{"id":{"repo_id":"uwtsd","oai_identifier":"oai:repository.uwtsd.ac.uk:3306"},"canonical_url":"https://search.dev.ndltd.org/etd/uwtsd/oai:repository.uwtsd.ac.uk:3306","repository":{"repo_id":"uwtsd","name":"University of Wales Trinity Saint David","base_url":"https://repository.uwtsd.ac.uk/cgi/oai2"},"display":{"title":"Detection of Fusarium Head Blight on Wheat Spikelets Using a Multi-Scale Feature Fusion CNN Model","abstract":"Fusarium head blight (FHB) is a globally significant fungal disease that severely impacts wheat crops, leading to reduced yields, degraded grain quality, and the accumulation of harmful mycotoxins. These mycotoxins pose severe threats to human and animal health and result in substantial economic losses. Accurate and efficient assessment of FHB disease phenotypes is crucial for developing resistant wheat varieties through breeding. However, current methods for collecting and analyzing wheat phenotypic data are time-consuming, labor-intensive, and often imprecise, particularly at the spikelet level, where disease symptoms are more nuanced and challenging to detect. Hence this study addresses these challenges by proposing a novel lightweight object detection model based on multi-scale feature fusion, specifically designed to detect FHB at the spikelet level in wheat. The proposed model leverages the advanced YOLOv9 framework, integrating the Multi-Scale Feature Enhancement and Fusion (MSFEF) module, to significantly enhance the accuracy of detecting small and subtle disease features in complex field environments. The performance of the developed model is evaluated using a self-constructed dataset comprising 620 annotated RGB images. Results show that the proposed model achieves a mean Average Precision (mAP) of 90.6%, outperforming state-of-the-art models such as YOLOv9-C and YOLOv10-S while maintaining real-time performance with 294 FPS. Additionally, the model excels in detecting diseased spikelets, achieving an Average Precision (AP) of 92%, and shows high robustness in recognizing healthy and infected spikelets. Regarding disease phenotype extraction, the model's correlation coefficients with manual detection for disease spikelet rate and FHB severity are 0.81 and 0.75, respectively, underscoring its significant potential for practical application in breeding and phenotype collection. This study introduces a novel, fine-grained detection method for wheat FHB disease and offers practical solutions to enhance the efficiency of wheat resistance breeding and plant disease management.","abstract_html":"Fusarium head blight (FHB) is a globally significant fungal disease that severely impacts wheat crops, leading to reduced yields, degraded grain quality, and the accumulation of harmful mycotoxins. These mycotoxins pose severe threats to human and animal health and result in substantial economic losses. Accurate and efficient assessment of FHB disease phenotypes is crucial for developing resistant wheat varieties through breeding. However, current methods for collecting and analyzing wheat phenotypic data are time-consuming, labor-intensive, and often imprecise, particularly at the spikelet level, where disease symptoms are more nuanced and challenging to detect. Hence this study addresses these challenges by proposing a novel lightweight object detection model based on multi-scale feature fusion, specifically designed to detect FHB at the spikelet level in wheat. The proposed model leverages the advanced YOLOv9 framework, integrating the Multi-Scale Feature Enhancement and Fusion (MSFEF) module, to significantly enhance the accuracy of detecting small and subtle disease features in complex field environments. The performance of the developed model is evaluated using a self-constructed dataset comprising 620 annotated RGB images. Results show that the proposed model achieves a mean Average Precision (mAP) of 90.6%, outperforming state-of-the-art models such as YOLOv9-C and YOLOv10-S while maintaining real-time performance with 294 FPS. Additionally, the model excels in detecting diseased spikelets, achieving an Average Precision (AP) of 92%, and shows high robustness in recognizing healthy and infected spikelets. Regarding disease phenotype extraction, the model's correlation coefficients with manual detection for disease spikelet rate and FHB severity are 0.81 and 0.75, respectively, underscoring its significant potential for practical application in breeding and phenotype collection. This study introduces a novel, fine-grained detection method for wheat FHB disease and offers practical solutions to enhance the efficiency of wheat resistance breeding and plant disease management.","abstract_has_math":false,"creators":["Wu, Ze"],"institution":"University of Wales Trinity Saint David","degree_name":"msc","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-11","date_published":"2024-11","updated_at":"2026-07-24T05:53:02Z","subjects":["TK Peirianneg drydanol. Peirianneg Niwclear Electronig"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.grantnumber","label":"Dc Identifier Grantnumber","values":["UWTSD"],"render_values":[{"text":"UWTSD","href":null,"code":true}]}]},"links":{"outbound_url":"https://doi.org/10.82227/repository.uwtsd.ac.uk.00003306","outbound_label":"DOI","outbound_source":"dc:identifier.doi"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.sponsor","label":"Sponsor","values":["University of Wales Trinity Saint David"]},{"key":"dc:creator","label":"Author","values":["Wu, Ze"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-11-11"]},{"key":"dc:date.issued","label":"Date","values":["2024-11"]},{"key":"dc:publisher.commercial","label":"Dc Publisher Commercial","values":["University of Wales Trinity Saint David"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Traethodau Meistr","Wales Institute for Science and Art: Applied Computing"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Wales Trinity Saint David"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://repository.uwtsd.ac.uk/id/eprint/3306/"]},{"key":"dc:type","label":"Dc Type","values":["Gosodiad"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["masters"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["msc"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["TK Peirianneg drydanol. Peirianneg Niwclear Electronig"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.82227/repository.uwtsd.ac.uk.00003306"]},{"key":"dc:identifier.grantnumber","label":"Dc Identifier Grantnumber","values":["UWTSD"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://repository.uwtsd.ac.uk/id/eprint/3306/1/Wu_Ze_MSc_Thesis.