智慧農業感測資料標準化在臺灣農業發展中的應用研究：數據互通與標準制定之挑戰及未來方向

智慧農業作為現代農業發展的重要方向，結合物聯網（IoT）、大數據、人工智慧（AI）等技術，旨在提升農業生產效率與資源利用率。然而，由於智慧農業中使用的感測器來自不同廠商，其量測範圍、頻譜範圍、數據格式及通訊協議存在差異，導致數據整合與互通的挑戰。本研究聚焦於智慧農業感測資料的標準化問題，特別是針對量測範圍與頻譜範圍差異的處理，提出適用於臺灣農業場域的感測數據比對與校準方法。本研究採用文獻分析法及案例分析法，深入分析智慧農業數據標準化的技術挑戰，並透過案例探討光照度感測器與光量子感測器的數據整合問題。研究結果顯示，透過資料標準化，可顯著提升數據整合效率，支援跨品牌感測器數據的即時整合與分析，並增強AI模型的應用價值。此外，標準化數據有助於支持智慧農業的多場景應用，包括精準灌溉、病蟲害預測及資源管理，進一步促進臺灣智慧農業生態圈的發展。

關鍵字

智慧農業；感測器數據；標準化

並列摘要

Smart agriculture, as a significant development strategy for modern agriculture, integrates technologies such as the Internet of Things (IoT), big data, and artificial intelligence (AI) to improve agricultural productivity and resource utilization efficiency. However, due to the variations in measurement range, spectral range, data format, and communication protocols of sensors from different manufacturers, data integration and interoperability pose significant challenges. This study focuses on the standardization of sensor data in smart agriculture, particularly addressing the issues of differences in measurement and spectral ranges. A method for sensor data comparison and calibration, suitable for Taiwan's agricultural context, is proposed. Using literature analysis and case studies, this research explores the technical challenges of data standardization in smart agriculture. Through case studies, the issues of data integration for illuminance sensors and quantum light sensors were investigated. The results indicate that data standardization significantly improves data integration efficiency, enables real-time integration and analysis of cross-brand sensor data, and enhances the application value of AI models. Additionally, standardized data facilitates multi-scenario applications in smart agriculture, including precision irrigation, pest prediction, and resource management, further promoting the development of Taiwan's smart agriculture ecosystem.