本論文的主要目的是建立一套泳姿自動分類系統，利用電腦視覺中影像處理技術，輔以資料分析技術，即時且自動的分析在一段從岸邊水上拍攝，包含單一泳者向著鏡頭方向游泳的彩色影片中，泳者所進行的姿勢。首先我們利用膚色模型分離出影像中屬於游泳者手臂的區域。接下來，我們分析這些區域的特徵，例如長寬比以及斜率、面積等。最後，基於連續影像中每個單一畫面的手臂區域特徵，利用一個評分系統來判斷該泳者目前所進行的是哪一種游泳姿勢，分為蝶式、仰式、蛙式、自由式等四種泳姿。在實驗結果中，我們所提出的方法對於從http://swim.ee網站上下載的50段影片中，四種泳姿的分類達到了百分之百的正確性。 / In this thesis, we present a robust method to classify swimming styles from live color video sequences based on the features extracted from the upper-body of the swimmer. In our approach, potential body parts are first extracted using a simple skin color model. Next, the dominant components in the segmented regions are selected according to quantitative measures such as the aspect ratio and the area. Regression analysis is then performed to calculate the relative position of the constituent body parts. Finally, a scoring system is constructed to carry out the classification of four strokes, including butterfly, backstroke, breast stroke, and freestyle. Experimental results demonstrate the validity and efficiency of our proposed approach.