计算机科学 ›› 2020, Vol. 47 ›› Issue (11): 148-158.doi: 10.11896/jsjkx.191000104
李渊彤1, 罗裕升2, 朱珍民2,3
LI Yuan-tong1, LUO Yu-sheng2, ZHU Zhen-min2,3
摘要: 中医舌诊因其直观稳定且易于观察的特点,以及具有较高的临床应用价值和快捷方便的实用性,成为了一个重要的研究课题。目前,将医学图像处理技术、人工智能技术和中医专家的临床经验相结合,实现了对中医舌诊的客观化、定量化和自动化,这是当前中医舌诊现代化研究的主流方向。文中研究了基于迁移学习和深度学习的舌体精确分割和舌象特征识别的关键技术,提出了一种基于区域关联的单像素损失函数的舌体分割方法,新的损失函数不仅考虑到了区域像素之间的关联关系,而且有效利用了像素标签语义的监督信息,能够更好地指导模型进行训练学习,在测试集上的MIoU指标达到了96.32%。然后,针对舌象几何特征提出了一个融合空间转换网络和VGG16模型的舌象几何特征分析模型,使用了空间转换网络来显式地学习空间不变性,并复用了VGG16模型的卷积部分,使得可以用舌体分割任务学习到的知识来进行参数迁移学习。通过两组对比实验,验证了空间转换网络对提高模型空间不变性的有效性,以及舌体分割的知识迁移能使模型更快、更平稳地收敛。同时,提出了一个融合深度纹理编码网络和VGG16模型的舌象纹理特征分类模型,使用深度纹理编码网络能将卷积得到的有序特征图编码成无序的纹理语义表示,以更有效地表达纹理信息。通过实验对比分析验证了深度纹理编码网络的无序编码对舌象纹理语义表示的有效性。
中图分类号:
[1] XU J T.Clinical map of traditional Chinese medicine tonguediagnosis[M].BeiJing:Chemical Industry Press,2017. [2] ZHOU Z,HUANG F.Application of mathematical morphology method on the segmentationof tongue images [J].Technological Development of Enterprise,2009,28(3):164-166. [3] ZHANG L,QIN J.Tongue-image segmentation based on grayprojection and threshold-adaptive method[J].Chinese Tissue Engineering Research and Clinical Rehabilitation,2010,14(9):1638-1641. [4] WU J,ZHANG Y H,BAI J,et al.Tongue contour image extraction using a watershed transform and an active contour model[J].Journal of Tsinghua University (Science and Technology),2008(6):1040-1043. [5] ZHANG X Y.Tongue image segmentation based on randomwalk algorithm [D].Beijing:Beijing Institute of Technology,2016. [6] HUANG Z P,HUANG Y S,YI F L,et al.An automatic tongue segmentation algorithm based on OTSU and region growing[J].Lishizhen Medicine and Materia Medica Research,2017,28(12):3062-3064. [7] LIU Y B,YANG S.New algorithm on image automatic segmentation of body of tongue[J].Journal of Shenyang Normal University (Natural Science Edition),2011,29(4):514-517. [8] SUN X L,PANG C Y.An improved snake model method on tongue segmentation[J].Journal of Changchun University of Science and Technology (Natural Science Edition),2013,36(5):154-156. [9] WEI B G,SHEN L S.Automatic analysis for plumpness and slenderness of tongue [J].Computer Engineering,2004(11):25-26. [10] XU J T,ZHANG Z F,REN H F,et al.An imaging diagnostic method about analyzing for plumpness and slenderness of tongue[J].Chinese Imaging Journal of Integrated Traditional and Western Medicine,2009,7(06):407-410. [11] LU P.Study of teeth-marked tongue based on image processing and pattern recognition [D].Shanghai:East China University of Science and Technology, 2014. [12] ZHAI T T,XIA C M,WANG Y Q.Recognition of Greasy or curdy tongue coating based on gabor wavelet transformation [J].Computer Applications and Software,2016,33(10):162-166. [13] YANG Z H.Dissertation for the doctoral degree in engineering [D].Harbin:Harbin Institute of Technology, 2010. [14] LIU G Z.Research on application of traditional Chinese medicine tongue images classification based om CNN [D].Jilin:Jilin University,2018. [15] HU J L,KAN H X.Tongue classification based on convolutional neural network[J].Journal of Anqing Normal University (Natural Science Edition),2018,24(4):44-49. [16] SUN L Y,CHENG Z,GAO F S,et al.Discussion on Objective Research of Tongue Diagnosis by Computer Image Recognition Technology[J].Journal of Anhui Traditional Chinese Medical College,1986(4):5-7. [17] WANG Y Q,WEI B G,CAI Y H,et al.A knowledge-basedarithmetic for automatic tongue segmentation[J].Acta Electronica Sinica,2004(3):489-491. [18] GAO L,LING X M.A tongue segmentation method based on improved fuzzy operator and morphology[J].Journal of Lanzhou Jiaotong University,2006(3):89-91. [19] LI Q L,XUE Y Q,WANG J Y,et al.Automated tongue segmentation algorithm based on hyperspectral image[J].Journal of Infrared and Millimeter Waves,2007(1):77-80. [20] LIU Z,CHEN J X,ZHAO Y M,et al.Automatic tongue image segmentation based on visual attention and support vector machine[J].Journal of Beijing University of Traditional Chinese Medicine,2013,36(1):18-20. [21] WANG P,YANG W C,SUN C K,et al.Tongue segmentation and tongue crack extraction of tongue 3D color point cloud[J].Infrared and Laser Engineering,2017,46(S1):88-95. [22] WANG L R,TANG Y P,CHEN P,et al.Two-phase convolutional neural network design for tongue segmentation[J].Journal of Image and Graphics,2018,23(10):1571-1581. [23] GONG Y P,CHEN S Z,LIAN Y S,et al.Quantity study on the pathological nature of the tongue fur[J].Chinese Journal of Information on Traditional Chinese Medicine,2006(11):28-29. [24] SHENG S Y,LI B,YUE X Q,et al.Color feature extraction of tongue image based on manifold learning[J].Space Medicine & Medical Engineering,2008(5):435-439. [25] HU S N.Research on color recognition for tongue image in traditional Chinese medicine [D].Huangzhou:Zhejiang Sci-Tech University,2010. [26] ZHANG J,QIAN J,DONG H Y,et al.Analysis of traditional Chinese medicine digital tongue texture based on fractal theory[J].China Journal of Traditional Chinese Medicine and Pharmacy,2016,31(1):104-106. [27] XIE T.A new approach to the tongue-image segmentation and moistening analysis based on image processing [D].Shanghai:East China University of Science and Technology, 2017. [28] WANG S,LIU K H,WANG L T.Tongue spots and petechiae recognition and extraction in tongue diagnosis images[J].Computer Engineering & Science,2017,39(6):1126-1132. [29] LIU B,HU G Q,ZHANG X F,et al.An improved automatic description method of tongue coating thickness in Chinese medicine[J].Beijing Biomedical Engineering,2018,37(2):157-163. [30] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//IEEE transactions on pattern analysis and machine intelligence.2018. [31] ZJADERBERG M, SIMONYAN K, ZISSERMAN A. Spatialtransformer networks[C]//Advances in Neural Information Processing Systems.2015. [32] ZHANG H, XUE J, DANA K. Deep TEN: Texture encoding Network[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2017. |
