臺灣博碩士論文加值系統

本研究為檢測化學纖維紡口板的各個紡嘴阻塞情況，並採用閃頻光源飛拍紡嘴做光學非接觸式檢測。待測物的紡口板為用於紡織業中，製作化學纖維時，將液態的高分子化學纖維原料經由紡口板擠出，抽拉成絲線後可製成化學纖維，此過程中原料會經過紡口板的紡嘴成型，紡嘴如果因為製作過程中的原料殘留使孔洞阻塞，則會影響化學纖維的品質。本研究採用機器視覺以自動化檢測取代人工檢測，減少人力成本與人工檢測時的不穩定性，並結合深度學習做辨識提升檢測系統的精確度。
本研究化學纖維紡口板的檢測系統利用XYZ三軸移動平台以龍門式的架構，將待測物的紡口板至於工作區中進行檢測，視覺系統的部分採用Basler ace工業相機至於Z軸上做檢測時的移動，並以GigE的方式連接至PC端傳輸影像，雙工業相機分別搭配低倍率(0.5 x)和高倍率(6.0 x)遠心鏡頭使用，分別用於低倍率做紡嘴的座標定位高倍率做各別紡嘴孔的拍攝，並以閃頻控制器控制光源和外部觸發工業相機，飛拍掃描紡嘴影像提升檢測速度，以連續拍攝方式完成一塊紡口板的紡嘴拍攝，將紡嘴影像利用深度學習模型進行檢測取得阻塞情況。本研究以光學非接觸式檢測紡口板上的紡嘴阻塞情況，紡口板上的紡嘴以直行排列共 96 個紡嘴孔，紡嘴為圓形孔洞半徑 85 μm，可檢測放置於檢測工作區之紡口板，紡嘴影像由語義分割的深度學習方式，標註出阻塞區域做模型的訓練，其模型可辨識出紡嘴內阻塞的區域，定義為瑕疵特徵提取出來，經由計算取得阻塞區域於紡嘴中的占比面積，判斷各個紡嘴的阻塞情況。

This study focuses on the optical inspection of nozzle blockages on a chemical fiber spinning plate, using the flash frequncy controller for optical non-contact inspection. During the production of chemical fibers, liquid cellulose are extruded through the chemical fiber spinnere to form fibers,by through the spinning nozzle on the spinnerets. Blockages in these nozzles due to residual materials can adversely affect fiber quality. This research utilizes machine vision to automate detection, replacing manual inspection to reduce labor costs and instability associated with manual methods. Deep learning is integrated to enhance the accuracy of the detection system.
The chemical fiber spinnerets inspection system uses XYZ three-axis moving stage, inspection object spinneret set in the work area. The vision system using Basler ace industrial cameras set on the Z-axis during inspection, with images transmitted to a PC via GigE. Two industrial cameras equipped with 0.5x and 6.0x telecentric lenses are utilized, the 0.5x telecentric lenses for nozzle coordinate positioning and the 6.0x telecentric lenses for capturing individual nozzle holes. The flash frequncy controller regulates the light source and triggers the industrial cameras for rapid image acquisition, facilitating continuous scanning of spinneret images to assess blockage conditions. This study focuses on optical non-contact inspection of spinneret nozzles,the inspection object spinnerets with 96 holes arranged linearly, each hole being circular with a radius of 85 μm. Images of the spinneret are processed using deep learning semantic segmentation to annotate blocked regions for model training. The trained model identifies blockages within the spinneret as defect features, calculating the percentage area occupied by blockages to assess the condition of each nozzle.

