Research on Classification of Corn Leaf Disease Image by Improved Residual Network

doi:10.3778/j.issn.1002-8331.2105-0321

Abstract

Abstract:

Aiming at the problems of low accuracy and slow speed of traditional corn leaf disease image recognition methods, a corn leaf image recognition algorithm based on improved deep residual network model is proposed. Here are the proposed improvement strategies：replacing the 7×7 convolution kernel in the first convolution layer of the traditional ResNet-50 model with three 3×3 convolution kernels; using the LeakyReLU activation function to replace the ReLU activation function; changing the order of the batch normalization layer, activation function and convolutional layer in the residual block. Firstly, data preprocessing is carried out, dividing the ratio of training set and test set to 4∶1, using data enhancement to expand the training set. Subsequently, the improved ResNet-50 model is subjected to transfer learning to obtain the weight parameters pre-trained on ImageNet. The experimental results show that the improved network has a 98.3% correct rate in corn leaf disease images classification. Compared with other network models, the accuracy rate is greatly improved, and the robustness is further enhanced, which can provide a reference for the recognition of corn leaf diseases.

Key words: corn leaf disease, image recognition, deep learning, residual network, transfer learning

摘要：

针对传统的玉米叶片病害图像识别方法正确率不高、速度慢等问题，提出一种基于改进深度残差网络模型的玉米叶片图像识别算法。提出的改进策略有：将传统的ResNet-50模型第一层卷积层中7×7卷积核替换为3个3×3的卷积核；使用LeakyReLU激活函数替代ReLU激活函数；改变残差块中批标准化层、激活函数与卷积层的排列顺序。进行数据预处理，将训练集与测试集的比例划分为4∶1，采用数据增强的方式对训练集进行扩充，将改进的ResNet-50模型经过迁移学习得到在ImageNet上预训练好的权重参数。实验结果表明，改进的网络在玉米叶片病害图像分类中得到了98.3%的正确率，与其他网络模型相比准确率大幅提升，鲁棒性进一步增强，可为玉米叶片病害的识别提供参考。

关键词: 玉米叶片病害, 图像识别, 深度学习, 残差网络, 迁移学习

HUANG Yinglai, AI Xin. Research on Classification of Corn Leaf Disease Image by Improved Residual Network[J]. Computer Engineering and Applications, 2021, 57(23): 178-184.

黄英来，艾昕. 改进残差网络在玉米叶片病害图像的分类研究[J]. 计算机工程与应用, 2021, 57(23): 178-184.

References

[1] 傅俊范，景殿玺，刘震，等.玉米叶部病害流行动态及预测预警研究进展[J].吉林农业大学学报，2016，38（6）：651-655.
FU J F，JING D X，LIU Z，et al.Review of epidemic dynamics and forecasting & warning of maize leaf disease[J].Journal of Jilin Agricultural University，2016，38（6）：651-655.
[2] 郑肖兰，徐春华，郑行恺，等.海南省南繁种植区玉米病害调查[J].热带农业科学，2019，39（6）：56-66.
ZHENG X L，XU C H，ZHENG X K，et al.A survey of maize diseases in the winter breeding area of Hainan province[J].Chinese Journal of Tropical Agriculture，2019，39（6）：56-66.
[3] 赵建敏，李艳，李琦，等.基于卷积神经网络的马铃薯叶片病害识别系统[J].江苏农业科学，2018，46（24）：251-255.
ZHAO J M，LI Y，LI Q，et al.Potato leaf disease identification system based on convolutional neural network[J].Jiangsu Agricultural Sciences，2018，46（24）：251-255.
[4] 吴开兴，苗雪菲，马文妙.基于卷积神经网络的番茄叶部病害识别研究[J].电脑知识与技术，2020，16（25）：25-27.
WU K X，MIAO X F，MA W M.Tomato leaf disease recognition research based on convolutional neural network[J].Computer Knowledge and Technology，2020，16（25）：25-27.
[5] SALIH T A，ALI A J，AHMED M N.Deep learning convolution neural network to detect and classify tomato plant leaf diseases[J].Open Access Library Journal，2020，7（5）：e6296.
[6] YADAV D，AKANKSHA，YADAV A K.A novel convolutional neural network based model for recognition and classification of apple leaf diseases[J].Traitement du Signal，2020，37（6）：1093-1101.
[7] 庄福振，罗平，何清，等.迁移学习研究进展[J].软件学报，2015，26（1）：26-39.
ZHUANG F Z，LUO P，HE Q，et al.Survey on transfer learning research[J].Journal of Software，2015，26（1）：26-39.
[8] 李彦冬，郝宗波，雷航.卷积神经网络研究综述[J].计算机应用，2016，36（9）：2508-2515.
LI Y D，HAO Z B，LEI H.Survey of convolutional neural network[J].Journal of Computer Applications，2016，36（9）：2508-2515.
[9] 韩猛.基于深度学习的花卉图像种类识别[D].晋中：山西农业大学，2017.
HAN M.A study based on deep convolution neural network for flower classification[D].Jinzhong：Shanxi Agricultural University，2017.
[10] ALSHARMAN N，JAWARNEH I.GoogleNet CNN neural network towards chest CT-coronavirus medical image classification[J].Journal of Computer Science，2020，16（5）：620-625.
[11] 倪黎，邹卫军.基于SE模块改进Xception的动物种类识别[J].导航与控制，2020，19（2）：106-111.
NI L，ZOU W J.Recognition of animal species based on improved Xception by SE module[J].Navigation and Control，2020，19（2）：106-111.
[12] 巨志勇，马素萍.改进的Inceptionv3果蔬识别算法[J].包装工程，2019，40（21）：30-35.
JU Z Y，MA S P.Fruit and vegetable recognition algorithm based on improved Inceptionv3[J].Packaging Engineering，2019，40（21）：30-35.
[13] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
[14] XU B，WANG N，CHEN T，et al.Empirical evaluation of rectified activations in convolutional network[J].arXiv：1505.00853，2015.
[15] IOFFE S，SZEGEDY C.Batch normalization：accelerating deep network training by reducing internal covariate shift[C]//International Conference on Machine Learning，2015：448-456.
[16] 关胤.基于残差网络迁移学习的花卉识别系统[J].计算机工程与应用，2019，55（1）：174-179.
GUAN Y.Flower species recognition system based on residual network transfer learning[J].Computer Engineering and Applications，2019，55（1）：174-179.
[17] 许景辉，邵明烨，王一琛，等.基于迁移学习的卷积神经网络玉米病害图像识别[J].农业机械学报，2020，51（2）：230-236.
XU J H，SHAO M Y，WANG Y C，et al.Recognition of corn leaf spot and rust based on transfer learning with convolutional neural network[J].Transactions of the Chinese Society for Agricultural Machinery，2020，51（2）：230-236.
[18] 孙俊，谭文军，毛罕平，等.基于改进卷积神经网络的多种植物叶片病害识别[J].农业工程学报，2017，33（19）：209-215.
SUN J，TAN W J，MAO H P，et al.Recognition of multiple plant leaf diseases based on improved convolutional neural network[J].Transactions of the Chinese Society of Agricultural Engineering，2017，33（19）：209-215.
[19] 吴宇豪，陈晓辉.基于改进的ResNet的人脸表情识别系统[J].信息通信，2020（7）：37-39.
WU Y H，CHEN X H.Facial expression recognition system based on improved ResNet[J].Information & Communications，2020（7）：37-39.