基于改进动态ReLU和注意力机制模型的中药材粉末显微图像识别研究  

Research on microscopic image recognition of Chinese medicinal materials powder based on improved dynamic ReLU and attention mechanism model

作　　者：王一丁[1] 姚毅 李耀利[2] 蔡少青[2] 袁媛[3] Wang Yiding;Yao Yi;Li Yaoli;Cai Shaoqing;Yuan Yuan(College of Information Science&Technology,North China University of Technology,Beijing 100144,China;School of Pharmaceutical Sciences,Peking University,Beijing 100191,China;National Resource Center for Chinese Materia Medica,China Academy of Chinese Medical Sciences,Beijing 100700,China)

机构地区：[1]北方工业大学信息学院,北京100144 [2]北京大学药学院,北京100191 [3]中国中医科学院中药资源中心,北京100700

出　　处：《计算机应用研究》2021年第9期2861-2865,2870,共6页Application Research of Computers

基　　金：中医药行业科研专项项目(201407003);中央本级重大增减支项目“名贵中药资源可持续利用能力建设项目”(2060302)。

摘　　要：中药材粉末显微特征图像数据量少、样本类别分布不均衡、类间差异小,传统的图像识别方法分类效果不佳。针对以上问题提出一种基于动态ReLU和注意力机制模型的深度卷积神经网络改进方法。首先,采用对小样本数据分类效果明显的Xception作为基础网络;其次,将网络中的静态ReLU激活函数替换为改进的动态ReLU函数,让每个样本具有自己独特的ReLU参数;最后,在网络中嵌入改进的SE模块,使网络能够更好地自动学习到每个特征通道的重要程度。以上方法可以使网络更加注重于图像中的细节信息,能很好地解决样本类别分布不均衡、类间差异小的问题。实验结果表明,对56种中药材粉末导管图像进行分类识别,其准确率提升了约1.5%,达到93.8%,证明了所提研究方法相比于其他图像分类方法具有一定的优越性。Due to the small amount of microscopic features image data of traditional Chinese medicinal materials powder,unbalanced distribution of sample classes and small difference between classes,it is difficult to achieve a satisfying classification effect through traditional image classification methods.To solve the above problems,this paper proposed an improved method of deep convolution neural network based on dynamic ReLU and attention mechanism model.Firstly,it used Xception as the basic network,which had an obvious effect on small sample data classification.Secondly,it replaced the static ReLU activation function in the network with the improved dynamic ReLU function,so that each sample had its own unique ReLU parameters.Finally,it embedded the improved SE module in the network to enable the network to learn the importance of each feature channel automatically.The proposed method can make the network pay more attention to the detailed information in the image,and can solve the problem of unbalanced distribution of sample classes and small differences between classes.The experimental results show that the image classification accuracy of 56 kinds of traditional Chinese medicinal materials powder vessel is increased by about 1.5%to 93.8%,which demonstrates that the proposed method is advantageous over other image classification methods.

关 键 词：卷积神经网络 中药材粉末显微特征图像识别 深度学习 动态ReLU函数 SE模块 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]