基于改进YOLOv8模型的玉米叶斑病快速识别方法

doi:10.14088/j.cnki.issn0439-8114.2025.05.025

湖北农业科学 ›› 2025, Vol. 64 ›› Issue (5): 160-166.doi: 10.14088/j.cnki.issn0439-8114.2025.05.025

基于改进YOLOv8模型的玉米叶斑病快速识别方法

张露, 吴雪莲

长江大学经济与管理学院,湖北荆州 434023

收稿日期:2024-10-31 出版日期:2025-05-25 发布日期:2025-06-11
通讯作者: 吴雪莲(1978-),女,湖北荆州人,教授,主要从事智慧农业研究,(电话)13554481394(电子信箱)21712215@qq.com。
作者简介:张露（1996-）,女,湖北恩施人,在读硕士研究生,主要从事农业可持续发展研究,（电话）18971870613（电子信箱）744459240@qq.com
基金资助:
湖北省教育厅科学研究计划项目（D20211302）

A rapid identification method for maize leaf spot disease based on the improved YOLOv8 model

ZHANG Lu, WU Xue-lian

Economics and Management School, Yangtze University, Jingzhou 434023, Hubei, China

Received:2024-10-31 Published:2025-05-25 Online:2025-06-11

摘要/Abstract

摘要： 为了实现玉米叶斑病的快速识别,通过集成全局注意力机制（Global attention module,GAM）、Slim-Neck轻量化模块及Inner-CIoU损失函数,优化YOLOv8模型的玉米叶斑病检测性能。与YOLOv8模型相比,改进YOLOv8模型（GAM+Slim-Neck+Inner-CIoU）的Precision、Recall、mAP@0.5和mAP@[0.5∶0.95]分别增加4.15%、5.51%、3.91%和11.35%,参数量和检测时间分别减少10.39%和3.42%。改进后的YOLOv8模型在Precision、Recall、mAP@0.5和mAP@[0.5∶0.95]方面普遍优于其他模型（YOLOv3、YOLOv5、YOLOv6及Faster R-CNN）,同时在参数量和检测时间上也表现出显著的优势,兼具高效性与轻量化特点。改进后的YOLOv8模型能够更高效地捕获关键信息,充分融合多维度特征,合理分配计算资源,从而提升识别准确率。

关键词: 玉米, 叶斑病, 改进, YOLOv8模型, 快速识别

Abstract: To achieve rapid identification of maize leaf spot disease, the detection performance of the improved YOLOv8 model was optimized by integrating the Global Attention Module (GAM), Slim-Neck lightweight module, and Inner-CIoU loss function. Compared with the original YOLOv8 model, the improved YOLOv8 model (GAM+Slim-Neck+Inner-CIoU) showed increases of 4.15% in Precision, 5.51% in Recall, 3.91% in mAP @0.5, and 11.35% in mAP @[0.5∶0.95], while the number of parameters and detection time decreased by 10.39% and 3.42%, respectively. The improved YOLOv8 model outperformed other models (YOLOv3, YOLOv5, YOLOv6, and Faster R-CNN) in Precision, Recall, mAP @0.5, and mAP @[0.5∶0.95], while also demonstrating significant advantages in parameter quantity and detection time, combining high efficiency with lightweight characteristics. The improved YOLOv8 model efficiently captured critical information, fully integrated multi-dimensional features, and rationally allocated computational resources, thereby enhancing recognition accuracy.

Key words: maize, leaf spot disease, improvement, YOLOv8 model, rapid identification

中图分类号:

张露, 吴雪莲. 基于改进YOLOv8模型的玉米叶斑病快速识别方法[J]. 湖北农业科学, 2025, 64(5): 160-166.

ZHANG Lu, WU Xue-lian. A rapid identification method for maize leaf spot disease based on the improved YOLOv8 model[J]. HUBEI AGRICULTURAL SCIENCES, 2025, 64(5): 160-166.

