Advanced Multi-Crop Disease Classification: A Stacking Ensemble Approach for Solanaceae and Malvaceae in Tirupattur District
Fuente:
Zenodo
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Recurso digital |
| Lenguaje: | inglés |
| Publicado: |
Zenodo
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901717933621248 |
|---|---|
| author | BASKAR M Dr.NAVEEN A |
| author_facet | BASKAR M Dr.NAVEEN A |
| contents | <h3><span lang="EN-US">Abstract</span></h3> <p><span lang="EN-US">Leaf diseases pose a severe and multifaceted threat to the productivity of key cash crops in the Tirupattur district, notably <span>Tomato</span> (<em>Solanum lycopersicum</em>), <span>Brinjal</span> (<em>Solanum melongena</em>), <span>Chilli</span> (<em>Capsicum annuum</em>), and <span>Lady's Finger</span> (<em>Abelmoschus esculentus</em>). Traditional detection methods fail to scale effectively across this crop diversity. This paper proposes a novel <span>Stacked Generalization Ensemble Deep Learning (DL) model</span> to establish a single, robust diagnostic architecture for simultaneously classifying multiple diseases across all four species. The architecture utilizes <span>Transfer Learning</span> on three distinct Convolutional Neural Network (CNN) feature extractors: <span>EfficientNetB3</span>, <span>ResNet101</span>, and <span>DenseNet169</span>. Their unique, high-dimensional feature vectors are concatenated and input to a non-linear <span>Support Vector Machine (SVM)</span>, which acts as the discriminative Meta-Learner. Tested on a comprehensive, locally augmented dataset of 18,000 images, the ensemble model achieved an outstanding <span>99.25% overall classification accuracy</span> and an <span>F1-Score of 99.23%</span>. This performance significantly surpassed the best individual baseline model, DenseNet169, by <span>0.87%</span>. Detailed feature visualization using t-SNE confirms the ensemble's ability to resolve inter-species symptom ambiguities, validating the method’s efficiency and superior generalization for complex agricultural environments.</span></p> <p><strong><span lang="EN-US">Keywords:</span></strong><span lang="EN-US"> </span><span lang="EN-US">Deep Learning, Ensemble Learning, Stacking Generalization, Multi-Crop, Solanaceae, Malvaceae, Support Vector Machine, Transfer Learning.</span></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_17762603 |
| institution | Zenodo |
| language | eng |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Advanced Multi-Crop Disease Classification: A Stacking Ensemble Approach for Solanaceae and Malvaceae in Tirupattur District BASKAR M Dr.NAVEEN A Deep Learning, Ensemble Learning, Stacking Generalization, Multi-Crop, Solanaceae, Malvaceae, Support Vector Machine, Transfer Learning. <h3><span lang="EN-US">Abstract</span></h3> <p><span lang="EN-US">Leaf diseases pose a severe and multifaceted threat to the productivity of key cash crops in the Tirupattur district, notably <span>Tomato</span> (<em>Solanum lycopersicum</em>), <span>Brinjal</span> (<em>Solanum melongena</em>), <span>Chilli</span> (<em>Capsicum annuum</em>), and <span>Lady's Finger</span> (<em>Abelmoschus esculentus</em>). Traditional detection methods fail to scale effectively across this crop diversity. This paper proposes a novel <span>Stacked Generalization Ensemble Deep Learning (DL) model</span> to establish a single, robust diagnostic architecture for simultaneously classifying multiple diseases across all four species. The architecture utilizes <span>Transfer Learning</span> on three distinct Convolutional Neural Network (CNN) feature extractors: <span>EfficientNetB3</span>, <span>ResNet101</span>, and <span>DenseNet169</span>. Their unique, high-dimensional feature vectors are concatenated and input to a non-linear <span>Support Vector Machine (SVM)</span>, which acts as the discriminative Meta-Learner. Tested on a comprehensive, locally augmented dataset of 18,000 images, the ensemble model achieved an outstanding <span>99.25% overall classification accuracy</span> and an <span>F1-Score of 99.23%</span>. This performance significantly surpassed the best individual baseline model, DenseNet169, by <span>0.87%</span>. Detailed feature visualization using t-SNE confirms the ensemble's ability to resolve inter-species symptom ambiguities, validating the method’s efficiency and superior generalization for complex agricultural environments.</span></p> <p><strong><span lang="EN-US">Keywords:</span></strong><span lang="EN-US"> </span><span lang="EN-US">Deep Learning, Ensemble Learning, Stacking Generalization, Multi-Crop, Solanaceae, Malvaceae, Support Vector Machine, Transfer Learning.</span></p> |
| title | Advanced Multi-Crop Disease Classification: A Stacking Ensemble Approach for Solanaceae and Malvaceae in Tirupattur District |
| topic | Deep Learning, Ensemble Learning, Stacking Generalization, Multi-Crop, Solanaceae, Malvaceae, Support Vector Machine, Transfer Learning. |
| url | https://doi.org/10.5281/zenodo.17762603 |