基于AlexNet深度学习模型的中草药植物图像分类研究与实现  

Research and Implementation of Chinese Herbal Medicine Plant Image Classification Based on AlexNet Deep Learning Model

作　　者：黄方亮 俞磊 沈同平 金力 许欢庆 黄星语 HUANG Fang-liang;YU Lei;SHEN Tong-ping;JIN Li;XU Huan-qing;HUANG Xing-yu(School of Medical Information Engineering,Anhui University of Chinese Medicine,Hefei 230012,China)

机构地区：[1]安徽中医药大学医药信息工程学院,合肥230012

出　　处：《齐鲁工业大学学报》2020年第2期44-49,共6页Journal of Qilu University of Technology

基　　金：国家自然科学基金项目(61701005);安徽省高校省级重点项目(KJ2019A0437,SK2019A0242);安徽省高校省级质量工程项目(2017mooc220,2018zhkt079);安徽中医药大学自然科学研究项目(2019zrzd11,2018zryb06);安徽中医药大学教学研究项目(2018xjjy_yb004);安徽中医药大学2019年度课程思政教学改革试点项目(课程序号31,软件工程)。

摘　　要：本文研究了卷积神经网络中AlexNet模型在5种中草药图像分类过程中的应用。通过Python爬虫算法爬取百度图片中5类中草药3000张图片,并通过数据增扩算法将数据集扩增到12000张,以满足模型训练的需求。为了提高训练的效率将数据集转换成LMDB格式,并采用减均值的方式加快模型收敛,以最大限度平衡用CPU训练带来的训练周期过长的影响。合理调整模型参数,经过300次迭代得到87.5%的分类准确率。The application of Alexnet model in convolutional neural network in the image classification of five kinds of Chinese herbal medicines is studied in this paper.In order to meet the needs of model training,the Python crawler algorithm is used to crawl 3000 pictures of five kinds of Chinese herbal medicines in Baidu pictures,and the data set is expanded to 12000 by data expansion algorithm.The data set is transformed into LMDB format to improve the efficiency of training,and the method of reducing the mean value is used to speed up the convergence of the model,so as to maximize the balance of the influence of too long training period brought by CPU training.After 300 iterations,87.5%classification accuracy was obtained.

关 键 词：AlexNet 深度学习 中草药 图像分类 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]