[1]杨如强,赵霞,张鑫.基于改进YOLOv11模型的柑橘叶片病害检测[J].江苏农业学报,2026,42(01):99-109.[doi:doi:10.3969/j.issn.1000-4440.2026.01.011]
　YANG Ruqiang,ZHAO Xia,ZHANG Xin.An improved YOLOv11 model for citrus leaf disease detection[J].,2026,42(01):99-109.[doi:doi:10.3969/j.issn.1000-4440.2026.01.011]
点击复制

基于改进YOLOv11模型的柑橘叶片病害检测()

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
42
期数:
2026年01期
页码:
99-109
栏目:
农业信息工程
出版日期:
2026-01-31

文章信息/Info

Title:
An improved YOLOv11 model for citrus leaf disease detection
作者:
杨如强; 赵霞; 张鑫
(甘肃农业大学信息科学技术学院,甘肃兰州730070)
Author(s):
YANG Ruqiang; ZHAO Xia; ZHANG Xin
(College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, China)
关键词:
柑橘; 病害检测; YOLOv11模型; 特征融合
Keywords:
citrus; disease detection; YOLOv11 model; feature fusion
分类号:
S127
DOI:
doi:10.3969/j.issn.1000-4440.2026.01.011
文献标志码:
A
摘要:
为提高果园场景下柑橘叶片病害的检测精度,本研究以YOLOv11模型为基准模型,引入GhostHGNetV2网络替换原有的主干网络,利用C3k2_MSEIS模块(C3k2_MutilScaleEdgeInformationSelect)和ADown模块替换原有模型颈部网络的C3k2模块和Conv模块,并利用MultiSEAMHead替换原有检测头,以增强模型对小目标的检测能力,实现多尺度特征融合与融合路径优化,提出一种改进的YOLOv11模型——YOLOv11-Citrus,最后利用柑橘园获取的4种常见叶片病害影像对模型的检测性能进行分析。结果表明,改进后的模型YOLOv11-Citrus对柑橘叶片病害的检测精确率、召回率和平均精度均值mAP50分别比基准模型提高3.4个百分点、1.1个百分点和2.6个百分点。与RE-DETR、YOLOv5、YOLOv8、YOLOv10n等主流模型相比,YOLOv11-Citrus模型对柑橘叶片病害的识别具有更高的检测精确率和平均精度均值mAP50。本研究结果为柑橘叶片病害的精准识别和有效防治提供了技术支持。
Abstract:
To improve the detection accuracy of citrus leaf diseases in orchard scenes, this study used the YOLOv11 model as the baseline model. The original backbone network was replaced with the GhostHGNetV2 network. The C3k2 and Conv modules in the original model’s neck network were replaced with the C3k2_MSEIS module (C3k2_MutilScaleEdgeInformationSelect) and the ADown module, respectively. Furthermore, the original detection head was substituted with the MultiSEAMHead to enhance the model’s ability to detect small targets, achieving multi-scale feature fusion and fusion path optimization. An improved YOLOv11 model (YOLOv11-Citrus) was proposed. Finally, the detection performance of the model was analyzed using images of four common leaf diseases collected from citrus orchards. The results showed that the improved YOLOv11-Citrus model achieved increases of 3.4 percentage points in detection accuracy, 1.1 percentage points in recall rate, and 2.6 percentage points in mean average precision (mAP50) compared with the baseline model. Compared with mainstream models such as RE-DETR, YOLOv5, YOLOv8, and YOLOv10n, the YOLOv11-Citrus model demonstrated higher detection accuracy and mean average precision (mAP50) in detecting citrus leaf diseases. The results of this study provide technical support for the accurate identification and effective control of citrus leaf diseases.

参考文献/References:

