
浏览全部资源
扫码关注微信
1.四川大学 视觉合成图形图像技术国防重点学科实验室,成都 610065
2.成都中医药大学,成都 611137
3.成都中医药大学 附属医院,成都 610075
吴冲,硕士,从事图像识别研究,E-mail:chongwuscu@outlook.com
* 陈虎,博士,讲师,从事深度学习研究,E-mail:huchen@scu.edu.cn
收稿日期:2020-01-16,
网络出版日期:2020-03-10,
纸质出版日期:2020-11-05
移动端阅览
吴冲,谭超群,黄永亮等.基于深度学习算法的川贝母、山楂及半夏饮片的智能鉴别[J].中国实验方剂学杂志,2020,26(21):195-201.
WU Chong,TAN Chao-qun,HUANG Yong-liang,et al.Intelligent Identification of Fritillariae Cirrhosae Bulbus,Crataegi Fructus and Pinelliae RhizomaBased on Deep Learning Algorithms[J].Chinese Journal of Experimental Traditional Medical Formulae,2020,26(21):195-201.
吴冲,谭超群,黄永亮等.基于深度学习算法的川贝母、山楂及半夏饮片的智能鉴别[J].中国实验方剂学杂志,2020,26(21):195-201. DOI: 10.13422/j.cnki.syfjx.20201152.
WU Chong,TAN Chao-qun,HUANG Yong-liang,et al.Intelligent Identification of Fritillariae Cirrhosae Bulbus,Crataegi Fructus and Pinelliae RhizomaBased on Deep Learning Algorithms[J].Chinese Journal of Experimental Traditional Medical Formulae,2020,26(21):195-201. DOI: 10.13422/j.cnki.syfjx.20201152.
目的
2
利用人工智能和机器视觉技术,提出一种检测与评价中药材的新方法。
方法
2
以川贝母、山楂及半夏饮片为研究对象,通过机器视觉采集图片大数据,建立图像数据库;通过对中药外在性状特征的智能分析,以深度学习为手段,研究建立深度卷积神经网络模型来实现定位检测、品种识别等功能,以显著提高中药快速鉴别的准确率。
结果
2
测试的11种饮片(生山楂、炒山楂、焦山楂、山楂炭、松贝、青贝、炉贝、生半夏、姜半夏、法半夏、清半夏)图像品种分类准确度可达99%以上,具体类别的平均识别准确度可达到97%以上。
结论
2
通过深度学习算法实现的中药饮片智能鉴别技术具有简洁、快速、精度高、可批量化检测的优点,可为中药质量检测与评价提供技术支持,并丰富了中药品质评价的研究思路。
Objective
2
To propose a new method for detecting and evaluating traditional Chinese medicine (TCM) by artificial intelligence and machine vision technology.
Method
2
Taking Fritillariae Cirrhosae Bulbus, Crataegi Fructus and Pinelliae Rhizoma as the research objects, big data of pictures was collected by machine vision and the image database was established. Through the intelligent analysis of the external characteristics of TCM, the deep convolutional neural network model was established to realize the functions of location detection and variety identification by means of deep learning, so as to significantly improve the accuracy of rapid identification of TCM.
Result
2
The classification accuracy of 11 kinds of Chinese herbal pieces (raw, fried, parched and charred products of Crataegi Fructus, Pinelliae Rhizoma, Pinelliae Rhizoma Praeparatum Cum Zingibere et Alumine, Pinelliae Rhizoma Praeparatum, Pinelliae Rhizoma
Praeparatum Cum Alumine and three products of Fritillariae Cirrhosae Bulbus) could be more than 99%, and the average recognition accuracy of specific categories could reach more than 97%.
Conclusion
2
The intelligent identification technology of TCM decoction pieces realized by deep learning algorithms has the advantages of simplicity, rapidity, high precision and quantifiable detection, which can provide technical support for the quality detection and evaluation of TCM, and enrich the research ideas of quality evaluation of TCM.
汪劲 , 程存归 . 傅立叶变换红外光谱的SVM快速中药鉴别 [J]. 仪器仪表学报 , 2005 , 26 ( 8 ): 710 - 711,715 .
蔡从中 , 袁前飞 , 肖汉光 , 等 . 中药组方的计算机辅助分类与识别 [J]. 重庆大学学报:自然科学版 , 2006 , 29 ( 10 ): 42 - 46 .
WANG J F , CAI C Z , KONG C Y , et al . A computer method for validating traditional Chinese medicine herbal prescriptions [J]. Am J Chin Med , 2005 , 33 ( 2 ): 281 - 297 .
