隨著工業發展，扣件產業所製造的產品種類不僅日漸複雜，尺寸依照銷售國家有不同的規範標準。扣件零售雖然在台灣賣場或零售店多以小批量貼上條碼販售，若在歐美大商場則都是以單件零售，結帳時會由販售員現場以專用尺規量測再查詢扣件的型號與價格，為了一支小小扣件的販售，手續複雜結帳效率差外，人為查詢常會販售價格錯誤造成庫存記錄問題。台灣外銷至歐美的製造商想服務銷售端改善這樣的困擾，提供額外的服務，故而起心動念有開發行動式扣件自動辨識系統的想法。
本研究所開發的行動式扣件辨識雛型系統，設計光箱透過兩台手機相機擷取扣件影像，提供穩定的拍攝環境。以USB連接手機與電腦，傳輸影像至電腦的扣件辨識軟體處理。
所開發的扣件辨識軟體先以蔡氏相機校正取得座標轉換關係，再以影像處理技術分離出扣件輪廓，透過演算法辨識扣件幾個特徵尺寸，推導幾何修正公式解決量測點偏移問題，計得更精準的量測數據，同時建置扣件資料庫與查詢功能，以數據自動辨識扣件型號。以市面販售包含公制、英制、美規標準規格的100支扣件樣本進行辨識實驗，重複進行三次，得到平均辨識成功率約85.33%。透過二元分類指標分析辨識系統在判定是否為資料庫內扣件的誤判機率。扣件辨識耗時實驗，測得從扣件置入光學辨識箱、辨識、取出平均需時約35.25秒，其中辨識耗時3.96秒，置入取出各7秒，拍攝耗時17.29秒，拍攝需時佔去一半時間，原因是手機與電腦影像傳輸與溝通方法過慢造成，未來改用工業相機這問題可解決。

The retail stores of fasteners in Taiwan are mostly sold in batches with barcodes, but in a huge western retail stores, fasteners are sold in single pieces. The sales staff will use special ruler to measure characteristic dimensions of fastener in order to find out the product model and the price of each piece. These procedures are too complicated and inefficient. Selling with wrong price and inventory recording mistake often occurs. In order to solve these problems and provide additional services for retail stores, we have an idea of developing a portable fasteners recognition system.

The portable fasteners recognition system includes hardware and software. We use two cameras of two mobile phone to picture images of fasteners, and design a light box to provide a stable environment. The fasteners recognition software is implemented with Tsai’s camera calibration method, image processing and some method developed by us to recognize several characteristic dimensions of the fastener to measure. The software can automatically find the model name of fastener by the result of measurement data and the built fastener database.

The recognition experiment was carried out three times with 100 fastener samples and the recognition accuracy was about 85.33%. Analyze the probability of mistakes by the system in determining whether a fastener in the database through binary classification indicators, and conduct a time-consuming experiment on fastener recognition. The average recognition time from putting a fastener in the light box to taking out is measured to be 35.25s.

