MI-YOLO多光谱无人机图像中松材线虫病轻度变色木检测

邵欣欣; 刘文萍; 王晗; 宗世祥; 袁博

doi:10.12302/j.issn.1000-2006.202501038

邵欣欣, 刘文萍, 王晗, 宗世祥, 袁博

南京林业大学学报（自然科学版） ›› 2026, Vol. 50 ›› Issue (2) : 19-28.

PDF(4476 KB)

国家林草科技领军期刊
中国精品科技期刊
中国高校百佳科技期刊
江苏省新闻出版政府奖期刊奖
RCCSE林学权威期刊（A+）
CSCD核心期刊
Scopus数据库收录期刊
中文核心期刊
SCD核心期刊

作者加群：102861116

微信公众号：南京林业大学学报

高级检索

PDF(4476 KB)

南京林业大学学报（自然科学版） ›› 2026, Vol. 50 ›› Issue (2) : 19-28. DOI: 10.12302/j.issn.1000-2006.202501038

专题报道（Ⅰ）：森林资源智能感知与精准监测（执行主编李凤日曹林张怀清）

MI-YOLO多光谱无人机图像中松材线虫病轻度变色木检测

邵欣欣 ¹ ,
刘文萍 ¹^,^* ,
王晗 ¹ ,
宗世祥 ² ,
袁博 ¹

作者信息 +

MI-YOLO model for detecting mildly discolored pine trees infected with pine wilt disease with multispectral UAV images

SHAO Xinxin ¹ ,
LIU Wenping ¹^,^* ,
WANG Han ¹ ,
ZONG Shixiang ² ,
YUAN Bo ¹

Author information +

文章历史 +

摘要

【目的】基于YOLOv8n提出一种多光谱图像目标检测算法(multispectral images YOLO,MI-YOLO),以准确、实时检测多光谱无人机图像中松材线虫病(pine wilt disease, PWD)轻度变色木。【方法】利用频域互功率谱快速实现光谱图像配准;针对轻度变色木与健康木之间差异小而难检测的问题,使用轻量版多分支辅助特征金字塔结构,增强模型颈部网络特征利用和融合;最后采用轻量化模块C2f-Faster替换YOLOv8n中的特征融合模块C2f以减少冗余计算。【结果】所提出的MI-YOLO目标检测算法在IoU为0.5时的平均精度(average precision at IoU 0.5,AP₅₀)达到84.5%,模型参数量(params)为2.1 MB,浮点运算量(floating point operations,FLOPs)为7.25 GB,推理速度为50帧/s。与YOLOv8n相比,AP₅₀提高10.5个百分点,参数量降低30%,浮点运算数降低13%。【结论】MI-YOLO目标检测模型准确度高且具有轻量化的优点,可实时应用于多光谱无人机图像检测松材线虫病轻度变色木。

Abstract

【Objective】This research aims to accurately and efficiently detect mildly discolored pine trees infected with pine wilt disease in real time using multispectral unmanned aerial vehicle (UAV) remote sensing images, a Multispectral Images YOLO (MI-YOLO) model is proposed based on YOLOv8n.【Method】First, multispectral images are rapidly aligned using the cross power spectrum in the frequency domain. Second, a tiny multi-branch auxiliary feature pyramid network is introduced as the neck network to enhance feature utilization while maintaining model lightweight. Finally, the original C2f feature fusion module in YOLOv8n is replaced with a lightweight C2f-Faster module to reduce redundant computation.【Result】The proposed MI-YOLO model achieves an average precision of 84.5% at an IoU threshold of 0.5 (AP₅₀), with 2.1 MB parameters and 7.25 GB floating point operations (FLOPs). Compared with YOLOv8n, AP₅₀ is improved by 10.5 percentage points, while the number of parameters and FLOPs are reduced by 30% and 13%, respectively.【Conclusion】The MI-YOLO object detection model exhibits high accuracy and a lightweight structure, enabling real-time detection of mildly discolored pine trees infected with pine wilt disease in multispectral images.

