MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions
Abstract 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 f...
Saved in:
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2025-01-01
|
Series: | Scientific Reports |
Subjects: | |
Online Access: | https://doi.org/10.1038/s41598-025-87128-y |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832585889212006400 |
---|---|
author | Zili Chen Yilong Peng Jiadong Jiao Aiguo Wang Laigang Wang Wei Lin Yan Guo |
author_facet | Zili Chen Yilong Peng Jiadong Jiao Aiguo Wang Laigang Wang Wei Lin Yan Guo |
author_sort | Zili Chen |
collection | DOAJ |
description | Abstract 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 × 107, 2.3392 × 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. |
format | Article |
id | doaj-art-fa37eb85948d4c479f7fbb112d0bb1fc |
institution | Kabale University |
issn | 2045-2322 |
language | English |
publishDate | 2025-01-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Scientific Reports |
spelling | doaj-art-fa37eb85948d4c479f7fbb112d0bb1fc2025-01-26T12:26:27ZengNature PortfolioScientific Reports2045-23222025-01-0115111610.1038/s41598-025-87128-yMD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutionsZili Chen0Yilong Peng1Jiadong Jiao2Aiguo Wang3Laigang Wang4Wei Lin5Yan Guo6Institute of Agricultural Information Technology, Henan Academy of Agricultural SciencesInstitute of Agricultural Information Technology, Henan Academy of Agricultural SciencesCollege of Computer and Information Engineering, Henan Normal University/Henan Provincial Key Laboratory of Educational Artificial Intelligence and Personalized LearningZhengzhou Tobacco Research Institute of CNTCInstitute of Agricultural Information Technology, Henan Academy of Agricultural SciencesCollege of Computer and Information Engineering, Henan Normal University/Henan Provincial Key Laboratory of Educational Artificial Intelligence and Personalized LearningInstitute of Agricultural Information Technology, Henan Academy of Agricultural SciencesAbstract 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 × 107, 2.3392 × 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.https://doi.org/10.1038/s41598-025-87128-yDeep learningTobacco leaf diseasesLesion segmentationConvolutional neural networks |
spellingShingle | Zili Chen Yilong Peng Jiadong Jiao Aiguo Wang Laigang Wang Wei Lin Yan Guo MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions Scientific Reports Deep learning Tobacco leaf diseases Lesion segmentation Convolutional neural networks |
title | MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions |
title_full | MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions |
title_fullStr | MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions |
title_full_unstemmed | MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions |
title_short | MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions |
title_sort | md unet for tobacco leaf disease spot segmentation based on multi scale residual dilated convolutions |
topic | Deep learning Tobacco leaf diseases Lesion segmentation Convolutional neural networks |
url | https://doi.org/10.1038/s41598-025-87128-y |
work_keys_str_mv | AT zilichen mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT yilongpeng mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT jiadongjiao mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT aiguowang mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT laigangwang mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT weilin mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions AT yanguo mdunetfortobaccoleafdiseasespotsegmentationbasedonmultiscaleresidualdilatedconvolutions |