[1]董虎胜,鲜学丰,孙逊,等.基于多阶段特征融合的MobileNet v3ULAM茶叶病害识别[J].江苏农业科学,2025,53(15):200-211.
 Dong Husheng,et al.Identification of tea leaf disease based on MobileNet v3ULAM with multistage feature fusion[J].Jiangsu Agricultural Sciences,2025,53(15):200-211.
点击复制

基于多阶段特征融合的MobileNet v3ULAM茶叶病害识别()

《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]

卷:
第53卷
期数:
2025年第15期
页码:
200-211
栏目:
农业工程与信息技术
出版日期:
2025-08-05

文章信息/Info

Title:
Identification of tea leaf disease based on MobileNet v3ULAM with multistage feature fusion
作者:
董虎胜12鲜学丰1孙逊1杨元峰1
1.苏州市职业大学计算机工程学院,江苏苏州 215104; 2.江苏省现代企业信息化应用支撑软件工程技术研发中心,江苏苏州 215104
Author(s):
Dong Hushenget al
关键词:
茶叶病害识别MobileNet v3注意力机制边缘计算卷积神经网络
Keywords:
-
分类号:
S126;TP391.41
DOI:
-
文献标志码:
A
摘要:
为了实现对茶园种植的智能化与精准化管理,针对自然场景下茶叶病害的快速与准确识别问题,提出了应用多阶段特征融合策略与ULAM 超轻量注意力对MobileNet v3网络作改进的轻量级茶叶病害识别模型。该模型在MobileNet v3网络中新增了1个分支,对网络各中间阶段的特征作拼接与变换处理,然后与原主干网络提取的特征融合,实现对中间阶段特征的复用,有效增强特征的判别力。为了改进MobileNet v3网络中SE注意力对空间信息处理不足和运算量较大的问题,还设计了一款ULAM注意力模块;ULAM不仅实现了对空间与通道信息的协同处理,还借助转置卷积运算显著降低了运算量,在该注意力模块中只有54个需要学习的参数,具有极为轻量的优势。在CIFAR10通用图像分类任务上,改进后的模型达到94.18%的识别准确率,比原MobileNet v3提高4.13百分点,综合性能优于常见CNN模型。在自建的茶叶常见病害数据集上,直接使用原始数据训练本研究模型可达到95.46%的平均识别准确率,相较于MobileNet v3提升3.34百分点。进一步采用数据平衡处理后,模型平均识别精度提升至9888%,能够准确地识别白星病、藻斑病、炭疽病等茶叶病害。本研究模型在参数量上比原始MobileNet v3进一步降低,因而适合在移动设备与农机等资源受限的场景中部署。
Abstract:
-

参考文献/References:

[1]He K M,Zhang X Y,Ren S Q,et al. Deep residual learning for image recognition[C]//2016 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE,2016:770-778.
[2]Szegedy C,Vanhoucke V,Ioffe S,et al. Rethinking the inceptionarchitecture for computer vision[C]//2016 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Las Vegas:IEEE,2016:2818-2826.
[3]Simonyan K,Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2014-09-04)[2024-07-09]. https://arxiv.org/abs/1409.1556.
[4]Huang G,Liu Z,van der Maaten L,et al. Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu:IEEE,2017:2261-2269.
[5]翟肇裕,曹益飞,徐焕良,等. 农作物病虫害识别关键技术研究综述[J]. 农业机械学报,2021,52(7):1-18.
[6]邵明月,张建华,冯全,等. 深度学习在植物叶部病害检测与识别的研究进展[J]. 智慧农业(中英文),2022,4(1):29-46.
[7]Zeng W H,Li H D,Hu G S,et al. Lightweight dense-scale network (LDSNet) for corn leaf disease identification[J]. Computers and Electronics in Agriculture,2022,197:106943.
[8]崔金荣,魏文钊,赵敏. 基于改进MobileNet v3的水稻病害识别模型[J]. 农业机械学报,2023,54(11):217-224,276.
[9]任守纲,贾馥玮,顾兴健,等. 反卷积引导的番茄叶部病害识别及病斑分割模型[J]. 农业工程学报,2020,36(12):186-195.
[10]赵越,赵辉,姜永成,等. 基于深度学习的马铃薯叶片病害检测方法[J]. 中国农机化学报,2022,43(10):183-189.
[11]孙道宗,刘欢,刘锦源,等. 基于改进YOLO v4模型的茶叶病害识别[J]. 西北农林科技大学学报(自然科学版),2023,51(9):145-154.
[12]李子茂,徐杰,郑禄,等. 基于改进DenseNet的茶叶病害小样本识别方法[J]. 农业工程学报,2022,38(10):182-190.
[13]Hu G S,Wu H Y,Zhang Y,et al. A low shot learning method for tea leafs disease identification[J]. Computers and Electronics in Agriculture,2019,163:104852.
[14]Howard A G,Zhu M L,Chen B,et al. MobileNets:efficient convolutional neural networks for mobile vision applications[EB/OL]. (2017-04-17)[2024-07-09]. https://arxiv.org/abs/1704.04861.
[15]Tan M X,Le Q V. EfficientNet:rethinking model scaling for convolutional neural networks[C]//2019 International Conference on Machine Learning. Long Beach:PMLR,2019:6105-6114.
[16]Zhang X Y,Zhou X Y,Lin M X,et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:6848-6856.
[17]Han K,Wang Y H,Tian Q,et al. GhostNet:more features from cheap operations[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle:IEEE,2020:1577-1586.
[18]施杰,林双双,张威,等. 基于轻量化改进型YOLO v5s的玉米病虫害检测方法[J]. 江苏农业学报,2024,40(3):427-437.
[19]Howard A,Sandler M,Chen B,et al. Searching for MobileNet v3[C]//2019 IEEE/CVF International Conference on Computer Vision. Seoul:IEEE,2019:1314-1324.
[20]李博,江朝晖,洪石兰,等. 基于边缘智能的茶叶病害识别[J]. 中国农机化学报,2022,43(6):175-180.
[21]Sandler M,Howard A,Zhu M L,et al. MobileNet v2:inverted residuals and linear bottlenecks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE,2018:4510-4520.
[22]谭彬,蔡健荣,许骞,等. 基于注意力机制改进卷积神经网络的柑橘病虫害识别[J]. 江苏农业科学,2024,52(8):176-182.
[23]Li L T,Zhao Y D. Tea disease identification based on ECA attention mechanism ResNet50 network [J]. Frontiers in Plant Science,2025,16:1489655.
[24]Wang Q L,Wu B G,Zhu P F,et al. ECA-Net:Efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle:IEEE,2020:11531-11539.
[25]Bao W X,Zhu Z Q,Hu G S,et al. UAV remote sensing detection of tea leaf blight based on DDMA-YOLO[J]. Computers and Electronics in Agriculture,2023,205:107637.
[26]Xiao Y,Wu J,Ma C. GA-GhostNet:a lightweight CNN model for identifying pests and diseases using a gated multi-scale coordinate attention mechanism [J]. Engineering Letters,2024,32(7):1281-1290.
[27]Hou Q B,Zhou D Q,Feng J S. Coordinate attention for efficient mobile network design[C]// IEEE.2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Nashville,2021:13708-13717.
[28]Bochkovskiy A,Wang C Y,Liao H M. YOLO v4:optimal speed and accuracy of object detection[EB/OL]. (2020-04-23)[2024-07-09]. https://arxiv.org/abs/2004.10934.
[29]Hu J,Shen L,Sun G. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE,2018:7132-7141.
[30]Woo S,Park J,Lee J Y,et al. CBAM:convolutional block attention module[C]//Computer Vision-ECCV 2018.Cham:Springer International Publishing,2018:3-19.
[31]Wang X L,Girshick R,Gupta A,et al. Non-local neural networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE,2018:7794-7803.
[32]Hughes D P,Salathe M. An open access repository of images on plant health to enable the development of mobile disease diagnostics[EB/OL]. (2015-11-25)[2024-07-09]. https://arxiv.org/abs/1511.08060.

