• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

用于柑橘疾病分类的深度监督多尺度特征网络（DS-MENet）

DS-MENet for the classification of citrus disease.

作者信息

Liu Xuyao, Hu Yaowen, Zhou Guoxiong, Cai Weiwei, He Mingfang, Zhan Jialei, Hu Yahui, Li Liujun

机构信息

College of Computer & Information Engineering, Central South University of Forestry and Technology, Changsha, China.

Plant Protection Research Institute, Hunan Academy of Agricultural Sciences, Changsha, China.

出版信息

Front Plant Sci. 2022 Jul 22;13:884464. doi: 10.3389/fpls.2022.884464. eCollection 2022.

DOI:10.3389/fpls.2022.884464
PMID:35937334
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC9355402/
Abstract

Affected by various environmental factors, citrus will frequently suffer from diseases during the growth process, which has brought huge obstacles to the development of agriculture. This paper proposes a new method for identifying and classifying citrus diseases. Firstly, this paper designs an image enhancement method based on the MSRCR algorithm and homomorphic filtering algorithm optimized by Laplacian (HFLF-MS) to highlight the disease characteristics of citrus. Secondly, we designed a new neural network DS-MENet based on the DenseNet-121 backbone structure. In DS-MENet, the regular convolution in Dense Block is replaced with depthwise separable convolution, which reduces the network parameters. The ReMish activation function is used to alleviate the neuron death problem caused by the ReLU function and improve the robustness of the model. To further enhance the attention to citrus disease information and the ability to extract feature information, a multi-channel fusion backbone enhancement method (MCF) was designed in this work to process Dense Block. We use the 10-fold cross-validation method to conduct experiments. The average classification accuracy of DS-MENet on the dataset after adding noise can reach 95.02%. This shows that the method has good performance and has certain feasibility for the classification of citrus diseases in real life.

摘要

受各种环境因素影响，柑橘在生长过程中经常会遭受病害，这给农业发展带来了巨大障碍。本文提出了一种柑橘病害识别与分类的新方法。首先，本文设计了一种基于MSRCR算法和经拉普拉斯优化的同态滤波算法（HFLF-MS）的图像增强方法，以突出柑橘的病害特征。其次，我们基于DenseNet-121骨干结构设计了一种新的神经网络DS-MENet。在DS-MENet中，Dense Block中的常规卷积被深度可分离卷积所取代，从而减少了网络参数。使用ReMish激活函数来缓解ReLU函数导致的神经元死亡问题，并提高模型的鲁棒性。为了进一步增强对柑橘病害信息的关注和提取特征信息的能力，本文设计了一种多通道融合骨干增强方法（MCF）来处理Dense Block。我们使用10折交叉验证方法进行实验。DS-MENet在添加噪声后的数据集上的平均分类准确率可达95.02%。这表明该方法具有良好的性能，在实际生活中对柑橘病害分类具有一定的可行性。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/03a67e2aa412/fpls-13-884464-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/5cf9eb59da68/fpls-13-884464-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/51c54367dd8e/fpls-13-884464-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/3077e99579cd/fpls-13-884464-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/587aa28edb11/fpls-13-884464-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/7e2341d8de45/fpls-13-884464-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/bf46fd191471/fpls-13-884464-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/8d3998f55ea8/fpls-13-884464-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/6d2b8e82b1d2/fpls-13-884464-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/7b550e4e34d5/fpls-13-884464-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/deceee334b10/fpls-13-884464-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/0ead84110c8d/fpls-13-884464-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/9e4e682ab4c2/fpls-13-884464-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/a1452b8fbe67/fpls-13-884464-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/03a67e2aa412/fpls-13-884464-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/5cf9eb59da68/fpls-13-884464-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/51c54367dd8e/fpls-13-884464-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/3077e99579cd/fpls-13-884464-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/587aa28edb11/fpls-13-884464-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/7e2341d8de45/fpls-13-884464-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/bf46fd191471/fpls-13-884464-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/8d3998f55ea8/fpls-13-884464-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/6d2b8e82b1d2/fpls-13-884464-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/7b550e4e34d5/fpls-13-884464-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/deceee334b10/fpls-13-884464-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/0ead84110c8d/fpls-13-884464-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/9e4e682ab4c2/fpls-13-884464-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/a1452b8fbe67/fpls-13-884464-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8da9/9355402/03a67e2aa412/fpls-13-884464-g014.jpg

