[1]夏顺兴,倪铭,罗友璐,等.基于改进YOLOv8n的草莓叶片病害检测方法[J].江苏农业学报,2025,(04):664-675.[doi:doi:10.3969/j.issn.1000-4440.2025.04.005]
　XIA Shunxing,NI Ming,LUO Youlu,et al.Strawberry leaf disease detection method based on improved YOLOv8n[J].,2025,(04):664-675.[doi:doi:10.3969/j.issn.1000-4440.2025.04.005]
点击复制

基于改进YOLOv8n的草莓叶片病害检测方法()

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年04期
页码:
664-675
栏目:
植物保护
出版日期:
2025-04-30

文章信息/Info

Title:
Strawberry leaf disease detection method based on improved YOLOv8n
作者:
夏顺兴; 倪铭; 罗友璐; 贺英豪; 赵涛涛
(四川农业大学信息工程学院,四川雅安625014)
Author(s):
XIA Shunxing; NI Ming; LUO Youlu; HE Yinghao; ZHAO Taotao
(College of Information Engineering, Sichuan Agricultural University, Ya’an 625014,China)
关键词:
草莓; 叶片病害; 目标检测; YOLOv8n; 动态卷积; GSConv; CARAFE
Keywords:
strawberry; leaf diseases; object detection; YOLOv8n; dynamic convolution; GSConv; CARAFE
分类号:
TP391.4
DOI:
doi:10.3969/j.issn.1000-4440.2025.04.005
文献标志码:
A
摘要:
为提高目标检测模型在果园环境中对草莓叶片病害的检测能力,本研究以YOLOv8n模型为基准模型,引入动态卷积(Dynamic convolution)模块替换骨干网络的第3个卷积层和颈部网络的部分C2f模块,引入GSConv和Slim-neck模块替换颈部网络的卷积层和部分C2f模块,并引入内容感知特征重组上采样（CARAFE）算子替换上采样中的最近邻插值法,提出了一种改进的YOLOv8n模型——YOLOv8n-DGC,以期在保持模型轻量化的同时提高模型对草莓叶片病害的检测精度。结果表明,改进后的模型YOLOv8n-DGC对草莓叶片病害检测的交并比（IoU）阈值为0.50时的平均精度均值（mAP50）、IoU阈值范围为0.50～0.95的平均精度均值（mAP50∶95）、精确率和召回率分别比基准模型提高2.5个百分点、1.5个百分点、1.6个百分点和1.6个百分点,模型大小和参数量分别增加3.2%和3.3%,而浮点运算量减少8.6%。与Faster R-CNN、SSD、YOLOv5s、YOLOv7-tiny等模型相比,YOLOv8n-DGC模型能更好地实现检测精度与效率的平衡,更适合布置到轻量化的检测设备或终端中。
Abstract:
In order to improve the detection ability of target detection models for strawberry leaf diseases in orchard environment, this study used the YOLOv8n model as the baseline model, introduced the dynamic convolution module to replace the third convolution layer of the backbone network and part of the C2f module of the neck network, introduced the GSConv and Slim-neck module to replace the convolution layer and part of the C2f module of the neck network, and introduced the content-aware reassembly of features (CARAFE) operator to replace the nearest neighbor interpolation method in upsampling. An improved YOLOv8n model named YOLOv8n-DGC was proposed to improve the detection accuracy of strawberry leaf diseases while maintaining the lightweight of the model. The results showed that the mean average precision when the intersection over union (IoU) threshold was 0.50 (mAP50), the mean average precision when the IoU was between 0.50 and 0.95 (mAP50∶95), precision and recall rate of the improved model YOLOv8n-DGC for strawberry leaf disease detection were 2.5 percentage points, 1.5 percentage points, 1.6 percentage points and 1.6 percentage points higher than those of the baseline model, respectively. The model size and parameter quantity increased by 3.2% and 3.3%, respectively, while the number of floating point of operations decreased by 8.6%. Compared with models such as Faster R-CNN, SSD, YOLOv5s, and YOLOv7-tiny, the YOLOv8n-DGC model better achieved a balance between detection accuracy and efficiency, and was more suitable for placement in lightweight detection equipment or terminals.

参考文献/References:

[1]曹丹,张明,王竞,等. 土壤改良剂对连作草莓生长及品质的影响[J]. 现代农业科技,2024(3):52-55.
[2]潘少香,刘雪梅,郑晓冬,等. 基于质构分析的鲜食草莓质地感官品质综合评价[J]. 食品科技,2023,48(8):37-43.
[3]唐歆玥,曹叶婷,邓明益,等. 柠檬醛对草莓贮藏期间品质和抗氧化能力的影响[J]. 食品安全质量检测学报,2024,15(2):131-140.
[4]杨海艳,王洪玲,钟国跃,等. 草莓属植物资源分布、化学成分、药理活性研究进展[J]. 中成药,2022,44(2):510-518.
[5]联合国粮食及农业组织（FAO）. 粮农组织统计数据库——粮食和农业数据[DB/OL].
[2024-10-01]. http://www.fao.org/faostat/zh/#Data/QCL.
[6]刘玉娟,刘颜达,闫振,等. 注意力机制的混合卷积高光谱图像分类方法[J]. 光谱学与光谱分析,2024,44(10):2916-2922.
[7]GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//IEEE. 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus,OH,USA:IEEE,2014:580-587.
[8]GIRSHICK R. Fast R-CNN[C]//IEEE. 2015 IEEE International Conference on Computer Vision (ICCV). Santiago,Chile:IEEE,2015:1440-1448.
[9]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.
[10]REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified,real-time object detection[C]//IEEE. 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,NV,USA:IEEE,2016:779-788.
[11]REDMON J, FARHADI A. YOLO9000:better,faster,stronger[C]//IEEE:2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI,USA:IEEE,2017:6517-6525.
[12]REDMON J, FARHADI A. YOLOv3:an incremental improvement[EB/OL]. (2018-04-08). https://arxiv.org/abs/1804.02767v1.
[13]BOCHKOVSKIY A, WANG C Y, LIAO H M. YOLOv4:optimal speed and accuracy of object detection[EB/OL]. (2020-04-23). https://arxiv. org/abs/2004.10934v1.
[14]LIU W, ANGUELOV D, ERHAN D, et al. SSD:single shot MultiBox detector[M]//ECCV. Computer Vision-ECCV 2016. Cham:Springer International Publishing,2016:21-37.
[15] 杨宇游, 潘文林. 基于改进YOLOv5的草莓病害检测[J]. 云南民族大学学报(自然科学版),2024,33(5):624-629.
[16]LI Y, WANG J C, WU H R, et al. Detection of powdery mildew on strawberry leaves based on DAC-YOLOv4 model[J]. Computers and Electronics in Agriculture,2022,202:107418.
[17]公徐路,张淑娟. 基于改进YOLOv5s的苹果叶片小目标病害轻量化检测方法[J]. 农业工程学报,2023,39(19):175-184.
[18]时雷,杨程凯,雷镜楷,等. 基于改进YOLOv8s的小麦小穗赤霉病检测研究[J]. 农业机械学报,2024,55(7):280-289.
[19]胡根生,谢一帆,鲍文霞,等. 基于轻量型网络的无人机遥感图像中茶叶枯病检测方法[J]. 农业机械学报,2024,55(4):165-175.
[20]罗友璐,潘勇浩,夏顺兴,等. 基于改进YOLOv8的苹果叶病害轻量化检测算法[J]. 智慧农业(中英文),2024,6(5):128-138.
[21]贺英豪,唐德钊,倪铭,等. 基于改进YOLOv5对果园环境中李的识别[J]. 华中农业大学学报,2024,43(5):31-40.
[22]HAN K, WANG Y H, GUO J Y, et al. ParameterNet:parameters are all you need for large-scale visual pretraining of mobile networks[C]//IEEE. 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle,WA,USA:IEEE,2024:15751-15761.
[23]涂智荣,凌海英,李帼,等. 基于改进YOLOv7-Tiny的轻量化百香果检测方法[J]. 广西师范大学学报(自然科学版),2024,42(5):79-90.
[24]CHOLLET F. Xception:deep learning with depthwise separable convolutions[C]//IEEE:2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,HI,USA:IEEE,2017:1800-1807.
[25]LI H L, LI J, WEI H B, et al. Slim-neck by GSConv:a lightweight-design for real-time detector architectures[J]. Journal of Real-Time Image Processing,2024,21(3):62.
[26]张佳承,韦锦,陈义时. 改进YOLOv8的实时轻量化鲁棒绿篱检测算法[J/OL]. 计算机工程
[2024-10-01]. https://doi.org/10.19678/j.issn.1000-3428.0069524.
[27]ZHANG X Y, ZHOU X Y, LIN M X, et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[C]//IEEE. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City,UT,USA:IEEE,2018:6848-6856.
[28]WANG J Q, CHEN K, XU R, et al. CARAFE:content-aware ReAssembly of FEatures[C]//IEEE. 2019 IEEE/CVF International Conference on Computer Vision. Seoul,Korea(South):IEEE,2019:3007-3016.
[29]张顺,龚怡宏,王进军. 深度卷积神经网络的发展及其在计算机视觉领域的应用[J]. 计算机学报,2019,42(3):453-482.