相似文献/References:

[1]严陈慧子,田芳明,谭峰,等.基于改进YOLOv4的水稻病害快速检测方法[J].江苏农业科学,2023,51(6):187.
 Yanchen Huizi,et al.Rapid detection method of rice diseases based on improved YOLOv4[J].Jiangsu Agricultural Sciences,2023,51(15):187.
[2]张风伟,朱成杰,朱洪波.基于改进MobileNet v3的苹果叶片病害识别方法及移动端应用[J].江苏农业科学,2024,52(7):205.
 Zhang Fengwei,et al.Apple leaf disease recognition method based on improved MobileNet v3 and its mobile terminal application[J].Jiangsu Agricultural Sciences,2024,52(15):205.
[3]郑超杰,李少波,蒲睿强,等.基于轻量化卷积神经网络的番茄叶片病害识别[J].江苏农业科学,2024,52(11):225.
 Zheng Chaojie,et al.Tomato leaf disease recognition based on lightweight convolutional neural network[J].Jiangsu Agricultural Sciences,2024,52(15):225.
[4]李豫晋,沈陆明,何少芳,等.基于改进MobileNet v3的苹果叶片病害识别研究[J].江苏农业科学,2024,52(12):224.
 Li Yujin,et al.Identification of apple leaf diseases based on improved MobileNet v3[J].Jiangsu Agricultural Sciences,2024,52(15):224.
[5]周江龙,王天一,李论,等.基于深度学习的轻量化农作物叶片病害识别模型[J].江苏农业科学,2024,52(16):230.
 Zhou Jianglong,et al.A lightweight model for crop leaf disease identification based on deep learning[J].Jiangsu Agricultural Sciences,2024,52(15):230.
[6]陈雪,谭峰,宁韶瞳,等.基于轻量级网络的玉米田间杂草识别方法[J].江苏农业科学,2025,53(20):194.
 Chen Xue,et al.A weed identification method for corn field based on lightweight network[J].Jiangsu Agricultural Sciences,2025,53(15):194.
[7]王彩萍,卢雯皎,吴燕.轻量级深度学习模型在草莓病害识别中的应用——以改进MobileNet v3Small为例[J].江苏农业科学,2026,54(4):222.
 Wang Caiping,et al.Application of lightweight deep learning models in strawberry disease recognition: taking improved MobileNet v3small as an example[J].Jiangsu Agricultural Sciences,2026,54(15):222.

备注/Memo

备注/Memo:
收稿日期:2024-07-09
基金项目:苏州市科技计划(编号:SS202151、SNG2021037、SNGD202307);2023年江苏省高校优秀科技创新团队项目。
作者简介:董虎胜(1981—),男,江苏泗洪人,博士,副教授,主要从事深度学习与人工智能研究。E-mail:hsdong2012@QQ.com。
更新日期/Last Update: 2025-08-05