Peanut Leaf Wilting Estimation From RGB Color Indices and Logistic Models
| dc.contributor.author | Sarkar, Sayantan | en |
| dc.contributor.author | Ramsey, A. Ford | en |
| dc.contributor.author | Cazenave, Alexandre-Brice | en |
| dc.contributor.author | Balota, Maria | en |
| dc.date.accessioned | 2022-04-19T14:42:14Z | en |
| dc.date.available | 2022-04-19T14:42:14Z | en |
| dc.date.issued | 2021-06-18 | en |
| dc.description.abstract | Peanut (Arachis hypogaea L.) is an important crop for United States agriculture and worldwide. Low soil moisture is a major constraint for production in all peanut growing regions with negative effects on yield quantity and quality. Leaf wilting is a visual symptom of low moisture stress used in breeding to improve stress tolerance, but visual rating is slow when thousands of breeding lines are evaluated and can be subject to personnel scoring bias. Photogrammetry might be used instead. The objective of this article is to determine if color space indices derived from red-green-blue (RGB) images can accurately estimate leaf wilting for breeding selection and irrigation triggering in peanut production. RGB images were collected with a digital camera proximally and aerially by a unmanned aerial vehicle during 2018 and 2019. Visual rating was performed on the same days as image collection. Vegetation indices were intensity, hue, saturation, lightness, a*, b*, u*, v*, green area (GA), greener area (GGA), and crop senescence index (CSI). In particular, hue, a*, u*, GA, GGA, and CSI were significantly (p <= 0.0001) associated with leaf wilting. These indices were further used to train an ordinal logistic regression model for wilting estimation. This model had 90% accuracy when images were taken aerially and 99% when images were taken proximally. This article reports on a simple yet key aspect of peanut screening for tolerance to low soil moisture stress and uses novel, fast, cost-effective, and accurate RGB-derived models to estimate leaf wilting. | en |
| dc.description.notes | This study was funded by USDA NIFA-CARE and NIFA-AFRI grant (grant no.-2017-67013-26193) and the Virginia Crop Improvement Association (VCIA). | en |
| dc.description.sponsorship | USDA NIFA-CAREUnited States Department of Agriculture (USDA); Virginia Crop Improvement Association (VCIA); NIFA-AFRI grant [2017-67013-26193] | en |
| dc.description.version | Published version | en |
| dc.format.mimetype | application/pdf | en |
| dc.identifier.doi | https://doi.org/10.3389/fpls.2021.658621 | en |
| dc.identifier.issn | 1664-462X | en |
| dc.identifier.other | 658621 | en |
| dc.identifier.pmid | 34220885 | en |
| dc.identifier.uri | http://hdl.handle.net/10919/109697 | en |
| dc.identifier.volume | 12 | en |
| dc.language.iso | en | en |
| dc.rights | Creative Commons Attribution 4.0 International | en |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | en |
| dc.subject | peanut leaf wilting | en |
| dc.subject | RGB color space indices | en |
| dc.subject | logistic regression | en |
| dc.subject | machine learning | en |
| dc.subject | high-throughput phenotyping | en |
| dc.title | Peanut Leaf Wilting Estimation From RGB Color Indices and Logistic Models | en |
| dc.title.serial | Frontiers in Plant Science | en |
| dc.type | Article - Refereed | en |
| dc.type.dcmitype | Text | en |