第3期 马岭，等：基于小样本学习的LCD产品缺陷自动检测方法 ·567· [6]马岭，蒋慧琴，刘玉敏.基于局部特征的驾驶证自动识别 ment Analysis and Recognition.Tunis,Tunisia,2015: 系统.郑州大学学报（工学版），2017,38(5)13-17,22. 846-850 MA Ling,JIANG Huiqin,LIU Yumin.Automatic recogni- [16]SZEGEDY C.IOFFE S,VANHOUCKE V,et al.Incep- tion system of driver's license based on local features[J]. tion-V4,inception-ResNet and the impact of residual con- Journal of Zhengzhou University (engineering science edi- nections on learning[C]//Proceedings of the Thirty-First tion,2017,38(5):13-17,22. AAAI Conference on Artificial Intelligence.San Fran- [7]LANGKVIST M.KARLSSON L.LOUTFI A.A review of cisco.USA,2017. unsupervised feature learning and deep learning for time- [17]Lin M,Chen Q,Yan S.Network In Network[Cl//Proceed- series modeling[J].Pattern recognition letters,2014,42: ings of the 2th International Conference on Learning Rep- 11-24. resentations.Banff,Canada,2016. [8]KRIZHEVSKY A,SUTSKEVER I.HINTON G E.Im- [18]SZEGEDY C,LIU Wei,JIA Yangqing,et al.Going deep- ageNet classification with deep convolutional neural net- er with convolutions[C]//Proceedings of 2015 IEEE Con- works[C]//Proceedings of the 25th International Confer- ference on Computer Vision and Pattern Recognition.Bo- ence on Neural Information Processing Systems.Siem ston.USA,2015:1-9. Reap,Cambodia,2012:1097-1105. [19]Radford A,Metz L,Chintala S.Unsupervised Representa- [9]BENGIO Y,COURVILLE A,VINCENT P.Representa- tion Learning with Deep Convolutional Generative Ad- tion learning:a review and new perspectives[J].IEEE versarial Networks[Cl//Proceedings of the 4th Internation- transactions on pattern analysis and machine intelligence, al Conference on Learning Representations.San Juan,Pu- 2013,35(8):1798-1828. erto Rico.2016. [10]DENG Jia,BERG A,SATHEESH S,et al.Large scale [20]GOODFELLOW I J.POUGET-ABADIE J.MIRZA M. visual recognition challenge[EB/OL].[2013-11-14]. et al.Generative adversarial nets[C]//Proceedings of the http://image-net.org/challenges/LSVRC/2013/. 27th International Conference on Neural Information Pro- [11]EVERINGHAM M,ALI ESLAMI S M,VAN GOOL L, et al.The PASCAL visual object classes challenge:a ret- cessing Systems.Montreal,Canada,2014:2672-2680. rospective[J].International journal of computer vision, 作者简介： 2015,111(1:98-136. 马岭，教授.博士，主要研究方向 [12]TAJBAKHSH N,SUZUKI K.Comparing two classes of 为深度学习和机器视觉。主持NSFC end-to-end machine-learning models in lung nodule de- 河南联合基金重点项目1项，获河南 tection and classification:MTANNs vs.CNNs[J].Pattern 省科技进步一等奖1项，获发明专利 授权5项。发表学术论文30余篇，出 recognition,2017,63:476-486. 版专著1部。 [13]郑胤，陈权崎，章毓晋.深度学习及其在日标和行为识 别中的新进展[.中国图象图形学报，2014,19(2)： 175-184 鲁越，硕士研究生，主要研究方向 ZHENG Yin,CHEN Quanqi,ZHANG Yujin.Deep learn- 为深度学习和机器视觉。 ing and its new progress in object and behavior recogni- tion[J].Journal of image and graphics,2014,19(2): 175-184 [14]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Im- ageNet classification with deep convolutional neural net- works[J].Communications of the ACM,2017,60(6): 蒋慧琴，教授，博士，主要研究方 向为深度学习和医疗人工智能。主持 84-90. 和参与完成国家自然科学基金面上项 [15]ZHONG Zhuoyao,JIN Lianwen,XIE Zecheng.High per- 目2项、省部级项目4项，获发明专利 formance offline handwritten chinese character recogni- 授权3项。发表学术论文50余篇。 tion using googlenet and directional feature maps[C//Pro- ceedings of 2015 13th International Conference on Docu-马岭, 蒋慧琴, 刘玉敏. 