Two-input stream deep deconvolution neural network for interpolation and recognition

doi:10.11772/j.issn.1001-9081.2018122555

Journal of Computer Applications ›› 2019, Vol. 39 ›› Issue (8): 2271-2275.DOI: 10.11772/j.issn.1001-9081.2018122555

• Artificial intelligence • Previous Articles Next Articles

Two-input stream deep deconvolution neural network for interpolation and recognition

ZHANG Qiang¹, YANG Jian^1,2, FU Lizhen¹

1. School of Software, North University of China, Taiyuan Shanxi 030051, China;
2. Key Laboratory of Electromagnetic Wave Information of Ministry of Education(Fudan University), Shanghai 200433, China

Received:2018-12-27 Revised:2019-04-09 Online:2019-08-10 Published:2019-04-18
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61602427).

双输入流深度反卷积的插值神经网络

张强¹, 杨剑^1,2, 富丽贞¹

1. 中北大学软件学院, 太原 030051;
2. 电磁波信息科学教育部重点实验室(复旦大学), 上海 200433

通讯作者: 杨剑
作者简介:张强(1993-),男,山西忻州人,硕士研究生,主要研究方向:计算机视觉、图像处理;杨剑(1979-),男,山西临汾人,讲师,博士,CCF会员,主要研究方向:计算机视觉、无线通信;富丽贞(1984-),女,山西大同人,讲师,博士,CCF会员,主要研究方向:人工智能、数据挖掘。
基金资助:
国家自然科学基金资助项目（61602427）。

Abstract

Abstract: It is impractical to have a large size of training dataset in real work for neural network training, so a two-input stream generative neural network which can generate a new image with the given parameters was proposed, hence to augment the training dataset. The framework of the proposed neural network consists of a two-input steam convolution network and a deconvolution network. The two-input steam network has two convolution networks to extract features, and the deconvolution network is connected to the end. Two images with different angle were input into the convolution network to get high-level description, then an interpolation target image with a new perspectives was generated by using the deconvolution network with the above high-level description and set parameters. The experiment results on ShapeNetCore show that on the same dataset, the recognition rate of the proposed network has increased by 20% than the common network framework. This method can enlarge the size of the training dataset and is useful for multi-angle recognition.

Key words: deep learning, artificial intelligence, generative neural network, deconvolution, two-input stream

摘要： 在实际工作中深度学习方法通常不具备大量的训练样本，因此提出了双输入流深度反卷积生成神经网络的构架，依据给定的条件产生新的目标图像，从而扩充训练样本集。该神经网络的整体架构由双输入的卷积网络和一个反卷积网络输出构成，其中双输入卷积网络接收目标物体不同视角的两张图片并提取抽象特征，而反卷积网络则利用抽象特征和设定的参数产生新的插值目标图像。在ShapeNetCore数据集上的实验结果显示，在相同数量的训练样本空间中，与未扩展数据集的卷积网络相比，双输入流深度反卷积生成神经网络的识别率提高了20%左右。结果表明，双输入流深度反卷积生成神经网络无需输入目标物类别，可生成新参数条件下的目标图像，扩充训练样本空间，从而提高识别率，可用于少样本的目标物多角度识别。

关键词: 深度学习, 人工智能, 生成神经网络, 反卷积, 双输入流

CLC Number:

TP183

ZHANG Qiang, YANG Jian, FU Lizhen. Two-input stream deep deconvolution neural network for interpolation and recognition[J]. Journal of Computer Applications, 2019, 39(8): 2271-2275.

张强, 杨剑, 富丽贞. 双输入流深度反卷积的插值神经网络[J]. 计算机应用, 2019, 39(8): 2271-2275.