摘要.........i
Abstract.........ii
誌謝.........iv
目錄.........v
表目錄.........vii
圖目錄.........viii
符號說明.........xi
第一章 緒論.........1
1.1 前言.........1
1.2 研究目的.........2
1.3 國內外相關文獻.........3
1.4 論文架構.........21
第二章 系統介紹.........22
2.1 硬體架構.........22
2.2 系統架構.........23
2.3 設備元件架構及介紹.........24
2.3.1 運動控制軸卡.........25
2.3.2 AC伺服馬達驅動器.........25
2.3.3 AC伺服馬達.........27
2.3.4 PoE網路擴充卡.........28
2.3.5 Arduino UNO板.........29
2.3.6 PWM調光驅動器.........29
2.3.7 高角度環形光.........31
2.3.8 LED燈板.........31
2.3.9 灰階攝影機.........32
2.3.10 低倍率與高倍率遠心鏡頭.........33
2.3.11 閃頻控制器.........34
2.3.12 檢測平台配線.........37
2.3.13 閃頻控制器配線.........39
2.3.14 圖形處理器.........39
2.3.15 人機介面及影像處理軟體.........41
第三章 阻塞瑕疵檢測系統之建置.........42
3.1 移動平台校正.........42
3.2 機器視覺檢測架構.........46
3.2.1 軟體自動對焦.........46
3.2.2 中值濾波.........49
3.2.3 深度學習.........49
3.3 單位像素轉換.........57
3.4 鏡頭校正與阻塞分類.........57
第四章 實驗結果與討論.........59
4.1 程式介面及檢測操作.........59
4.1.1 人機介面(Human-Machine Interface, HMI).........59
4.1.2 XYZ三軸智慧自動檢測平台.........60
4.1.3 紡嘴檢測結果顯示介面.........62
4.2 深度學習模型的設定參數與訓練結果.........63
4.3 實驗待測物檢測流程與結果.........72
4.3.1 紡口板之紡嘴掃描及阻塞檢測.........73
4.4 結果討論.........79
4.5 不確定因素分析.........80
4.5.1 電控系統誤差.........80
4.5.2 視覺系統誤差.........80
第五章 結論與未來展望.........81
5.1 結論.........81
5.2 未來展望.........81
參考文獻.........82
Extended Abstract.........85

[1] 105年度人造纖維業之製造業原物料耗用通常水準調查報告https://www.ntbk.gov.tw/
[2] Fayao Liu , Guosheng Lin , Chunhua Shen , " CRF learning with CNN features for image segmentation " , Pattern Recognition Volume 48, Issue 10, October 2015, Pages 2983-2992
[3] Malay Kishore Dutta , Anushikha Singh , Sabari Ghosal , "A computer vision based technique for identification of acrylamide in potato chips", Computers and Electronics in Agriculture Volume 119, November 2015, Pages 40-50
[4] Kah Phooi Seng , Li-Minn Ang; Leigh M. Schmidtke , Suzy Y. Rogiers, "Computer Vision and Machine Learning for Viticulture Technology", IEEE Access ( Volume: 6) 22 October 2018, Pages: 67494 - 67510
[5] Zelin Zhang , Yang Liu , Qi Hu , Zhiwei Zhang , Lei Wang , Xiang Liu , Xuhui Xia , "Multi-information online detection of coal quality based on machine vision" Powder Technology Volume 374, September 2020, Pages 250-262
[6] Changsheng Li , Xianmin Zhang , Yanjiang Huang , Chuangang Tang , Sergej Fatikow , "A novel algorithm for defect extraction and classification of mobile phone screen based on machine vision" , Computers & Industrial EngineeringVolume 146, August 2020, 106530
[7] Wenbo Hu, Weidong Wang, Chengbo Ai , Jin Wang, Wenjuan Wang, Xuefei Meng, Jun Liu, Haowen Tao, Shi Qiu "Machine vision-based surface crack analysis for transportation infrastructure" Automation in Construction Volume 132, December 2021, 103973
[8] Núria Banús , Imma Boada , Anton Bardera , Pol Toldrà, "A Deep-Learning Based Solution to Automatically Control Closure and Seal of Pizza Packages", IEEE Access ( Volume: 9) 15 December 2021 , Page(s): 167267 - 167281
[9] Pablo Malvido Fresnillo , Saigopal Vasudevan , Wael M. Mohammed , Jose L. Martinez Lastra , Jose A. Perez Garcia, "An approach based on machine vision for the identification and shape estimation of deformable linear objects" Mechatronics Volume 96, December 2023, 103085
[10] Naveen Venkatesh S, Sugumaran V , "Machine vision based fault diagnosis of photovoltaic modules using lazy learning approach" , Measurement Volume 191, 15 March 2022, 110786
[11] Nadia Moradi, Saeideh Gorji Kandi, Hossein Yahyaei , "A new approach for detecting and grading blistering defect of coatings using a machine vision system" , Measurement Volume 203, 15 November 2022, 111954
[12] Ruichen LI, Lei MA, Qichao TANG, Yongkui SUN, Bin SU, "Quality Inspection for the Body-in-white Assembly of High-Speed Train Carriage Based on Machine Vision", 6th IFAC Symposium on Telematics Applications TA 2022, Pages 13-18
[13] Eldar Rahimov, Michael Watson, Andreas Hadjisoteriou, Matthew Marshall, "Investigation of wear mechanisms in AlSi-polyester abradable - Ti(6Al4V) blade contacts using stroboscopic imaging", Wear Volumes 494–495, 15 April 2022, 204207
[14] Marco Carratù , Vincenzo Gallo , Consolatina Liguori , Jan Lundgren , Mattias O’Nils , Antonio Pietrosanto, "Vision-Based System for Measuring the Diameter of Wood Logs", IEEE Open Journal of Instrumentation and Measurement ( Volume: 2) ,03 April 2023, 5000112
