Jul 2026· 2026 International Conference on Emerging Trends in Information, Communication & Systems (ICETICS)· pp. 1-11· 0 citations· 31 references
Abstract
Agriculture plays a great role in ensuring food security in the world but there are serious threats in rice production due to different types of bacterial and fungal diseases. Manual diagnosis is subjective and labor intensive and may be delayed. The proposed paper suggests a multi-class identification of ten different pathologies related to rice leaves through a multi-class Artificial Intelligence framework using the dataset of 15,023 images. Three methodological paradigms that use mobile net version two with XGBoost, a custom Convolutional Neural Network (CNN), and a novel Hybrid CNN-LSTM model that views patterns of diseases as feature sequences are benchmarked. Whereas the XGBoost classifier has the highest accuracy of 91.75 percent, the deep learning models have high feature extraction capacity. The proposed Ensemble model that incorporates both spatial and sequential learning achieves 97 percent accuracy. This paper supports the hypothesis that deep learning hybrid models have a great impact on diagnostic accuracy, which can be successfully used as an automated instrument to provide disease control in precision agriculture.
Performance evaluation using accuracy, model size, time per image, and number of parameters showed that the proposed model achieved high accuracy and provided better discrimination between visually similar disease classes.
Plant diseases have long been considered a major threat to global food production systems. Therefore, early diagnosis is vital to mitigate the risk of these diseases. This task can be challenging, as the number of harmful diseases is substantial. One technology that has gained widespread interest is artificial intelligence, specifically deep learning, which is used to identify plant diseases using leaf patterns. This paper presents two deep learning models, a custom CNN model and a transfer learning model based on the DenseNet-121 architecture. Experiments were carried out using the PlantCity dataset, which consists of twelve subsets of diverse crop species representing fruits, vegetables, and grains with variations among the subsets, including the number of classes, the subset sizes, class distribution, and visual complexity of disease symptoms. The two models were evaluated using multiple metrics, including accuracy, loss, precision, recall, and F1-score. Explainable AI using the LIME technique was deployed to better interpret the acquired results. Results showed that the developed transfer learning model based on DenseNet-121 had superior performance over the CNN model, with accuracies ranging from 86% to 99% across eleven experimented crops. In order to perform an independent experimental validation for the developed model, future work will focus on constructing a local crop dataset captured from Iraqi fields to evaluate the developed models based on the local environment.
H. Jeiad, S. Samaan, Omar Janeh et al.· Automation· 0 citations
Agriculture is a key component in ensuring food security in the world and one of the most common staple foods in the world is rice. But there are a number of leaf diseases affecting rice, like bacterial leaf blight, brown spot, and rice blast, which cause lower production yield. In order to limit economic losses, it is necessary to be able to diagnose these diseases on time. This paper presents an intelligent prediction and classification of rice leaf disease prediction and classification using deep learning and transfer learning approaches. For that purpose, the proposed model uses two pre-trained transfer learning models, DenseNet201 and EfficientNetB3, and a conventional CNN, to classify healthy and diseased rice leaves from a set of image data. The performances of the three models are measured using the following performance measures: accuracy, precision, recall, and F1-score. From the experiments conducted, it can be noted that the best performing model among all three models is the EfficientNetB3, which shows an excellent classification accuracy of 95%. It performs better than both DenseNet201, whose accuracy is 92% and CNN model, whose accuracy is 83.5%. The reason behind the excellent performance of the EfficientNetB3 is that the architecture of the model is optimized in such a way that it is able to extract the most discriminative features with less computational complexity. The proposed framework can offer an efficient, reliable, and automated solution for the early detection of the diseases in the leaves of rice crops.
Vandana Samanthula, N. Valli, Sultanov Sirojiddin et al.· 2026 6th International Confe...· 0 citations
Agricultural productivity and food security are heavily impacted by plant diseases, and thus there is a high demand for accurate and automated plant disease detection that can be achieved by applying deep learning techniques. This research proposes a Multi-Model Ensemble Method Based on Deep Learning for multi-plant disease detection using ResNet50 to improve classification performance across multiple crop varieties. The proposed framework takes five important categories of plants into consideration including tomato, potato, grape, apple and maize, and 10 classes of healthy and diseased plants are generated from the PlantSeg dataset. The Anaconda platform was used along with Python to create a development environment that allows data preprocessing, augmentation, training and testing to be implemented efficiently. The proposed ensemble framework combines the feature extraction power of ResNet50 with several deep learning classifiers so as to obtain a good identification performance at different resolutions and environments. The proposed model performance is tested with the following metrics Accuracy, Precision, Inference Time, and Resolution quality and compared with MobileNetV2, YOLOv8 and the baseline CNN models. Experimental results show that the proposed ensemble ResNet50 framework achieves an accuracy of 98.7% and precision of 98.3%, which is about 6.4%, 4.8%, and 9.2% higher than that of MobileNetV2, YOLOv8, and CNN respectively. Moreover, the proposed method achieves high resolution disease localization capability with an inference time improvement of almost 18% compared with YOLOv8. The proposed system greatly improves the detection accuracy of the early stage and the calculation speed of the system, which is very suitable for smart agriculture applications and real-time monitoring of the health status of crops.
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
—Agricultural production has become a key factor for the well-being of citizens and stability of the world economy. However, pest diseases in the fields will reduce the production of crops and pose a threat to food security. In order to solve this problem, farmers have to use more advanced pest recognition systems to identify the types of pest diseases quickly rather than making judgement based on their own observation. This paper presents an artificial intelligence recognition model based on multi-scale convolutional neural network called LH-DenseNet. In this model, the advanced deep convolution network model DenseNet is used as the basic framework, and the large-scale public dataset ImageNet is used to train the convolution neural network's powerful feature extraction ability by applying various data enhancement strategies. Extracting and integrating the global and detailed features of images can improve the accuracy of pest classification. Thus, it can enhance the potential of deep learning in the field of agriculture, allowing more autonomic and systematic systems emerge.
Sihai Li· International Journal of Fut...· 0 citations