應用影像處理的2.5 軸微操縱系統之研究

A Study on the 2.5 DOF Micro-manipulatorBaed on Image Processing

廖振益(Chen-Yi Liao)

指導教授 ： 葉隆吉

大同大學/工程學院/機械工程學系所/碩士(2004年)

摘要

本研究透過CCD的影像擷取、VB程式的影像處理及軌跡規劃與雙軸微步進馬達平台，建構出一套’’應用影像處理的2.5軸微操縱系統’’。研究中，CCD攝影機擷取到的影像資料，由自行撰寫的影像處理程式，經灰階化、取閥值、二值化、細線化等一連串的處理之後，將影像轉換為一骨架形式的運動軌跡。系統程式再將所得到的運動軌跡，轉換為刻痕器的運動控制，如此能將影像轉刻到工件上。
刻痕實驗的結果顯示，可以將影像的軌跡實現在刻痕上，但是因為刻痕寬度為0.1mm（鑽石筆尖），在刻痕與刻痕之間的距離小於0.1mm時，只能從CCD擷取的影像觀察到兩道刻痕重疊在一起。因此必須尋找刻痕寬度更小的刻痕器，方能使影像與刻痕結果更加相似。

關鍵字

細線化 ； 二值化 ； 影像處理

並列摘要

In this thesis, a 2.5 DOF micromanipulator based on image processing was proposed. The system includes an image acquisition and processing system, a two axles platform driven by micro stepper motors and a path planning method which is developed in this study. The image processing was carried out by using VB program, which is created by authors. A series of image processing was carried out step by step. There are gray-level transformation, threshold decision method, binarization and line-thinning process. Finally, the acquitted image was transferred to a set of motion paths. According to the result of image processing, the operating program of system drives the maker to trace the motion paths and cut mark on the workpiece.

Experimental results shown that the system can map an image to a cutting mark on the surface of workpiece. Due to the width of cutting marker is 0.1 mm, the marks will overlap if the distance between paths smaller than 0.1 mm. It is necessary to find out a marker can make mark thinner than 3 μm, so that the created marks can more similar to original image.

並列關鍵字

binarization ； image processing ； thinning

參考文獻

[1] N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Trans. Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 377-393, 1979.

Manufacturing

[2]、連國珍,2003,8,數位影像處理，儒林圖書公司

[3]、江東毅,2002,5,由影像輸入之機械臂書法系統

[4]、郭盈顯,2002,02,Visual Basic與電腦I/O控制實務