Research on crop pathological image classification algorithm based on convolutional neural network

doi:10.14088/j.cnki.issn0439-8114.2021.09.027

Abstract

Abstract: The rapid and effective detection of crop pathology is of great significance to agriculture. It can not only improve the efficiency of automated pathology recognition, but also increase crop yields. This paper takes potato, tomato and other crops as the pathological research object, and proposes a crop pathological classification model based on convolutional neural network MFCPNet. First, build a deep convolutional neural network model, which is constructed through a convolutional layer, an activation layer, and a fully connected layer of a pooling layer. Then, the extracted image pathological features are fused with multiple features to effectively enhance the feature richness of crop pathology.At the same time, the original data set is enhanced to eliminate the problem of uneven sample distribution. The results show that the standards of the proposed crop pathology classification model are better than the AlexNet, VGG16 and VGG19 models, reaching an accuracy of 94.92%. At the same time, it eliminates the need for manual construction of complex feature projects, which has a certain value in promoting agricultural automation.

Key words: crop pathology classification, convolutional neural network, feature fusion

CLC Number:

TP399

LIU Shuai-jun, KOU Xu-peng, HE Ying, MO Xue-feng. Research on crop pathological image classification algorithm based on convolutional neural network[J]. HUBEI AGRICULTURAL SCIENCES, 2021, 60(9): 131-134.

References

[1] FANG Y, RAMASAMY R P.Current and prospective methods for plant disease detection[J]. Biosensors, 2015, 5(3): 537-561.
[2] MARTINELLI F, SCALENGHE R, DAVINO S, et al.Advanced methods of plant disease detection. A review[J]. Agronomy for sustainable development, 2015, 35(1): 1-25.
[3] 张善文,张晴晴,李萍.基于改进深度卷积神经网络的苹果的病害识别[J].林业工程学报,2019,4(4):107-112.
[4] 李先锋. 基于特征优化和多特征融合的杂草识别方法研究[D].江苏镇江:江苏大学,2010.
[5] 赵川源,何东健,乔永亮. 基于多光谱图像和数据挖掘的多特征杂草识别方法[J].农业工程学报,2013,29(2):192-198.
[6] 张恒,陈丽娟,张岩.模糊植物病虫害图像的检测[J].计算机仿真,2012,29( 1):199-201,220.
[7] 李先锋,朱伟兴,纪滨,等.基于特征优化和LS-SVM的棉田杂草识别[J].农业机械学报,2010,41(11):168-172.
[8] BRAHIMI M, ARSENOVIC M, LARABA S, et al.Deep learning for plant diseases: Detection and saliency map visualisation[J]. Human and machine learning, 2018,8:93-117.
[9] 何晓萍,沈雅云.深度学习的研究现状与发展[J].现代情报,2017,37( 2):163-170.
[10] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenet classification with deep convolutional neural networks[J].Advances in neural information processing systems,2012,25(2):1097-1105.
[11] SZEGEDY C,LIU W,JIA Y,et al.Going deeper with convolutions[A].Proceedings of the IEEE conference on computer vision and pattern recognition[C].2015.1-9.
[12] SIMONYAN K, ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[J]. Preprint arXiv,2014:1-4.
[13] HE K, ZHANG X, REN S, et al.Deep residual learning for image recognition[A].Proceedings of the IEEE conference on computer vision and pattern recognition[C].2016.770-778.
[14] 王健,刘雪花.基于深度可分离卷积的苹果叶病理识别[J].计算机系统应用,2020,29(11):190-195.
[15] 刘阗宇,冯全,杨森. 基于卷积神经网络的葡萄叶片病害检测方法[J].东北农业大学学报,2018,49(3):73-83.
[16] 曹鹏. 不均衡数据分类方法的研究[D].沈阳:东北大学,2014.
[17] 吴国琴. 迁移学习在图像分类中的应用研究[D].合肥:安徽大学,2017.
[18] 龙明盛. 迁移学习问题与方法研究[D].北京:清华大学,2014.
[19] GONG Y,WANG L, GUO R, et al.“Multi-scale orderless pooling of deep convolutional activation features,” in Proc[J]. Eur Conf Comput Vis, 2014,8695:392-407.
[20] 刘扬. 基于MNCC模型的高分辨率遥感影像目标识别[D].郑州:河南大学,2016.
[21] 王阳,刘立波.面向细粒度图像分类的双线性残差注意力网络[J].激光与光电子学进展,2020,57(12):171-180.
[22] 黄斌,卢金金,王建华,等.基于深度卷积神经网络的物体识别算法[J].计算机应用,2016,36(12):3333-3340,3346.
[23] 郭树旭,马树志,李晶,等.基于全卷积神经网络的肝脏CT影像分割研究[J].计算机工程与应用,2017,53(18):126-131.