Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches
Early detection of drought stress is essential for preventing permanent plant damage and minimizing yield loss. This study utilized hyperspectral imaging at the leaf level to visualize drought stress in safflower plants (Carthamus tinctorius L.). Three safflower genotypes, Palenus, A82, and IL-111,...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Plant Stress |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667064X24003063 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850054863886483456 |
|---|---|
| author | Fatemeh Salek Seyed Ahmad Mireei Abbas Hemmat Mehrnoosh Jafari Mohammad R. Sabzalian Majid Nazeri Wouter Saeys |
| author_facet | Fatemeh Salek Seyed Ahmad Mireei Abbas Hemmat Mehrnoosh Jafari Mohammad R. Sabzalian Majid Nazeri Wouter Saeys |
| author_sort | Fatemeh Salek |
| collection | DOAJ |
| description | Early detection of drought stress is essential for preventing permanent plant damage and minimizing yield loss. This study utilized hyperspectral imaging at the leaf level to visualize drought stress in safflower plants (Carthamus tinctorius L.). Three safflower genotypes, Palenus, A82, and IL-111, were cultivated under three irrigation levels. Stress conditions were simulated by depleting 50%, 70%, and 90% of soil water content, representing unstressed (US), mild stress (MS), and severe stress (SS) conditions, respectively. Hyperspectral images of leaf samples were captured before any visible signs of water scarcity emerged. Classification analysis was performed using the full mean spectral data with partial least squares discriminant analysis, soft independent modeling of class analogy (SIMCA), support vector machines, and artificial neural network (ANN) classifiers. Feature selection methods were applied to extract the most informative wavebands, and ANN was used to build predictive models. Spatial analysis involved pixel-wise classification using both unsupervised (k-means clustering) and supervised (best classifiers) approaches. ANN outperformed other classifiers using the entire spectral data, effectively distinguishing US, MS, and SS classes in the Palenus, A82, and IL-111 genotypes, achieving F1-scores of 92.22%, 96.01%, and 96.47%, respectively. Among the feature selection methods, SIMCA-based features excelled in monitoring stress conditions in the Palenus and A82 genotypes. In supervised spatial analysis, ANN models clearly depicted the progression of stress in leaves across different genotypes. This study demonstrates the potential of hyperspectral imaging to differentiate various levels of drought stress in safflower, an important oilseed crop. |
| format | Article |
| id | doaj-art-59d882995b7c4ae79cdcaaa191d4f86a |
| institution | DOAJ |
| issn | 2667-064X |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Plant Stress |
| spelling | doaj-art-59d882995b7c4ae79cdcaaa191d4f86a2025-08-20T02:52:08ZengElsevierPlant Stress2667-064X2024-12-011410065310.1016/j.stress.2024.100653Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approachesFatemeh Salek0Seyed Ahmad Mireei1Abbas Hemmat2Mehrnoosh Jafari3Mohammad R. Sabzalian4Majid Nazeri5Wouter Saeys6Department of Biosystems Engineering, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, IranDepartment of Biosystems Engineering, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, Iran; Corresponding author.Department of Biosystems Engineering, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, IranDepartment of Biosystems Engineering, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, IranDepartment of Agronomy and Plant Breeding, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, IranDepartment of Laser and Photonics, Faculty of Physics, University of Kashan, Kashan 87317-53153, IranKU Leuven, Department of Biosystems, MeBioS, Kasteelpark Arenberg 30, 3001 Leuven, BelgiumEarly detection of drought stress is essential for preventing permanent plant damage and minimizing yield loss. This study utilized hyperspectral imaging at the leaf level to visualize drought stress in safflower plants (Carthamus tinctorius L.). Three safflower genotypes, Palenus, A82, and IL-111, were cultivated under three irrigation levels. Stress conditions were simulated by depleting 50%, 70%, and 90% of soil water content, representing unstressed (US), mild stress (MS), and severe stress (SS) conditions, respectively. Hyperspectral images of leaf samples were captured before any visible signs of water scarcity emerged. Classification analysis was performed using the full mean spectral data with partial least squares discriminant analysis, soft independent modeling of class analogy (SIMCA), support vector machines, and artificial neural network (ANN) classifiers. Feature selection methods were applied to extract the most informative wavebands, and ANN was used to build predictive models. Spatial analysis involved pixel-wise classification using both unsupervised (k-means clustering) and supervised (best classifiers) approaches. ANN outperformed other classifiers using the entire spectral data, effectively distinguishing US, MS, and SS classes in the Palenus, A82, and IL-111 genotypes, achieving F1-scores of 92.22%, 96.01%, and 96.47%, respectively. Among the feature selection methods, SIMCA-based features excelled in monitoring stress conditions in the Palenus and A82 genotypes. In supervised spatial analysis, ANN models clearly depicted the progression of stress in leaves across different genotypes. This study demonstrates the potential of hyperspectral imaging to differentiate various levels of drought stress in safflower, an important oilseed crop.http://www.sciencedirect.com/science/article/pii/S2667064X24003063Crop yieldK-means clusteringPartial least squares discriminant analysis (PLS-DA)Pixel-wise classificationSoft independent modeling of class analogy (SIMCA) |
| spellingShingle | Fatemeh Salek Seyed Ahmad Mireei Abbas Hemmat Mehrnoosh Jafari Mohammad R. Sabzalian Majid Nazeri Wouter Saeys Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches Plant Stress Crop yield K-means clustering Partial least squares discriminant analysis (PLS-DA) Pixel-wise classification Soft independent modeling of class analogy (SIMCA) |
| title | Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches |
| title_full | Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches |
| title_fullStr | Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches |
| title_full_unstemmed | Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches |
| title_short | Early monitoring of drought stress in safflower (Carthamus tinctorius L.) using hyperspectral imaging: A comparison of machine learning tools and feature selection approaches |
| title_sort | early monitoring of drought stress in safflower carthamus tinctorius l using hyperspectral imaging a comparison of machine learning tools and feature selection approaches |
| topic | Crop yield K-means clustering Partial least squares discriminant analysis (PLS-DA) Pixel-wise classification Soft independent modeling of class analogy (SIMCA) |
| url | http://www.sciencedirect.com/science/article/pii/S2667064X24003063 |
| work_keys_str_mv | AT fatemehsalek earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT seyedahmadmireei earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT abbashemmat earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT mehrnooshjafari earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT mohammadrsabzalian earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT majidnazeri earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches AT woutersaeys earlymonitoringofdroughtstressinsafflowercarthamustinctoriuslusinghyperspectralimagingacomparisonofmachinelearningtoolsandfeatureselectionapproaches |