[1]帖军,赵捷,郑禄,等. 改进YOLOv5模型在自然环境下柑橘识别的应用[J]. 中国农业科技导报,2024,26(7):111-120.
[2]周秀珊,文露婷,介百飞,等. 改进YOLOv11的水面膨化饲料颗粒图像实时检测算法[J]. 智慧农业(中英文),2024,6(6):155-167.
[3]王会征,孙良晨,李新龙,等. 基于改进YOLOv7-tiny的番茄叶片病虫害检测方法[J]. 农业工程学报,2024,40(10):194-202.
[4]周维,牛永真,王亚炜,等. 基于改进的YOLOv4-GhostNet水稻病虫害识别方法[J]. 江苏农业学报,2022,38(3):685-695.
[5]徐彦威,李军,董元方,等. YOLO系列目标检测算法综述[J]. 计算机科学与探索,2024,18(9):2221-2238.
[6]LIU J, WANG X W. Tomato diseases and pests detection based on improved yolo V3 convolutional neural network[J]. Frontiers in Plant Science,2020,11:898.
[7]周俊昌,曾维,彭鹏,等. 苹果叶片病害检测的轻量化YOLOv5研究与实现[J]. 中国农机化学报,2025,46(4):126-132.
[8]曾晏林,贺壹婷,蔺瑶,等. 基于BCE-YOLOv5的苹果叶部病害检测方法[J]. 江苏农业科学,2023,51(15):155-163.
[9]郭嘉璇,王蓉芳,南江华,等. 融入全局相应归一化注意力机制的YOLOv5农作物害虫识别模型[J]. 农业工程学报,2024,40(8):159-170.
[10]郭文娟,冯全. 基于改进YOLO的棉花叶片病害检测[J]. 干旱地区农业研究,2024,42(6):195-205.
[11]施国英,纪嘉鹏,李天华,等. 基于改进YOLOv8s的杭白菊检测与花期分类[J]. 农业工程学报,2025,41(7):192-199.
[12]曾亮,彭龑. 基于改进YOLOv8的马铃薯叶片病害检测算法[J]. 洛阳理工学院学报(自然科学版),2024,34(3):62-69.
[13]党珊珊,白明宇,路扬,等. 基于改进YOLOv10n的玉米叶片病害检测研究[J]. 内蒙古民族大学学报(自然科学版),2025,40(2):69-76.
[14]靳学萌,梁西银,邓鹏飞. 基于改进YOLOv10的轻量级黄花菜分级检测模型[J]. 智慧农业(中英文),2024,6(5):108-118.
[15]CHEN J D, WANG W H, ZHANG D F, et al. Attention embedded lightweight network for maize disease recognition[J]. Plant Pathology,2021,70(3):630-642.
[16]GONG H, LIU T H, LUO T Y, et al. Based on FCN and DenseNet framework for the research of rice pest identification methods[J]. Agronomy,2023,13(2):410.
[17]NIE X, WANG L Y, DING H X, et al. Strawberry Verticillium wilt detection network based on multi-task learning and attention[J]. IEEE Access,2019,7:170003-170011.
[18]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.
[19]蒋龙泉,鲁帅,冯瑞,等. 基于多特征融合和SVM分类器的植物病虫害检测方法[J]. 计算机应用与软件,2014,31(12):186-190.
[20]CHAUDHARI D J, MALATHI K. Detection and prediction of rice leaf disease using a hybrid CNN-SVM model[J]. Optical Memory and Neural Networks,2023,32(1):39-57.
[21]韩宇,齐康康,郑纪业,等. 基于改进YOLOv11的轻量化肉牛面部识别方法[J]. 智慧农业(中英文),2025,7(3):173-184.
[22]杨森,陈义明. 基于YOLOv11-PMS的草莓目标检测方法[J]. 农业与技术,2025,45(16):1-7.
[23]刘鹏,张天翼,冉鑫,等. 基于PBM-YOLOv8的水稻病虫害检测[J]. 农业工程学报,2024,40(20):147-156.
[24]李彬,李生林. 改进YOLOv11n的无人机小目标检测算法[J]. 计算机工程与应用,2025,61(7):96-104.
[25]朱康凯,侯红玲,杨澳,等. 干涉配合铆接缺陷检测方法研究[J]. 塑性工程学报,2025,32(2):96-104.
[26]孙月平,孟祥汶,郭佩璇,等. 基于改进YOLOv8n的渔光互补池塘作业船导航中线提取方法[J]. 农业工程学报,2024,40(22):173-182.
[27]YUE M, ZHANG L Q, HUANG J, et al. Lightweight and efficient tiny-object detection based on improved YOLOv8n for UAV aerial images[J]. Drones,2024,8(7):276.
[28]张磊,李熙尉,燕倩如,等. 基于改进YOLOv5s的综采工作面人员检测算法[J]. 中国安全科学学报,2023,33(7):82-89.
[29]LI M, ZHANG J C, LIU H B, et al. A lightweight method for apple-on-tree detection based on improved YOLOv5[J]. Signal,Image and Video Processing,2024,18(10):6713-6727.
[30]许英超,刘书玮,王相友,等. 基于改进YOLOv8n的轻量化马铃薯表面缺陷在线检测方法[J]. 农业工程学报,2025,41(5):135-144.
[31]董一兵,曾辉,侯少杰. LMUAV-YOLOv8:低空无人机视觉目标检测轻量化网络[J]. 计算机工程与应用,2025,61(3):94-110.
[32]WANG Y, RONG Q J, HU C H. Ripe tomato detection algorithm based on improved YOLOv9[J]. Plants,2024,13(22):3253.
[33]YAN C M, XU E E. ECM-YOLO:a real-time detection method of steel surface defects based on multiscale convolution[J]. Journal of the Optical Society of America A,2024,41(10):1905-1914.
[34]李芳,危疆树,王玉超,等. 基于改进YOLOv8n模型的辣椒病害检测方法[J]. 江苏农业学报,2025,41(2):323-334.

相似文献/References:

[1]沈丹,杨莉,胡威,等.柑橘胁迫响应基因WRKY47的克隆与表达分析[J].江苏农业学报,2021,(01):129.[doi:doi:10.3969/j.issn.1000-4440.2021.01.017]
　SHEN Dan,YANG Li,HU Wei,et al.Cloning and expression analysis of stress response gene WRKY47 in citrus[J].,2021,(01):129.[doi:doi:10.3969/j.issn.1000-4440.2021.01.017]
[2]李仁杰,宋涛,高婕,等.基于改进YOLOv5的自然环境下番茄患病叶片检测模型[J].江苏农业学报,2024,(06):1028.[doi:doi:10.3969/j.issn.1000-4440.2024.06.009]
　LI Renjie,SONG Tao,GAO Jie,et al.Tomato diseased leaf detection model based on improved YOLOv5 in natural environment[J].,2024,(01):1028.[doi:doi:10.3969/j.issn.1000-4440.2024.06.009]
[3]赵玉清,贾奥莹,王白娟,等.柑橘果实采摘机器人目标识别算法研究进展[J].江苏农业学报,2024,(08):1552.[doi:doi:10.3969/j.issn.1000-4440.2024.08.019]
　ZHAO Yuqing,JIA Aoying,WANG Baijuan,et al.Research progress on target recognition algorithms for citrus fruit picking robots[J].,2024,(01):1552.[doi:doi:10.3969/j.issn.1000-4440.2024.08.019]

备注/Memo

备注/Memo:
收稿日期：2025-04-16基金项目：甘肃省自然科学基金项目(24JRRA656) 作者简介：杨如强(2001-),男,甘肃宕昌人,硕士研究生,研究方向为智慧农业和目标检测。(E-mail)2818300843@qq.com通讯作者：赵霞,(E-mail)58892778@qq.com
更新日期/Last Update: 2026-02-09