[1]杨玉青,朱德泉,刘凯旋,等.基于改进的LSN-YOLOv8模型和无人机遥感图像的水稻稻曲病检测方法[J].江苏农业学报,2025,(05):905-915.[doi:doi:10.3969/j.issn.1000-4440.2025.05.009]
　YANG Yuqing,ZHU Dequan,LIU Kaixuan,et al.A method for rice false smut detection based on improved LSN-YOLOv8 model and unmanned aerial vehicle remote sensing images[J].,2025,(05):905-915.[doi:doi:10.3969/j.issn.1000-4440.2025.05.009]
点击复制

基于改进的LSN-YOLOv8模型和无人机遥感图像的水稻稻曲病检测方法()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年05期
页码:
905-915
栏目:
农业信息工程
出版日期:
2025-05-31

文章信息/Info

Title:
A method for rice false smut detection based on improved LSN-YOLOv8 model and unmanned aerial vehicle remote sensing images
作者:
杨玉青1; 朱德泉1; 刘凯旋1; 严从宽1; 孟凡凯1; 唐七星1; 廖娟1; 2
(1.安徽农业大学工学院,安徽合肥230036;2.安徽农业大学新农村发展研究院皖东综合试验站,安徽明光239400)
Author(s):
YANG Yuqing1; ZHU Dequan1; LIU Kaixuan1; YAN Congkuan1; MENG Fankai1; TANG Qixing1; LIAO Juan1; 2
(1.School of Engineering, Anhui Agricultural University, Hefei 230036, China;2.Wan-Dong Comprehensive Experimental Station, New Rural Development Research Institute, Anhui Agricultural University, Mingguang 239400, China)
关键词:
稻曲病; 病害识别; 无人机; YOLOv8模型; 大选择性核网络(LSKNet); 坐标注意力机制(CA)
Keywords:
rice false smut; disease identification; unmanned aerial vehicle; YOLOv8 model; large selective kernel network (LSKNet); coordinate attention mechanism (CA)
分类号:
S435.115
DOI:
doi:10.3969/j.issn.1000-4440.2025.05.009
文献标志码:
A
摘要:
本研究针对无人机采集的水稻稻曲病图像中存在的背景复杂、病斑目标小且与背景表征相似等问题,构建了一种水稻稻曲病检测模型LSN-YOLOv8。该模型以YOLOv8模型为基本框架,在骨干网络中融入大选择性核网络(LSKNet),通过动态调整感受野范围增强模型对小目标的特征提取能力;在骨干网络中加入坐标注意力机制（CA）模块, 将病斑空间位置信息与通道注意力相结合,增强模型对关键区域的关注度同时减少背景干扰;利用梯度加权类激活映射(Grad-CAM)技术实现检测过程的可视化分析,为模型决策提供直观解释。为验证模型性能,利用无人机拍摄不同发病时期、不同背景条件下的水稻稻曲病图像,构建水稻稻曲病数据集,用于模型训练与测试。试验结果表明,本研究提出的LSN-YOLOv8模型精准度、召回率和交并比阈值为0.50时的平均精度值均值(mAP50)分别为94.8%、87.3%和92.3%,均高于YOLOv5、YOLOv7、YOLOv8、Faster R-CNN模型等经典目标检测模型。梯度加权类激活映射(Grad-CAM)技术可视化分析结果表明,LSN-YOLOv8模型能够更准确地聚焦于图像中的病害区域。本研究提出的LSN-YOLOv8模型可为稻曲病监测、病害防治和水稻抗病性鉴定提供技术支持。
Abstract:
To address the challenges of complex backgrounds, small lesion targets, and the similarity between lesion targets and background features in rice false smut images collected by unmanned aerial vehicles (UAVs), we proposed the LSN-YOLOv8 detection model. The model was based on the YOLOv8 framework, and the large selective kernel network (LSKNet) was incorporated into the backbone network. By dynamically adjusting the receptive field range, the model enhanced its ability to extract features of small targets. Additionally, a coordinate attention mechanism (CA) module was integrated into the backbone network to combine the spatial location information of lesions with channel attention, thereby enhancing the model’s focus on key regions while reducing background interference. The detection process was visualized and analyzed using the gradient-weighted class activation mapping (Grad-CAM) technique, thereby providing intuitive explanations for the model’s decision-making. To verify the model’s performance, rice false smut images captured by UAVs at different disease stages and under various background conditions were used to construct a rice false smut dataset. This dataset was utilized for model training and testing. The experimental results indicated that the precision, recall, and mean average precision at an intersection over union threshold of 0.50 (mAP50) of the LSN-YOLOv8 model proposed in this study were 94.8%, 87.3%, and 92.3%, respectively. These indices were all higher than those of classic object detection models such as YOLOv5, YOLOv7, YOLOv8 and Faster R-CNN. The visualization analysis results using Grad-CAM technology indicated that the LSN-YOLOv8 model was capable of more accurately focusing on the diseased regions in the images. The LSN-YOLOv8 model proposed in this study can provide technical support for the monitoring of rice false smut, disease control and prevention, and the identification of rice disease resistance.

