本系列前期研究已開發一套可量測混凝土表面裂紋寬度的Android App,本論文接續前期研究,發展以手機執行此App進行裂縫寬度實測實驗之裝置與方法,以驗證此App之裂紋寬度量測準確度。研究結果顯示,此App裂縫寬度量測誤差多數落於-0.5 mm到+0.6 mm之間,但仍有部份量測誤差較大。 上述實測實驗裝置與方法,包括兩種測定手機至受測面空間距離的量測方法,分別是「Aruco Marker光學追蹤」方法與「雷射測距儀與不銹鋼鈑固定架」方法。研究結果顯示,「Aruco Marker光學追蹤」方法的量測誤差稍偏大;而後者「雷射測距儀與不銹鋼鈑固定架」方法的空間距離量測誤差僅為-5 mm至+3 mm,可做為測定手機與受測面距離的主要量測方法。 本研究並改良前版App之中的自動定位功能程式碼。此自動定位功能可自動搜尋定位出最靠近點擊位置的裂縫,再量測該處裂縫寬度。本研究改良該自動定位功能程式,使程式碼更簡潔、易維護,並改進自動定位搜尋的正確性。
An Android app that can measure widths of concrete-surface cracks has been developed in a previous study. Following the previous study, the study presented in this thesis develops experimental devices and methods to test the measurement accuracy of this app running on smartphones. The experimental results show that the width-measurement deviations of the app are mostly between -0.5 mm and +0.6 mm, whereas a small portion of the test results is relatively more devious. The experimental devices and methods developed in this study include two methods for measuring the distance between the smartphone and the measured surface, which are the "Aruco Marker Optical Tracking" method and the "Laser Rangers Attached to Fixing Steel Plate" method. The results show that the measurement accuracy of the "Aruco Marker Optical Tracking" method is relatively inadequate, and that the measurement deviations of the "Laser Rangers Attached to Fixing Steel Plate" method are between -5 mm and +3 mm and can serve the purpose well. This study also improves the program code for the automatic positioning function in the app. The automatic positioning function in the app can automatically locate the crack closest to the clicked point for the crack-width measurement. This study improves the program code to make it more concise and maintainable, and also improves its degree of correctness.