相似文献

1
DS-MENet for the classification of citrus disease.用于柑橘疾病分类的深度监督多尺度特征网络（DS-MENet）
Front Plant Sci. 2022 Jul 22;13:884464. doi: 10.3389/fpls.2022.884464. eCollection 2022.
2
Identification of citrus diseases based on AMSR and MF-RANet.基于先进微波扫描辐射计（AMSR）和多特征融合残差注意力网络（MF-RANet）的柑橘病害识别
Plant Methods. 2022 Sep 24;18(1):113. doi: 10.1186/s13007-022-00945-4.
3
An Accurate Classification of Rice Diseases Based on ICAI-V4.基于ICAI-V4的水稻病害精确分类
Plants (Basel). 2023 Jun 5;12(11):2225. doi: 10.3390/plants12112225.
4
A depthwise separable dense convolutional network with convolution block attention module for COVID-19 diagnosis on CT scans.基于卷积块注意力模块的深度可分离稠密卷积网络用于 CT 扫描的 COVID-19 诊断。
Comput Biol Med. 2021 Oct;137:104837. doi: 10.1016/j.compbiomed.2021.104837. Epub 2021 Sep 8.
5
Real-Time Target Detection Method Based on Lightweight Convolutional Neural Network.基于轻量级卷积神经网络的实时目标检测方法
Front Bioeng Biotechnol. 2022 Aug 16;10:861286. doi: 10.3389/fbioe.2022.861286. eCollection 2022.
6
YOLOv5s-SA: Light-Weighted and Improved YOLOv5s for Sperm Detection.YOLOv5s-SA：用于精子检测的轻量级改进型YOLOv5s
Diagnostics (Basel). 2023 Mar 14;13(6):1100. doi: 10.3390/diagnostics13061100.
7
Facial expression recognition based on improved depthwise separable convolutional network.基于改进的深度可分离卷积网络的面部表情识别
Multimed Tools Appl. 2023;82(12):18635-18652. doi: 10.1007/s11042-022-14066-6. Epub 2022 Nov 23.
8
Citrus Pests and Diseases Recognition Model Using Weakly Dense Connected Convolution Network.基于弱密集连接卷积网络的柑橘病虫害识别模型
Sensors (Basel). 2019 Jul 19;19(14):3195. doi: 10.3390/s19143195.
9
A double-channel multiscale depthwise separable convolutional neural network for abnormal gait recognition.双通道多尺度深度可分离卷积神经网络用于异常步态识别。
Math Biosci Eng. 2023 Feb 23;20(5):8049-8067. doi: 10.3934/mbe.2023349.
10
Sonar Objective Detection Based on Dilated Separable Densely Connected CNNs and Quantum-Behaved PSO Algorithm.基于扩张可分离密集连接卷积神经网络和量子行为粒子群算法的声纳目标检测
Comput Intell Neurosci. 2021 Jan 18;2021:6235319. doi: 10.1155/2021/6235319. eCollection 2021.

引用本文的文献

1
Segmentation detection method in tree-shaded environment for road cracks collected by inspection vehicle on WFU-Unet.基于WFU-Unet的巡检车辆采集的道路裂缝在树荫环境下的分割检测方法
Sci Rep. 2025 Apr 6;15(1):11760. doi: 10.1038/s41598-025-96219-9.
2
A Precise Segmentation Algorithm of Pumpkin Seedling Point Cloud Stem Based on CPHNet.一种基于CPHNet的南瓜幼苗点云茎精确分割算法。
Plants (Basel). 2024 Aug 18;13(16):2300. doi: 10.3390/plants13162300.
3
Diagnosis model of early Pneumocystis jirovecii pneumonia based on convolutional neural network: a comparison with traditional PCR diagnostic method.

本文引用的文献

1
Classification of Citrus Diseases Using Optimization Deep Learning Approach.基于优化深度学习方法的柑橘病害分类。
Comput Intell Neurosci. 2022 Feb 10;2022:9153207. doi: 10.1155/2022/9153207. eCollection 2022.
2
Diagnosis of Typical Apple Diseases: A Deep Learning Method Based on Multi-Scale Dense Classification Network.典型苹果病害诊断：一种基于多尺度密集分类网络的深度学习方法
Front Plant Sci. 2021 Oct 1;12:698474. doi: 10.3389/fpls.2021.698474. eCollection 2021.
3
Diagnosing Covid-19 chest x-rays with a lightweight truncated DenseNet with partial layer freezing and feature fusion.
基于卷积神经网络的早期肺孢子菌肺炎诊断模型：与传统 PCR 诊断方法的比较。
BMC Pulm Med. 2024 Apr 25;24(1):205. doi: 10.1186/s12890-024-02987-x.
4
An Accurate Classification of Rice Diseases Based on ICAI-V4.基于ICAI-V4的水稻病害精确分类
Plants (Basel). 2023 Jun 5;12(11):2225. doi: 10.3390/plants12112225.
使用具有部分层冻结和特征融合的轻量级截断密集连接网络诊断新冠肺炎胸部X光片。
Biomed Signal Process Control. 2021 Jul;68:102583. doi: 10.1016/j.bspc.2021.102583. Epub 2021 Apr 1.
4
Citrus Pests and Diseases Recognition Model Using Weakly Dense Connected Convolution Network.基于弱密集连接卷积网络的柑橘病虫害识别模型
Sensors (Basel). 2019 Jul 19;19(14):3195. doi: 10.3390/s19143195.
5
Squeeze-and-Excitation Networks.挤压激励网络。
IEEE Trans Pattern Anal Mach Intell. 2020 Aug;42(8):2011-2023. doi: 10.1109/TPAMI.2019.2913372. Epub 2019 Apr 29.
6
Detection of citrus canker and Huanglongbing using fluorescence imaging spectroscopy and support vector machine technique.利用荧光成像光谱技术和支持向量机技术检测柑橘溃疡病和黄龙病。
Appl Opt. 2016 Jan 10;55(2):400-7. doi: 10.1364/AO.55.000400.
7
The retinex theory of color vision.颜色视觉的视网膜皮层理论。
Sci Am. 1977 Dec;237(6):108-28. doi: 10.1038/scientificamerican1277-108.