2026· ITM Web of Conferences· Vol 87, pp. 01010· 0 citations· 6 references
TL;DR
The proposed work suggests an intelligent Edge-AI-based system of detecting any disease in maize employing deep Convolutional Neural Network (CNN) models to provide the opportunity to detect diseases in real-time with minimum latency and minimum reliance on the cloud.
Abstract
Maize is the most common cereal crops cultivated and early identification of leaf diseases is the primary measure to avoid yield of the crop and food insecurity. The conventional ways of manual inspections are time consuming, subjective and expert based. The proposed work suggests an intelligent Edge-AI-based system of detecting any disease in maize employing deep Convolutional Neural Network (CNN) models. The proposed work combines image acquisition with mobile or edge sensors and on-device deep learning inference to provide the opportunity to detect diseases in real-time with minimum latency and minimum reliance on the cloud. An edge hardware is used to perform a lightweight, but high-precision CNN network optimized by both transfer learning and data augmentation methods. The model is conditioned to identify various maize leaf diseases and healthy samples with greater resilience to different environmental factors. The experimental results show that the tools are highly classifiable, inference time is very low and resource use is efficient. The proposed design provides scalable, affordable, and smart solution to precision agriculture and smart farming implementation.
Conventional plant disease detection is time-consuming and prone to human error. The purpose of this study is to propose an edge artificial intelligence (AI)-based deep learning framework for plant disease detection under real-field conditions. The model integrates convolutional neural networks (CNNs) with a Sliding Window Mean Absolute Deviation (SWMAD) preprocessing technique to address illumination variability and complex background conditions.
A dual-layer CNN model is developed for plant disease classification using HSV segmentation, flood-fill segmentation and SWMAD preprocessing to handle real-field variations. The model is trained on PlantVillage and real farm images and deployed via TensorFlow Lite for real-time offline detection. TensorFlow Lite is used to enable efficient on-device inference for deployment in resource-constrained environments.
Benchmark experiments conducted on the PlantVillage dataset achieved a classification accuracy of 99.91% under controlled conditions. On the hybrid dataset comprising PlantVillage and real-field images, the optimized CNN framework achieved a validation accuracy of 95.01% following extensive evaluation of optimizers, layer architectures and worker configurations. The integration of the proposed SWMAD preprocessing technique into the finalized architecture further improved the validation accuracy to 97.38%, demonstrating enhanced robustness and classification performance under practical agricultural conditions in the final deployed model.
The originality of this study lies in several novel contributions. First, we introduce an SWMAD-based preprocessing technique, which enhances local statistical variations in leaf images by capturing pixel-level deviations from neighborhood intensity means. Unlike, conventional preprocessing methods, SWMAD is specifically designed to handle real-field challenges such as illumination variation, noise and complex backgrounds. The improvement in validation accuracy demonstrates the effectiveness of the proposed approach in capturing more discriminative features compared to existing techniques, thereby improving overall model robustness and reliability in practical agricultural environments.
S. Trivedi, Neha Sharma· International Journal of Int...· 0 citations
Agriculture is a key sector in developing economies, but crop diseases significantly impact productivity, food security, and farmers’ livelihoods. Early detection is crucial to minimize losses, yet traditional methods are slow, error-prone, and depend heavily on human expertise. Recent advancements in Artificial Intelligence (AI), particularly machine learning (ML) and deep learning (DL), have enabled more efficient automated crop disease detection. This study reviews pre-2018 AI-based approaches, focusing on techniques such as image processing, feature extraction, and classification methods. It highlights models like Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and hybrid systems combining traditional and modern techniques. The proposed approach includes preprocessing, segmentation, color transformation, and extraction of texture, color, and shape features, followed by supervised learning for classification. AI systems can detect subtle disease symptoms early, achieving over 90% accuracy under controlled conditions. Integration with mobile and IoT technologies enables real-time monitoring and decision support for farmers. However, challenges such as limited datasets, environmental variability, and computational constraints remain. Future work should focus on developing scalable, robust, and field-deployable solutions for diverse agricultural conditions.
J. Rogers, Brandon Truaxe· International Journal of Mod...· 0 citations
A scalable image-based framework for automatic rice leaf disease detection and monitoring using deep learning and cloud-enabled analytics and enables deployment on smartphones, drones, and edge devices for smart agriculture applications.
B. Venkata Ramulu, Geeta Tripathi· International Journal of Eng...· 0 citations
The general health condition of the crop is very essential for increasing the agricultural production and global food security. Fungal diseases in leaves may spread rapidly and result in yield reduction if not detected timely. To this end, an intelligent system is constructed based on deep convolutional neural network to classify the leaf diseases of plants using images. The CNN model can be further qualified by a diverse dataset coupled with preprocess and data augmentation for higher generalization capability. The CNN architecture drawn learns highly packed visual feature from the plant leaf image and can classify several types of disease with very high accuracy. Our experiment results verified the efficacy of the package with 92.23% classification accuracy which outperformed traditional image processing combined with classical machine learning approaches in the past. It can effectively be used for actual farming in the field with positive contribution to sustainable farming practices.
D.P. Rohitha, A.T.Vishaka, K. Subhashini· International Conference Com...· 0 citations
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.
Ejamandla Anuradha, Vijaya Chandra Jadala, P. Ramanjineyul· 2026 International Conferenc...· 0 citations
The Plant Disease Detecting System leverages advances in artificial intelligence and deep learning to provide an automated, efficient, and reliable solution for identifying plant diseases at an early stage and contributes to increased crop productivity, reduced chemical usage, and sustainable farming practices.
K. Maithili· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
