隨著精密加工領域的行業日漸普及，在銑削工件時也需要注意刀具的磨耗狀態。而許多公司較不願意購買昂貴的高精密檢測儀器去檢測刀具的使用狀況，往往都是憑著經驗去推估當下的刀具還能使用多久。而用估計的方式是不準確的，若磨耗過多的刀具銑削出的成品尺寸誤差會很大，影響產品良率也浪費時間成本，也因此如何檢測並正確判別刀具磨耗程度是很重要的課題。
本研究的目的是開發一個低成本的刀具磨耗檢測裝置，取代使用較昂貴的高精密儀器。本研究所使用的刀具是半徑0.75 mm的微球頭銑刀進行銑削，並對不同銑削路徑長度的刀具做紀錄。此檢測裝置透過機器視覺與影像處理的方式去檢測刀具磨耗，視覺系統主要以樹莓派為核心，透過樹莓派內建Thonny Python軟體進行程式撰寫達到拍照與影像處理目的。其中影像處理使用了歸一化處理、旋轉影像、手動ROI提取與Otsu二值化法將計算磨耗區的像素面積。結果可發現磨耗最低的刀具(銑削路徑長度: 47.5 mm)的像素面積為7 pixels，磨耗最多的刀具(銑削路徑長度: 570 mm) 的像素面積為73 pixels。另外透過肉眼觀察並計算未經影像處理的灰階影像磨耗區以驗證此檢測結果，得到磨耗最低的刀具像素面積為9 pixels，磨耗最多的刀具像素面積為79 pixels。由此結果可得知此低成本的刀具磨耗檢測裝置可以檢測到微球頭銑刀的磨耗面積變化，提供使用者了解磨耗資訊。

With the increasing popularity of the precision manufacturing industry, it is necessary to pay attention to the wear of the tool when milling the workpiece. However, many companies are less willing to buy expensive instruments to detect the status of tools. It’s often based on experience to estimate how long the current tool can be used. However, the estimate is usually not accurate. If the tool with too much wear is used, the size of the finished workpiece will be different from the designed value, which will make the completed workpiece useless. Therefore, how to determine the status of tool wear is very important.
The objective of this research was to develop a low-cost tool wear detection device. The tool used in this research was a micro ball-end milling cutter with a radius of 0.75 mm. This detection device uses machine vision and image processing to detect tool wear. The vision system is based on the Raspberry Pi, through the built-in Thonny Python software for programming to take pictures and perform image processing. The image processing uses normalization, manual ROI definition, and the Otsu method to calculate the area of the tool wear. As a result, it can be found that the area of the tool wear with the lowest wear (milling path length 47.5 mm) was 7 pixels and that with the most wear (milling path length 570 mm) was 73 pixels. In addition, the gray-scale image wear area without image processing was observed with human vision and calculated to verify the capability of the proposed vision system. The observation results show that the area of the tool wear with the lowest wear was 9 pixels and that with the most wear was 79 pixels. By comparing both results, it can be understood that the proposed low-cost tool wear detection device can detect the wear area of the micro-ball end milling cutter and provide users with reliable information about wear.

[1] M. C. Gomes, L. C. Brito, M. Bacci da Silva, and M. A. Viana Duarte, "Tool wear monitoring in micromilling using Support Vector Machine with vibration and sound sensors," Precision Engineering, Article vol. 67, pp. 137-151, 2021.
[2] K. M. Fong, X. Wang, S. Kamaruddin, and M. Z. Ismadi, "Investigation on universal tool wear measurement technique using image-based cross-correlation analysis," Measurement: Journal of the International Measurement Confederation, Article vol. 169, 2021, Art. no. 108489.
[3] W. Weis, "Tool wear measurement on basis of optical sensors, vision systems and neuronal networks (application milling)," in Proceedings of WESCON 1993 Conference Record, 1993, pp. 134-138.