[15] Van Doi Truong , Jiaping Xia , YuHyeong Jeong , Jonghun Yoon , "An automatic machine vision-based algorithm for inspection of hardwood flooring defects during manufacturing" Engineering Applications of Artificial Intelligence Volume 123, Part A, August 2023, 106268
[16] Márton Tomin , Dániel Török , Tamás Pászthy , Ákos Kmetty, "Deformation analysis in impact testing of functionally graded foams by the image processing of high-speed camera recordings", Polymer Testing Volume 122, May 2023, 108014
[17] Lei Dong, Weifang Chen, Shuying Yang, Houyun Yu, "Automated detection of gear tooth flank surface integrity: A cascade detection approach using machine vision", Measurement Volume 220, October 2023, 113375
[18] Shoufeng Jin , Zixuan Yang , Grzegorz Królczykg , Xinying Liu , Paolo Gardoni , Zhixiong Li, "Garbage detection and classification using a new deep learning-based machine vision system as a tool for sustainable waste recycling" Waste Management Volume 162, 1 May 2023, Pages 123-130
[19] Hao Li , Ruoyu Zhang , Wanhuai Zhou , Xiang Liu , Kai Wang , Mengyun Zhang , Qingxu Li, "A novel method for seed cotton color measurement based on machine vision technology", Computers and Electronics in Agriculture Volume 215, December 2023, 108381
[20] Omeed Mirbod , Daeun Choi , Paul H. Heinemann , Richard P. Marini , Long He, "On-tree apple fruit size estimation using stereo vision with deep learning-based occlusion handling" Biosystems Engineering Volume 226, February 2023, Pages 27-42
[21] Boxuan Gao, Hong Zhao, Xingyuan Miao, "A novel multi-model cascade framework for pipeline defects detection based on machine vision ", Measurement Volume 220, October 2023, 113374
[22] Yi Chen , Jun Bin , Chao Kang , "Application of machine vision and convolutional neural networks in discriminating tobacco leaf maturity on mobile devices", Smart Agricultural Technology Volume 5, October 2023, 100322
[23] Swarit Anand Singh, Aitha Sudheer Kumar, K.A. Desai, "Comparative assessment of common pre-trained CNNs for vision-based surface defect detection of machined components", Expert Systems with ApplicationsVolume 218, 15 May 2023, 119623
[24] Yang Zhang , Yi-Qing Ni , Xinyu Jia , You-Wu Wang , "Identification of concrete surface damage based on probabilistic deep learning of images", Automation in Construction Volume 156, December 2023, 105141
[25] Liu Xiaohang , Zhang Zhao , C. Igathinathane , Flores Paulo , Zhang Man , Li Han , Han Xiongzhe , Ha Tuan , Ampatzidis Yiannis , Kim Hak-Jin, "Infield corn kernel detection using image processing, machine learning, and deep learning methodologies under natural lighting", Expert Systems with ApplicationsVolume 238, Part E, 15 March 2024, 122278
[26] Fei Zhou , Weiyue Xie , Ming Lin b, Longfei Ye , Chu Zhang , Zhangfeng Zhao , Fei Liu , Jiyu Peng , Wenwen Kong , "Rapid authentication of geographical origins of Baishao (Radix Paeoniae Alba) slices with laser-induced breakdown spectroscopy based on conventional machine learning and deep learning", Spectrochimica Acta Part B: Atomic Spectroscopy Volume 212, February 2024, 106852
[27] Rasoul Ameri , Chung-Chian Hsu , Shahab S. Band , "A systematic review of deep learning approaches for surface defect detection in industrial applications" , Engineering Applications of Artificial Intelligence Volume 130, April 2024, 107717
[28] Evgin Goceri, "Vision transformer based classification of gliomas from histopathological images", Expert Systems with Applications Volume 241, 1 May 2024, 122672
[29] 林伯軒，應用深度學習與機器視覺於化學纖維紡嘴智慧自動檢測系統開發，碩士論文，國立虎尾科技大學自動化工程所碩士班，台灣，2020。
[30] 陳冠霖，整合水平關節機器人與深度學習於化學纖維紡嘴智慧自動檢測系統開發，碩士論文，國立虎尾科技大學自動化工程所碩士班，台灣，2022。
[31] Delta, http://www.delta.com.tw
[32]ADLINK Technology, https://www.adlinktech.com
[33]Arduino, https://www.arduino.cc
[34]MEAN WELL, https://www.meanwell.com
[35]Chen Hsuan, https://www.chtco.com.tw
[36]Basler, https://www.baslerweb.com
[37]Myutron, https://www.myutron.com
[38]JM Vistec Sysem Pte Ltd, https://www.jm-vistec.com/
[39]ASUS, https://www.asus.com/tw
[40]Keysight, https://www.keysight.com
[41]鍾國亮，影像處理與電腦視覺第六版，東華書局，台灣
[42]IThelp, https://ithelp.ithome.com.tw/
[43] EETESMEDN, https://mbb.eet-china.com/