[1]徐振南,王建坤,胡益嘉,等.基于MobileNetV3的马铃薯病害识别[J].江苏农业科学,2022,50(10):176-182.
 Xu Zhennan,et al.Potato disease recognition based on MobileNetV3[J].Jiangsu Agricultural Sciences,2022,50(10):176-182.
点击复制

基于MobileNetV3的马铃薯病害识别()

《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]

卷:
第50卷
期数:
2022年第10期
页码:
176-182
栏目:
农业工程与信息技术
出版日期:
2022-05-20

文章信息/Info

Title:
Potato disease recognition based on MobileNetV3
作者:
徐振南1王建坤1胡益嘉1张智聪1赵旭东1杨琳琳12李文峰12
1.云南农业大学机电工程学院,云南昆明 650201; 2.云南省作物生产与智慧农业重点实验室,云南昆明 650201
Author(s):
Xu Zhennanet al
关键词:
马铃薯叶部病害GrabCutMobileNetV3迁移学习小样本
Keywords:
-
分类号:
TP391.41
DOI:
-
文献标志码:
A
摘要:
针对传统卷积神经网络在马铃薯叶部病害识别中结构复杂、参数庞大,难以实现在移动设备上的良好应用的问题,提出一种基于轻量级卷积神经网络和迁移学习的马铃薯叶部病害识别方法。首先,采集马铃薯叶部病害图像样本,再运用GrabCut算法进行图像分割;再基于 MobileNetV3构建病害识别基础模型,并通过调整模型结构及宽度系数α等方式对模型进行优化,最后运用迁移学习的方式将预训练参数迁移至优化模型进行训练。结果表明,该方法对马铃薯健康、晚疫病、早疫病、炭疽病及其他病害叶部图像识别准确率为 98.00%,模型权重仅为0.68 MB,识别速率为0.014 s/幅。本研究结果可为马铃薯叶部病害识别在移动设备上应用的实现提供理论支持。
Abstract:
-

参考文献/References:

[1]肖金钟. 马铃薯生产中的常见病害[J]. 现代园艺,2020,43(18):37-40.
[2]范子耀,王文桥,孟润杰,等. 7种杀菌剂对马铃薯早疫病菌室内毒力及田间防效[J]. 农药,2011,50(7):531-533.
[3]王秀鹏. 马铃薯常见病害识别及防治技术[J]. 种子科技,2018,36(6):101,103.
[4]边柯橙,杨海军,路永华. 深度学习在农业病虫害检测识别中的应用综述[J]. 软件导刊,2021,20(3):26-33.
[5]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.
[6]孙文杰,牟少敏,董萌萍,等. 基于卷积循环神经网络的桃树叶部病害图像识别[J]. 山东农业大学学报(自然科学版),2020,51(6):998-1003.
[7]谢裕睿,苗晟,张铄,等. 基于残差神经网络的烟草病害识别研究[J]. 现代计算机,2020(30):27-31.
[8]杨森,冯全,张建华,等. 基于轻量卷积网络的马铃薯外部缺陷无损分级[J]. 食品科学,2021,42(10):284-289.
[9]刘洋,冯全,王书志. 基于轻量级CNN的植物病害识别方法及移动端应用[J]. 农业工程学报,2019,35(17):194-204.
[10]Rother C,Kolmogorov V,Blake A.“GrabCut”:interactive foreground extraction using iterated graph cuts[J]. ACM Transactions on Graphics,2004,23(3):309-314.
[11]鲍文霞,吴刚,胡根生,等. 基于改进卷积神经网络的苹果叶部病害识别[J]. 安徽大学学报(自然科学版),2021,45(1):53-59.
[12]贾鹤鸣,郎春博,姜子超. 基于轻量级卷积神经网络的植物叶片病害识别方法[J]. 计算机应用,2021,41(6):1812-1819.
[13]Krizhevsky A,Sutskever I,Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM,2017,60(6):84-90.
[14]Simonyan K,Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. Computer Science,2014,52(3):1-14.
[15]Szegedy C,Liu W,Jia Y Q,et al. Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition.June 7-12,2015,Boston,MA.IEEE,2015:1-9.
[16]He K M,Zhang X Y,Ren S Q,et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition,June 27-30,2016,Las Vegas,NV,USA.Las Vegas:IEEE,2016:770-778.
[17]Howard A G,Zhu M L,Chen B,et al. MobileNets:efficient convolutional neural networks for mobile vision applications[EB/OL]. [2021-08-17]. https://arxiv.org/abs/1704.04861.
[18]Howard A,Sandler M,Chen B,et al. Searching for MobileNetV3[C]//Proceeding of the IEEE/CVF International Conference on Computer Vision (ICCV). Seoul:IEEE,2019:1314-1324.
[19]Sandler M,Howard A,Zhu M,et al. MobileNetV2 inverted residuals and linear bottlenecks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE,2018:4510-4520.
[20]刘梦伦,赵希梅,魏宾. 基于MobileNet V2-ELM的肝硬化识别[J]. 青岛大学学报(自然科学版),2019,32(4):17-21.
[21]Tan M X,Chen B,Pang R M,et al. MnasNet:platform-aware neural architecture search for mobile[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).June 15-20,2019,Long Beach,CA,USA. IEEE,2019:2815-2823.
[22]Hu J,Shen L,Sun G,et al. Squeeze-and-excitation networks[C]//Proceeding of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Piscataway:IEEE,2018:7132-7141.
[23]Tan M X,Le Q V. EfficientNet:rethinking model scaling for convolutional neural networks[C]//36th International Conference on Machine Learning(ICML 2019).Los Angeles:Proceeding of Machine Learning Research,2019:6105-6114.
[24]米硕,孙瑞彬,李欣,等. 迁移学习在卷积神经网络中的应用[J]. 中国新通信,2018,20(7):121.
[25]冯晓,李丹丹,王文君,等. 基于轻量级卷积神经网络和迁移学习的小麦叶部病害图像识别[J]. 河南农业科学,2021,50(4):174-180.

