Scholarlink

In the age of digital transformation, agriculture is harnessing the potential of advanced technologies to address global challenges related to food security and environmental sustainability. While Smart Agriculture and Precision Agriculture lay the foundation, there remains a pressing need for real-time, granular data processing to realize their full potential. Edge computing, coupled with Machine Intelligence (MI), promises a solution, yet existing models often lack the adaptability and real-time efficiency demanded by diverse agricultural scenarios. Addressing these challenges, this paper proposes the Precision Intensive Crop Health Observer (PICHO). PICHO is an innovative three-tier architecture comprising the Field Data Interface Unit (FDIU) for seamless sensor data collection, the Edge Analytics Processor (EAP) for real-time edge-based analytics, and the Cloud Coordination Hub (CCH) for data storage and in-depth analysis. Central to the EAP's efficacy is the lightweight learning model LENet, which is designed with dual pipelines for processing sensor and image data. Specifically crafted to be trained on resource-constrained edge devices, LENet ensures that PICHO remains adaptable, efficient, and poised to redefine the future of precision agriculture.

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-Driven Agricultural Monitoring Architecture with Novel Light Machine Learning for Enhanced Precision Crop Health Analysis

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hari Krishnan Andi,
  • Lokeshkumar Ramasamy,
  • Senthil Kumar Nramban Kannan,
  • Sasikumar Murugesan,
  • Bejjam S. N. Benarji,
  • Umamaheswari Manogaran,
  • Sudhakar Sengan

摘要

In the age of digital transformation, agriculture is harnessing the potential of advanced technologies to address global challenges related to food security and environmental sustainability. While Smart Agriculture and Precision Agriculture lay the foundation, there remains a pressing need for real-time, granular data processing to realize their full potential. Edge computing, coupled with Machine Intelligence (MI), promises a solution, yet existing models often lack the adaptability and real-time efficiency demanded by diverse agricultural scenarios. Addressing these challenges, this paper proposes the Precision Intensive Crop Health Observer (PICHO). PICHO is an innovative three-tier architecture comprising the Field Data Interface Unit (FDIU) for seamless sensor data collection, the Edge Analytics Processor (EAP) for real-time edge-based analytics, and the Cloud Coordination Hub (CCH) for data storage and in-depth analysis. Central to the EAP's efficacy is the lightweight learning model LENet, which is designed with dual pipelines for processing sensor and image data. Specifically crafted to be trained on resource-constrained edge devices, LENet ensures that PICHO remains adaptable, efficient, and poised to redefine the future of precision agriculture.