本研究為探討利用MATLAB及視訊裝置進行即時目標物的辨識。機器視覺近年的研究有增加的趨勢，各式各樣的應用也都與機器視覺有關。因此在本文中，將使用MATLAB並透過視訊進行即時的目標物選取與辨視。MATLAB是一個高階程式語言，以矩陣運算聞名。數位影像在電腦中即是以矩陣來組成，因此MATLAB在影像處理上是相當理想的一種選擇。本研究主要將針對即時選定目標物發展適用的演算法，並且以實際結果驗證其演算法效率與成功率。 This paper investigates the application of machine vision to the real-time target recognition in visible spectrum using MATLAB as platform. Machine vision is applied to wider and wider areas in recent years. Therefore, more and more researches have been done to investigate the applications. In this thesis we use MATLAB as the platform, and aim to recognize any target using machine vision in real-time operations. MATLAB is a well-known software, expert in matrix operations. Since digital images are represented as arrays in computer, MATLAB should be a good choice to deal with this sort of data. This research aims to develop an algorithm that allows us to pick up a target at any time and the computer can analyze the features of the target in real-time. Then the target can be recognized with the obtained features. Experiments are designed to examine the feasibility and efficiency of our algorithm.
