透水混凝土因為堅固且好修補的特性而被廣泛的用於道路鋪面的材料上，而孔隙率是用來預測混凝土性能的一項重要指標，雖然傳統量測混凝土孔隙率的方法很準確，但是所花的時間卻很冗長，不過有了X光電腦斷層掃描技術後，量測混凝土孔隙率的時間將大為縮短。在這個研究裡，我們掃描了混凝土柱以量測它們的孔隙率，掃描過後會產出一系列的圖片檔案，如何處理掃描後的圖片檔案也是自動化很重要的一環。本研究的目的是建立一個自動的模式，以便於輔助傳統量測方法偵測混凝土內部結構，幫助鋪面管理更加的迅速有效率。 本研究使用了Somatom Emotion電腦斷層掃描於孔隙混凝土，在取得掃描圖片檔後，影像處理技術有被使用於取得圖片檔的混凝土屬性資訊，影像處理的結果有著高準確率且整個過程都非常的有效率，本研究產出的模式將輔助傳統量測孔隙率方法，提供圖形化的圖片給決策者能對未來問題做出更正確的決策。影像化的資訊將提供更清楚的孔隙混凝土內部結構資訊以研究混凝土柱屬性之間的關聯性。 ;Permeable Concrete has been widely used as the main road surface material due to its sturdiness and ability to be repaired quickly. The porosity of concrete has been used for the prediction of the properties of concrete. Traditional method of measuring the porosity of concrete is accurate, however it takes a lot of time to measure; but with X-ray computerized tomography (CT scan), measuring porosity would become much faster. In this study, we used CT to scan a concrete cylinder for measuring the its porosity. After the scanning, the results are a series of pictures; therefore, how we process the series of pictures is also an important part to automation. The study aims to create an automatic way to detect the internal structure of the pavement material instead of using the traditional way, and at the same time helping the pavement management to be quicker and more efficient. In this research, we used Somatom Emotion CT scanner to scan the pervious concretes. After obtaining the images from CT scanning, the image processing method is applied in order to get the properties of the concrete through the images from CT scan. The final results after the image processing method are pretty accurate, and the whole process is very efficient. The model created in the study will assist traditional method ways for examining porosity by providing visualized images for decision makers to make correct decisions for future problems. Also, the visualized images will provide a better understanding for the inner structure of the pervious concrete to study the correlations between the properties of the concrete cylinders.
