MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions
文献类型： 外文期刊
作者： Chen, Zili 1 ; Peng, Yilong 1 ; Jiao, Jiadong 2 ; Wang, Aiguo 4 ; Wang, Laigang 1 ; Lin, Wei 2 ; Guo, Yan 1 ;
作者机构： 1.Henan Acad Agr Sci, Inst Agr Informat Technol, Zhengzhou 450002, Peoples R China
2.Henan Normal Univ, Coll Comp & Informat Engn, Henan Prov Key Lab Educ Artificial Intelligence &, Xinxiang 453007, Henan, Peoples R China
3.Minist Agr & Rural Affairs, Key Lab Huang Huai Hai Smart Agr Technol, Zhengzhou 450002, Peoples R China
4.Zhengzhou Tobacco Res Inst CNTC, Zhengzhou 450001, Peoples R China
关键词： Deep learning; Tobacco leaf diseases; Lesion segmentation; Convolutional neural networks
期刊名称：SCIENTIFIC REPORTS （ 影响因子：3.9； 五年影响因子：4.3 ）
ISSN： 2045-2322
年卷期： 2025 年 15 卷 1 期
页码：
收录情况： SCI
摘要： Identification and diagnosis of tobacco diseases are prerequisites for the scientific prevention and control of these ailments. To address the limitations of traditional methods, such as weak generalization and sensitivity to noise in segmenting tobacco leaf lesions, this study focused on four tobacco diseases: angular leaf spot, brown spot, wildfire disease, and frog eye disease. Building upon the Unet architecture, we developed the Multi-scale Residual Dilated Segmentation Model (MD-Unet) by enhancing the feature extraction module and integrating attention mechanisms. The results demonstrated that MD-Unet achieved 92.75%, 90.94%, 84.93%, and 91.81% for the lesion CPA, recall, IoU, and F1 metrics, respectively, with an overall Dice score of 94.67%. Furthermore, the model parameters, floating-point operations, and inference time per single image for MD-Unet were 4.65 x 107, 2.3392 x 1011, and 65.096 ms, respectively. Compared to Unet, PSP, DeepLab v3+, FCN, SegNet, UNET++, and DoubleU-Net, MD-Unet significantly improved accuracy while effectively managing model complexity, achieving optimal overall performance. This work provides the theoretical foundations and technical support for precise segmentation of tobacco lesions, with potential applications in the segmentation of other plant diseases.
- 相关文献
作者其他论文 更多>>
-
Crop sample prediction and early mapping based on historical data: Exploration of an explainable FKAN framework
作者：Cheng, Feifei;Qiu, Bingwen;Yang, Peng;Wu, Wenbin;Yu, Qiangyi;Qian, Jianping;Wu, Bingfang;Chen, Jin;Chen, Xuehong;Tubiello, Francesco N.;Tryjanowski, Piotr;Takacs, Viktoria;Duan, Yuanlin;Lin, Lihui;Wang, Laigang;Zhang, Jianyang;Dong, Zhanjie
关键词：Historical Data; Sample generation; Crop mapping; Interpretability; Google Earth Engine
-
A robust framework for mapping complex cropping patterns: The first national-scale 10 m map with 10 crops in China using Sentinel 1/2 images
作者：Qiu, Bingwen;Wu, Fangzheng;Hu, Xiang;Yang, Peng;Wu, Wenbin;Qian, Jianping;Chen, Jin;Chen, Xuehong;He, Liyin;Joe, Berry;Tubiello, Francesco N.;Wang, Laigang
关键词：Cropping patterns mapping; Model generalization; Dual-driven models; Crop diversity; Sentinel-1/2
-
Improving the Accuracy of Soil Classification by Using Vis-NIR, MIR, and Their Spectra Fusion
作者：Li, Shuo;Shen, Xinru;Shen, Xue;Li, Shuo;Shen, Xinru;Guo, Yan;Shen, Xue;Cheng, Jun;Xu, Dongyun;Makar, Randa S.;Guo, Yan;Hu, Bifeng;Chen, Songchao;Shi, Zhou;Chen, Songchao;Hong, Yongsheng;Peng, Jie
关键词：spectroscopy; proximal sensing; random forest; partial least-squares discriminant analysis (PLSDA); outer product analysis (OPA)
-
Automatic Rice Early-Season Mapping Based on Simple Non-Iterative Clustering and Multi-Source Remote Sensing Images
作者：Wang, Gengze;Chen, Riqiang;Yang, Guijun;Feng, Haikuan;Wang, Gengze;Chen, Riqiang;Yang, Guijun;Feng, Haikuan;Meng, Di;Jin, Hailiang;Ge, Xiaosan;Wang, Laigang;Feng, Haikuan
关键词：early-season rice mapping; spectral index (SI); synthetic aperture radar (SAR); Simple Non-Iterative Clustering (SNIC); time series filtering; K-Means; Jeffries-Matusita (JM) distance
-
Improvement of Winter Wheat Aboveground Biomass Estimation Using Digital Surface Model Information Extracted from Unmanned-Aerial-Vehicle-Based Multispectral Images
作者：Guo, Yan;He, Jia;Zhang, Huifang;Wei, Panpan;Jing, Yuhang;Yang, Xiuzhong;Zhang, Yan;Wang, Laigang;Zheng, Guoqing;Guo, Yan;He, Jia;Zhang, Huifang;Wei, Panpan;Jing, Yuhang;Yang, Xiuzhong;Zhang, Yan;Zheng, Guoqing;Guo, Yan;Yang, Xiuzhong;Zhang, Yan;Zheng, Guoqing;Shi, Zhou;Wang, Laigang
关键词：aboveground biomass; UAV; height; transferability; BP neural network; machine learning
-
National-scale 10-m maps of cropland use intensity in China during 2018-2023
作者：Qiu, Bingwen;Liu, Baoli;Xu, Weiming;Liang, Juanzhu;Chen, Nan;Li, Zhengrong;Wu, Fangzheng;Tang, Zhenghong;Dong, Jinwei;Chen, Jiangping;Wang, Laigang;Zhang, Chengming
关键词：
-
Where is tea grown in the world: A robust mapping framework for agroforestry crop with knowledge graph and sentinels images
作者：Peng, Yufeng;Qiu, Bingwen;Xu, Weiming;Li, Mengmeng;Liang, Juanzhu;Huang, Yingze;Cheng, Feifei;Chen, Jianfeng;Wu, Fangzheng;Jian, Zeyu;Li, Zhengrong;Tang, Zhenghong;Yang, Peng;Wu, Wenbin;Chen, Xuehong;Zhu, Xiaolin;Zhu, Peng;Zhu, Peng;Zhang, Xin;Wang, Xinshuang;Zhang, Chengming;Wang, Laigang;Qiu, Bingwen
关键词：Special cash crop; Agroforestry crop mapping; Tea plantation; Phenology-based algorithm; Sentinel-1/2
