台灣的工具機產業逐年進步,製造出更高精密的工具機為重要目標,而在精密機械技術中,鏟花(Scraping)是一種重要的手工加工技術,在機械的組裝過程中,鏟花可以對機械的零組件進行微量的鏟削,用來改善組合件的密合情形或滑動件的平面度及潤滑情形。而目前鏟花加工及油溝承斑的合格與否往往是由鏟花師傅獨自進行,不僅耗時費力,也可能會導致鏟花工件的品質標準不一致的情形。 本研究嘗試開發一自動鏟花系統,希望能減輕鏟花師傅的負擔。此一系統包含相機鏡頭、壓力感測器及能夠進行三軸運動的機構。利用相機鏡頭擷取鏟花表面紋路並撰寫Matlab程式進行影像處理與分析,藉此檢測鏟花的品質和分析待鏟位置,再使用所設計之鏟花機構與已知之待鏟位置對鏟花工件進行加工。加工過程中,利用壓力感測器將自動檢測與鏟花加工結合形成一自動鏟花系統。
Scraping is a fundamental technique for the production of precision machine tools. The accuracy of machined components is lower than the accuracy of machine used to make the components. The scraping techniques were used to improve component surface flatness or to develop proper oil pockets for slider and thus the machine tool accuracy and performance can be improved. However, scraping is a time consuming process. The purpose of this study is to develop an auto-scraping system which can be used to identify the tinted area and scrape the tinted area. Basically, the developed system includes a machine vision system and a mechanism with tri-axial movement for scraping. With this system, the image of the work surface captured by the vision system was processed and analyzed to identify areas by specific color space. The scraping mechanism is then directed to the desired area for scraping automatically.