参考文献/References:

[1]BIN RAHMAN A N M R, ZHANG J H. Trends in rice research:2030 and beyond[J]. Food and Energy Security,2023,12(2):e390.
[2]SUN W X, FAN J, FANG A F, et al. Ustilaginoidea virens:insights into an emerging rice pathogen[J]. Annual Review of Phytopathology,2020,58:363-385.
[3]QIU J, MENG S, DENG Y, et al. Ustilaginoidea virens:a fungus infects rice flower and threats world rice production[J]. Rice Science,2019,26(4):199-206.
[4]ZHOU L, MUBEEN M, IFTIKHAR Y, et al. Rice false smut pathogen:implications for mycotoxin contamination,current status,and future perspectives[J]. Frontiers in Microbiology,2024,15:1344831.
[5]ROY A, SAHU P K, DAS C, et al. Conventional and new-breeding technologies for improving disease resistance in lentil (Lens culinaris Medik)[J]. Frontiers in Plant Science,2023,13:1001682.
[6]陆煜,俞经虎,朱行飞,等. 基于卷积神经网络的轻量级水稻叶片病害识别模型[J]. 江苏农业学报,2024,40(2):312-319.
[7]BUJA I, SABELLA E, MONTEDURO A G, et al. Advances in plant disease detection and monitoring:from traditional assays to in-field diagnostics[J]. Sensors,2021,21(6):2129.
[8]ZENG W, LI M. Crop leaf disease recognition based on self-attention convolutional neural network[J]. Computers and Electronics in Agriculture,2020,172:105341.
[9]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.
[10]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,2024,40(6):1028-1037.
[11]WANG J, WANG P X, TIAN H R, et al. A deep learning framework combining CNN and GRU for improving wheat yield estimates using time series remotely sensed multi-variables[J]. Computers and Electronics in Agriculture,2023,206:107705.
[12]鲍文霞,吴育桉,胡根生,等. 基于改进RDN网络的无人机茶叶图像超分辨率重建[J]. 农业机械学报,2023,54(4):241-249.
[13]HU G S, YAO P, WAN M Z, et al. Detection and classification of diseased pine trees with different levels of severity from UAV remote sensing images[J]. Ecological Informatics,2022,72:101844.
[14]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.
[15]TETILA E C, MACHADO B B, ASTOLFI G, et al. Detection and classification of soybean pests using deep learning with UAV images[J]. Computers and Electronics in Agriculture,2020,179:105836.
[16]孙钰,周焱,袁明帅,等. 基于深度学习的森林虫害无人机实时监测方法[J]. 农业工程学报,2018,34(21):74-81.
[17]KERKECH M, HAFIANE A, CANALS R. Vine disease detection in UAV multispectral images using optimized image registration and deep learning segmentation approach[J]. Computers and Electronics in Agriculture,2020,174:105446.
[18]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.
[19]LI Y X, LI X, DAI Y M, et al. Lsknet:a foundation lightweight backbone for remote sensing[J]. International Journal of Computer Vision,2024,133:1410-1431.
[20]SONG Y F, ZHANG Z, SHAN C F, et al. Constructing stronger and faster baselines for skeleton-based action recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,45(2):1474-1488.
[21]JAHMUNAH V, NG E Y K, AN R S, et al. Explainable detection of myocardial infarction using deep learning models with Grad-CAM technique on ECG signals[J]. Computers in Biology and Medicine,2022,146:105550.
[22]SHEN L Y, LANG B H, SONG Z X. DS-YOLOv8-Based object detection method for remote sensing images[J]. IEEE Access,2023,11:125122-125137.
[23]SOLIMANI F, CARDELLICCHIO A, DIMAURO G, et al. Optimizing tomato plant phenoty detection:boosting YOLOv8 architecture to tackle data complexity[J]. Computers and Electronics in Agriculture,2024,218:108728.
[24]YANG S Z, WANG W, GAO S, et al. Strawberry ripeness detection based on YOLOv8 algorithm fused with LW-Swin Transformer[J]. Computers and Electronics in Agriculture,2023,215:108360.
[25]PAN P, GUO W L, ZHENG X M, et al. Xoo-YOLO:a detection method for wild rice bacterial blight in the field from the perspective of unmanned aerial vehicles[J]. Frontiers in Plant Science,2023,14:1256545.