基于局部特征的驾驶证自动识别 系统 [J]. 郑州大学学报(工学版), 2017, 38(5): 13–17, 22. MA Ling, JIANG Huiqin, LIU Yumin. Automatic recogni￾tion system of driver’s license based on local features[J]. Journal of Zhengzhou University (engineering science edi￾tion), 2017, 38(5): 13–17, 22. [6] LÄNGKVIST M, KARLSSON L, LOUTFI A. A review of unsupervised feature learning and deep learning for time￾series modeling[J]. Pattern recognition letters, 2014, 42: 11–24. [7] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Im￾ageNet classification with deep convolutional neural net￾works[C]//Proceedings of the 25th International Confer￾ence on Neural Information Processing Systems. Siem Reap, Cambodia, 2012: 1097−1105. [8] BENGIO Y, COURVILLE A, VINCENT P. Representa￾tion learning: a review and new perspectives[J]. IEEE transactions on pattern analysis and machine intelligence, 2013, 35(8): 1798–1828. [9] DENG Jia, BERG A, SATHEESH S, et al. Large scale visual recognition challenge[EB/OL]. [2013-11-14]. http://image-net.org/challenges/LSVRC/2013/. [10] EVERINGHAM M, ALI ESLAMI S M, VAN GOOL L, et al. The PASCAL visual object classes challenge: a ret￾rospective[J]. International journal of computer vision, 2015, 111(1): 98–136. [11] TAJBAKHSH N, SUZUKI K. Comparing two classes of end-to-end machine-learning models in lung nodule de￾tection and classification: MTANNs vs. CNNs[J]. Pattern recognition, 2017, 63: 476–486. [12] 郑胤, 陈权崎, 章毓晋. 深度学习及其在目标和行为识 别中的新进展 [J]. 中国图象图形学报, 2014, 19(2): 175–184. ZHENG Yin, CHEN Quanqi, ZHANG Yujin. Deep learn￾ing and its new progress in object and behavior recogni￾tion[J]. Journal of image and graphics, 2014, 19(2): 175–184. [13] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Im￾ageNet classification with deep convolutional neural net￾works[J]. Communications of the ACM, 2017, 60(6): 84–90. [14] ZHONG Zhuoyao, JIN Lianwen, XIE Zecheng. High per￾formance offline handwritten chinese character recogni￾tion using googlenet and directional feature maps[C]//Pro￾ceedings of 2015 13th International Conference on Docu- [15] ment Analysis and Recognition. Tunis, Tunisia, 2015: 846−850. SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Incep￾tion-V4, inception-ResNet and the impact of residual con￾nections on learning[C]//Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Fran￾cisco, USA, 2017. [16] Lin M, Chen Q, Yan S. Network In Network[C]//Proceed￾ings of the 2th International Conference on Learning Rep￾resentations. Banff, Canada, 2016. [17] SZEGEDY C, LIU Wei, JIA Yangqing, et al. Going deep￾er with convolutions[C]//Proceedings of 2015 IEEE Con￾ference on Computer Vision and Pattern Recognition. Bo￾ston, USA, 2015: 1−9. [18] Radford A, Metz L, Chintala S. Unsupervised Representa￾tion Learning with Deep Convolutional Generative Ad￾versarial Networks[C]//Proceedings of the 4th Internation￾al Conference on Learning Representations. San Juan, Pu￾erto Rico, 2016. [19] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Pro￾cessing Systems. Montreal, Canada, 2014: 2672–2680. [20] 作者简介： 马岭，教授，博士，主要研究方向 为深度学习和机器视觉。主持 NSFC- 河南联合基金重点项目 1 项，获河南 省科技进步一等奖 1 项，获发明专利 授权 5 项。发表学术论文 30 余篇，出 版专著 1 部。 鲁越，硕士研究生，主要研究方向 为深度学习和机器视觉。 蒋慧琴，教授，博士，主要研究方 向为深度学习和医疗人工智能。主持 和参与完成国家自然科学基金面上项 目 2 项、省部级项目 4 项，获发明专利 授权 3 项。发表学术论文 50 余篇。 第 3 期 马岭，等：基于小样本学习的 LCD 产品缺陷自动检测方法 ·567·