銑刀特徵值之機器視覺量測

Abstract

微型銑刀為一種切削PCB電路板之重要工具，生產業者根據不同的客戶需求，生產符合該客戶需求規格之銑刀。目前品管作業是以人工進行量測，在工業科技日益精密的情況下，微型銑刀之直徑越來越小，也增加人工量測之困難度。 本論文運用電腦視覺方法，發展出一套自動光學量測系統，設計並運用影像處理方法擷取銑刀魚尾端面上之特徵點，基於這些特徵點計算出銑刀特徵值，實驗結果顯示，能快速而客觀有效的量測出微型銑刀成品魚尾端面上之特徵值，避免人工量測所造成之誤差。若量測結果偏離所設計之銑刀規格，使銑刀加工生產之作業員能即時調校加工機台，進而確保銑刀成品之品質。本論文所研發完成之機台，可針對銑刀魚尾端面特徵值包括右刃溝深、前周角與間隙進行精確且穩健的自動量測。

The microrouter is an important tool used to cut a printed circuit boards (PCB). The microrouter manufacture products a variety of microrouters to fulfill distinct customer’s requirements. The traditional way to control the quality of microrouter depends on human inspectors. As the diameter of the microrouter gets smaller, it is more difficult for human inspectors to examine the microrouter’s quality. In this thesis, we proposed an auto-optical measurement system by using machine vision method. We designed and applied digital image processing to capture the feature points on microrouter fishtail and then calculate the characteristics of the microrouter including right hook deep (RHD), the rake and the gap. Experimentation results revealed that the developed measurement system can objectively and effectively measures the characteristics of microrouter in different sizes and different types. In addition, the system can avoid measurement error caused by the human inspectors due to long-term working. The accurate and robust assistant remind producer to immediately fine-tune the processing machine if the results of the measurement deviate from the designed specifications.