[1]范明超,高欣峰.基于改进YOLOv11的葡萄叶片害虫检测[J].江苏农业科学,2026,54(4):289-300.
Fan Mingchao,et al.Detection of grape leaf pests based on improved YOLO v11[J].Jiangsu Agricultural Sciences,2026,54(4):289-300.
点击复制
基于改进YOLOv11的葡萄叶片害虫检测(
)
《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]
- 卷:
-
第54卷
- 期数:
-
2026年第4期
- 页码:
-
289-300
- 栏目:
-
病虫害智能检测
- 出版日期:
-
2026-02-20
文章信息/Info
- Title:
-
Detection of grape leaf pests based on improved YOLO v11
- 作者:
-
范明超; 高欣峰
-
青岛农业大学动漫与传媒学院,山东青岛 266109
- Author(s):
-
Fan Mingchao; et al
-
-
- 关键词:
-
葡萄叶片害虫; C3K2_PConvX; EUCB; S_MCA; MPDIoU
- Keywords:
-
-
- 分类号:
-
S126;TP391.41
- DOI:
-
-
- 文献标志码:
-
A
- 摘要:
-
葡萄是全球重要的果树作物,害虫高发已成为限制其稳产优质的关键因素。传统人工识别方法存在效率低、主观性强等不足,难以满足精准防控需求。为实现葡萄叶片害虫的快速与高精度检测,提出改进模型YOLO v11-GrapePest。首先构建覆盖大青叶蝉、十星叶甲、绿盲蝽等7类典型害虫的数据集,并结合传统数据增强与Gemini-2.5-Flash-Image模型生成雾天、夜天与雨天等复杂环境样本,以提升数据多样性与模型泛化能力。在模型结构上,以C3K2多支路并行架构为基础,嵌入PConv并融合深度可分离卷积、ECA注意力与轻量化MLP,形成C3K2_PConvX模块,从而替代原生C3K2;采用EUCB模块优化颈部上采样结构;在MCA注意力机制基础上,引入稀疏约束与Softmax自适应加权,提出改进的S_MCA注意力机制;将CIoU损失替换为MPDIoU,以提升边界框回归精度与收敛速度。结果表明,YOLO v11-GrapePest的精确率、召回率和mAP@0.5分别达到95.1%、90.5%与 95.0%,相较原始YOLO v11分别提升2.4、3.8、2.4百分点,同时降低了模型参数和计算量,推理速度提升至85.2帧/s,在检测精度、轻量化与实时性方面均优于RT-DETR-R18、Faster R-CNN、YOLO v8n等主流模型。YOLO v11-GrapePest能在复杂背景和多变天气条件下稳定识别葡萄害虫,为葡萄害虫的智能监测与精准防控提供有效技术支撑。
- Abstract:
-
-
参考文献/References:
[1]吕书平,郭荣荣,再吐娜·买买提,等.鲜食葡萄贮藏保鲜技术应用的思考[J]. 中国农技推广,2019,35(9):24-27.
[2]吕科,张建海,董雪洁,等.促早栽培措施对葡萄果实品质影响的研究进展[J]. 中外葡萄与葡萄酒,2020,45(2):64-67.
[3]马雯轩,屠超凡,张浩源,等. 我国葡萄常见病虫害及防治措施[J]. 现代农业科技,2025,54(20):52-56.
[4]杨勇,杨俊祥,宫霞,等. 葡萄及葡萄属植物中的天然活性物质研究与利用现状[J]. 酿酒科技,2011,35(6):75-79.
[5]王灿,杨晨露,王华,等.葡萄籽油生理功能及应用综述[J]. 中外葡萄与葡萄酒,2020,45(6):66-71.
[6]徐敏. 葡萄的病虫害防治与田间管理技术[J]. 农业灾害研究,2022,12(1):10-12.
[7]韦峰. 葡萄栽培管理和病虫害防治技术[J]. 种子科技,2025,43(11):135-137.
[8]夏如兵. 建国以来中国昆虫学的主要成就及其发展动因[D]. 南京:南京农业大学,2002:68.
[9]潘鑫. 基于空间金字塔注意力的水稻虫害分类方法研究[D]. 长春:吉林农业大学,2024:3-4.
[10]Girshick R,Donahue J,Darrell T,et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus,OH,USA:IEEE,2014:580-587.
[11]Ren S Q,He K M,Girshick R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[12]Redmon J,Divvala S,Girshick R,et al. You only look once:unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,NV,USA:IEEE,2016:779-788.
