Dataset Information

Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset.

ABSTRACT: 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.

SUBMITTER: Sapoukhina N

PROVIDER: S-EPMC9685808 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset.

Sapoukhina Natalia N Boureau Tristan T Rousseau David D

Frontiers in plant science 20221110

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 traine ...[more]

PMID: 36438124

Similar Datasets

Project description:Duckweeds (Lemnaceae species) are extensively used models in ecotoxicology, and chlorophyll fluorescence imaging offers a sensitive and high throughput platform for phytotoxicity assays with these tiny plants. However, the vast number of potentially applicable chlorophyll fluorescence-based test endpoints makes comparison and generalization of results hard among different studies. The present study aimed to jointly measure and compare the sensitivity of various chlorophyll fluorescence parameters in Spirodela polyrhiza (giant duckweed) plants exposed to nickel, chromate (hexavalent chromium) and sodium chloride for 72 h, respectively. The photochemistry of Photosystem II in both dark- and light-adapted states of plants was assessed via in vivo chlorophyll fluorescence imaging method. Our results indicated that the studied parameters responded with very divergent sensitivity, highlighting the importance of parallelly assessing several chlorophyll fluorescence parameters. Generally, the light-adapted parameters were more sensitive than the dark-adapted ones. Thus, the former ones might be the preferred endpoints in phytotoxicity assays. Fv/Fm, i.e., the most extensively reported parameter literature-wise, proved to be the least sensitive endpoint; therefore, future studies might also consider reporting Fv/Fo, as its more responsive analogue. The tested toxicants induced different trends in the basic chlorophyll fluorescence parameters and, at least partly, in relative proportions of different quenching processes, suggesting that a basic distinction of water pollutants with different modes of action might be achievable by this method. We found definite hormetic patterns in responses to several endpoints. Hormesis occurred in the concentration ranges where the applied toxicants resulted in strong growth inhibition in longer-term exposures of the same duckweed clone in previous studies. These findings indicate that changes in the photochemical efficiency of plants do not necessarily go hand in hand with growth responses, and care should be taken when one exclusively interprets chlorophyll fluorescence-based endpoints as general proxies for phytotoxic effects.

Project description:BackgroundIn order to select for quantitative plant resistance to pathogens, high throughput approaches that can precisely quantify disease severity are needed. Automation and use of calibrated image analysis should provide more accurate, objective and faster analyses than visual assessments. In contrast to conventional visible imaging, chlorophyll fluorescence imaging is not sensitive to environmental light variations and provides single-channel images prone to a segmentation analysis by simple thresholding approaches. Among the various parameters used in chlorophyll fluorescence imaging, the maximum quantum yield of photosystem II photochemistry (Fv/Fm) is well adapted to phenotyping disease severity. Fv/Fm is an indicator of plant stress that displays a robust contrast between infected and healthy tissues. In the present paper, we aimed at the segmentation of Fv/Fm images to quantify disease severity.ResultsBased on the Fv/Fm values of each pixel of the image, a thresholding approach was developed to delimit diseased areas. A first step consisted in setting up thresholds to reproduce visual observations by trained raters of symptoms caused by Xanthomonas fuscans subsp. fuscans (Xff) CFBP4834-R on Phaseolus vulgaris cv. Flavert. In order to develop a thresholding approach valuable on any cultivars or species, a second step was based on modeling pixel-wise Fv/Fm-distributions as mixtures of Gaussian distributions. Such a modeling may discriminate various stages of the symptom development but over-weights artifacts that can occur on mock-inoculated samples. Therefore, we developed a thresholding approach based on the probability of misclassification of a healthy pixel. Then, a clustering step is performed on the diseased areas to discriminate between various stages of alteration of plant tissues. Notably, the use of chlorophyll fluorescence imaging could detect pre-symptomatic area. The interest of this image analysis procedure for assessing the levels of quantitative resistance is illustrated with the quantitation of disease severity on five commercial varieties of bean inoculated with Xff CFBP4834-R.ConclusionsIn this paper, we describe an image analysis procedure for quantifying the leaf area impacted by the pathogen. In a perspective of high throughput phenotyping, the procedure was automated with the software R downloadable at http://www.r-project.org/. The R script is available at http://lisa.univ-angers.fr/PHENOTIC/telechargements.html.

Dataset Information

Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset.

Publications

Plant disease symptom segmentation in chlorophyll fluorescence imaging with a synthetic dataset.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets