A Lightweight Single-Token CNN-Transformer Architecture for Robust Multi-Crop Plant Disease Classification

A biologically structured hierarchical vision transformer-CNN framework for robust tomato leaf disease classification

Precise and reliable diagnosis of leaf diseases in tomato is essential for enhancing crop cultivation and minimizing agricultural losses. While deep learning models have performed well on benchmark datasets, the majority of present techniques rely on flat multi-class classification, which predicts all disease categories simultaneously. Such formulations promotes inter-class confusion, particularly when biologically different diseases with similar visual symptoms are learned within a same model. To overcome this constraint, we propose a biologically structured hierarchical deep learning framework in this study. Instead of directly classifying 10 disease classes, the proposed method first classifies leaf images into meaningful biological groups such as bacterial, fungal, pest-associated and healthy using a vision transformer (ViT) model. Then, specialized convolutional neural network (CNN) experts perform fine-grained classification within each category. The proposed hierarchical model shows an overall accuracy of 97.8%, when validated on PlantVillage tomato dataset. A flat ViT model trained with class-weighted loss obtained 96.3% accuracy, whereas a flat CNN model reached 99.3% under clean conditions but decreased sharply to 41% under Gaussian perturbation (σ = 0.05). On the other hand, the hierarchical model performed steadily under noise with 97.4% accuracy at the same perturbation level. These results indicate that adding biological structure to model design reduces confusion, helps prevent imbalance effects and increases robustness, providing a more trustworthy and interpretable solution for real-world agricultural disease diagnosis.

A Deep Hybrid Convolutional Neural Network (CNN)–Transformer Approach for Early Detection of Tomato Leaf Diseases

The early and effective diagnosis of tomato leaf diseases is very important to enhance crop yield and reduce economic loss in precision agriculture. The conventional image-based methods are typically based on single architecture model, which cannot capture fine-grained lesion details and global contextual patterns simultaneously in the real-field. To this end, we introduce a deep hybrid Convolutional Neural Network (CNN) –Transformer architecture by combining ConvNeXt Large (ConvNeXt-L) (as local feature extractor) and Swin Transformer (as global context encoder). The concatenated features vector is then fed to a shallow classifier to predict the disease. The model was tested on two datasets, namely a field dataset in agriculture areas from Madhya Pradesh (India) and a benchmark tomato leaf dataset. Experimental results revealed that the proposed scheme achieved accuracy of 92.83% on a primary dataset, and performance was significantly high with an accuracy of up to 95.65% in terms of generalization rate for computing technique models from various environmental conditions.

A Hybrid Transfer Learning Framework for Corn Crop Detection Using Deep Convolutional Networks

Corn is a staple crop of global significance; however, foliar diseases may lead to 30–60% yield loss if not detected at an early stage. Conventional visual inspection is time-consuming, subjective, and difficult to scale for smallholder farmers worldwide. To overcome these issues, we present Corn Transfer Learning Network (CTL-Net), an end-to-end hybrid deep learning model for corn leaf disease identification. CTL-Net, which combines Inception-ResNet-v2 as the backbone and MobileNetV3 as a feature extractor in parallel convolutional streams, simultaneously learns diverse scales of texture information from low-level textures, mid-level structural patterns, and high-level disease semantics of RGB leaf images. Adaptive feature fusion is formulated through learnable weighting coefficients and bi-directional spatial–channel attention mechanisms, which further enhance feature discriminability and robustness. The proposed approach is tested on an extensive dataset of 12,456 images from 10 corn diseases, including Northern Leaf Blight, Common Rust, Gray Leaf Spot, and Cercospora Leaf Spot, acquired under controlled and real-field conditions. CTL-Net attains the highest classification accuracy of 99.42%, outperforming DenseNet121 (97.92%), EfficientNetB3 (97.35%), and general stacking models (97.89%). Robustness experiments demonstrate the effectiveness of the proposed method against illumination variations, additive noise, and partial occlusions. CTL-Net enables real-time inference with a latency of 42 ms on an NVIDIA RTX 3090 GPU. Gradient-weighted Class Activation Mapping++ (Grad-CAM++)-based interpretability analysis results in a mean Intersection over Union (IoU) of 87.6% with expert-annotated disease regions. Five-fold cross-validation, ablation studies, and statistical significance testing (p < 0.001) demonstrated that the contribution of each architectural component is significant. To summarize, CTL-Net allows accurate and interpretable predictions in a computationally efficient manner, making it an excellent candidate for mobile and resource-limited applications in precision agriculture.

From Handcrafted Features to Transformers: A Hybrid CNN–Vision Transformer Framework for Cherry Leaf Disease Detection

An Intelligent Hybrid Deep Learning Model Integrating CNN, Transformer, and LSTM for Precision Cotton Disease Diagnosis

Cotton leaf diseases significantly affect crop yield and quality, leading to substantial economic losses in agricultural regions. Early and accurate detection of these diseases is essential for sustainable crop management and precision agriculture. Existing studies primarily rely on conventional Convolutional Neural Networks (CNNs) or object detection models that focus on static image classification, often lacking global contextual learning and temporal disease progression modeling. To address these limitations, this study proposes a novel hybrid deep learning framework integrating Convolutional Neural Networks (CNN), Transformer-based attention mechanisms, and Long Short-Term Memory (LSTM) networks for spatio-temporal cotton leaf disease detection and classification.In the proposed architecture, a pre-trained EfficientNet backbone is employed for spatial feature extraction, followed by a multi-head self-attention Transformer block to capture global contextual relationships within leaf patterns. An LSTM layer is incorporated to enable temporal sequence modeling, allowing the framework to be extended toward disease progression analysis and forecasting. The model is trained and evaluated on a structured cotton leaf dataset with data augmentation and fine-tuning strategies to enhance generalization performance.Experimental results demonstrate that the proposed hybrid model achieves superior classification accuracy, improved robustness, and enhanced feature representation compared to conventional CNN-based approaches. Confusion matrix analysis and ROC curve evaluation confirm high discriminative capability across multiple disease classes. The integration of spatial, global attention, and temporal learning components makes the proposed framework suitable for intelligent precision agriculture systems and real-time disease monitoring applications.The developed model provides a scalable foundation for future integration with IoT-based environmental sensing systems, enabling predictive crop health management and early warning advisory mechanisms in smart farming environments.

A Compact Deep Learning Framework for Potato Leaf Disease Classification Across Controlled/Uncontrolled Environments

Potato leaf disease poses a significant threat to global food security, causing substantial crop losses that jeopardise agricultural productivity and farmers’ livelihoods worldwide. Existing automated detection frameworks suffer from several persistent limitations, including over-reliance on controlled benchmark datasets, narrow disease class coverage, exclusive use of spatial feature representations, absence of feature selection, and dependence on single-architecture end-to-end pipelines. To address these limitations, this paper proposes ComPo-Net, a novel lightweight ensemble framework that integrates three efficient CNN architectures—ResNet18, ShuffleNet, and MobileNetV2—for nine-class potato leaf disease detection and classification. Deep features are extracted from three intermediate layers of each network, with the Discrete Wavelet Transform applied for dimensionality reduction and cross-network fusion of the higher-dimensional layer features, capturing spectral–spatial information that purely spatial approaches cannot provide, while the remaining layer features are directly concatenated across networks. One-way Analysis of Variance (ANOVA) feature selection is subsequently applied to retain the most statistically significant features from the combined multi-scale, multi-network representation, and seven machine learning classifiers are systematically evaluated to identify the optimal classification strategy. The framework is assessed on a merged dataset of three publicly available benchmarks spanning both controlled and uncontrolled imaging environments, constituting a nine-class evaluation setting not previously addressed at this scale in the literature. ComPo-Net achieves an accuracy of 96.32%, an F1-score of 93.87%, an MCC of 0.9350, and AUC values exceeding 0.993 across all nine classes with Cubic SVM as the best-performing classifier. When compared against methods evaluated on the seven-class uncontrolled-environment dataset—the closest available task setting to ComPo-Net’s nine-class merged benchmark—ComPo-Net surpasses the best-performing comparable method by a margin of 6.45 percentage points, demonstrating the effectiveness of multi-scale ensemble feature extraction combined with spectral–spatial representation and principled feature selection for robust potato leaf disease detection under diverse real-world conditions.