相似文献/References:

[1]巫建华,贾思振,冯英娜,等.茉莉酸合成关键酶基因FaOPR3调控草莓果实成熟[J].江苏农业学报,2016,(05):1148.[doi:10.3969/j.issn.1000-4440.2016.05.031]
　WU Jian-hua,JIA Si-zhen,FENG Ying-na,et al.Regulation of strawberry fruit ripening by jasmonic acid synthesis gene FaOPR3[J].,2016,(04):1148.[doi:10.3969/j.issn.1000-4440.2016.05.031]
[2]赵密珍,庞夫花,袁华招,等.不同栽培条件下草莓品种宁玉花序分化进程[J].江苏农业学报,2016,(01):196.[doi:10.3969/j.issn.1000-4440.2016.01.030 ]
　ZHAO Mi-zhen,PANG Fu-hua,YUAN Hua-zhao,et al.Floral bud differentiation of strawberry Ningyu under conditions of forcing cultivation and open field cultivation[J].,2016,(04):196.[doi:10.3969/j.issn.1000-4440.2016.01.030 ]
[3]吴祥,姚克兵,吉沐祥,等.句容地区草莓枯萎病病原菌的分离鉴定及田间防治[J].江苏农业学报,2015,(04):764.[doi:10.3969/j.issn.1000-4440.2015.04.009]
　WU Xiang,YAO Ke-bing,JI Mu-xiang,et al.Isolation and identification and field control of strawberry fusarium wilt in Jurong, Jiangsu province[J].,2015,(04):764.[doi:10.3969/j.issn.1000-4440.2015.04.009]
[4]周游,李海梅,赵金山,等.乳酸菌对草莓生长和品质性状的影响[J].江苏农业学报,2017,(05):1124.[doi:doi:10.3969/j.issn.1000-4440.2017.05.025]
　ZHOU You,LI Hai-mei,ZHAO Jin-shan,et al.Effects of lactic acid bacteria on the growth and quality characters of strawberry[J].,2017,(04):1124.[doi:doi:10.3969/j.issn.1000-4440.2017.05.025]
[5]江景勇,王会福,何玲玲,等.台州草莓农药残留风险评估[J].江苏农业学报,2017,(06):1408.[doi:doi:10.3969/j.issn.1000-4440.2017.06.030]
　JIANG Jing-yong,WANG Hui-fu,HE Ling-ling,et al.Risk assessment of pesticide residues in strawberry from Taizhou city[J].,2017,(04):1408.[doi:doi:10.3969/j.issn.1000-4440.2017.06.030]
[6]丛嘉昕,宋江峰,李大婧,等.基于RSM和MLP-ANN的草莓果浆超声酶解参数优化[J].江苏农业学报,2018,(06):1354.[doi:doi:10.3969/j.issn.1000-4440.2018.06.022]
　CONG Jia-xin,SONG Jiang-feng,LI Da-jing,et al.Optimization of ultrasonic enzymolysis of strawberry pulp based on RSM and MLP-ANN[J].,2018,(04):1354.[doi:doi:10.3969/j.issn.1000-4440.2018.06.022]
[7]姜玉素,李珍,王庆莲,等.草莓KEA家族基因的克隆、鉴定及表达分析[J].江苏农业学报,2019,(02):391.[doi:doi:10.3969/j.issn.1000-4440.2019.02.021]
　JIANG Yu-su,LI Zhen,WANG Qing-lian,et al.Cloning, characterization and expression analysis of KEA family genes in strawberry[J].,2019,(04):391.[doi:doi:10.3969/j.issn.1000-4440.2019.02.021]
[8]王廷峰,赵密珍,关玲,等.玉米套作及秸秆还田对草莓连作土壤养分及微生物区系的影响[J].江苏农业学报,2019,(06):1421.[doi:doi:10.3969/j.issn.1000-4440.2019.06.022]
　WANG Ting-feng,ZHAO Mi-zhen,GUAN-Ling,et al.Effects of intercropping with corn and straw returning on nutrients and microflora in strawberry continuous cropping soil[J].,2019,(04):1421.[doi:doi:10.3969/j.issn.1000-4440.2019.06.022]
[9]张宽朝,陈杰,马伟,等.响应面法优化草莓总黄酮提取工艺[J].江苏农业学报,2019,(06):1450.[doi:doi:10.3969/j.issn.1000-4440.2019.06.026]
　ZHANG Kuan-chao,CHEN Jie,MA Wei,et al.Optimization on extraction technology of total flavonoids from strawberry by response surface methodology[J].,2019,(04):1450.[doi:doi:10.3969/j.issn.1000-4440.2019.06.026]
[10]李梦雅,陈莎莎,王世梅.连作草莓低发病土壤与高发病土壤理化性质及生物学特征差异比较[J].江苏农业学报,2021,(04):910.[doi:doi:10.3969/j.issn.1000-4440.2021.04.013]
　LI Meng-ya,CHEN Sha-sha,WANG Shi-mei.Contrast of physicochemical properties and biological characteristics of low disease and high disease soils of continuous cropping strawberry[J].,2021,(04):910.[doi:doi:10.3969/j.issn.1000-4440.2021.04.013]

备注/Memo

备注/Memo:
收稿日期：2024-10-11基金项目：四川省自然科学基金面上项目(2022NSFSC0172)作者简介：夏顺兴(2001-),男,四川眉山人,硕士研究生,主要从事深度学习和目标检测研究。(E-mail)xsxaca@163.com通讯作者：倪铭,(E-mail)nm@sicau.edu.cn
更新日期/Last Update: 2025-05-26