相似文献/References:

[1]宋益民,丛国林.稻曲病、水稻穗颈瘟防治药剂筛选及其协同防治技术[J].江苏农业学报,2016,(06):1268.[doi:doi:10.3969/j.issn.1000-4440.2016.06.012]
　SONG Yi-min,CONG Guo-lin.Fungicide screen on rice neck blast, rice false smut and preliminary study on its synergistic technique[J].,2016,(05):1268.[doi:doi:10.3969/j.issn.1000-4440.2016.06.012]
[2]张俊喜,成晓松,宋益民,等.中国水稻稻曲病研究进展[J].江苏农业学报,2016,(01):234.[doi:10.3969/j.issn.1000-4440.2016.01.037 ]
　ZHANG Jun-xi,CHENG Xiao-song,SONG Yi-min,et al.Research advances in rice false smut (Ustilaginoidea virens) in China[J].,2016,(05):234.[doi:10.3969/j.issn.1000-4440.2016.01.037 ]
[3]张善文,谢泽奇,张晴晴.卷积神经网络在黄瓜叶部病害识别中的应用[J].江苏农业学报,2018,(01):56.[doi:doi:10.3969/j.issn.1000-4440.2018.01.008]
　ZHANG Shan-wen,XIE Ze-qi,ZHANG Qing-qing.Application research on convolutional neural network for cucumber leaf disease recognition[J].,2018,(05):56.[doi:doi:10.3969/j.issn.1000-4440.2018.01.008]
[4]杨晋丹,杨涛,苗腾,等.基于卷积神经网络的草莓叶部白粉病病害识别[J].江苏农业学报,2018,(03):527.[doi:doi:10.3969/j.issn.1000-4440.2018.03.007]
　YANG Jin-dan,YANG Tao,MIAO Teng,et al.Recognition of powdery mildew disease of strawberry leaves based on convolutional neural network[J].,2018,(05):527.[doi:doi:10.3969/j.issn.1000-4440.2018.03.007]
[5]胡玲艳,周婷,刘艳,等.基于轻量级网络自适应特征提取的番茄病害识别[J].江苏农业学报,2022,38(03):696.[doi:doi:10.3969/j.issn.1000-4440.2022.03.015]
　HU Ling-yan,ZHOU Ting,LIU Yan,et al.Tomato disease recognition based on lightweight network auto-adaptive feature extraction[J].,2022,38(05):696.[doi:doi:10.3969/j.issn.1000-4440.2022.03.015]

备注/Memo

备注/Memo:
收稿日期：2024-07-22基金项目：国家重点研发计划项目(2022YFD2001801-3);国家自然科学基金项目(32201665)作者简介：杨玉青(2001-),男,安徽六安人,硕士研究生,研究方向为农作物病害识别。(E-mail)23720754@stu.ahau.edu.cn通讯作者：廖娟,(E-mail)liaojuan@ahau.edu.cn
更新日期/Last Update: 2025-06-24