Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset
Despite the wide use of computer vision methods in plant health monitoring, little attention is paid to segmenting the diseased leaf area at its early stages. It can be explained by the lack of datasets of plant images with annotated disease lesions. We propose a novel methodology to generate fluore...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-11-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2022.969205/full |
| _version_ | 1811232743055425536 |
|---|---|
| author | Natalia Sapoukhina Tristan Boureau David Rousseau David Rousseau |
| author_facet | Natalia Sapoukhina Tristan Boureau David Rousseau David Rousseau |
| author_sort | Natalia Sapoukhina |
| collection | DOAJ |
| description | Despite the wide use of computer vision methods in plant health monitoring, little attention is paid to segmenting the diseased leaf area at its early stages. It can be explained by the lack of datasets of plant images with annotated disease lesions. We propose a novel methodology to generate fluorescent images of diseased plants with an automated lesion annotation. We demonstrate that a U-Net model aiming to segment disease lesions on fluorescent images of plant leaves can be efficiently trained purely by a synthetically generated dataset. The trained model showed 0.793% recall and 0.723% average precision against an empirical fluorescent test dataset. Creating and using such synthetic data can be a powerful technique to facilitate the application of deep learning methods in precision crop protection. Moreover, our method of generating synthetic fluorescent images is a way to improve the generalization ability of deep learning models. |
| first_indexed | 2024-04-12T11:09:22Z |
| format | Article |
| id | doaj.art-9e1a06b2213e4fb383c8c107461029fa |
| institution | Directory Open Access Journal |
| issn | 1664-462X |
| language | English |
| last_indexed | 2024-04-12T11:09:22Z |
| publishDate | 2022-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj.art-9e1a06b2213e4fb383c8c107461029fa2022-12-22T03:35:40ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2022-11-011310.3389/fpls.2022.969205969205Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic datasetNatalia Sapoukhina0Tristan Boureau1David Rousseau2David Rousseau3Univ Angers, Institut Agro, INRAE, IRHS, SFR QUASAV, Angers, FrancePhenotic Platform, Univ Angers, Institut Agro, INRAE, IRHS, SFR QUASAV, Angers, FranceUniv Angers, Institut Agro, INRAE, IRHS, SFR QUASAV, Angers, FranceLaboratoire Angevine de Recherche en Ingénierie des Systèmes (LARIS), Université d’Angers, Angers, FranceDespite the wide use of computer vision methods in plant health monitoring, little attention is paid to segmenting the diseased leaf area at its early stages. It can be explained by the lack of datasets of plant images with annotated disease lesions. We propose a novel methodology to generate fluorescent images of diseased plants with an automated lesion annotation. We demonstrate that a U-Net model aiming to segment disease lesions on fluorescent images of plant leaves can be efficiently trained purely by a synthetically generated dataset. The trained model showed 0.793% recall and 0.723% average precision against an empirical fluorescent test dataset. Creating and using such synthetic data can be a powerful technique to facilitate the application of deep learning methods in precision crop protection. Moreover, our method of generating synthetic fluorescent images is a way to improve the generalization ability of deep learning models.https://www.frontiersin.org/articles/10.3389/fpls.2022.969205/fullsynthetic datasemantic segmentationplant diseaseprecision agriculturedeep learningcomputer vision |
| spellingShingle | Natalia Sapoukhina Tristan Boureau David Rousseau David Rousseau Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset Frontiers in Plant Science synthetic data semantic segmentation plant disease precision agriculture deep learning computer vision |
| title | Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| title_full | Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| title_fullStr | Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| title_full_unstemmed | Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| title_short | Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| title_sort | plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset |
| topic | synthetic data semantic segmentation plant disease precision agriculture deep learning computer vision |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2022.969205/full |
| work_keys_str_mv | AT nataliasapoukhina plantdiseasesymptomsegmentationinchlorophyllfluorescenceimagingwithasyntheticdataset AT tristanboureau plantdiseasesymptomsegmentationinchlorophyllfluorescenceimagingwithasyntheticdataset AT davidrousseau plantdiseasesymptomsegmentationinchlorophyllfluorescenceimagingwithasyntheticdataset AT davidrousseau plantdiseasesymptomsegmentationinchlorophyllfluorescenceimagingwithasyntheticdataset |