导出引用

邵欣欣, 刘文萍, 王晗, 等. MI-YOLO多光谱无人机图像中松材线虫病轻度变色木检测[J]. 南京林业大学学报（自然科学版）. 2026, 50(2): 19-28 https://doi.org/10.12302/j.issn.1000-2006.202501038

SHAO Xinxin, LIU Wenping, WANG Han, et al. MI-YOLO model for detecting mildly discolored pine trees infected with pine wilt disease with multispectral UAV images[J]. Journal of Nanjing Forestry University (Natural Sciences Edition）. 2026, 50(2): 19-28 https://doi.org/10.12302/j.issn.1000-2006.202501038

中图分类号： S763.7;TP391.4

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	朱宁溪. 宁波市松材线虫病疫情继续“双下降”[J]. 浙江林业, 2019(12):27. ZHU N X. The epidemic situation of Bursaphelenchus xylophilus in Ningbo continued to “double decline”[J]. Zhejiang Forestry, 2019(12):27. 本文引用 [1]

[2]	叶建仁, 吴小芹. 松材线虫病研究进展[J]. 中国森林病虫, 2022, 41(3):1-10. YE J R, WU X Q. Research progress of pine wilt disease[J]. Forest Pest and Disease, 2022, 41(3):1-10. DOI: 10.19688/j.cnki.issn1671-0886.20220026. 本文引用 [1]

[3]

徐华潮, 骆有庆, 张廷廷, 等. 松材线虫自然侵染后松树不同感病阶段针叶光谱特征变化[J]. 光谱学与光谱分析, 2011, 31(5):1352-1356.

H C

, LUO

Y Q

, ZHANG

T T

, et al. Changes of reflectance spectra of pine needles in different stage after being infected by pine wood nematode[J]. Spectroscopy and Spectral Analysis, 2011, 31(5):1352-1356. DOI: 10.3964/j.issn.1000-0593(2011)05-1352-05.

本文引用 [1]

[4]	WU D W, YU L F, YU R, et al. Detection of the monitoring window for pine wilt disease using multi-temporal UAV-based multispectral imagery and machine learning algorithms[J]. Remote Sensing, 2023, 15(2):444. DOI: 10.3390/rs15020444. 本文引用 [1]

[5]	IORDACHE M D, MANTAS V, BALTAZAR E, et al. A machine learning approach to detecting pine wilt disease using airborne spectral imagery[J]. Remote Sensing, 2020, 12(14):2280. DOI: 10.3390/rs12142280. 本文引用 [1]

[6]	WANG S K, CAO X W, WU M Q, et al. Detection of pine wilt disease using drone remote sensing imagery and improved YOLOv8 algorithm:a case study in Weihai,China[J]. Forests, 2023, 14(10):2052. DOI: 10.3390/f14102052. 本文引用 [1]

[7]	YUAN Q B, ZOU S H, WANG H J, et al. A lightweight pine wilt disease detection method based on vision transformer-enhanced YOLO[J]. Forests, 2024, 15(6):1050. DOI: 10.3390/f15061050. 本文引用 [1]

[8]	YE X Q, PAN J, LIU G S, et al. Exploring the close-range detection of UAV-based images on pine wilt disease by an improved deep learning method[J]. Plant Phenomics, 2023, 5:0129. DOI: 10.34133/plantphenomics.0129. 本文引用 [1]

[9]	DU Z J, WU S F, WEN Q Q, et al. Pine wilt disease detection algorithm based on improved YOLOv5[J]. Frontiers in Plant Science, 2024, 15:1302361. DOI: 10.3389/fpls.2024.1302361. 本文引用 [1]

[10]	姚博文. 可见光-近红外图像参数化配准及光流计算提升方法[D]. 杭州: 浙江大学, 2023. YAO B W. Boosting parametric registration and optical flow computation of RGB-NIR images[D]. Hangzhou: Zhejiang University, 2023. DOI: 10.27461/d.cnki.gzjdx.2023.003790. 本文引用 [1]

[11]	BIAN C F, SHI H T, WU S Q, et al. Prediction of field-scale wheat yield using machine learning method and multi-spectral UAV data[J]. Remote Sensing, 2022, 14(6):1474. DOI: 10.3390/rs14061474. 本文引用 [1]

[12]	FEI S P, HASSAN M A, HE Z H, et al. Assessment of ensemble learning to predict wheat grain yield based on UAV-multispectral reflectance[J]. Remote Sensing, 2021, 13(12):2338. DOI: 10.3390/rs13122338. 本文引用 [1]

