XMACNet: An Explainable Lightweight Attention based CNN with Multi Modal Fusion for Chili Disease Classification

TL;DR

XMACNet is a lightweight, explainable CNN that fuses multi-modal data for accurate chili disease classification, suitable for edge devices, and incorporates novel dataset augmentation and interpretability techniques.

Contribution

The paper introduces XMACNet, a novel CNN architecture combining self-attention and multi-modal fusion for plant disease detection, with a new dataset and explainability features.

Findings

XMACNet outperforms baseline models in accuracy, F1-score, and AUC.

The model is compact and suitable for real-time edge deployment.

Explainability methods effectively highlight disease features.

Abstract

Plant disease classification via imaging is a critical task in precision agriculture. We propose XMACNet, a novel light-weight Convolutional Neural Network (CNN) that integrates self-attention and multi-modal fusion of visible imagery and vegetation indices for chili disease detection. XMACNet uses an EfficientNetV2S backbone enhanced by a self-attention module and a fusion branch that processes both RGB images and computed vegetation index maps (NDVI, NPCI, MCARI). We curated a new dataset of 12,000 chili leaf images across six classes (five disease types plus healthy), augmented synthetically via StyleGAN to mitigate data scarcity. Trained on this dataset, XMACNet achieves high accuracy, F1-score, and AUC, outperforming baseline models such as ResNet-50, MobileNetV2, and a Swin Transformer variant. Crucially, XMACNet is explainable: we use Grad-CAM++ and SHAP to visualize and quantify the models focus on disease features. The models compact size and fast inference make it suitable for edge deployment in real-world farming scenarios.