[1] | 徐涌鑫, 赵俊峰, 王亚沙, 谢冰, 杨恺. 时序知识图谱表示学习 Temporal Knowledge Graph Representation Learning 计算机科学, 2022, 49(9): 162-171. https://doi.org/10.11896/jsjkx.220500204 |
[2] | 饶志双, 贾真, 张凡, 李天瑞. 基于Key-Value关联记忆网络的知识图谱问答方法 Key-Value Relational Memory Networks for Question Answering over Knowledge Graph 计算机科学, 2022, 49(9): 202-207. https://doi.org/10.11896/jsjkx.220300277 |
[3] | 汤凌韬, 王迪, 张鲁飞, 刘盛云. 基于安全多方计算和差分隐私的联邦学习方案 Federated Learning Scheme Based on Secure Multi-party Computation and Differential Privacy 计算机科学, 2022, 49(9): 297-305. https://doi.org/10.11896/jsjkx.210800108 |
[4] | 方义秋, 张震坤, 葛君伟. 基于自注意力机制和迁移学习的跨领域推荐算法 Cross-domain Recommendation Algorithm Based on Self-attention Mechanism and Transfer Learning 计算机科学, 2022, 49(8): 70-77. https://doi.org/10.11896/jsjkx.210600011 |
[5] | 孙奇, 吉根林, 张杰. 基于非局部注意力生成对抗网络的视频异常事件检测方法 Non-local Attention Based Generative Adversarial Network for Video Abnormal Event Detection 计算机科学, 2022, 49(8): 172-177. https://doi.org/10.11896/jsjkx.210600061 |
[6] | 王剑, 彭雨琦, 赵宇斐, 杨健. 基于深度学习的社交网络舆情信息抽取方法综述 Survey of Social Network Public Opinion Information Extraction Based on Deep Learning 计算机科学, 2022, 49(8): 279-293. https://doi.org/10.11896/jsjkx.220300099 |
[7] | 郝志荣, 陈龙, 黄嘉成. 面向文本分类的类别区分式通用对抗攻击方法 Class Discriminative Universal Adversarial Attack for Text Classification 计算机科学, 2022, 49(8): 323-329. https://doi.org/10.11896/jsjkx.220200077 |
[8] | 姜梦函, 李邵梅, 郑洪浩, 张建朋. 基于改进位置编码的谣言检测模型 Rumor Detection Model Based on Improved Position Embedding 计算机科学, 2022, 49(8): 330-335. https://doi.org/10.11896/jsjkx.210600046 |
[9] | 胡艳羽, 赵龙, 董祥军. 一种用于癌症分类的两阶段深度特征选择提取算法 Two-stage Deep Feature Selection Extraction Algorithm for Cancer Classification 计算机科学, 2022, 49(7): 73-78. https://doi.org/10.11896/jsjkx.210500092 |
[10] | 程成, 降爱莲. 基于多路径特征提取的实时语义分割方法 Real-time Semantic Segmentation Method Based on Multi-path Feature Extraction 计算机科学, 2022, 49(7): 120-126. https://doi.org/10.11896/jsjkx.210500157 |
[11] | 侯钰涛, 阿布都克力木·阿布力孜, 哈里旦木·阿布都克里木. 中文预训练模型研究进展 Advances in Chinese Pre-training Models 计算机科学, 2022, 49(7): 148-163. https://doi.org/10.11896/jsjkx.211200018 |
[12] | 周慧, 施皓晨, 屠要峰, 黄圣君. 基于主动采样的深度鲁棒神经网络学习 Robust Deep Neural Network Learning Based on Active Sampling 计算机科学, 2022, 49(7): 164-169. https://doi.org/10.11896/jsjkx.210600044 |
[13] | 苏丹宁, 曹桂涛, 王燕楠, 王宏, 任赫. 小样本雷达辐射源识别的深度学习方法综述 Survey of Deep Learning for Radar Emitter Identification Based on Small Sample 计算机科学, 2022, 49(7): 226-235. https://doi.org/10.11896/jsjkx.210600138 |
[14] | 祝文韬, 兰先超, 罗唤霖, 岳彬, 汪洋. 改进Faster R-CNN的光学遥感飞机目标检测 Remote Sensing Aircraft Target Detection Based on Improved Faster R-CNN 计算机科学, 2022, 49(6A): 378-383. https://doi.org/10.11896/jsjkx.210300121 |
[15] | 王建明, 陈响育, 杨自忠, 史晨阳, 张宇航, 钱正坤. 不同数据增强方法对模型识别精度的影响 Influence of Different Data Augmentation Methods on Model Recognition Accuracy 计算机科学, 2022, 49(6A): 418-423. https://doi.org/10.11896/jsjkx.210700210 |
|