相似文献/References:

[1]王朝海,陈春艳,顾尚敬,等.不同覆土高度对马铃薯产量及其构成的影响[J].江苏农业科学,2013,41(04):101.
[2]石虎,杨永智,周云,等.马铃薯新品种青薯9号高效再生体系的建立[J].江苏农业科学,2013,41(05):14.
 Shi Hu,et al.Establishment of efficient regeneration system of new potato cultivar “Qingshu No.9”[J].Jiangsu Agricultural Sciences,2013,41(10):14.
[3]李珺,马力通.马铃薯淀粉体表达载体的构建及其转基因植物的培养[J].江苏农业科学,2013,41(08):22.
 Li Jun,et al.Construction of potato amyloplast expression vector and breeding of its genetically modified plants[J].Jiangsu Agricultural Sciences,2013,41(10):22.
[4]陈春艳,王朝海,白永生,等.不同稀释倍数代森锰锌防治马铃薯晚疫病的药效试验[J].江苏农业科学,2013,41(05):106.
 Chen Chunyan,et al.Control effect of different dilution multiple of mancozeb on potato late blight[J].Jiangsu Agricultural Sciences,2013,41(10):106.
[5]李葵花,高玉亮,吴京姬.转P5CS基因马铃薯“东农303”耐盐、抗旱性研究[J].江苏农业科学,2014,42(11):131.
 Li Kuihua,et al().Study on salt resistance and drought resistance of P5CS transgenic potato cultivar “Dongnong 303”[J].Jiangsu Agricultural Sciences,2014,42(10):131.
[6]李成松,冯玉磊,坎杂,等.单行悬挂式马铃薯施肥种植机的研制[J].江苏农业科学,2013,41(06):369.
 Li Chengsong,et al.Development of potato fertilizing and planting machine of single-line and suspension type[J].Jiangsu Agricultural Sciences,2013,41(10):369.
[7]陈建保,段伟伟.马铃薯加工专用薯脱毒种薯的生产现状及改进措施——以乌兰察布地区夏波蒂原种(G2)生产为例[J].江苏农业科学,2015,43(12):117.
 Chen Jianbao,et al.Production status investigation and improvement measures of virus-free seeds of processing potatoes—Taking production of Shepody (G2) as an example in Wulanchabu area[J].Jiangsu Agricultural Sciences,2015,43(10):117.
[8]郭成瑾,张丽荣,沈瑞清.土壤消毒对马铃薯连作田土壤微生物数量的影响[J].江苏农业科学,2014,42(10):367.
 Guo Chengjin,et al.Effects of soil sterilization on soil microbial quantity in potato continuous cropping land[J].Jiangsu Agricultural Sciences,2014,42(10):367.
[9]张海颖,郭凤柳,许华民,等.河北省张北地区马铃薯疮痂病的病菌鉴定[J].江苏农业科学,2014,42(10):131.
 Zhang Haiying,et al.Identification of potato scab pathogen in Zhangbei,Hebei Province[J].Jiangsu Agricultural Sciences,2014,42(10):131.
[10]贺苗苗.不同基因型马铃薯的花药培养研究[J].江苏农业科学,2014,42(09):90.
 He Miaomiao.Study on anther culture of different genotypes of potatoes[J].Jiangsu Agricultural Sciences,2014,42(10):90.

备注/Memo

备注/Memo:
收稿日期:2021-08-17
基金项目:国家自然科学基金(编号:31860331);云南省科技厅重大专项(编号:2018ZI001);云南省智慧农业重点实验室基金。
作者简介:徐振南(1997—),男,浙江金华人,硕士研究生,研究方向为智能检测及自动控制。E-mail:872137530@qq.com。
通信作者:杨琳琳,博士,副教授,研究方向为智能检测及自动控制。E-mail:29545343@qq.com。
更新日期/Last Update: 2022-05-20