[13]Redmon J,Farhadi A. YOLO9000:better,faster,stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI:IEEE,2017:7263-7271.
[14]Zhao Y A,Lv W Y,Xu S L,et al. DETRs beat YOLOs on real-time object detection[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle,WA,USA:IEEE,2024:16965-16974.
[15]Vaswani A,Shazeer N,Parmar N,et al. Attention is all you need[EB/OL]. arXiv preprint(2023-08-02)[2025-05-08]. https://arxiv.org/abs/1706.03762.
[16]李颖多,孙国玺,李新超. 基于机器视觉的轴承表面缺陷检测与分拣系统设计[J/OL]. 机床与液压,2025:1-8(2025-10-30)[2025-11-17]. https://link.cnki.net/urlid/44.1259.th.20251029.1713.070.
[17]曹亮,肖伟,李湘丽. YOLO算法及其在农作物识别及病虫害检测应用综述[J]. 仲恺农业工程学院学报,2025,38(6):60-71.
[18]周月. 基于深度学习的葡萄病虫害识别[D]. 杭州:浙江农林大学,2024:1-58.
[19]刘港. 基于深度学习的葡萄虫害识别算法研究[D]. 雅安:四川农业大学,2024:1-68.
[20]祁昕. 基于深度学习的葡萄病虫害识别模型的构建及应用研究[D]. 西安:西安建筑科技大学,2022:1-52.
[21]Wang C X,Wang L Y,Ma G,et al. CSF-YOLO:a lightweight model for detecting grape leafhopper damage levels[J]. Agronomy,2025,15(3):741.
[22]Wang C X,Tian W X,Ma G,et al. Accurate identification of grade of grape damage by Brevipalpus spp.based on the improved YOLO v8n model[J]. Applied Sciences,2025,15(5):2712.
[23]Khanam R,Hussain M. YOLO v11:an overview of the key architectural enhancements[J]. arXiv preprint(2024-10-23)[2025-05-08]. https://arxiv.org/abs/2410.17725.
[24]Chen J R,Kao S H,He H,et al. Run,dont walk:chasing higher FLOPS for faster neural networks[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Vancouver,BC,Canada:IEEE,2023:12021-12031.
[25]吴晗. 基于红外与可见光图像融合的光伏组件热斑检测方法研究[D]. 合肥:安徽建筑大学,2025:43.
[26]姜晟久,钟国韵. 基于可分离扩张卷积和通道剪枝的番茄病害分类方法[J]. 江苏农业科学,2024,52(2):182-189.
[27]Balasundaram A,Mohanty A,Shaik A,et al. Zero-DCE++ inspired object detection in less illuminated environment using improved YOLOv5[J]. Computers,Materials & Continua,2023,77(3):2751-2769.
[28]Rukundo O,Cao H. Nearest neighbor value interpolation[J]. arXiv preprint(2019-05-04)[2025-05-08]. https://arxiv.org/abs/1211.1768.
[29]谢从珍,黄梦成,黄奕琅,等. 基于WMEF-YOLO的绝缘子缺陷图像检测方法[J/OL]. 高电压技术,2025:1-13(2025-07-14)[202 5-11-19]. https://doi.org/10.13336/j.1003-6520.hve.20250429.
[30]Rahman M M,Munir M,Marculescu R. EMCAD:efficient multi-scale convolutional attention decoding for medical image segmentation[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle,WA,USA:IEEE,2024:11769-11779.
[31]Wang C Y,Bochkovskiy A,Liao H M. YOLOv7:trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Vancouver,BC,Canada:IEEE,2023:7464-7475.
[32]Yu Y,Zhang Y,Cheng Z Y,et al. MCA:multidimensional collaborative attention in deep convolutional neural networks for image recognition[J]. Engineering Applications of Artificial Intelligence,2023,126:107079.
[33]Ma S,Xu Y. Mpdiou:a loss for efficient and accurate bounding box regression[J]. arXiv preprint (2023-07-14)[2025-05-08]. https://arxiv.org/abs/2307.07662.
备注/Memo
- 备注/Memo:
-
收稿日期:2025-12-22
基金项目:2024年度青岛市科技惠民示范专项(编号:24-1-8-xdny-12-nsh)。
作者简介:范明超(2001—),女,山东青岛人,硕士研究生,主要从事农业工程与信息技术研究。E-mail:shun960707@163.com。
通信作者:高欣峰,博士,教授,主要从事农业传播研究。E-mail:xfgao@qau.edu.cn。
更新日期/Last Update:
2026-02-20