Aug 2026· International Journal of Advances in Data and Information Systems· 0 citations· 36 references
TL;DR
A hybrid framework integrating a Convolutional Neural Network with a Large Language Model to perform image-based plant leaf disease classification accompanied by interpretable diagnostic explanations demonstrated that combining high-performance visual recognition with language-based reasoning enhanced both diagnostic reliability and interpretability in digital agriculture applications.
Abstract
Plant diseases have continued to threaten agricultural productivity, while manual inspection methods have remained inefficient and prone to subjectivity. This study proposed and assessed a hybrid framework integrating a Convolutional Neural Network (CNN) with a Large Language Model (LLM) to perform image-based plant leaf disease classification accompanied by interpretable diagnostic explanations. EfficientNetV2-M was employed as the visual backbone and trained on 11 selected classes of apple, grape, and potato leaf images derived from the PlantVillage dataset. A structured data splitting strategy was applied to ensure reliable model validation and unbiased testing. The classification capability of the CNN component was examined through standard multi-class evaluation indicators, including class-wise predictive consistency and error distribution analysis. Experimental results indicated that the model delivered highly consistent predictions, reaching a peak test accuracy of 99.79%, reflecting its robustness in distinguishing visually similar disease patterns. To overcome the black-box limitation, prediction outputs were transformed into structured prompts and processed by GPT-4o to generate contextual explanations. The generated narratives systematically described observable symptoms, highlighted distinguishing characteristics, and suggested initial management actions. Overall, the proposed hybrid system demonstrated that combining high-performance visual recognition with language-based reasoning enhanced both diagnostic reliability and interpretability in digital agriculture applications.
This study presents a novel CNN for multi-class classification of 38 diseases, demonstrating an effective balance between predictive performance and computational efficiency, positioning the model as a promising tool for real-world agricultural deployment.
Image-based techniques are widely used in plant disease classification to support agricultural productivity and facilitate early detection. This study presents a comparative analysis between a baseline convolutional neural network (CNN) and a ResNet50-based transfer learning model for tomato leaf disease classification. A publicly available dataset containing five categories—tomato bacterial spot, tomato late blight, tomato septoria leaf spot, tomato yellow leaf curl virus, and healthy leaves—was used in the experiments. Model performance was evaluated using several standard classification metrics, with emphasis on overall accuracy. The baseline CNN achieved an accuracy of 97.0%, whereas the ResNet50 model reached 99.6%. The results demonstrate that the ResNet50 model produces more stable and reliable predictions, particularly when distinguishing between visually similar disease classes. These findings confirm that transfer learning can effectively improve classification performance in plant disease recognition tasks.
Sumana Budsabok, Wachiraporn Polpanumas, Piyanan Khongphai· International Journal of Ele...· 0 citations
Findings confirm that the hybrid CNN–Transformer architecture effectively enhances classification performance, robustness, and generalization in coffee leaf disease classification, with potential applications in precision agriculture and data-driven crop management.
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.
Aekkarat Suksukont, Ekachai Naowanich· Journal of Advances in Infor...· 0 citations
An extensive set of experiments was conducted to evaluate the performance of the proposed model for plant disease detection, and it is demonstrated that the model achieves highly reliable results, with an accuracy of 97.13%.
Hayat Meddeber, M. Meddeber· ITEGAM- Journal of Engineeri...· 0 citations
— Plant leaf disease classification involves identification and classification of different diseases based on indicators of plant leaves, which plays a crucial role in managing crop health. However, classifying plant leaf diseases is challenging due to wide variation in leaf shape, texture, and color, which leads to overlapping symptoms and inaccurate classification. In this research, the Wombat Escape Strategy-Hippopotamus Optimization Algorithm-based Recalibrated Multi-Scale Squeeze and Excitation Convolutional Neural Network (WES-HOA-based ReScaleX-CNN) is proposed to classify plant leaf disease accurately. In HOA, WES is incorporated to select the most appropriate features that enhance global searchability by guiding agents away from local optima. ReScaleX-CNN enhances the model’s ability to concentrate on informative features by emphasizing significant spatial and channel-wise information. The multiscale approach captures disease patterns at various resolutions, leading to robust performance. Hence, the proposed method obtains a high accuracy of 99.92% on PlantVillage dataset in comparison with existing methods, such as DeepPlantNet.
Spoorthi Pothaganahalli Akkalappa, Shivaputra Shivaputra, Meenakshi Laxman Rathod et al.· Journal of Communications So...· 0 citations