臺灣博碩士論文加值系統

本研究檢測的陶瓷基板是一種脆弱的待測物，而本系統為非接觸式量測，可避免物件的表面磨損或破損的情況發生，若以人工方式檢測容易產生人為造成的損傷，嚴重則會毀損陶瓷基板，反而增加陶瓷基板的瑕疵。
利用自動光學檢測的應用逐漸普及，本研究以面掃描相機為架構，在高精度的取像下，由於解析度高而可視範圍縮小，整體的檢測速度被限制。因此本研究將針對面掃描相機應用在陶瓷基板檢測中，克服相機取像時間過長的問題，開發一套應用外部觸發面掃描相機與閃頻器控制光源於陶瓷基板表面瑕疵的檢測系統。
此陶瓷基板表面瑕疵檢測系統是以XYZ三軸移動平台，結合工業用高解析度面掃描相機與遠心鏡頭，相機解析度為3.45 μm，視野範圍(Field of View, FOV)為14.1 mm ×10.3 mm，再搭配0.5倍遠心鏡頭後，視野範圍變為28.2 mm × 20.6 mm。利用閃頻控制器控制高亮度低角度環形LED光源拍攝大小為128 mm × 128 mm與180 mm × 139 mm的陶瓷基板。透過影像處理軟體Halcon對拍攝的影像做處理，再利用灰階演算法將拍攝的影像做拼接，以利於陶瓷基板影像方便觀察缺陷位置。相機接收來自三軸控制平台的訊號並且與光源做延遲100 ms的觸發，本系統為硬體觸發，在三軸平台移動時，相機與閃頻控制器接收到來自平台的觸發訊號，使相機做拍攝取像並且閃頻控制器控制LED(light-emitting diode)光源在定點做光源的開關。
本研究以光學式非接觸量測的方式，透過拼接演算檢測大尺寸待測物，利用閃頻控制器在觸發的瞬間突破光源的額定電流，使光源輸出至極限以補強相機的曝光時間，使用多執行續的方式，將影像拍攝、影像處理以及影像拼接三者同時進行，大幅減少整體的檢測時間，5吋陶瓷基板檢測結果誤判率為2.3 %，5 × 7吋陶瓷基板檢測結果誤判率為2.1 %。

The ceramic substrate tested in this study is a fragile test object. This system is a non-contact measurement, which can avoid wear or damage to the surface of the object. If it is detected manually, it will be easy to cause personal factor damage, and in severe cases, the ceramic substrate will be damaged, and the defects of the ceramic substrate will be increased instead.
The application of automatic optical inspection is becoming more and more popular, in this study, the area scan camera is used as the framework. Under the high-precision image acquisition, the overall detection speed is limited due to the high resolution and the narrowing of the visible range.
Therefore, this research will focus on the application of the area scan camera in the inspection of ceramic substrates, overcome the excessively long time in the camera to take images, and develop a detection system that uses an externally triggered area scan camera and a flash frequency controller control the light source on the surface of ceramic substrates.
The ceramic substrate surface inspection system is based on a three-axis moving platform, combined with industrial high-resolution area scan camera and telecentric lens. The resolution of the camera is 3.45 μm, and field of view (FOV) is 14.1 mm × 10.3 mm. When the camera integrates with 0.5x telecentric lens, the field of view will become 28.2 mm × 20.6 mm. Using the flash controller to control the high-brightness and low-angle ring LED(light-emitting diode) light source to shoot ceramic substrates with sizes of 128 mm × 128 mm and 180 mm × 139 mm, the captured images are processed through the image processing software Halcon. Then, the gray-scale algorithm is used to stitch the captured images, so as to facilitate the observation of the defect position in the image of the ceramic substrate. The camera receives the signal from the three-axis control platform and triggers the light source with a delay of 100 ms. This system is triggered by hardware. When the three-axis platform moves, the camera and the flash frequency controller receive the trigger signal from the platform to make the camera take pictures and the flash frequency controller controls the LED light source to switch on and off the light source at a fixed point.
Inspect large-size objects through image stitching by non-contact measurement. Use the flash controller to break through the rated current of the light source at the moment of triggering, the light source out of limit to compensate the exposure time of the camera. the image capturing, image processing and image stitching are carried out at the same time by using the multi-execution continuation method,which reduce experiment time, the misjudgment rate of the 5-inch ceramic substrate inspecting results is 2.3%, and the f misjudgment of the 5 × 7-inch ceramic substrate inspecting results is 2.1%.

