Skip to content
Open access

A controlled analysis of lightweight CNN variants for chilli leaf disease classification with class weighting, label smoothing and class imbalance handling

Aug 2026 · Discover Artificial Intelligence · Vol 6 · 0 citations · 60 references
Computer Science

TL;DR

Dense256Net provides the strongest performance–efficiency balance among the proposed variants and remains competitive with the evaluated baselines and remains competitive with the evaluated baselines.

Abstract

Accurate and timely recognition of chilli leaf disorders is essential for reducing crop loss and supporting efficient disease management in practical agricultural environments. However, multi-class chilli leaf disease classification remains challenging because visually similar symptoms can arise from fungal infection, pest damage, nutritional deficiency, and normal leaf appearance, while class imbalance and limited interpretability often reduce the practical reliability of deep learning systems. To address these issues, this study proposes a unified multi-variant lightweight convolutional neural network (CNN) framework for five-class chilli leaf disease classification, comprising cercospora, healthy, mites_and_trips, nutritional, and powdery mildew. Four custom CNN variants, namely Conv2DNet, MaxPoolNet, GAPNet, and Dense256Net, were evaluated under identical conditions to provide a controlled architectural comparison. The framework combines imbalance-aware training through data augmentation, class weighting, label smoothing, and validation Macro-F1-guided checkpointing; comprehensive reliability assessment through class-wise metrics, confusion matrices, ROC and precision–recall curves, confidence analysis, Grad-CAM, cross-validation, statistical testing, ablation analysis, and external validation; and deployment-oriented comparison using computational complexity and inference measures. A focused head ablation isolates the contribution of the Dense(256) projection, while the baseline evaluation is extended to EfficientNetV2-B0 and ConvNeXt-Tiny in addition to the existing standard and lightweight models. Dense256Net achieved a test accuracy of 0.928, a Macro-F1 score of 0.918, and a weighted F1-score of 0.925. Five-fold cross-validation confirmed stable performance with a mean accuracy of (0.930 ± 0.003) and Macro-F1 score of (0.920 ± 0.004). The model retained a favorable profile with 2.12M parameters, 0.28G FLOPs, an 8.1 MB model size, and 4.8 ms/image GPU inference time. External validation on a publicly available independent chilli leaf dataset achieved 0.904 accuracy and 0.891 Macro-F1 without retraining. Overall, Dense256Net provides the strongest performance–efficiency balance among the proposed variants and remains competitive with the evaluated baselines.

Read PDF

Similar papers

Open access Aug 2026

A Lightweight Convolutional Neural Network for Apple Leaf Disease Classification Using Data Augmentation

A lightweight, effectual CNN model trained from scratch for classifying apple leaf images into four categories: healthy, rust, scab, and multiple diseases achieves good classification performance while maintaining a simple architecture suitable for identifying apple leaf diseases in the context of precision agriculture...

V. Devi, Pardeep Kumar · 0 citations
Open access Aug 2026

A comparative benchmark of vision transformer architectures for chili leaf disease classification

Chili plant disease detection represents a critical component for enhancing agricultural productivity. Although Convolutional Neural Networks (CNN) have demonstrated promising results, they encounter limitations in capturing global contextual relationships within images. However, existing Vision Transformer studies on...

Acihmah Sidauruk, Danang Wijayanto, I. M. D. Agastya et al. · 0 citations
Open access Sep 2026

Deep Learning-based Bean Leaf Disease Classification: A Comparison of ResNet50 and VGG19

This work demonstrates ResNet50’s performance for bean leaf disease classification tasks, providing important insights for future research and practical applications in agricultural disease control.

Yu-Yan Xu, H. Chen, Qing-Mei Lin · 0 citations
Open access Aug 2026

Comparative Analysis of Convolutional Neural Network, MobileNetV2, and EfficientNet for Tomato Leaf Disease Classification

The experimental results demonstrate that the transfer learning approach significantly outperformed the custom CNN, with EfficientNet-B0 achieving the highest classification accuracy of 95.73% and improved class separability across disease categories.

Guntur Guntur, Abdul Latief Arda, A. Affandy et al. · 0 citations
Open access Sep 2026

A Lightweight RepViT-M1 Framework for Tea Leaf Disease Classification: Benchmarking Against VGG16 for Efficient Edge Deployment

The findings indicate that compact reparameterizable backbones can match far larger networks on tea-disease recognition, offering a practical route to on-device agricultural diagnostics.

Sarada R, Juvvala Bala Ambedkar, Praveen Kumar Nalli et al. · 0 citations
Open access Aug 2026

A Two-Step CNN-Based Approach for Banana Leaf Disease Classification Using ResNet-50 Optimization with Weight Adjustment and Class Balancing

The proposed study introduces a novel two-step convolutional neural network (CNN)-based classification approach using ResNet-50 with class weighting and Synthetic Minority Over-sampling Technique (SMOTE) augmentation to improve banana leaf disease classification accuracy while addressing class imbalance issues. Unlike...

Divya Rohatgi, Miguel Villagómez-Galindo, Ana Beatriz Martinez-Valencia et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.