Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security

<p dir="ltr">Wheat is an essential crop that plays a vital role in global food security, but is susceptible to a variety of diseases, which have the potential to drastically decrease crop productivity. Detection of disease at an early stage and in an accurate manner is crucial to min...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Muhammad Khubaib (22927822) (author)
مؤلفون آخرون: Tanzila Kehkashan (20748842) (author), Maha Abdelhaq (735574) (author), Muhammad Asghar Khan (13411285) (author), Muhammad Zaman (66868) (author), Imran Ashraf (7370771) (author), Abdul Rehman (604331) (author), Adnan Akhunzada (20151648) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513523555500032
author Muhammad Khubaib (22927822)
author2 Tanzila Kehkashan (20748842)
Maha Abdelhaq (735574)
Muhammad Asghar Khan (13411285)
Muhammad Zaman (66868)
Imran Ashraf (7370771)
Abdul Rehman (604331)
Adnan Akhunzada (20151648)
author2_role author
author
author
author
author
author
author
author_facet Muhammad Khubaib (22927822)
Tanzila Kehkashan (20748842)
Maha Abdelhaq (735574)
Muhammad Asghar Khan (13411285)
Muhammad Zaman (66868)
Imran Ashraf (7370771)
Abdul Rehman (604331)
Adnan Akhunzada (20151648)
author_role author
dc.creator.none.fl_str_mv Muhammad Khubaib (22927822)
Tanzila Kehkashan (20748842)
Maha Abdelhaq (735574)
Muhammad Asghar Khan (13411285)
Muhammad Zaman (66868)
Imran Ashraf (7370771)
Abdul Rehman (604331)
Adnan Akhunzada (20151648)
dc.date.none.fl_str_mv 2025-06-23T03:00:00Z
dc.identifier.none.fl_str_mv 10.1109/tce.2025.3582267
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Data-Efficient_Wheat_Disease_Detection_Using_Shifted_Window_Transformer_Enhancing_Accuracy_Sustainability_and_Global_Food_Security/31289212
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Agricultural, veterinary and food sciences
Crop and pasture production
Information and computing sciences
Computer vision and multimedia computation
Machine learning
Wheat disease detection
Deep learning
Swin-transformer
Image classification
Precision agriculture
dc.title.none.fl_str_mv Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Wheat is an essential crop that plays a vital role in global food security, but is susceptible to a variety of diseases, which have the potential to drastically decrease crop productivity. Detection of disease at an early stage and in an accurate manner is crucial to minimize crop losses. This research presents a deep learning technique based on the Shifted Window (Swin) Transformer, a powerful attention-based model that effectively captures both local and global information for enhanced classification output. Unlike conventional CNN-based methods, which often face limitations in feature extraction, the Swin Transformer utilizes hierarchical attention mechanisms to improve disease detection accuracy. The proposed model is trained on a dataset of 9,346 wheat leaf images, categorized into eight disease classes and one healthy class. Using Bayesian hyperparameter optimization, we tuned key parameters such as learning rates, batch sizes, and dropout rates, and achieved an accuracy of 99.3%. We also conducted comparative analyses with baseline self-attention, CNN-based feature extraction, and hybrid attention layers. To enhance interpretability, Grad-CAM visualization techniques were applied, confirming model reliability. This research advances precision agriculture by improving the efficiency of wheat disease identification and supporting sustainable farming through state-of-the-art deep learning methodologies.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Consumer Electronics<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" rel="noreferrer noopener" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tce.2025.3582267" target="_blank">https://dx.doi.org/10.1109/tce.2025.3582267</a></p>
eu_rights_str_mv openAccess
id Manara2_abe3f41dd57b52da099c091f391c59fe
identifier_str_mv 10.1109/tce.2025.3582267
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/31289212
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food SecurityMuhammad Khubaib (22927822)Tanzila Kehkashan (20748842)Maha Abdelhaq (735574)Muhammad Asghar Khan (13411285)Muhammad Zaman (66868)Imran Ashraf (7370771)Abdul Rehman (604331)Adnan Akhunzada (20151648)Agricultural, veterinary and food sciencesCrop and pasture productionInformation and computing sciencesComputer vision and multimedia computationMachine learningWheat disease detectionDeep learningSwin-transformerImage classificationPrecision agriculture<p dir="ltr">Wheat is an essential crop that plays a vital role in global food security, but is susceptible to a variety of diseases, which have the potential to drastically decrease crop productivity. Detection of disease at an early stage and in an accurate manner is crucial to minimize crop losses. This research presents a deep learning technique based on the Shifted Window (Swin) Transformer, a powerful attention-based model that effectively captures both local and global information for enhanced classification output. Unlike conventional CNN-based methods, which often face limitations in feature extraction, the Swin Transformer utilizes hierarchical attention mechanisms to improve disease detection accuracy. The proposed model is trained on a dataset of 9,346 wheat leaf images, categorized into eight disease classes and one healthy class. Using Bayesian hyperparameter optimization, we tuned key parameters such as learning rates, batch sizes, and dropout rates, and achieved an accuracy of 99.3%. We also conducted comparative analyses with baseline self-attention, CNN-based feature extraction, and hybrid attention layers. To enhance interpretability, Grad-CAM visualization techniques were applied, confirming model reliability. This research advances precision agriculture by improving the efficiency of wheat disease identification and supporting sustainable farming through state-of-the-art deep learning methodologies.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Consumer Electronics<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" rel="noreferrer noopener" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tce.2025.3582267" target="_blank">https://dx.doi.org/10.1109/tce.2025.3582267</a></p>2025-06-23T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/tce.2025.3582267https://figshare.com/articles/journal_contribution/Data-Efficient_Wheat_Disease_Detection_Using_Shifted_Window_Transformer_Enhancing_Accuracy_Sustainability_and_Global_Food_Security/31289212CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/312892122025-06-23T03:00:00Z
spellingShingle Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
Muhammad Khubaib (22927822)
Agricultural, veterinary and food sciences
Crop and pasture production
Information and computing sciences
Computer vision and multimedia computation
Machine learning
Wheat disease detection
Deep learning
Swin-transformer
Image classification
Precision agriculture
status_str publishedVersion
title Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
title_full Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
title_fullStr Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
title_full_unstemmed Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
title_short Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
title_sort Data-Efficient Wheat Disease Detection Using Shifted Window Transformer: Enhancing Accuracy, Sustainability, and Global Food Security
topic Agricultural, veterinary and food sciences
Crop and pasture production
Information and computing sciences
Computer vision and multimedia computation
Machine learning
Wheat disease detection
Deep learning
Swin-transformer
Image classification
Precision agriculture