自适应深度卷积神经网络模型构建方法

doi:10.13190/j.jbupt.2017.04.016

北京邮电大学学报 ›› 2017, Vol. 40 ›› Issue (4): 98-103.doi: 10.13190/j.jbupt.2017.04.016

自适应深度卷积神经网络模型构建方法

邹国锋¹, 傅桂霞¹, 王科俊², 高明亮¹, 申晋¹

1. 山东理工大学电气与电子工程学院, 山东淄博 255049;
2. 哈尔滨工程大学自动化学院, 哈尔滨 150001

收稿日期:2017-03-19 出版日期:2017-08-28 发布日期:2017-07-10
作者简介:邹国锋(1984-),男,讲师,E-mail:zgf841122@163.com.
基金资助:
山东省自然科研基金项目（ZR2015FL029，ZR2016FL14）；国家自然科学基金项目（61601266）；中国博士后科学基金项目（2017M612306）

Construction Method of Adaptive Deep Convolutional Neural Network Model

ZOU Guo-feng¹, FU Gui-xia¹, WANG Ke-jun², GAO Ming-liang¹, SHEN Jin¹

1. College of Electrical and Electronic Engineering, Shandong University of Technology, Shandong Zibo 255049, China;
2. College of Automation, Harbin Engineering University, Harbin 150001, China

Received:2017-03-19 Online:2017-08-28 Published:2017-07-10

摘要/Abstract

摘要： 针对传统卷积神经网络（CNN）模型构建过度依赖经验知识、参数多、训练难度大等缺点，同时鉴于复杂多类问题的CNN模型构建策略的重要价值，提出一种自适应深度CNN模型构建方法.首先，将初始网络模型的卷积层和池化层设置为仅含一幅特征图；然后，以网络收敛速度为评价指标，对网络进行全局扩展，全局扩展后，根据交叉验证样本识别率控制网络展开局部扩展，直到识别率达到预设期望值后停止局部网络学习；最后，针对新增训练样本，通过拓展新支路实现网络结构的自适应增量学习.通过图像识别实验验证了所提算法在网络训练时间和识别效果上的优越性.

关键词: 深度卷积神经网络, 自适应模型构建, 深度学习, 图像识别

Abstract: The construction process of traditional convolutional neural network (CNN) model has many shortcomings, such as over reliance on experience knowledge, a lot of parameters and difficult to training. At the same time, in view of the important value of constructing strategy of CNN model in complex multi-class problems, a new construction method of adaptive deep CNN model was proposed. First, the convolution layer and pooling layer of the initial CNN model are set only to include one feature map; and then, the convergence rate of CNN is used as evaluation index, the global expansion of network is carried out. After global expansion, the CNN is controlled to local expansion according to the recognition rate of cross validation samples. The local network learning is stopped until the recognition rate reaches the expected value. Finally, the training process for new samples, the adaptive incremental learning of network structure is realized by expanding some new branches. The superiority of proposed algorithm in network training time and recognition effect is verified through some image recognition experiments.

Key words: deep convolutional neural network, adaptive model constructing, deep learning, image recognition

中图分类号:

TP391.41

邹国锋, 傅桂霞, 王科俊, 高明亮, 申晋. 自适应深度卷积神经网络模型构建方法[J]. 北京邮电大学学报, 2017, 40(4): 98-103.

ZOU Guo-feng, FU Gui-xia, WANG Ke-jun, GAO Ming-liang, SHEN Jin. Construction Method of Adaptive Deep Convolutional Neural Network Model[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2017, 40(4): 98-103.