[4] Y. Dai and K. Zhu, "A machine vision system for micro-milling tool condition monitoring," Precision Engineering, Article vol. 52, pp. 183-191, 2018.
[5] M. Szydłowski, B. Powałka, M. Matuszak, and P. Kochmański, "Machine vision micro-milling tool wear inspection by image reconstruction and light reflectance," Precision Engineering, vol. 44, pp. 236-244, 2016.
[6] L. Yu-Teng and C. Yih-Chih, "Vision-based automatic tool wear monitoring system," in 2008 7th World Congress on Intelligent Control and Automation, 2008, pp. 6031-6035.
[7] R. G. Lins, B. Guerreiro, P. R. M. d. Araujo, and R. Schmitt, "In-Process Tool Wear Measurement System Based on Image Analysis for CNC Drilling Machines," IEEE Transactions on Instrumentation and Measurement, vol. 69, no. 8, pp. 5579-5588, 2020.
[8] J. Zhang, C. Zhang, S. Guo, and L. Zhou, "Research on tool wear detection based on machine vision in end milling process," Production Engineering, Article vol. 6, no. 4-5, pp. 431-437, 2012.
[9] 李明哲, "影像處理輔助微銑刀以及鑽石刀具之刀具磨耗檢測," 碩士, 機械工程系, 國立臺灣科技大學, 台北市, 2011.
[10] J. Zhou and J. Yu, "Chisel edge wear measurement of high-speed steel twist drills based on machine vision," Computers in Industry, Article vol. 128, 2021, Art. no. 103436.
[11] W. J. Zhang, D. Li, F. Ye, and H. Sun, "Automatic Optical Defect Inspection and Dimension Measurement of Drill Bit," in 2006 International Conference on Mechatronics and Automation, 2006, pp. 95-100.
[12] J. Loizou, W. Tian, J. Robertson, and J. Camelio, "Automated wear characterization for broaching tools based on machine vision systems," Journal of Manufacturing Systems, Article vol. 37, pp. 558-563, 2015.
[13] A. Qin, L. Guo, Z. You, H. Gao, X. Wu, and S. Xiang, "Research on automatic monitoring method of face milling cutter wear based on dynamic image sequence," International Journal of Advanced Manufacturing Technology, Article vol. 110, no. 11-12, pp. 3365-3376, 2020.
[14] M. Sortino, "Application of statistical filtering for optical detection of tool wear," International Journal of Machine Tools and Manufacture, Article vol. 43, no. 5, pp. 493-497, 2003.
[15] G. Xiong, J. Liu, and A. Avila, "Cutting tool wear measurement by using active contour model based image processing," in 2011 IEEE International Conference on Mechatronics and Automation, 2011, pp. 670-675.
[16] A. Bieniek and A. Moga, "An efficient watershed algorithm based on connected components," Pattern Recognition, Article vol. 33, no. 6, pp. 907-916, 2000.
[17] F. Bai, M. Fan, H. Yang, and L. Dong, "Image segmentation method for coal particle size distribution analysis," Particuology, Article vol. 56, pp. 163-170, 2021.
[18] K. Zhu and X. Yu, "The monitoring of micro milling tool wear conditions by wear area estimation," Mechanical Systems and Signal Processing, Article vol. 93, pp. 80-91, 2017.
[19] L. Li and Q. An, "An in-depth study of tool wear monitoring technique based on image segmentation and texture analysis," Measurement: Journal of the International Measurement Confederation, Article vol. 79, pp. 44-52, 2016.
[20] (2011). [OpenCV] 轉換影像為灰階. Available: https://cg2010studio.com/2011/06/06/opencv-%E8%BD%89%E6%8F%9B%E5%BD%B1%E5%83%8F%E7%82%BA%E7%81%B0%E9%9A%8E-transform-image-to-gray-level/
[21] (2020). [OpenCV] Otsu's Threshold 大津二值化. Available: https://www.wongwonggoods.com/python/python_opencv/opencv-threshold-all-otsu/