Scholarlink

<p>Tree-based horticulture sustains regional economies, food systems, and ecological harmony. Apple orchards are a&#xa0;mainstay of livelihood and horticultural production in the Kashmir Valley. Unfortunately, the productivity and health of apple trees are often challenged by various fungal, bacterial, and environmental diseases. The conventional diagnostic methods for diseases that depend on visual observation and clinical expertise face challenges of high turnaround time, low objectivity, and are not suitable for co-infections and early stages of disease. As climate change and ecological shifts drive disease outbreaks at an accelerating pace, the global health community is demanding intelligent, scalable, real-time systems of disease risk management. Our work aims to fill this gap with the proposed integrated architecture of Internet of Things (IoT) and machine learning (ML) for multi-disease detection and management of apple orchards in Kashmir. The proposed multi-disease management system targets nine major apple diseases, namely, apple scab (<i>Venturia inaequalis</i>), Marssonina leaf blotch, black rot canker, collar rot, powdery mildew, Alternaria leaf spot/blight, core rot, white rot/root rot, and seedling blight. Datasets of infected apple leaves were acquired using a&#xa0;multimodal approach. This work used five ML algorithms (random forest, support vector machine [SVM], k‑nearest neighbors [KNN], convolutional neural networks [CNN], and XGBoost) to analyze the processed data. The results show that the proposed technique is robust and accurate in identifying multiple diseases, with CNN and XGBoost models offering the better performance for complex disease feature classification, while random forest and SVM generalized better for early detection. The proposed system also offers scalable deployment, receives alerts in real time, and delivers actionable insights to farmers via mobile or Web dashboards.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

IoT-ML Based Multi-Disease Management System: A Case Study of Apple Trees

  • 期刊论文

  • 发表日期：

    2026年2月16日

查看全文
  • Saniya Zahoor

摘要

Tree-based horticulture sustains regional economies, food systems, and ecological harmony. Apple orchards are a mainstay of livelihood and horticultural production in the Kashmir Valley. Unfortunately, the productivity and health of apple trees are often challenged by various fungal, bacterial, and environmental diseases. The conventional diagnostic methods for diseases that depend on visual observation and clinical expertise face challenges of high turnaround time, low objectivity, and are not suitable for co-infections and early stages of disease. As climate change and ecological shifts drive disease outbreaks at an accelerating pace, the global health community is demanding intelligent, scalable, real-time systems of disease risk management. Our work aims to fill this gap with the proposed integrated architecture of Internet of Things (IoT) and machine learning (ML) for multi-disease detection and management of apple orchards in Kashmir. The proposed multi-disease management system targets nine major apple diseases, namely, apple scab (Venturia inaequalis), Marssonina leaf blotch, black rot canker, collar rot, powdery mildew, Alternaria leaf spot/blight, core rot, white rot/root rot, and seedling blight. Datasets of infected apple leaves were acquired using a multimodal approach. This work used five ML algorithms (random forest, support vector machine [SVM], k‑nearest neighbors [KNN], convolutional neural networks [CNN], and XGBoost) to analyze the processed data. The results show that the proposed technique is robust and accurate in identifying multiple diseases, with CNN and XGBoost models offering the better performance for complex disease feature classification, while random forest and SVM generalized better for early detection. The proposed system also offers scalable deployment, receives alerts in real time, and delivers actionable insights to farmers via mobile or Web dashboards.