第2期 毛鸴池，等：基于Faster R-CNN的多任务增强裂缝图像检测方法 ·293· Communications and Microphone Arrays.San Francisco, International Seminar on Application for Technology of USA,2017:46-50. Information and Communication.Semarang,Indonesia, [5]GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich 2018:26-30 feature hierarchies for accurate object detection and se- [15]AKCAY S.KUNDEGORSKI ME,WILLCOCKS C G, mantic segmentation[C]//Proceedings of 2014 IEEE Con- et al.Using deep convolutional neural network architec- ference on Computer Vision and Pattern Recognition(CV- tures for object classification and detection within X-ray PR).Columbus,USA,2014:580-587. baggage security imagery[J].IEEE transactions on in- [6]GIRSHICK R.Fast R-CNN[C]//Proceedings of the 2015 formation forensics and security,2018,13(9):2203-2215. IEEE International Conference on Computer Vision.Santi- [16]RAHMAN M A,WANG Yang.Optimizing intersection- ago,Chile,2015:1440-1448. over-union in deep neural networks for image segmenta- [7]LIU WEI,ANGUELOV D,ERHAN D,et al.SSD:single tion[C]//Proceedings of 12th International Symposium on shot MultiBox detector[C]//Proceedings of the 14th Advances in Visual Computing.Las Vegas,USA,2016: European Conference on Computer Vision.Amsterdam, 234-244. the Netherlands,2016:21-37. [17]ZEILER M D,FERGUS R.Visualizing and understand- [8]REDMON J,FARHADI A.YOLO9000:better,faster, ing convolutional networks[C]//Proceedings of 13th stronger[C]//Proceedings of 2017 IEEE Conference on European Conference on Computer Vision.Zurich, Computer Vision and Pattern Recognition.Honolulu,USA, Switzerland,2014:818-833. 2017:6517-6525 [18]SIMONYAN K,ZISSERMAN A.Very deep convolu- [9]KANG HH,LEE S W.YOU S H,et al.Novel vehicle de- tional networks for large-scale image recognition[Cl//Pro- tection system based on stacked DoG kernel and Ada- ceedings of 3rd International Conference on Learning Boost[J].PLoS one,2018,13(3):e0193733. Representations.San Diego,USA,2015. [10]DAI Wenyuan,YANG Qiang,XUE Guirong,et al. 作者简介： Boosting for transfer learning[C]//Proceedings of the 24th 毛莺池，教授，博士，博士生导师 International Conference on Machine Learning.New 主要研究方向为云计算和边缘计算、 York.USA.2007:193-200. 分布式技术和物联网数据分析。曾获 [11]AL-STOUHI S,REDDY C K.Adaptive boosting for 大禹水利科学技术奖一等奖：华能集 团科技进步奖二等奖：江苏省科学技 transfer learning using dynamic updates[C]//Joint 术奖三等奖：2018年度江苏省计算机 European Conference on Machine Learning and Know- 学会优秀科技工作者。发表学术论文 ledge Discovery in Databases.Berlin,Germany,2011: 50余篇。 60-75. [12]郭勇.基于单源及多源的迁移学习方法研究D1.西安： 唐江红，硕士研究生，主要研究方 西安电子科技大学，2013. 向为图像处理。 GUO Yong.Research of transfer learning based on single-source and multi-source[D].Xi'an:Xidian Uni- versity,2013. [13]HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Deep residual learning for image recognition[C]//Proceed- ings of the 2016 IEEE Conference on Computer Vision 王静，硕士研究生，主要研究方向 and Pattern Recognition.Las Vegas,USA,2016: 为图像处理。 770-778. [14]WICAKSONO Y A,RIZALDY A,FAHRIAH S,et al. Improve image segmentation based on closed form mat- ting using K-means clustering[C]//Proceedings of 2017Communications and Microphone Arrays. San Francisco, USA, 2017: 46−50. GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and se￾mantic segmentation[C]//Proceedings of 2014 IEEE Con￾ference on Computer Vision and Pattern Recognition (CV￾PR). Columbus, USA, 2014: 580−587. [5] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Santi￾ago, Chile, 2015: 1440−1448. [6] LIU WEI, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Amsterdam, the Netherlands, 2016: 21−37. [7] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 6517−6525. [8] KANG H H, LEE S W, YOU S H, et al. Novel vehicle de￾tection system based on stacked DoG kernel and Ada￾Boost[J]. PLoS one, 2018, 13(3): e0193733. [9] DAI Wenyuan, YANG Qiang, XUE Guirong, et al. Boosting for transfer learning[C]//Proceedings of the 24th International Conference on Machine Learning. New York, USA, 2007: 193−200. [10] AL-STOUHI S, REDDY C K. Adaptive boosting for transfer learning using dynamic updates[C]//Joint European Conference on Machine Learning and Know￾ledge Discovery in Databases. Berlin, Germany, 2011: 60−75. [11] 郭勇. 基于单源及多源的迁移学习方法研究 [D]. 西安: 西安电子科技大学, 2013. GUO Yong. Research of transfer learning based on single-source and multi-source[D]. Xi’an: Xidian Uni￾versity, 2013. [12] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceed￾ings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA, 2016: 770−778. [13] WICAKSONO Y A, RIZALDY A, FAHRIAH S, et al. Improve image segmentation based on closed form mat￾ting using K-means clustering[C]//Proceedings of 2017 [14] International Seminar on Application for Technology of Information and Communication. Semarang, Indonesia, 2018: 26−30. AKCAY S, KUNDEGORSKI M E, WILLCOCKS C G, et al. Using deep convolutional neural network architec￾tures for object classification and detection within X-ray baggage security imagery[J]. IEEE transactions on in￾formation forensics and security, 2018, 13(9): 2203–2215. [15] RAHMAN M A, WANG Yang. Optimizing intersection￾over-union in deep neural networks for image segmenta￾tion[C]//Proceedings of 12th International Symposium on Advances in Visual Computing. Las Vegas, USA, 2016: 234−244. [16] ZEILER M D, FERGUS R. Visualizing and understand￾ing convolutional networks[C]//Proceedings of 13th European Conference on Computer Vision. Zurich, Switzerland, 2014: 818−833. [17] SIMONYAN K, ZISSERMAN A. Very deep convolu￾tional networks for large-scale image recognition[C]//Pro￾ceedings of 3rd International Conference on Learning Representations. San Diego, USA, 2015. [18] 作者简介： 毛莺池，教授，博士，博士生导师， 主要研究方向为云计算和边缘计算、 分布式技术和物联网数据分析。曾获 大禹水利科学技术奖一等奖；华能集 团科技进步奖二等奖；江苏省科学技 术奖三等奖；2018 年度江苏省计算机 学会优秀科技工作者。发表学术论文 50 余篇。 唐江红，硕士研究生，主要研究方 向为图像处理。 王静，硕士研究生，主要研究方向 为图像处理。 第 2 期 毛莺池，等：基于 Faster R-CNN 的多任务增强裂缝图像检测方法 ·293·