AgroXAI: a multi-XAI deep learning system for explainable multi-species plant disease classification
- 1School of Computer Science and Engineering, Vellore Institute of Technology
Abstract
Automated plant disease classification can benefit from models that provide both predictive performance and interpretable evidence for their decisions. This study presents AgroXAI, a Multi-XAI deep learning framework for explainable multi-species plant disease classification under a controlled, leakage-aware evaluation protocol. A publicly available dataset containing 2,815 leaf images from seven plant species was organized into 14 joint species–health classes and partitioned into fixed group-disjoint training, validation, and held-out test sets containing 1,970, 422, and 423 images, respectively. Preprocessing and architecture selection were conducted using only the training and validation partitions, leading to the AgroX-Swin classifier based on Swin Transformer-Tiny with 384 × 384 Original RGB inputs. Explainability was quantitatively evaluated using Grad-CAM++, SHAP, LIME, and their complementary fusions on a predefined class-balanced subset of 42 held-out images. Across five independently initialized training runs, AgroX-Swin achieved a mean held-out accuracy of 87.19% and macro-F1 of 0.8300 ± 0.0122, while the five-model probability ensemble achieved 87.94% accuracy and 0.8355 macro-F1. Deterministic aggregation of the same 14-class predictions yielded 99.29% species-recognition accuracy and 88.42% binary health-status accuracy, with a diseased-class F1-score of 0.9160. Temperature scaling reduced the representative model's expected calibration error from 0.0792 to 0.0310. At the 20% perturbation level, Grad-CAM++ + LIME produced the largest mean confidence drop (0.2540), while Grad-CAM++ showed the highest transformation stability (0.8371) and SHAP + LIME showed the highest attribution-burden stability (0.7954), demonstrating complementary rather than uniformly superior behavior among individual and fused explanation methods. Model-randomization analysis further examined the dependence of Grad-CAM++ explanations on learned model parameters. AgroXAI provides an integrated framework combining leakage-aware model development, multi-seed training-stability assessment, held-out classification, calibration, statistical uncertainty analysis, and quantitative evaluation of complementary XAI methods. The resulting attribution maps are interpreted as prediction-relevant visual explanations rather than validated lesion localization or quantitative disease-severity estimates. External robustness and real-world agricultural generalization remain to be established using independently acquired field datasets.
Related
- A hybrid SE-ResNet50 deep learning framework for high-accuracy and explainable cotton leaf disease classification
- Explainable plant disease classification and severity assessment using MobileNetV3-large for crop health monitoring
- Transfer Learning-Driven Convolutional Neural Networks for Classification of Plant Species from Leaf Vein Patterns
- Explainable computational framework for subtype classification and biomarker identification in head and neck squamous cell carcinoma using machine learning
- MetaMamba: Meta-learning with mamba for few-shot vegetation species classification using UAV-based hyperspectral imagery in desert rangeland
BibTeX
@article{kumar2026agroxai,
title = {{AgroXAI: a multi-XAI deep learning system for explainable multi-species plant disease classification}},
author = {Sathish Kumar K. and G. N. Balaji},
journal = {Frontiers in Plant Science},
year = {2026},
month = oct,
doi = {10.3389/fpls.2026.1965848},
url = {https://www.frontiersin.org/articles/10.3389/fpls.2026.1965848/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - AgroXAI: a multi-XAI deep learning system for explainable multi-species plant disease classification
AU - Sathish Kumar K.
AU - G. N. Balaji
PY - 2026
DA - 2026/10/14
JO - Frontiers in Plant Science
DO - 10.3389/fpls.2026.1965848
UR - https://www.frontiersin.org/articles/10.3389/fpls.2026.1965848/full
AB - Automated plant disease classification can benefit from models that provide both predictive performance and interpretable evidence for their decisions. This study presents AgroXAI, a Multi-XAI deep learning framework for explainable multi-species plant disease classification under a controlled, leakage-aware evaluation protocol. A publicly available dataset containing 2,815 leaf images from seven plant species was organized into 14 joint species–health classes and partitioned into fixed group-disjoint training, validation, and held-out test sets containing 1,970, 422, and 423 images, respectively. Preprocessing and architecture selection were conducted using only the training and validation partitions, leading to the AgroX-Swin classifier based on Swin Transformer-Tiny with 384 × 384 Original RGB inputs. Explainability was quantitatively evaluated using Grad-CAM++, SHAP, LIME, and their complementary fusions on a predefined class-balanced subset of 42 held-out images. Across five independently initialized training runs, AgroX-Swin achieved a mean held-out accuracy of 87.19% and macro-F1 of 0.8300 ± 0.0122, while the five-model probability ensemble achieved 87.94% accuracy and 0.8355 macro-F1. Deterministic aggregation of the same 14-class predictions yielded 99.29% species-recognition accuracy and 88.42% binary health-status accuracy, with a diseased-class F1-score of 0.9160. Temperature scaling reduced the representative model's expected calibration error from 0.0792 to 0.0310. At the 20% perturbation level, Grad-CAM++ + LIME produced the largest mean confidence drop (0.2540), while Grad-CAM++ showed the highest transformation stability (0.8371) and SHAP + LIME showed the highest attribution-burden stability (0.7954), demonstrating complementary rather than uniformly superior behavior among individual and fused explanation methods. Model-randomization analysis further examined the dependence of Grad-CAM++ explanations on learned model parameters. AgroXAI provides an integrated framework combining leakage-aware model development, multi-seed training-stability assessment, held-out classification, calibration, statistical uncertainty analysis, and quantitative evaluation of complementary XAI methods. The resulting attribution maps are interpreted as prediction-relevant visual explanations rather than validated lesion localization or quantitative disease-severity estimates. External robustness and real-world agricultural generalization remain to be established using independently acquired field datasets.
SN - 1664-462X
ER -