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Fusarium head blight (FHB) is a globally significant fungal disease that severely impacts wheat crops, leading to reduced yields, degraded grain quality, and the accumulation of harmful mycotoxins. These mycotoxins pose severe threats to human and animal health and result in substantial economic losses. Accurate and efficient assessment of FHB disease phenotypes is crucial for developing resistant wheat varieties through breeding. However, current methods for collecting and analyzing wheat phenotypic data are time-consuming, labor-intensive, and often imprecise, particularly at the spikelet level, where disease symptoms are more nuanced and challenging to detect. Hence this study addresses these challenges by proposing a novel lightweight object detection model based on multi-scale feature fusion, specifically designed to detect FHB at the spikelet level in wheat. The proposed model leverages the advanced YOLOv9 framework, integrating the Multi-Scale Feature Enhancement and Fusion (MSFEF) module, to significantly enhance the accuracy of detecting small and subtle disease features in complex field environments. The performance of the developed model is evaluated using a self-constructed dataset comprising 620 annotated RGB images. Results show that the proposed model achieves a mean Average Precision (mAP) of 90.6%, outperforming state-of-the-art models such as YOLOv9-C and YOLOv10-S while maintaining real-time performance with 294 FPS. Additionally, the model excels in detecting diseased spikelets, achieving an Average Precision (AP) of 92%, and shows high robustness in recognizing healthy and infected spikelets. Regarding disease phenotype extraction, the model's correlation coefficients with manual detection for disease spikelet rate and FHB severity are 0.81 and 0.75, respectively, underscoring its significant potential for practical application in breeding and phenotype collection. This study introduces a novel, fine-grained detection method for wheat FHB disease and offers practical solutions to enhance the efficiency of wheat resistance breeding and plant disease management."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dc:title","label":"Title","values":["Detection of Fusarium Head Blight on Wheat Spikelets Using a Multi-Scale Feature Fusion CNN Model"]}]}],"canonical_facts":{"dc:contributor.sponsor":["University of Wales Trinity Saint David"],"dc:creator":["Wu, Ze"],"dc:date":["2024-11-11"],"dc:date.issued":["2024-11"],"dc:description.abstract":["Fusarium head blight (FHB) is a globally significant fungal disease that severely impacts wheat crops, leading to reduced yields, degraded grain quality, and the accumulation of harmful mycotoxins. These mycotoxins pose severe threats to human and animal health and result in substantial economic losses. Accurate and efficient assessment of FHB disease phenotypes is crucial for developing resistant wheat varieties through breeding. However, current methods for collecting and analyzing wheat phenotypic data are time-consuming, labor-intensive, and often imprecise, particularly at the spikelet level, where disease symptoms are more nuanced and challenging to detect. Hence this study addresses these challenges by proposing a novel lightweight object detection model based on multi-scale feature fusion, specifically designed to detect FHB at the spikelet level in wheat. The proposed model leverages the advanced YOLOv9 framework, integrating the Multi-Scale Feature Enhancement and Fusion (MSFEF) module, to significantly enhance the accuracy of detecting small and subtle disease features in complex field environments. The performance of the developed model is evaluated using a self-constructed dataset comprising 620 annotated RGB images. Results show that the proposed model achieves a mean Average Precision (mAP) of 90.6%, outperforming state-of-the-art models such as YOLOv9-C and YOLOv10-S while maintaining real-time performance with 294 FPS. Additionally, the model excels in detecting diseased spikelets, achieving an Average Precision (AP) of 92%, and shows high robustness in recognizing healthy and infected spikelets. Regarding disease phenotype extraction, the model's correlation coefficients with manual detection for disease spikelet rate and FHB severity are 0.81 and 0.75, respectively, underscoring its significant potential for practical application in breeding and phenotype collection. This study introduces a novel, fine-grained detection method for wheat FHB disease and offers practical solutions to enhance the efficiency of wheat resistance breeding and plant disease management."],"dc:format":["text"],"dc:identifier.doi":["10.82227/repository.uwtsd.ac.uk.00003306"],"dc:identifier.grantnumber":["UWTSD"],"dc:identifier.uri":["https://repository.uwtsd.ac.uk/id/eprint/3306/1/Wu_Ze_MSc_Thesis.pdf"],"dc:publisher.commercial":["University of Wales Trinity Saint David"],"dc:publisher.department":["Traethodau Meistr","Wales Institute for Science and Art: Applied Computing"],"dc:publisher.institution":["University of Wales Trinity Saint David"],"dc:relation.isreferencedby":["https://repository.uwtsd.ac.uk/id/eprint/3306/"],"dc:subject":["TK Peirianneg drydanol. Peirianneg Niwclear Electronig"],"dc:title":["Detection of Fusarium Head Blight on Wheat Spikelets Using a Multi-Scale Feature Fusion CNN Model"],"dc:type":["Gosodiad"],"dc:type.qualificationlevel":["masters"],"dc:type.qualificationname":["msc"]},"updated_at":"2026-07-24T05:53:02Z"}