参考文献

[1] 赵玉霞. 基于图像识别的玉米叶部病害诊断技术研究[D].北京:北京邮电大学,2007.
[2] 樊湘鹏,周建平,许燕.基于改进区域卷积神经网络的田间玉米叶部病害识别[J].华南农业大学学报, 2020, 41(6):10.
[3] 慕君林,马博,王云飞,等.基于深度学习的农作物病虫害检测算法综述[J].农业机械学报, 2023, 54(S2):301-313.
[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] 杨锋,姚晓通.基于改进YOLOv8的小麦叶片病虫害检测轻量化模型[J].智慧农业中英文, 2024, 6(1):147-157.
[6] 刘润飞. 基于改进YOLOv8的棉花虫害检测算法[J].农业工程, 2024, 14(7):42-47.
[7] 朱婷茹. 基于无人机影像的松林变色立木识别方法研究[D].南京:南京林业大学,2023.
[8] 高伟锋. 基于YOLOv8的柑橘病虫害识别系统研究与设计[J].智慧农业导刊, 2023, 3(15):27-30.
[9] 张书贵,陈书理,赵展.改进YOLOv8的农作物叶片病虫害识别算法[J].中国农机化学报, 2024,45(7):255-260.
[10] 李志良,李梦霞,董勇,等.基于改进YOLO v8的轻量化玉米害虫识别方法[J].江苏农业科学, 2024, 52(14):196-206.
[11] SHRIVASTAVA A, GUPTA A.Contextual priming and feedback for faster R-CNN[A].Computer vision-ECCV 2016: 14th European conference[C].Amsterdam, Netherlands:Springer, 2016.
[12] HE K, ZHANG X, REN S, et al.Spatial pyramid pooling in deep convolutional networks for visual recognition[J].IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9):1904-1916.
[13] DEVRIES T, TAYLOR G W.Improved regularization of convolutional neural networks with cutout[J].arXiv preprint arXiv:170804552. 2017.
[14] ZHANG H, CISSE M, DAUPHIN Y N,et al.Mixup: Beyond empirical risk minimization[J].arXiv preprint arXiv:171009412. 2017.
[15] HENDRYCKS D, MU N, CUBUK E D, et al.Augmix: A simple data processing method to improve robustness and uncertainty[J].arXiv preprint arXiv:191202781. 2019.
[16] ROY A M, BHADURI J.DenseSPH-YOLOv5: An automated damage detection model based on denseNet and swin-transformer prediction head-enabled YOLOv5 with attention mechanism[J].Advanced engineering informatics,2023,56:102007.
[17] ROY A M, BHADURI J, KUMAR T, et al.WilDect-YOLO: An efficient and robust computer vision-based accurate object localization model for automated endangered wildlife detection[J].Ecological informatics,2023,75:101919.
[18] WU D, JIANG S, ZHAO E, et al.Detection of Camellia oleifera fruit in complex scenes by using YOLOv7 and data augmentation[J].Applied sciences,2022,12(22):11318.
[19] ZHAO Q, YANG L, LYU N.A driver stress detection model via data augmentation based on deep convolutional recurrent neural network[J].Expert systems with applications,2024, 238:122056.
[20] LI F,ZHANG H,XU H,et al.Mask dino: Towards a unified transformer-based framework for object detection and segmentation[A].IEEE conference on computer vision and pattern recognition[C].IEEE,2023. 3041-3050.
[21] YE Z, GUO Q, WEI J,et al.Recognition of terminal buds of densely-planted Chinese fir seedlings using improved YOLOv5 by integrating attention mechanism[J].Frontiers in plant science,2022,13:991929.
[22] ZHENG Z, HU Y, QIAO Y, et al.Real-time detection of winter jujubes based on improved YOLOX-nano network[J].Remote sensing,2022,14(19):4833.
[23] HU J, SHEN L, SUN G.Squeeze-and-excitation networks[A].Proceedings of the IEEE conference on computer vision and pattern recognition[C].IEEE,2018. 7132-7141.

基于改进YOLOv8模型的玉米叶斑病快速识别方法

A rapid identification method for maize leaf spot disease based on the improved YOLOv8 model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刘玉耀, 彭琼尹. 基于改进Faster R-CNN模型的丁岙杨梅成熟度检测方法[J]. 湖北农业科学, 2025, 64(4): 7-13.
[2]	王明辉, 夏振洲, 颜鸿远, 谈仲川, 林田生, 陈志刚, 王斌, 李宁, 熊飞, 丁凤菊, 邹彩琼. 湖北省道地药材射干生态育苗套种玉米技术[J]. 湖北农业科学, 2025, 64(4): 42-45.
[3]	周洋, 张强, 高士伟, 崔清梅, 梁金波, 黄忠浩. 恩施玉露茶精揉机的搓板改进及测试[J]. 湖北农业科学, 2025, 64(3): 162-166.
[4]	赵小平, 董众祥, 冯瑾霞, 闫学龙, 胡晶洁. 基于改进NanoDet-Plus网络的树木小目标病虫害识别[J]. 湖北农业科学, 2025, 64(2): 192-196.
[5]	常晓, 王小博, 刘柏林, 杨兆生, 李健. 遮阴处理对玉米子粒败育及产量的影响[J]. 湖北农业科学, 2025, 64(1): 1-6.
[6]	余长平, 周华平, 叶青松, 吴承国, 肖能武, 秦光明, 李永学, 王致云, 陈强, 刘永忠. 高产优质抗病玉米新品种达美5号的选育、特性分析及栽培技术[J]. 湖北农业科学, 2024, 63(9): 1-5.
[7]	黄珍, 景月楼. 基于改进后局部搜索算法的猕猴桃预估产量在种植区域的分配性能[J]. 湖北农业科学, 2024, 63(8): 92-95.
[8]	侯晨连, 张吴平, 王国芳, 李富忠. 基于WOFOST模型与遥感数据的旱作玉米估产及精度评价[J]. 湖北农业科学, 2024, 63(8): 209-215.
[9]	王宦臣, 张吴平, 李富忠. 复合种植方式下大豆和玉米株高的动态反演[J]. 湖北农业科学, 2024, 63(8): 231-235.
[10]	李淑君, 付忠军, 祁志云, 袁亮. 子粒镉低积累型甜糯玉米材料的筛选[J]. 湖北农业科学, 2024, 63(7): 1-5.
[11]	彭晓丹, 孔令成. 黄河生态经济带玉米用水效率地区差异及空间收敛性分析——基于环境规制视角[J]. 湖北农业科学, 2024, 63(7): 192-199.
[12]	冯亮, 施爱平. 农业废弃物好氧发酵高温阶段工艺[J]. 湖北农业科学, 2024, 63(5): 17-21.
[13]	柴凤兰, 张帆, 王文辉, 许兆斐. 秸秆木醋液的制备及其斑蝥复合剂抗菌活性[J]. 湖北农业科学, 2024, 63(5): 56-60.
[14]	于德涵, 朱颖怡, 黎莉, 武海玲. 超声辅助有机酸预处理玉米芯复合酶解工艺优化[J]. 湖北农业科学, 2024, 63(5): 168-174.
[15]	宋振帅, 宋龙, 周艳, 何磊, 朱贺, 王治民, 韩大龙. 基于改进U-Net的冬季休眠期矮化苹果树修剪枝条分割方法[J]. 湖北农业科学, 2024, 63(5): 194-200.