應用影像疊圖技術之刀具磨耗檢測

Application of Photo-overlapping Technique on Tool Wear Detection

蔡孟勳 ； 林俊佑 ； 連震杰 ； 麥朝創

《機械工業雜誌》 432期 (2019/03) Pp. 54-58

摘要

現今工業都朝向自動化與智能化發展，各家工具機廠商亦致力於提升產品的品質及自動化生產的目標，刀具磨耗的情況常常會對產品的品質、自動化生產的效率產生重大影響。本研究以機械視覺為基礎開發線上刀具磨耗檢測方法。由於端銑刀的複數刀刃與螺旋形刀具形狀不同於一般車刀，加上金屬的反光的效應，增加的端銑刀在影像拍攝困難，本研究利用刀具影像疊圖技術將立體的端銑刀展開成平面圖面，並使用影像處理的方法對展開後的刀具圖片進行磨耗檢測。實際以工具機機台進行切削加工，定時進行刀具影像疊圖並檢測刀具磨耗狀態。機械視覺系統被驗證可實際應用於刀具磨耗檢測。

關鍵字

刀具磨耗 ； 刀具狀態監控 ； 機械視覺 ； 數位影像處理

並列摘要

In today's time, industries are moving toward intensive automated intelligent systems. Various machine tool manufacturers have devoted to improve quality of products to reach the goal of automated production. Tool wear has significant impact on quality of products during automated production. This research is focused on development of an online tool wear inspection method based on machine vision technology. Since the number of blades and the geometric shape of the end mills are different from turning tool, affected by add-in metallic surface reflection effects, the image of end mills are more difficult to capture. This study developed a method using tool image overlap stitch on to expand the three-dimensional end mill into a plane view. Using the image processing method, we can perform wear detection on the overlapping stitched tool image. A practical system with regular tool wear monitoring functions was developed and deployed for an actual milling process; the feasibility of the automatic online inspection system of end mill has been verified.

並列關鍵字

Tool wear ； Tool condition monitoring ； Machine vision ； Digital image processing