A deep semantic network-based image segmentation of soybean rust pathogens
IntroductionAsian soybean rust is a highly aggressive leaf-based disease triggered by the obligate biotrophic fungus Phakopsora pachyrhizi which can cause up to 80% yield loss in soybean. The precise image segmentation of fungus can characterize fungal phenotype transitions during growth and help to...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-03-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2024.1340584/full |
| _version_ | 1827307839124668416 |
|---|---|
| author | Yalin Wu Zhuobin Xi Fen Liu Weiming Hu Hongjuan Feng Qinjian Zhang |
| author_facet | Yalin Wu Zhuobin Xi Fen Liu Weiming Hu Hongjuan Feng Qinjian Zhang |
| author_sort | Yalin Wu |
| collection | DOAJ |
| description | IntroductionAsian soybean rust is a highly aggressive leaf-based disease triggered by the obligate biotrophic fungus Phakopsora pachyrhizi which can cause up to 80% yield loss in soybean. The precise image segmentation of fungus can characterize fungal phenotype transitions during growth and help to discover new medicines and agricultural biocides using large-scale phenotypic screens.MethodsThe improved Mask R-CNN method is proposed to accomplish the segmentation of densely distributed, overlapping and intersecting microimages. First, Res2net is utilized to layer the residual connections in a single residual block to replace the backbone of the original Mask R-CNN, which is then combined with FPG to enhance the feature extraction capability of the network model. Secondly, the loss function is optimized and the CIoU loss function is adopted as the loss function for boundary box regression prediction, which accelerates the convergence speed of the model and meets the accurate classification of high-density spore images.ResultsThe experimental results show that the mAP for detection and segmentation, accuracy of the improved algorithm is improved by 6.4%, 12.3% and 2.2% respectively over the original Mask R-CNN algorithm.DiscussionThis method is more suitable for the segmentation of fungi images and provide an effective tool for large-scale phenotypic screens of plant fungal pathogens. |
| first_indexed | 2024-04-24T18:46:21Z |
| format | Article |
| id | doaj.art-70eb8bbb521b4e2bbb22b656b08935fa |
| institution | Directory Open Access Journal |
| issn | 1664-462X |
| language | English |
| last_indexed | 2024-04-24T18:46:21Z |
| publishDate | 2024-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj.art-70eb8bbb521b4e2bbb22b656b08935fa2024-03-27T05:09:25ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2024-03-011510.3389/fpls.2024.13405841340584A deep semantic network-based image segmentation of soybean rust pathogensYalin Wu0Zhuobin Xi1Fen Liu2Weiming Hu3Hongjuan Feng4Qinjian Zhang5Lushan Botanical Garden, Jiangxi Province and Chinese Academy of Sciences, Jiujiang, ChinaMechanical Electrical Engineering School, Beijing Information Science & Technology University, Beijing, ChinaLushan Botanical Garden, Jiangxi Province and Chinese Academy of Sciences, Jiujiang, ChinaLushan Botanical Garden, Jiangxi Province and Chinese Academy of Sciences, Jiujiang, ChinaZhongzhen Kejian (ShenZhen) Holdings Co., Ltd, Shenzhen, ChinaMechanical Electrical Engineering School, Beijing Information Science & Technology University, Beijing, ChinaIntroductionAsian soybean rust is a highly aggressive leaf-based disease triggered by the obligate biotrophic fungus Phakopsora pachyrhizi which can cause up to 80% yield loss in soybean. The precise image segmentation of fungus can characterize fungal phenotype transitions during growth and help to discover new medicines and agricultural biocides using large-scale phenotypic screens.MethodsThe improved Mask R-CNN method is proposed to accomplish the segmentation of densely distributed, overlapping and intersecting microimages. First, Res2net is utilized to layer the residual connections in a single residual block to replace the backbone of the original Mask R-CNN, which is then combined with FPG to enhance the feature extraction capability of the network model. Secondly, the loss function is optimized and the CIoU loss function is adopted as the loss function for boundary box regression prediction, which accelerates the convergence speed of the model and meets the accurate classification of high-density spore images.ResultsThe experimental results show that the mAP for detection and segmentation, accuracy of the improved algorithm is improved by 6.4%, 12.3% and 2.2% respectively over the original Mask R-CNN algorithm.DiscussionThis method is more suitable for the segmentation of fungi images and provide an effective tool for large-scale phenotypic screens of plant fungal pathogens.https://www.frontiersin.org/articles/10.3389/fpls.2024.1340584/fullAsian soybean rustPhakopsora pachyrhizideep learninginstance segmentationmask R-CNN |
| spellingShingle | Yalin Wu Zhuobin Xi Fen Liu Weiming Hu Hongjuan Feng Qinjian Zhang A deep semantic network-based image segmentation of soybean rust pathogens Frontiers in Plant Science Asian soybean rust Phakopsora pachyrhizi deep learning instance segmentation mask R-CNN |
| title | A deep semantic network-based image segmentation of soybean rust pathogens |
| title_full | A deep semantic network-based image segmentation of soybean rust pathogens |
| title_fullStr | A deep semantic network-based image segmentation of soybean rust pathogens |
| title_full_unstemmed | A deep semantic network-based image segmentation of soybean rust pathogens |
| title_short | A deep semantic network-based image segmentation of soybean rust pathogens |
| title_sort | deep semantic network based image segmentation of soybean rust pathogens |
| topic | Asian soybean rust Phakopsora pachyrhizi deep learning instance segmentation mask R-CNN |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2024.1340584/full |
| work_keys_str_mv | AT yalinwu adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT zhuobinxi adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT fenliu adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT weiminghu adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT hongjuanfeng adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT qinjianzhang adeepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT yalinwu deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT zhuobinxi deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT fenliu deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT weiminghu deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT hongjuanfeng deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens AT qinjianzhang deepsemanticnetworkbasedimagesegmentationofsoybeanrustpathogens |