應用機器視覺於自動化檢測電表圓盤系統

本研究應用機器視覺量測電錶中轉盤的波擺，即針對所擷取的影像進行邊緣偵測(canny edge detect)運算來抽取目標的特徵(位置、厚度)，再根據預設的允許度輸出結果，實現自動識別功能。系統是用Visual C++2005之MFC程式寫成視窗化操控介面，並與改良的調整機台、制氣壓系統(敲擊缸、爪缸)、進馬達以及光纖感測器整合成為一套自動調整電錶圓盤的波擺度達到合格範圍的系統，來提高生產效率和生產的自動化程度。本論文將對「調整機台」、「檢測方法」及「視窗化操控軟體」之原理、製作、操作程序進行說明，並分析證明之效能。

關鍵字

機器視覺；邊緣偵測；電錶圓盤

並列摘要

This research applies machine vision to measure the disk wave pendulum of kilowatt meter. In another word, this research operate canny edge detection algorithm (position, thickness) to the grabbed images. Furthermore, according to the expected permission output result to achieve the automatic diagnosis function. The system is composed by MFC in Visual C++ 2005 to build a window-based control interface. Meanwhile integrated a system with an improved mechanical adjustment device, control pneumatic system (punched cylinder & gripping jaw), stepper motor and optical fiber sensor to accomplish an automatic system which controls the wave throw of the disk in a kilowatt meter in an acceptable rang in order to raise the manufacture efficiency and automatized level. This research will discuss the fundamental principle, manufacturing and the operate sequence of the improvement mechanical adjustment device, examination method and window-based operate software, and analyzed, proofed the efficiency of the potency.

並列關鍵字

machine vision ； canny edge detection ； disk of kilowatt meter

參考文獻

[8] Canny, J.F., “ A computational approachto edge detection”, IEEE Trans Pattern Analysis and Machine Intelligence, 8(6): 679-698, Nov 1986

[1] 葉隆吉“ 圓盤波罷自動調整機”,大同公司生產技術研究中心,民81.12

Google Scholar

[2] Jozˇe Derganc, Bosˇtjan Likar, Franjo Pernus “A machine vision system for measuring theeccentricity of bearings“ University of Ljubljana, Faculty of Electrical Engineering, accepted 2 September