参考文献

[1] Hinton G, Salakhutdinov R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 5786(313):504-507.
[2] Arel I, Rose D C, Karnowski T P. Deep machine learning:a new frontier in artificial intelligence research[J]. IEEE Computational Intelligence Magazine, 2010, 5(4):13-18.
[3] He YunLong, Kavukcuoglu K, Wang Yun, et al. Unsupervised feature learning by deep sparse coding[C]//2014 SIAM International Conference on Data Mining, Philadelphia, SIAM. 2014:902-910.
[4] Kim L W, Asaad S, Linsker R. A fully pipelined FPGA architecture of a factored restricted boltzmann machine artificial neural network[J]. ACM Transactions on Reconfigurable Technology and Systems, 2014, 7(1):1-23.
[5] Hinton G, Osindero S, Teh Y. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7):1527-1554.
[6] Pinheiro P, Collobert R. Recurrent convolutional neural networks for scene labeling[C]//Proceedings of The 31^st International Conference on Machine Learning. Beijing:International Machine Learning Society, 2014:151-159.
[7] Huang Yongzhen, Huang Kaiqi, Tao Dacheng, et al. Enhanced biologically inspired model for object recognition[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B:Cybernetics, 2011, 41(6):1668-1680.
[8] LeCun Y, Boser B, Denker J S, et al. Back propagation applied to hand written zip code recognition[J]. Neural Computation, 1989, 1(4):541-551.
[9] Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the Advances in Neural Information Processing Systems. New York, NY, ACM, 2012:1097-1105.
[10] Abdel-Hamid O, Mohamed A R, Jiang Hui, et al. Convolutional neural networks for speech recognition[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014, 22(10):1533-1545.
[11] Hamester D, Barros P, Wermter S. Face Expression recognition with a 2-channel convolutional neural network[C]//2015 International Joint Conference on Neural Networks. Piscataway, NJ:IEEE, 2015:1-8.
[12] Xiong Chao, Zhao Xiaowei, Tang Danhang, et al. Conditional convolutional neural network for modality-aware face recognition[C]//2015 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2015:3667-3675.
[13] Pattabhi Ramaiah N, Ijjina E P, Mohan C K. Illumination invariant face recognition using convolutional neural network[C]//2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems. Piscataway, NJ:IEEE, 2015:1-4.
[14] Wolfshaar J V D, Karaaba M F, Wiering M A. Deep convolutional neural networks and support vector machines for gender recognition[C]//2015 IEEE Symposium Series on Computational Intelligence. Piscataway, NJ:IEEE, 2015:188-195.
[15] Li Haoxiang, Lin Zhe, Shen Xiaohui, et al. A Convolutional neural network cascade for face detection[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2015:5325-5334.
[16] Zhang Shaohua, Yang Hua, Yin Zhouping. Multiple deep convolutional neural networks averaging for face alignment[J]. Journal of Electronic Imaging, 2015, 24(3):1-13.
[17] Yun Dong, Yue Wu. Adaptive cascade deep convolutional neural networks for face alignment[J]. Computer Standards & Interfaces, 2015(42):105-112.
[18] Zhang Yuanyuan, Zhao Dong, Sun Jiande, et al. Adaptive convolutional neural network and its application in face recognition[J]. Neural Processing Letters, 2016, 43(2):389-399.
[19] 连自锋, 景晓军, 孙松林, 等. 一种多尺度嵌套卷积神经网络模型[J]. 北京邮电大学学报, 2016, 39(5):1-5. Lian Zifeng, Jing Xiaojun, Sun Songlin, et al. Multi-Scale convolutional neural network model with multilayer maxout networks[J]. Journal of Beijing University of Posts and Telecommunications, 2016, 39(5):1-5.

自适应深度卷积神经网络模型构建方法

Construction Method of Adaptive Deep Convolutional Neural Network Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刘阳, 滕颖蕾, 牛涛, 郅佳琳. 基于深度强化学习的滤波器剪枝方案[J]. 北京邮电大学学报, 2023, 46(3): 31-36.
[2]	靳梦凡, 黄智濒, 储志强. 复杂环境下多模态指导的点云补全方法[J]. 北京邮电大学学报, 2023, 46(3): 103-108.
[3]	荣震宇刘建毅. 基于Transformer和MLP的眼底血管分割算法[J]. 北京邮电大学学报, 2023, 46(1): 26-31.
[4]	刘金桐, 杨国兴, 刘晓鸿, 王光宇. 针对心电图分类模型的平滑攻击算法[J]. 北京邮电大学学报, 2022, 45(4): 44-50.
[5]	黄柳婷, 刘可欣, 牛凯, 常春, 贺志强. 基于深度学习的哮喘患者CT影像黏液栓自动识别[J]. 北京邮电大学学报, 2022, 45(4): 58-63.
[6]	柳长源, 虎浩媛, 毕晓君. 双线性融合网络的驾驶员分心行为识别[J]. 北京邮电大学学报, 2022, 45(2): 79-84.
[7]	赵海英, 朱会, 侯小刚. 基于改进EMA单元的传统服饰图像语义分割[J]. 北京邮电大学学报, 2022, 45(1): 69-74.
[8]	张滨宇, 赵衍运, 杜昀昊, 万俊峰, 佟知航. 一种基于深度学习的PCB图像字符检测方法[J]. 北京邮电大学学报, 2022, 45(1): 108-114.
[9]	贾珺, 冯春燕, 夏海轮, 张天魁, 李成钢. 基于样本均衡与特征交互的通信网络故障预测方法[J]. 北京邮电大学学报, 2021, 44(6): 59-66.
[10]	普运伟, 郭江, 刘涛涛, 吴海潇. 基于多学习单元卷积神经网络的雷达辐射源信号识别[J]. 北京邮电大学学报, 2021, 44(6): 74-82.
[11]	王艺霏, 莫爽, 吴文睿, 范少华, 肖丁. 基于内外卷积网络的网络入侵检测[J]. 北京邮电大学学报, 2021, 44(5): 94-100.
[12]	高慧, 张继威, 来扬, 王文东. 深度学习的人体图像半自动标注系统[J]. 北京邮电大学学报, 2021, 44(1): 104-109.
[13]	蒲悦逸, 王文涵, 朱强, 陈朋朋. 基于CNN-ResNet-LSTM模型的城市短时交通流量预测算法[J]. 北京邮电大学学报, 2020, 43(5): 9-14.
[14]	韦娟, 顾兴权, 宁方立. 一种基于EEMD的异常声音识别方法[J]. 北京邮电大学学报, 2020, 43(5): 112-117.
[15]	马向亮, 李冰, 杨丹, 黄克振, 段晓毅. 基于深度学习的类SM4算法S盒逆向分析[J]. 北京邮电大学学报, 2020, 43(5): 118-124.