[13]	SALGADO L, LÓPEZ-SÁNCHEZ C A, COLINA A, et al. Hg and as pollution in the soil-plant system evaluated by combining multispectral UAV-RS, geochemical survey and machine learning[J]. Environmental Pollution, 2023, 333:122066. DOI: 10.1016/j.envpol.2023.122066. 本文引用 [1]

[14]	XIAO D Q, PAN Y Q, Feng J Z, et al. Remote sensing detection algorithm for apple fire blight based on UAV multispectral image[J]. Computers and Electronics in Agriculture, 2022, 199:107137. DOI: 10.1016/j.compag.2022.107137. 本文引用 [1]

[15]	FERRO M V, SØRENSEN C G, CATANIA P. Comparison of different computer vision methods for vineyard canopy detection using UAV multispectral images[J]. Computers and Electronics in Agriculture, 2024, 225:109277. DOI: 10.1016/j.compag.2024.109277. 本文引用 [1]

[16]	蔡佩华. 多源遥感影像松树林区域提取与松材线虫病变色木识别[D]. 荆州: 长江大学, 2023. CAI P H. Extraction of pine forest area and identification of pine wilt disease colored trees based on multi-source remote sensing imagery[D]. Jingzhou: Yangtze University, 2023. DOI: 10.26981/d.cnki.gjhsc.2023.000297. 本文引用 [1]

[17]

谢雨涵, 史建康, 孙晓慧, 等. 基于大疆精灵4无人机多光谱影像的西沙植被监测[J]. 遥感技术与应用, 2022, 37(5):1170-1178.

XIE

Y H

, SHI

J K

, SUN

X H

, et al. Xisha vegetation monitoring based on UAV multispectral images obtained with the DJI phantom 4 platform[J]. Remote Sensing Technology and Application, 2022, 37(5):1170-1178. DOI: 10.11873/j.issn.1004-0323.2022.5.1170.

本文引用 [1]

[18]	王德传. 利用U-Net神经网络的多光谱图像草地特征提取[J]. 地理空间信息, 2024, 22(8):41-44. WANG D C. Grass feature extraction for multi-spectral images based on U-Net neural network[J]. Geospatial Information, 2024, 22(8):41-44. DOI: 10.3969/j.issn.1672-4623.2024.08.009. 本文引用 [1]

[19]	孙昆, 武一, 牛雅睿, 等. 基于多阶段交叉信息融合的多光谱行人检测[J]. 电子测量技术, 2023, 46(15):118-125. SUN K, WU Y, NIU Y R, et al. Multi-spectral pedestrian detection based on multi-stage cross information fusion[J]. Electronic Measurement Technology, 2023, 46(15):118-125. DOI: 10.19651/j.cnki.emt.2211842. 本文引用 [1]

[20]	朱宇. 基于多光谱信息融合的行人检测研究[D]. 无锡: 江南大学, 2021. ZHU Y. Research on pedestrian detection based on multispectral information fusion[D]. Wuxi: Jiangnan University, 2021. DOI: 10.27169/d.cnki.gwqgu.2021.000554. 本文引用 [1]

[21]	ZHAO T Y, YUAN M X, JIANG F, et al. Removal then selection: a coarse-to-fine fusion perspective for RGB-infrared object detection[EB/OL]. (2024-01-19). [2025-01-10]. 2024: arXiv: 2401.10731. https://arxiv.org/abs/2401.10731. https://arxiv.org/abs/2401.10731 本文引用 [1]

[22]

刘世川, 王庆, 唐晴, 等. 基于多特征提取与注意力机制深度学习的高分辨率影像松材线虫病树识别[J]. 林业工程学报, 2022, 7(1):177-184.

LIU

S C

, WANG

, TANG

, et al. High-resolution image identification of trees with pinewood nematode disease based on multi-feature extraction and deep learning of attention mechanism[J]. Journal of Forestry Engineering, 2022, 7(1):177-184. DOI: 10.13360/j.issn.2096-1359.202211015.