References

[1] DING J, CHEN B, LIU H, et al. Convolutional neural network with data augmentation for SAR target recognition[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(3):364-368.
[2] 林懿伦,戴星原,李力,等.人工智能研究的新前线:生成式对抗网络[J].自动化学报,2018,44(5):775-792. (LIN Y L, DAI X Y, LI L, et al. The new frontier of AI research:generative adversarial networks[J]. Acta Automatica Sinica, 2018, 44(5):775-792.)
[3] 朱俊鹏, 赵洪利, 杨海涛. 基于卷积神经网络的视差图生成技术[J]. 计算机应用, 2018, 38(1):255-259. (ZHU J P, ZHAO H L, YANG H T. Disparity map generation technology based on convolutional neural network[J]. Journal of Computer Applications, 2018, 38(1):255-259.)
[4] 陈文兵,管正雄,陈允杰.基于条件生成式对抗网络的数据增强方法[J].计算机应用,2018,38(11):3305-3311. (CHEN W B, GUAN Z X, CHEN Y J. Data augmentation method based on generative adversarial network model[J]. Journal of Computer Applications, 2018, 38(11):3305-3311.)
[5] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[6] SALAKHUTDINOV R R, HINTON G E. Deep Boltzmann machines[C/OL]//Proceedings of the 12th International Conference on Artificial Intelligence and Statistics, 2009[2018-12-09]. http://proceedings.mlr.press/v5/salakhutdinov09a/salakhutdinov09a.pdf.
[7] BENGIO Y, THIBODEAU-LAUFER É, ALAIN G, et al. Deep generative stochastic networks trainable by backprop[C/OL]//Proceedings of the 31st International Conference on Machine Learning, 2014[2014-05-24]. https://arxiv.org/abs/1306.1091.
[8] REZENDE D J, MOHAMED S, WIERSTRA D. Stochastic backpropagation and approximate inference in deep generative models[J]. arXiv E-print, 2014:arXiv:1401.4082.
[9] KINGMA D P, WELLING M. Auto-encoding variational Bayes[J]. arXiv E-print, 2014:arXiv:1312.6114.
[10] DOSOVITSKIY A, SPRINGENBERG J T, TATARCHENKO M, et al. Learning to generate chairs, tables and cars with convolutional networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):692-705.
[11] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge, MA:MIT Press, 2014:2672-2680.
[12] ZHU Z, LUO P, WANG X, et al. Multi-view perceptron:a deep model for learning face identity and view representations[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge, MA:MIT Press, 2014:217-225.
[13] LI F-F. Knowledge transfer in learning to recognize visual objects classes[C]//Proceedings of the 2006 International Conference on Development and Learning. Washington, DC:IEEE Computational Intelligence Society, 2006:1-51.
[14] LI F, ROB F, PIETRO P. One-Shot learning of object categories[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(4):594-611.
[15] SANTORO A, BARTUNOV S, BOTVINICK M, et al. Meta-learning with memory-augmented neural networks[C]//Proceedings of the 33rd International Conference on Machine Learning. New York:International Machine Learning Society, 2016, 48:1842-1850.
[16] ANDRYCHOWICZ M, DENIL M, COLMENAREJO S G, et al. Learning to learn by gradient descent by gradient descent[C]//Proceedings of the 30th Conference on Neural Information Processing Systems. La Jolla, CA:Neural Information Processing Systems Foundation, 2016:3981-3989.
[17] RAVI S, LAROCHELLE H. Optimization as a model for few-shot learning[C/OL]//Proceedings of the 5nd International Conference on Learning Representations. 2017[2018-10-24]. https://openreview.net/pdf?id=rJY0-Kcll.

Two-input stream deep deconvolution neural network for interpolation and recognition

双输入流深度反卷积的插值神经网络

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

[1]	CHEN Chengrui, SUN Ning, HE Shibiao, LIAO Yong. Deep learning-based joint channel estimation and equalization algorithm for C-V2X communications [J]. Journal of Computer Applications, 2021, 41(9): 2687-2693.
[2]	XIE Defeng, JI Jianmin. Syntax-enhanced semantic parsing with syntax-aware representation [J]. Journal of Computer Applications, 2021, 41(9): 2489-2495.
[3]	DAI Yurou, YANG Qing, ZHANG Fengli, ZHOU Fan. Trajectory prediction model of social network users based on self-supervised learning [J]. Journal of Computer Applications, 2021, 41(9): 2545-2551.
[4]	LIU Zichen, LI Xiaojuan, WEI Wei. Automatic patent price evaluation based on recurrent neural network [J]. Journal of Computer Applications, 2021, 41(9): 2532-2538.
[5]	ZHAO Hong, KONG Dongyi. Chinese description of image content based on fusion of image feature attention and adaptive attention [J]. Journal of Computer Applications, 2021, 41(9): 2496-2503.
[6]	XU Jianglang, LI Linyan, WAN Xinjun, HU Fuyuan. Indoor scene recognition method combined with object detection [J]. Journal of Computer Applications, 2021, 41(9): 2720-2725.
[7]	ZHENG Zhiqiang, HU Xin, WENG Zhi, WANG Yuhe, CHENG Xi. Cattle eye image feature extraction method based on improved DenseNet [J]. Journal of Computer Applications, 2021, 41(9): 2780-2784.
[8]	CAO Yuhong, XU Hai, LIU Sun'ao, WANG Zixiao, LI Hongliang. Review of deep learning-based medical image segmentation [J]. Journal of Computer Applications, 2021, 41(8): 2273-2287.
[9]	DING Yin, SANG Nan, LI Xiaoyu, WU Feizhou. Prediction method of capacity data in telecom industry based on recurrent neural network [J]. Journal of Computer Applications, 2021, 41(8): 2373-2378.
[10]	QIN Binbin, PENG Liangkang, LU Xiangming, QIAN Jiangbo. Research progress on driver distracted driving detection [J]. Journal of Computer Applications, 2021, 41(8): 2330-2337.
[11]	HE Zhenghai, XIAN Yantuan, WANG Meng, YU Zhengtao. Case reading comprehension method combining syntactic guidance and character attention mechanism [J]. Journal of Computer Applications, 2021, 41(8): 2427-2431.
[12]	LI Yafang, LIANG Ye, FENG Weiwei, ZU Baokai, KANG Yujian. Deep network embedding method based on community optimization [J]. Journal of Computer Applications, 2021, 41(7): 1956-1963.
[13]	WANG Yue, JIANG Yiming, LAN Julong. Intrusion detection based on improved triplet network and K-nearest neighbor algorithm [J]. Journal of Computer Applications, 2021, 41(7): 1996-2002.
[14]	DU Yan, LYU Liangfu, JIAO Yichen. Fuzzy prototype network based on fuzzy reasoning [J]. Journal of Computer Applications, 2021, 41(7): 1885-1890.
[15]	HOU Xiaohan, JIN Guodong, TAN Lining, XUE Yuanliang. Synthetic aperture radar ship detection method based on self-adaptive and optimal features [J]. Journal of Computer Applications, 2021, 41(7): 2150-2155.