摘要
我国是农业大国,种植的农作物种类繁多,其中农作物病害是导致农业损失的主要原因。随着深度学习的发展,卷积神经网络作为深度学习中具有代表性的算法,因更好的特征学习和特征表达能力被广泛应用于农作物病害的识别研究中。针对已经发表的文献,对近些年基于卷积神经网络的农作物病害分类的研究进行综述,介绍几种经典且常用的卷积神经网络模型结构及优化方法,通过对农作物病害分类进行实验,对比总结不同算法的特点,同时对基于卷积神经网络的农作物病害图像分类的应用前景进行展望。
China is a big agricultural country with a wide variety of crops, among which crop diseases are the main cause of agricultural losses. With the development of deep learning, convolutional neural network, as a representative algorithm in deep learning, is widely used in the identification of crop diseases due to its better feature learning and feature expression capabilities. In view of the published literature, this paper reviews the research on crop disease classification based on convolutional neural network in recent years, introduces several classic and commonly used convolutional neural network model structures and optimization methods, and conducts experiments on crop disease classification. The characteristics of different algorithms are summarized, and the application prospect of crop disease image classification based on convolutional neural network is prospected.
作者
黄乾峰
董琴
HUANG Qianfeng;DONG Qin(School of Mechanical Engineering,Yancheng Institute of Technology,Yancheng Jiangsu 224001,China;School of Information Engineering,Yancheng Institute of Technology,Yancheng Jiangsu 224001,China)
出处
《信息与电脑》
2022年第5期138-142,146,共6页
Information & Computer
关键词
深度学习
图像分类
卷积神经网络
农作物病害
迁移学习
deep learning
image processing
convolutional neural network
crop disease
migration study
作者简介
通信作者:黄乾峰(1998-),男,江苏南通人,硕士研究生。研究方向:人工智能、图像处理。E-mail:1150379914@qq.com。