本文引用 [1]

[23]	TU Z Z, LI Z, LI C L, et al. Multi-interactive dual-decoder for RGB-thermal salient object detection[J]. IEEE Transactions on Image Processing, 2021, 30:5678-5691. DOI: 10.1109/TIP.2021.3087412. 本文引用 [1]

[24]	WANG L N, CAI J J, WANG T T, et al. Detection of pine wilt disease using AAV remote sensing with an improved YOLO model[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17:19230-19242. DOI: 10.1109/JSTARS.2024.3478333. 本文引用 [1]

[25]	谢婉滢, 刘文萍, 王晗. 无人机松林图像早期松材线虫病害检测[J]. 林业科学, 2024, 60(9):124-133. XIE W Y, LIU W P, WANG H. UAV images of pine forests for early detection of pine wood nematode infestation[J]. Scientia Silvae Sinicae, 2024, 60(9):124-133. DOI: 10.11707/j.1001-7488.LYKX20220898. 本文引用 [1]

[26]

奚小波, 丁杰源, 翁小祥, 等. 基于轻量化YOLO v5s-MCA的番茄成熟度检测方法[J]. 农业机械学报, 2025, 56(3):383-391,436.

X B

, DING

J Y

, WENG

X X

, et al. Tomato maturity detection method based on lightweight YOLO v5s-MCA[J]. Transactions of the Chinese Society for Agricultural Machinery, 2025, 56(3):383-391,436. DOI: 10.6041/j.issn.1000-1298.2025.03.038.

本文引用 [1]

[27]	P 4 Multispectral Image Processing Guide[EB/OL]. (2020-07-17). [2024-12-05] https://dl.djicdn.com/downloads/p4-multispectral/20200717/P4_Multispectral_Image_Processing_Guide_EN.pdf. https://dl.djicdn.com/downloads/p4-multispectral/20200717/P4_Multispectral_Image_Processing_Guide_EN.pdf 本文引用 [1]

[28]	KUGLIN C D, HINES D C. The phase correlation image alignment method[C]// IEEE International Conference on Cybernetics and Society, October 26-28, 1975, New York. New York: IEEE, 1975:163-165. 本文引用 [1]

[29]	SUN X, XIAO B, WEI F Y, et al. Integral human pose regression[C]// Proceedings of the European Conference on Computer Vision (ECCV). Cham, Switzerland: Springer International Publishing, 2018:536-553. DOI: 10.1007/978-3-030-01231-1_33. 本文引用 [1]

[30]

CHEN

G G

, DAI

K X

, YANG

K Z

, et al. Bracketing image restoration and enhancement with high-low frequency decomposition[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops. New York: IEEE, 2024:6097-6107. DOI: 10.1109/CVPRW63382.2024.00616.

本文引用 [1]

[31]	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2018:8759-8768. DOI: 10.1109/CVPR.2018.00913. 本文引用 [1]

[32]

YANG

Z Q

, GUAN

, ZHAO

K E

, et al. Multi-branch auxiliary fusion YOLO with re-parameterization heterogeneous convolutional for accurate object detection[C]// Pattern Recognition and Computer Vision: PRCV 2024. Singapore: Springer Nature Singapore, 2025:492-505. DOI: 10.1007/978-981-97-8858-3_34.

本文引用 [1]

[33]	TAN M X, PANG R M, LE Q V. EfficientDet: scalable and efficient object detection[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2020:10778-10787. DOI: 10.1109/CVPR42600.2020.01079. 本文引用 [2]

[34]	CHEN J R, KAO S H, HE H, et al. Run,don’t walk: chasing higher FLOPS for faster neural networks[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2023:12021-12031. DOI: 10.1109/CVPR52729.2023.01157. 本文引用 [1]

[35]	SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM:visual explanations from deep networks via gradient-based localization[C]// 2017 IEEE International Conference on Computer Vision (ICCV 2017). New York: IEEE, 2017:618-626. DOI: 10.1109/ICCV.2017.74. 本文引用 [1]

[36]	KANG M, TING C M, TING F F, et al. ASF-YOLO: a novel YOLO model with attentional scale sequence fusion for cell instance segmentation[J]. Image and Vision Computing, 2024, 147:105057. DOI: 10.1016/j.imavis.2024.105057. 本文引用 [1]

[37]	XU X Z, JIANG Y Q, CHEN W H, et al. DAMO-YOLO:a report on real-time object detection design[EB/OL]. (2022-11-23). [2025-01-10]. 2022: arXiv: 2211.15444. https://arxiv.org/abs/2211.15444. https://arxiv.org/abs/2211.15444 本文引用 [1]