基于卷积神经网络的图像生成方式分类方法

doi:10.11959/j.issn.2096-109x.2016.00096

Chinese Journal of Network and Information Security ›› 2016, Vol. 2 ›› Issue (9): 40-48.doi: 10.11959/j.issn.2096-109x.2016.00096

• academic paper • Previous Articles Next Articles

Image generation classification method based on convolution neural network

Qiao-ling LI^1,²(),Qing-xiao GUAN^1,²,Xian-feng ZHAO^1,²

¹ State Key Laboratory of Information Security,Institute of Information Engineering,Chinese Academy of Sciences,Beijing 100093,China
² University of Chinese Academy of Sciences,Beijing 100049,China

Revised:2016-08-09 Online:2016-09-15 Published:2020-03-26
Supported by:
The National Natural Science Foundation of China(61303259);The National Natural Science Foundation of China(U1536105);The Strategic Pilot Sci-ence and Technology Project of the Chinese Academy of Sciences(XDA06030600);The Key Project of Institute of Informa-tion Engineering,Chinese Academy of Sciences(Y5Z0131201)

Abstract

Abstract:

Using convolution neural network which though convolution and pooling extracting features of high dis-tinguish ability and then make fusion for classification of natural images and scanned documents.Experimental re-sults show that the classification accuracy of the proposed classification method is more than 93% on the SKL image database.The model is highly robust to font sizes and image formats.Through contrast experiment validated that preprocessing of image has a positive effect on the accuracy of the model and the time cost on training.

Key words: convolution neural network, image generation mode, content pattern classification, multimedia security

CLC Number:

TP37

Qiao-ling LI,Qing-xiao GUAN,Xian-feng ZHAO. Image generation classification method based on convolution neural network[J]. Chinese Journal of Network and Information Security, 2016, 2(9): 40-48.

Figures/Tables 10

References 12

[1]	WANG Y , MOULIN P . On discrimination between photorealistic and photographic images[C]// IEEE International Conference on Acoustics Speech and Signal Processing. 2006.
[2]	LYU S , FARID H . How realistic is photorealistic[J]. IEEE Interna-tional Conference on Signal Processing, 2005,53(2):845-850.
[3]	ZHU J Y , KRAHENBUHL P , SHECHTMAN E , et al. Learning a discriminative model for the perception of realism in composite images[C]// IEEE International Conference on Computer Vision. 2015:3943-3951.
[4]	KHANNA N , CHIU G T C , ALLEBACH J P , et al. Forensic tech-niques for classifying scanner,computer generated and digital cam-era images[C]// IEEE International Conference on Acoustics,Speech and Signal Processing. 2008:1653-1656.
[5]	MAHENDRAN A , VEDALDI A . Understanding deep image rep-resentations by inverting them[C]// IEEE Conference on Computer Vision and Pattern Recognition. 2015:5188-5196.
[6]	LECUN Y , BOTTOU L , BENGIO Y , et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998,86(11):2278-2324.
[7]	KUSSUL E , BAIDYK T , WUNSCH II D C . Permutation coding technique for image recognition system standard[M] Neural Networks and Micromechanics. 201047-73.
[8]	KAIMING H , XIANGYU Z , SHAOQING R , et al. Delving deep into rectifiers:surpassing human-level performance on imagenet classification[C]// IEEE International Conference on Computer Vi-sion. 2015:1026-1034.
[9]	NAIR V , HINTON G E . Rectified linear units improve restricted boltzmann machines[C]// The 27th International Conference on Machine Learning. 2010:807-814.
[10]	SRIVASTAVA N , HINTON G , KRIZHEVSKY A , et al. Dropout:a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014,15(1):1929-1958.
[11]	SUTSKEVER I , MARTENS J , DAHL G , et al. On the importance of initialization and momentum in deep learning[C]// The 30th In-ternational Conference on Machine Learning. 2013:1139-1147.
[12]	JIA Y , SHELHAMER E , DONAHUE J , et al. Caffe:convolutional architecture for fast feature embedding[C]// The ACM International Conference on Multimedia. 2014:675-678.

Metrics

Recommended 0

No Suggested Reading articles found!

图像类型	设备名称
自然图像	Canon M2、Canon 450D、Nikon D700、Panasonic GF5
扫描文档	Brother MFC-7860DN

类型	核尺寸/步长	输出尺寸
Convolution	12×12/5	101×101×96
Max pool	3×3/2	50×50×96
Convolution	5×5/1	50×50×256
Max pool	4×4/2	24×24×256
Convolution	3×3/1	24×24×256
Max pool	4×4/2	11×11×256
Fully connection	—	1×1×512
Fully connection	—	1×1×2
Softmax	—	1×1×2

		预测类型
实际类型
实际类型
	自然图像		扫描文档
自然图像	49		1
扫描文档	5		45

Image generation classification method based on convolution neural network

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 12

Related Articles 4

Metrics

Recommended 0

[1]	Rongna XIE, Zhuhong MA, Zongyu LI, Ye TIAN. Encrypted traffic classification method based on convolutional neural network [J]. Chinese Journal of Network and Information Security, 2022, 8(6): 84-91.
[2]	Tong QIAO, Hongwei YAO, Binmin PAN, Ming XU, Yanli CHEN. Research progress of digital image forensic techniques based on deep learning [J]. Chinese Journal of Network and Information Security, 2021, 7(5): 13-28.
[3]	Bo XIE,Guowei SHEN,Chun GUO,Yan ZHOU,Miao YU. Cyber security entity recognition method based on residual dilation convolution neural network [J]. Chinese Journal of Network and Information Security, 2020, 6(5): 126-138.
[4]	Xiaobin ZHANG, Fucai CHEN, Ruiyang HUANG. Relation extraction based on CNN and Bi-LSTM [J]. Chinese Journal of Network and Information Security, 2018, 4(9): 44-51.