程存归 , 田玉梅 , 张长江 . 基于FTIR-CWT及ANN分类法的中药材菟丝子与同科属金灯藤子的识别方法研究 [J]. 化学学报 , 2008 , 66 ( 7 ): 793 - 798 .
杨添钧 . 基于机器视觉技术的药材及饮片“辨色”研究 [D]. 成都 : 成都中医药大学 , 2014 .
陶欧 , 林兆洲 , 张宪宝 , 等 . 基于饮片切面图像纹理特征参数的中药辨识模型研究 [J]. 世界科学技术—中医药现代化 , 2014 , 16 ( 12 ): 2558 - 2562 .
王耐 , 卢文彪 , 凌秀华 , 等 . 牛膝和川牛膝药材的特征提取与图像识别 [J]. 中国药房 , 2017 , 28 ( 12 ): 1670 - 1673 .
FATANIYA B , JOSHI M , MODI U , et al . Automatic identification of licorice and rhubarb by microscopic image processing [J]. Procedia Comput Sci , 2015 , 58 : 723 - 730 .
王亚杰 , 徐心和 . 中药材粉末显微图像模式识别方法 [J]. 农机化研究 , 2006 ( 12 ): 180 - 183 .
李震 . 中药饮片特征提取和识别系统 [D]. 哈尔滨 : 哈尔滨工业大学 , 2013 .
刘冰 . 基于广义多核学习的类椭球形中药材分类算法研究 [D]. 成都 : 四川师范大学 , 2016 .
胡继礼 , 阚红星 , 丁桦 , 等 . 中药材微性状真伪自动识别系统设计与实现 [J]. 电脑知识与技术 , 2016 , 12 ( 36 ): 178 - 180 .
朱黎辉 , 李晓宁 , 张莹 , 等 . 基于形状特征及纹理特征的中药材检索方法 [J]. 计算机工程与设计 , 2014 , 35 ( 11 ): 3903 - 3907 .
HABIB T , INGLADA J , MERCIER G , et al . Speeding up support vector machine (SVM) image classification by a kernel series expansion [A]. Proceedings of 15th IEEE International Conference on Image Processing [C]. New York : IEEE , 2008 : 865 - 868 .
GIRSHICK R , DONAHUE J , DARRELL T , et al . Rich feature hierarchies for accurate object detection and semantic segmentation [A]. Proceedings of CVPR 2014 [C]. New York : IEEE , 2014 : 580 - 587 .
GIRSHICK R . Fast R-CNN [A]. Proceedings of ICCV 2015 [C]. New York : IEEE , 2016 : 1440 - 1448 .
REN S , HE K M , GIRSHICK R , et al . Faster R-CNN:towards real-time object detection with region proposal networks [J]. IEEE Trans Pattern Anal Mach Intell , 2017 , 39 ( 6 ): 1137 - 1149 .
REDMON J , DIVVALA S , GIRSHICK R , et al . You Only Look Once:unified,real-time object detection [A]. Proceedings of CVPR 2016 [C]. New York : IEEE , 2016 : 779 - 788 .
KRIZHEVSKY A , SUTSKEVER I , HINTON G . ImageNet classification with deep convolutional neural networks [A]. Proceedings of 25th International Conference on Neural Information Processing Systems [C]. Cambridge : MIT Press , 2012 : 1097 - 1105 .
GLOROT X , BORDES A , BENGIO Y . Deep sparse rectifier neural networks [J]. J Mach Learn Res , 2011 , 15 : 315 - 323 .
SIMONYAN K , ZISSERMAN A . Very deep convolutional networks for large-scale image recognition [A]. Proceedings of ICLR 2015 [C]. San Diego : Computational and Biological Learning Society , 2014 : 1 - 14 .
WEN Y D , ZHANG K P , LI Z F , et al . A discriminative feature learning approach for deep face recognition [A]. Proceedings of ECCV 2016 [C]. Berlin : Springer , 2016 : 499 - 515 .
HE K M , ZHANG X , REN S Q , et al . Deep residual learning for image recognition [A]. Proceedings of CVPR 2016 [C]. New York : IEEE , 2016 : 770 - 778 .
HUANG G , LIU Z , WEINBERGER K Q , et al . Densely connected convolutional networks [A]. Proceedings of CVPR 2017 [C]. New York : IEEE , 2017 : 2261 - 2269 .
FAWCETT T . An introduction to ROC analysis [J]. Pattern Recogn Lett , 2006 , 27 ( 8 ): 861 - 874 .
0
浏览量
20
下载量
9
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621