD  - School of CyberSpace, Hangzhou Dianzi University, Hangzhou 310018, China. 
      hwt@hdu.edu.cn.
FAU - Zhang, Wei
AU  - Zhang W
AD  - School of Computer Science and Technology, Hangzhou Dianzi University, Hangzhou 
      310018, China. magherozhw@hdu.edu.cn.
FAU - Wu, Ting
AU  - Wu T
AD  - School of CyberSpace, Hangzhou Dianzi University, Hangzhou 310018, China. 
      wuting@hdu.edu.cn.
FAU - Shi, Yun-Qing
AU  - Shi YQ
AD  - Department of Electrical and Computer Engineering, New Jersey Institute of 
      Technology, Newark, NJ 07102, USA. shi@njit.edu.
LA  - eng
PT  - Journal Article
DEP - 20180423
PL  - Switzerland
TA  - Sensors (Basel)
JT  - Sensors (Basel, Switzerland)
JID - 101204366
PMC - PMC5948567
OTO - NOTNLM
OT  - computer-generated graphics
OT  - convolutional neural network
OT  - image forensics
OT  - natural images
OT  - sensor pattern noise
COIS- The authors declare no conflict of interest.
EDAT- 2018/04/25 06:00
MHDA- 2018/04/25 06:01
PMCR- 2018/04/01
CRDT- 2018/04/26 06:00
PHST- 2018/03/07 00:00 [received]
PHST- 2018/04/19 00:00 [revised]
PHST- 2018/04/20 00:00 [accepted]
PHST- 2018/04/26 06:00 [entrez]
PHST- 2018/04/25 06:00 [pubmed]
PHST- 2018/04/25 06:01 [medline]
PHST- 2018/04/01 00:00 [pmc-release]
AID - s18041296 [pii]
AID - sensors-18-01296 [pii]
AID - 10.3390/s18041296 [doi]
PST - epublish
SO  - Sensors (Basel). 2018 Apr 23;18(4):1296. doi: 10.3390/s18041296.