AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro
Early diagnosis and preventive measures are necessary to mitigate diseases’ impact on the yield of Colocasia esculenta (Taro). This study addresses the challenges of Taro disease identification by employing two key strategies: integrating explainable artificial intelligence techniques to interpret d...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/1/77 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589436714483712 |
|---|---|
| author | Valeria Maeda-Gutiérrez Juan José Oropeza-Valdez Luis C. Reveles-Gómez Cristian Padron-Manrique Osbaldo Resendis-Antonio Luis Octavio Solís-Sánchez Hector A. Guerrero-Osuna Carlos Alberto Olvera Olvera |
| author_facet | Valeria Maeda-Gutiérrez Juan José Oropeza-Valdez Luis C. Reveles-Gómez Cristian Padron-Manrique Osbaldo Resendis-Antonio Luis Octavio Solís-Sánchez Hector A. Guerrero-Osuna Carlos Alberto Olvera Olvera |
| author_sort | Valeria Maeda-Gutiérrez |
| collection | DOAJ |
| description | Early diagnosis and preventive measures are necessary to mitigate diseases’ impact on the yield of Colocasia esculenta (Taro). This study addresses the challenges of Taro disease identification by employing two key strategies: integrating explainable artificial intelligence techniques to interpret deep learning models and conducting a comparative analysis of advanced architectures Inception V3, ResNet-50, and Vision Transformers for classifying common Taro diseases, including leaf blight and mosaic virus, as well as identifying healthy leaves. The novelty of this work lies in the first-ever integration of SHapley Additive exPlanations (SHAP) with deep learning architectures to enhance model interpretability while providing a comprehensive comparison of state-of-the-art methods for this underexplored crop. The proposed models significantly improve the ability to recognize complex patterns and features, achieving high accuracy and robust performance in disease classification. The model’s efficacy was evaluated through multi-class statistical metrics, including accuracy, precision, F1 score, recall, specificity, Chohen’s kappa, and area under the curve. Among the architectures, Inception V3 exhibited superior performance in accuracy (0.9985), F1 score (0.9985), recall (0.9985), and specificity (0.9992). The explainability of Inception V3 was further enhanced using SHAP, which provides insights by dissecting the contributions of individual features in Taro leaves to the model’s predictions. This approach facilitates a deeper understanding of the disease classification process and supports the development of effective disease management strategies, ultimately contributing to improved Taro cultivation practices. |
| format | Article |
| id | doaj-art-de9fb346fe61454e8928a21ce8851507 |
| institution | Kabale University |
| issn | 2073-4395 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-de9fb346fe61454e8928a21ce88515072025-01-24T13:16:37ZengMDPI AGAgronomy2073-43952024-12-011517710.3390/agronomy15010077AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in TaroValeria Maeda-Gutiérrez0Juan José Oropeza-Valdez1Luis C. Reveles-Gómez2Cristian Padron-Manrique3Osbaldo Resendis-Antonio4Luis Octavio Solís-Sánchez5Hector A. Guerrero-Osuna6Carlos Alberto Olvera Olvera7Unidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Jardín Juárez 147, Centro, Zacatecas 98000, MexicoCentro de Ciencias de la Complejidad, Universidad Nacional Autónoma de México (UNAM), Mexico City 04510, MexicoUnidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Jardín Juárez 147, Centro, Zacatecas 98000, MexicoPrograma de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México (UNAM), Mexico City 04510, MexicoCentro de Ciencias de la Complejidad, Universidad Nacional Autónoma de México (UNAM), Mexico City 04510, MexicoLaboratorio de Sistemas Inteligentes de Visión Artificial, Posgrado en Ingeniería y Tecnología Aplicada, Universidad Autónoma de Zacatecas, Zacatecas 98000, MexicoLaboratorio de Sistemas Inteligentes de Visión Artificial, Posgrado en Ingeniería y Tecnología Aplicada, Universidad Autónoma de Zacatecas, Zacatecas 98000, MexicoLaboratorio de Invenciones Aplicadas a la Industria (LIAI), Unidad Académica de Ingeniería Eléctrica, Universidad Autónoma de Zacatecas, Zacatecas 98000, MexicoEarly diagnosis and preventive measures are necessary to mitigate diseases’ impact on the yield of Colocasia esculenta (Taro). This study addresses the challenges of Taro disease identification by employing two key strategies: integrating explainable artificial intelligence techniques to interpret deep learning models and conducting a comparative analysis of advanced architectures Inception V3, ResNet-50, and Vision Transformers for classifying common Taro diseases, including leaf blight and mosaic virus, as well as identifying healthy leaves. The novelty of this work lies in the first-ever integration of SHapley Additive exPlanations (SHAP) with deep learning architectures to enhance model interpretability while providing a comprehensive comparison of state-of-the-art methods for this underexplored crop. The proposed models significantly improve the ability to recognize complex patterns and features, achieving high accuracy and robust performance in disease classification. The model’s efficacy was evaluated through multi-class statistical metrics, including accuracy, precision, F1 score, recall, specificity, Chohen’s kappa, and area under the curve. Among the architectures, Inception V3 exhibited superior performance in accuracy (0.9985), F1 score (0.9985), recall (0.9985), and specificity (0.9992). The explainability of Inception V3 was further enhanced using SHAP, which provides insights by dissecting the contributions of individual features in Taro leaves to the model’s predictions. This approach facilitates a deeper understanding of the disease classification process and supports the development of effective disease management strategies, ultimately contributing to improved Taro cultivation practices.https://www.mdpi.com/2073-4395/15/1/77Taroconvolutional neural networkimage classificationinception V3ResNet-50visual transformers |
| spellingShingle | Valeria Maeda-Gutiérrez Juan José Oropeza-Valdez Luis C. Reveles-Gómez Cristian Padron-Manrique Osbaldo Resendis-Antonio Luis Octavio Solís-Sánchez Hector A. Guerrero-Osuna Carlos Alberto Olvera Olvera AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro Agronomy Taro convolutional neural network image classification inception V3 ResNet-50 visual transformers |
| title | AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro |
| title_full | AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro |
| title_fullStr | AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro |
| title_full_unstemmed | AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro |
| title_short | AI-Driven Plant Health Assessment: A Comparative Analysis of Inception V3, ResNet-50 and ViT with SHAP for Accurate Disease Identification in Taro |
| title_sort | ai driven plant health assessment a comparative analysis of inception v3 resnet 50 and vit with shap for accurate disease identification in taro |
| topic | Taro convolutional neural network image classification inception V3 ResNet-50 visual transformers |
| url | https://www.mdpi.com/2073-4395/15/1/77 |
| work_keys_str_mv | AT valeriamaedagutierrez aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT juanjoseoropezavaldez aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT luiscrevelesgomez aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT cristianpadronmanrique aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT osbaldoresendisantonio aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT luisoctaviosolissanchez aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT hectoraguerreroosuna aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro AT carlosalbertoolveraolvera aidrivenplanthealthassessmentacomparativeanalysisofinceptionv3resnet50andvitwithshapforaccuratediseaseidentificationintaro |