Transformer-augmented EfficientNetV2B3 for robust plant disease identification

Cbam-augmented ResNet for high-accuracy grape leaf disease detection

EfficientNet-CBAM-prototype: an attention-guided EfficientNet framework with dynamic prototype representation for tomato disease classification

In the field of precision agriculture, one of the major hurdles is the early and accurate identification of plant diseases. Farmers may face serious irreversible loss in yield if there is a delay in diagnosis by even a few days. The CNN model has helped in improving the classification of plant diseases but faces difficulty in distinguishing fine-grained symptoms, has limited generalizability, and is not easily interpretable. Though the deep CNN model performs well in classifying plant diseases, the current methods have several flaws. Most importantly, existing algorithms misclassify visually complex samples due to their limited spatial differentiation of disease-specific morphological features, including lesion borders and necrotic regions. Classification reliability is further compromised by low inter-class embedding separability for visually comparable illness phenotypes. Apart from these representational problems, training instability is still a major problem for attention-based models. Combining randomly initialized attention modules with pretrained backbone networks causes this instability, which still limits practical application. We suggest EfficientNet-CBAM-Prototype (ECP-Net), a unique end-to-end deep learning architecture, to overcome these constraints. Three complementary techniques are combined into a single framework by ECP-Net. First, parameter-efficient multi-scale feature extraction is done using an EfficientNetB0 backbone. Second, joint channel-wise and spatial feature recalibration is performed using a stabilized convolutional block attention module (CBAM). Third, a dynamic prototype memory layer uses cosine similarity-based categorization and exponential moving average (EMA) updates to maintain class-representative embedding vectors. To overcome the instability caused by randomly initialized attention weights, we further propose a two-phase training strategy wherein CBAM is frozen during phase 1 to allow prototype stabilization and then jointly fine-tuned with the learning rate in phase 2. Evaluated on the PlantVillage tomato subset comprising 10 disease classes across a class-balanced split of 10,000 training, 500 validation, and 500 test samples, ECP-Net achieves 98.6% test accuracy, 98.59% F1-score, and 98.65% precision with only 4.80M parameters and 85.04 ms average inference time. These results outperformed baselines including VGG16 (97.00%), ResNet50 (81.20%), MobileNetV2 (81.20%), and CNN (70.00%). Generalization is further validated on 35 real-field tomato leaf images captured under natural, uncontrolled conditions, confirming practical deployment potential.

SAEFormer: Self-Supervised and Attention-Enhanced Efficient Transformer for Robust Tomato Leaf Disease Recognition

Tomato is a major global crop, yet foliar diseases seriously affect yield and quality. Accurate and efficient disease identification techniques are of great significance for ensuring agricultural production safety. To address the limitations of existing methods in feature extraction capability, model complexity, and dependence on large-scale labeled data, as well as the difficulty of recognizing early-stage diseases whose visual symptoms are not yet fully developed, this paper proposes SAEFormer, a lightweight and robust disease recognition model. The model integrates a Multi-scale Selective Fusion Attention Block to enhance the ability to model multi-scale semantic information in lesion areas. In addition, by leveraging a self-supervised loss derived from dense relative localization as an auxiliary regularization term, the model’s generalization ability under limited training data is notably enhanced. To optimize normalization and improve inference efficiency, the RepBN normalization strategy is further adopted, significantly reducing computational cost while maintaining model performance. Experimental results on Dataset A show that SAEFormer achieves a Top-1 accuracy of 87.86% with 24.14 M parameters and 5.35 GFLOPs, demonstrating a favorable balance between recognition accuracy and model complexity. The training curves further indicate stable convergence during model optimization. Ablation experiments validate the complementary contributions of the proposed modules. Moreover, SAEFormer achieves competitive performance in cross-dataset evaluation on Dataset B, indicating its potential adaptability to different data distributions. Overall, SAEFormer provides an efficient approach to tomato leaf disease recognition and shows potential for deployment in precision agriculture applications.

Deep Learning for Plant Disease Detection: A Systematic Review

It is suggested that in order to be implementable in the field, future intelligent agricultural diagnosis systems must be able to balance predictive accuracy, explainability, computational efficiency and field adaptability.

PHYTOASSIST: Deep Learning-Driven Leaf Disease Identification with Interactive Agricultural Knowledge Support

Early diagnosis of tree leaf diseases is crucial for ensuring ecological stability, conserving biodiversity and sustainable agriculture productivity. Manual inspection and traditional image processing methods are often subjective, time-consuming, and susceptible to environmental changes like illumination variations, complex background and orientation of leaves in the images, which makes them difficult to work with. In response to the above drawbacks, the present work aims at proposing a novel, intelligent deep learning-based framework for automated detection of tree leaf diseases and supporting tree Agri-knowledge, namely PHYTOASSIST. The proposed approach is based on YOLO (You Only Look Once) convolutional neural network (CNN) for real-time object recognition of disease region on diseased leaf using high resolution image. To ensure robustness and generalisation capabilities to field scenarios, image pre-processing is used such as resize, normalisation and augmentation. The framework also includes a fertilizer and treatment recommendation module as well as an interactive agricultural chatbot that offers contextually relevant information for agricultural practice and disease prevention. Experimental results have proven the robustness of the accuracy (96.2%), precision (95.4%), recall (94.8%), and F1 score (95.1%) with an inference speed of less than 10 frames per second. The outcomes demonstrate the effectiveness of the design of the framework supporting the scalable agricultural disease monitoring and intelligent decision support in sustainable agricultural environments.

A CNN-BiLSTM Hybrid for Plant Leaf Disease Classification: Comparative Performance Evaluation of Deep Learning Algorithms

This study introduces a hybrid deep learning architecture that integrates squeeze-and-excitation residual blocks, capsule networks, bidirectional long short-term memory, and attention mechanisms, enabling farmers to obtain rapid, reliable, and cost-effective field diagnoses, thereby improving agricultural productivity and sustainability.