以影像自動量測河川表面流速方法之研究

Abstract

近年來我們深受地球暖化及全球氣候異常變化的影響，伴隨著災害事件發生趨勢的提升，因此相關監測系統應用於水文防災領域的數量隨之成長；然而在傳統監測有幾個侷限，其通常乃透過人力針對相關災情逐一檢視，因此較無法傳遞即時災情之變化，再者一般利用接觸式儀器進行災害資訊之測量，可能造成儀器的損毀或量測資訊干擾，同時造成施測人員安全等問題。本論文針對上述問題，設計一套以影像自動量測河川表面流速方法之智慧型監測系統，主要目的在於降低人力負荷、儀器折損限制與施測之風險，同時提高監測系統的安全性與可靠性。 在此，本論文的貢獻有以下三點：第一，本論文提出一套河川表面漂流物體影像特徵追蹤方法，包含針對影像增強及追蹤技術之應用，能有效克服惡劣監測環境光影變動干擾，同時進行流速量測；第二，經由表面漂流物體影像特徵追蹤、大尺度質點影像量測及改良式資訊整合機制，實現自動數據校正及資訊調變，能有效提取河川水文資訊與降低雜訊誤差之成效；第三，完成一套自動非接觸式河川水流表面流速量測的監測系統雛形，並於若干不同大小尺度、天候流況、人工與自然場域等條件之流速案例進行實驗與分析，同時藉由相關專業儀器驗證及專業人員確認，證實本系統能達到良好的效果及提升量測資訊之穩健性與可靠性。

Nowadays, we are deeply impacted by global warming and extremely climate change. With the tendency to enhance disaster occurred, the number of relevant monitoring system is used along with the growth of the field of hydrological disaster prevention. However, there are several limitations in the system of conventional monitors, including need to check disaster events by people, the instrument damaged and security of surveying staff. In this thesis, we design an intelligent video surveillance to reduce the loading of human resource and improve the security of measurement. There are three contributions of this thesis. First, this thesis proposes a feature-based tracking of drifting objects algorithm, including Retinex enhancement and feature-based tracking to detect drifting objects accurately. Second, this thesis proposes data reconciliation mechanism to improve the accuracy of river flow velocity measurement. Finally, this thesis completes the prototype of system in automatic river flow velocity measurement to reduce the loading of employees and risk of measurement.