第1期 徐立芳，等：基于改进的YOLOv:3算法的乳腺超声肿瘤识别 ·29· 类D].重庆：重庆大学，2018 [20]REDMON J,FARHADI A.Yolov3:an incremental im- XIAO Ting.Breast ultrasound image classification on provement[J].ar Xiv preprint:arXiv:1804.02767,2018. deep feature based transfer learning and feature fusion[D]. [21]HU Jie,SHEN Li,SUN Gang.Squeeze-and-excitation Chongqing:Chongqing University,2018. networks[C]//Proceedings of the 2018 IEEE/CVF Confer- [14]HAN S,KANG H K,JEONG J Y,et al.A deep learning ence on Computer Vision and Pattern Recognition.Salt framework for supporting the classification of breast le- Lake City,USA,2018:7132-7141. sions in ultrasound imagesfJ].Physics in medicine and [22]GAO Shanghua,CHENG Mingming,ZHAO Kai,et al. biology,2017,62(19y:7714-7728. [15]梁舒.基于残差学习U型卷积神经网络的乳腺超声图 Res2Net:a new multi-scale backbone architecture[J]. 像肿瘤分割研究D1.广州：华南理工大学，2018 IEEE transactions on pattern analysis and machine intelli- LIANG Shu.Research on breast ultrasound image seg- gence,2019:l-10 mentaion based on residual U-shaped convolution neural 作者简介： network[D].Guangzhou:South China University of Tech- 徐立芳，讲师，博士，主要研究方 nology,2018. 向为智能控制、机器视觉与机器认知 [16]王恒立.基于全卷积网络的乳腺超声图像语义分割方 人机混合智能。主持、参与省部级科 法D].哈尔滨：哈尔滨工业大学，2018. 研项目10项，授权发明专利6项。发 WANG Hengli.Semantic segmentation method for breast 表学术论文20余篇。 ultrasound images based on fully convolutional network[D].Harbin:Harbin Institute of Technology, 2018. 傅智杰，硕士研究生，主要研究方 [17]YAP M H,GOYAL M,OSMAN F M,et al.End-to-end 向为深度学习、计算机视觉、医学 breast ultrasound lesions recognition with a deep learning 影像。 approach[C]//Proceedings Volume 10578,Medical Ima- ging 2018:Biomedical Applications in Molecular,Struc- tural,and Functional Imaging.Houston,Texas,United States,2018:1057819. [18]CHIAO J Y.CHEN K Y.LIAO K Y.et al.Detection and 莫宏伟，教授，博士生导师，主要 classification the breast tumors using mask R-CNN on 研究方向为类脑计算与人工智能、机 sonograms[J].Medicine,2019,98(19):e15200. 器视觉与机器认知、人机混合智能。 [19]SHIN S Y,LEE S,YUN I D,et al.Joint weakly and semi- 主持省部级科研项目24项，授权发明 supervised deep learning for localization and classifica- 专利10项。发表学术论文80余篇。 tion of masses in breast ultrasound images[J].IEEE trans- actions on medical imaging,2019,38(3):762-774类 [D]. 重庆：重庆大学, 2018. XIAO Ting. Breast ultrasound image classification on deep feature based transfer learning and feature fusion[D]. Chongqing: Chongqing University, 2018. HAN S, KANG H K, JEONG J Y, et al. A deep learning framework for supporting the classification of breast le￾sions in ultrasound images[J]. Physics in medicine and biology, 2017, 62(19): 7714–7728. [14] 梁舒. 基于残差学习 U 型卷积神经网络的乳腺超声图 像肿瘤分割研究 [D]. 广州：华南理工大学, 2018. LIANG Shu. Research on breast ultrasound image seg￾mentaion based on residual U-shaped convolution neural network[D]. Guangzhou: South China University of Tech￾nology, 2018. [15] 王恒立. 基于全卷积网络的乳腺超声图像语义分割方 法 [D]. 哈尔滨：哈尔滨工业大学, 2018. WANG Hengli. Semantic segmentation method for breast ultrasound images based on fully convolutional network[D]. Harbin: Harbin Institute of Technology, 2018. [16] YAP M H, GOYAL M, OSMAN F M, et al. End-to-end breast ultrasound lesions recognition with a deep learning approach[C]//Proceedings Volume 10578, Medical Ima￾ging 2018: Biomedical Applications in Molecular, Struc￾tural, and Functional Imaging. Houston, Texas, United States, 2018: 1057819. [17] CHIAO J Y, CHEN K Y, LIAO K Y, et al. Detection and classification the breast tumors using mask R-CNN on sonograms[J]. Medicine, 2019, 98(19): e15200. [18] SHIN S Y, LEE S, YUN I D, et al. Joint weakly and semi￾supervised deep learning for localization and classifica￾tion of masses in breast ultrasound images[J]. IEEE trans￾actions on medical imaging, 2019, 38(3): 762–774. [19] REDMON J, FARHADI A. Yolov3: an incremental im￾provement[J]. arXiv preprint: arXiv: 1804.02767, 2018. [20] HU Jie, SHEN Li, SUN Gang. Squeeze-and-excitation networks[C]//Proceedings of the 2018 IEEE/CVF Confer￾ence on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018: 7132–7141. [21] GAO Shanghua, CHENG Mingming, ZHAO Kai, et al. Res2Net: a new multi-scale backbone architecture[J]. IEEE transactions on pattern analysis and machine intelli￾gence, 2019:1–10. [22] 作者简介： 徐立芳，讲师，博士，主要研究方 向为智能控制、机器视觉与机器认知、 人机混合智能。主持、参与省部级科 研项目 10 项，授权发明专利 6 项。发 表学术论文 20 余篇。 傅智杰，硕士研究生，主要研究方 向为深度学习、计算机视觉、医学 影像。 莫宏伟，教授，博士生导师，主要 研究方向为类脑计算与人工智能、机 器视觉与机器认知、人机混合智能。 主持省部级科研项目 24 项，授权发明 专利 10 项。发表学术论文 80 余篇。 第 1 期 徐立芳，等：基于改进的 YOLOv3 算法的乳腺超声肿瘤识别 ·29·