[1]雪花新聞, "如何識別螺釘和螺栓，看了就明白！", 2018, <https://www.xuehua.us/a/5eb9bf6586ec4d48b7bd2165?lang=zh-tw>, accessed on 2021/2/15.
[2]軸承小白, "今天頭條", 2019, <https://twgreatdaily.com/uIMVV24BMH2_cNUgydPP.html>, accessed on 2021/2/15.
[3]Wikipedia, "ISO metric screw thread", <https://en.wikipedia.org/wiki/ISO_metric_screw_thread>, accessed on 2021/2/15.
[4]Wikipedia, "British Standard Whitworth", <https://en.wikipedia.org/wiki/British_Standard_Whitworth>, accessed on 2021/2/15.
[5]Oberg, E. and Jones, F. D., Machinery's handbook, pp. 1813-1864 Industrial press, 1916.
[6]Rybczynski, W., One good turn: a natural history of the screwdriver and the screw, Simon and Schuster, 2013.
[7]Skrabec Jr, Q. R., The metallurgic age: the victorian flowering of invention and industrial science, McFarland, 2015.
[8]伍全工業購, "Tooth Gauge", 2021, <http://www.indus-go.com/img/prod/5057f9f8-9c0f-4ae7-b9bc-cd90142eff54.gif>, accessed on 2021/2/15.
[9]Amazon, "Gauge", 2020, <https://images-na.ssl-images-amazon.com/images/I/71qYnLGx%2BxL._AC_SL1500_.jpg>, accessed on 2021/2/15.
[10] Zawada-Tomkiewicz, A. and Wierucka, I., "A case study in technological quality assurance of a metric screw thread", Measurement, Vol.114, pp.208-217, 2018.
[11] Liu, C.-C. and Chen, W.-Y., "Screw pitch precision measurement using simple linear regression and image analysis", Applied mathematics and computation, Vol.178, No.2, pp.390-404, 2006.
[12] Kosarevsky, S. and Latypov, V., "Development of an algorithm to detect screw threads in planar point clouds", Measurement Science Review, Vol.10, No.4, pp.136-141, 2010.
[13] Amin, S. S. and Raja, J., "Screw Thread Metrology", The University of North Carolina in Charlotte, USA, 2004.
[14] Gadelmawla, E., "A novel system for automatic measurement and inspection of parallel screw threads", Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Vol.218, No.5, pp.545-556, 2004.
[15] Tong, Q., Ding, Z., Chen, J., Ai, L., and Yuan, F., "The research of screw thread parameter measurement based on position sensitive detector and laser", in Journal of Physics: Conference Series, p.105, 2006.
[16] Hunsicker, R. J., Patten, J., Ledford, A., Ferman, C., Allen, M., and Ellis, C., "Automatic vision inspection and measurement system for external screw threads", Journal of Manufacturing Systems, Vol.13, No.5, pp.370-384, 1994.
[17] Chen, M., "Compensation of thread profile distortion in image measuring screw thread", Measurement, Vol.129, pp.582-588, 2018.
[18] Gadelmawla, E., "Computer vision algorithms for measurement and inspection of external screw threads", Measurement, Vol.100, pp.36-49, 2017.
[19] Min, J., "Comprehensive measurement system for screw thread parameter based on machine vision", in 2015 International Conference on Computer and Computational Sciences (ICCCS), pp.266-270, 2015.
[20] Makeyev, O., Sazonov, E., Moklyachuk, M., and Lopez-Meyer, P., "Hybrid evolutionary algorithm for microscrew thread parameter estimation", Engineering Applications of Artificial Intelligence, Vol.23, No.4, pp.446-452, 2010.
[21] Keyence, "Image Measurement System", 2021, <https://www.keyence.com.tw/products/measure-sys/image-measure/>, accessed on 2021/2/15.
[22] Sephus, N., Bhagavatula, S., Shastri, P., and Gabriel, E., "An Industrial-Strength Pipeline For Recognizing Fasteners", in 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA), pp.781-786, 2015.
[23] Kamal, N. D. M., "Fastener recognition using combination of colour features with shape-based features and Zernike moments", in 2016 IEEE Student Conference on Research and Development (SCOReD), pp.1-5, 2016.
[24] Thompson, D., Automated fastener inspection system. 1989, U.S. Patent No. 4,823,396. Patent Term 21 Jul, 1987 to 21 Jul 2007
[25] Ukida, H., "Visual defect inspection of rotating screw heads", in SICE Annual Conference 2007, pp.1478-1483, 2007.
[26] Yunxin, G., "Research and development of automatic screw-parameters tester", in Proceedings of 2011 International Conference on Electronics and Optoelectronics, pp.V1-320-V1-323, 2011.
[27] Chen, S.-H. and Perng, D.-B., "Automatic optical inspection system for IC molding surface", Journal of Intelligent Manufacturing, Vol.27, No.5, pp.915-926, 2016.
[28] Rau, H. and Wu, C.-H., "Automatic optical inspection for detecting defects on printed circuit board inner layers", The International Journal of Advanced Manufacturing Technology, Vol.25, No.9-10, pp.940-946, 2005.
[29] Tsai, R., "A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses", IEEE Journal on Robotics and Automation, Vol.3, No.4, pp.323-344, 1987.
[30] Zhang, Z., "A flexible new technique for camera calibration", IEEE Transactions on pattern analysis and machine intelligence, Vol.22, No.11, pp.1330-1334, 2000.
[31] Pertuz, S., Puig, D., and Garcia, M. A., "Analysis of focus measure operators for shape-from-focus", Pattern Recognition, Vol.46, No.5, pp.1415-1432, 2013.
[32] Pech-Pacheco, J. L., Cristóbal, G., Chamorro-Martinez, J., and Fernández-Valdivia, J., "Diatom autofocusing in brightfield microscopy: a comparative study", in Proceedings 15th International Conference on Pattern Recognition. ICPR-2000, pp.314-317, 2000.
[33] Weng, J., Cohen, P., and Herniou, M., "Camera calibration with distortion models and accuracy evaluation", IEEE Transactions on pattern analysis and machine intelligence, Vol.14, No.10, pp.965-980, 1992.
[34] Usamentiaga, R., Garcia, D., Ibarra-Castanedo, C., and Maldague, X., "Highly accurate geometric calibration for infrared cameras using inexpensive calibration targets", Measurement, Vol.112, pp.105-116, 2017.
[35] MISUMI, "Web產品型錄", 2021, <https://tw.misumi-ec.com/>, accessed on 2021/2/15.