Logistic Boosting of Leveraging SVM Machine Learning for IoT-Enhanced Anomaly Detection and Agricultural Disease Classification | ||||
Journal of the Egyptian Mathematical Society | ||||
Volume 32, Issue 1, 2024, Page 123-148 PDF (1.47 MB) | ||||
Document Type: Original Article | ||||
DOI: 10.21608/joems.2024.407555 | ||||
![]() | ||||
Author | ||||
Walid Dabour ![]() | ||||
1Department of Mathematics and Computer Science, Faculty of Science, Menoufia University, Egypt | ||||
2Faculty of Engineering, Menoufia National University, Menoufia, Egypt | ||||
Abstract | ||||
Agriculture is pivotal to global food security and economic stability. Efficient disease management and pest control are essential for maintaining crop yield and quality. Apple cultivation, in particular, faces persistent threats from diseases like apple rust and apple scab, which significantly impact productivity. This study presents a novel hybrid approach for disease classification within an Internet of Things (IoT)-enabled framework. Leveraging DenseNet121 for feature extraction and Support Vector Machine (SVM) for classification, the proposed model integrates transfer learning with a hinge-loss SVM classifier. The model, evaluated using the Plant Pathology 2020 dataset, achieved 99% accuracy, surpassing existing benchmarks in precision, recall, and Area Under the Curve (AUC). The Adam optimizer further optimized DenseNet121's performance. Future work will focus on expanding the dataset and incorporating additional disease categories, underscoring the potential of IoT-enabled hybrid models to transform agricultural disease management. | ||||
Keywords | ||||
SVM Machine Learning, Agriculture, apple diseases, AI; DenseNet121, IoT, disease classification | ||||
Statistics Article View: 123 PDF Download: 63 |
||||