摘要......i
Abstract......ii
誌謝......iv
目錄......v
表目錄......vii
圖目錄......viii
第一章 緒論......1
1.1 前言......1
1.2 研究目的......1
1.3 國內外相關文獻......1
1.4 論文架構......17
第二章 系統介紹......18
2.1 系統架構......18
2.2 表面瑕疵檢測機台架構......19
2.3 設備元件介紹......20
2.3.1 運動控制軸卡與端子板......20
2.3.2 AC伺服馬達......21
2.3.3 AC伺服馬達驅動器......23
2.4 視覺系統相關元件與架構......24
2.4.1 面掃描相機......25
2.4.2 遠心式鏡頭......26
2.4.3 閃頻控制器......27
2.4.4 高亮度低角度環型LED光源......29
2.5 系統架構整合與線配置......30
2.6 作業系統與影像處理軟體......31
第三章 表面瑕疵檢測方式與建置......32
3.1 系統校正......32
3.2 視覺檢測架構與演算......36
3.2.1 影像二值化......36
3.2.2 均值濾波......38
3.2.3 影像形態學......39
3.2.4 探討拼接緣由與產生問題......41
3.2.5 拼接演算......43
3.3 單位像素轉換......49
第四章 實驗結果與討論......51
4.1 人機介面 (Human-Machine Interface, HMI)......51
4.1.1 軸動平臺控制與相機對焦介面......52
4.1.2 軸動取像頁面......53
4.1.3 視覺辨識顯示頁面......54
4.2 閃頻控制器介紹與相機速度參數......55
4.2.1 閃頻控制器介紹......55
4.2.2 相機速度參數介紹......56
4.2.3 閃頻器介面與脈波寬度說明......56
4.3 表面瑕疵檢測系統介紹......57
4.3.1 物件掃描拍攝......58
4.3.2 表面瑕疵檢測......59
4.4 表面瑕疵檢測結果......66
4.4.1 表面瑕疵檢測結果與資料統整......66
4.4.2 表面瑕疵分析......70
4.5 結果討論......71
4.6 不確定因素分析......73
第五章 結論與未來展望......74
5.1 結論......74
5.2 未來展望......74
參考文獻......75
附錄一......78
附錄二......79
附錄三......80
附錄四......81
附錄五......82
附錄六......83
附錄七......84
附錄八......85
Extended Abstract......86

[1]Yingen Xiong, Kari Pulli, "Sequential image stitching for mobile panoramas", 2009 7th International Conference on Information, Communications and Signal Processing (ICICS).
[2]Sung-suk Choi, Oh-hoon Cho, Eung-kon Kim, "Design of Automated Screw Blade Welding Defect Detection System for Image Processing-Based Zero-Defect Mass-Production", 2015 International Conference on Computational Science and Computational Intelligence (CSCI).
[3]Chuan Lia, Zhi-YongLiu, Xu Yanga, Hong Qiao, Jian-Hua Su, "Stitching contaminated images", Neurocomputing, Volume 214, 19 November 2016, Pages 829-836.
[4]Pengpeng Zhang, Tengfei Liu, Zhong-Ping Jiang, "Event-triggered control subject to external disturbance", 2016 35th Chinese Control Conference (CCC).
[5]Xiaoyu Liang, Liangyan Dong, Youyu Wu, "Research on Surface Defect Detection Algorithm of Tube-Type Bottle Based on Machine Vision", 2017 10th International Conference on Intelligent Computation Technology and Automation (ICICTA).
[6]Shoulu Lv, Fuqiang Zhou, Zhenzhong Wei, "Train wheel tread defects detection based on image registration", 2017 IEEE International Conference on Imaging Systems and Techniques (IST).
[7]Ning Li, Jianyu Zhao, Ping Jiang, "Fabric defects detection via visual attention mechanism", 2017 Chinese Automation Congress (CAC).
[8]Mehmet BAYGIN, Mehmet KARAKOSE, Alisan SARIMADEN, Erhan AKIN, "Machine vision based defect detection approach using image processing", 2017 International Artificial Intelligence and Data Processing Symposium (IDAP).
[9]Nengsheng Bao, Xie Ran, Zhanfu Wu, Yanfen Xue, Keyan Wang, "Design of inspection system of glaze defect on the surface of ceramic pot based on machine vision", 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC).
[10]Faisal Riaz, Khurram Kamal, Tayyab Zafar, Rizwan Qayyum, "An inspection approach for casting defects detection using image segmentation", 2017 International Conference on Mechanical, System and Control Engineering (ICMSC).
[11]Wei Li, Yue Zheng, Xianmin Wang, "Study on End Face Defect Detection Method of Bearing Inner Ring", 2018 5th Asia-Pacific World Congress on Computer Science and Engineering (APWC on CSE).
[12]Du-Ming Tsai, Daniel E. Rivera Molina, "Morphology-based defect detection in machined surfaces with circular tool-mark patterns", Measurement, Volume 134, February 2019, Pages 209-217.
[13]Li Chen, Ji Deng, "Research on Surface Defects Detection of Stainless Steel Spoon Based on Machine Vision", 2018 Chinese Automation Congress (CAC).
[14]Kaiyue Li, Guangtai Ding, "A Novel Automatic Image Stitching Algorithm for Ceramic Microscopic Images", 2018 International Conference on Audio, Language and Image Processing (ICALIP).
[15]Li Fu, Shuai Zhang, Yu Gong, Quanjun Huang, "Medicine Glass Bottle Defect Detection Based on Machine Vision", 2019 Chinese Control And Decision Conference (CCDC).
[16]Naijian Chen, Jianbo Sun, Xu Wang, Yulin Huang, Yingjun Li, Chao Guo, "Research on surface defect detection and grinding path planning of steel plate based on machine vision", 2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA).
[17]Fan Dahuang, Ding Lei, Deng Jiehang, "Automatic Detection and Localization of Surface Defects for Whole Piece of Ultrahigh-definition Leather Images", 2019 IEEE 4th International Conference on Computer and Communication Systems (ICCCS).
[18]Yuzhen Lu, Renfu Lu, "Structured-illumination reflectance imaging for the detection of defects in fruit: Analysis of resolution, contrast and depth-resolving features", Biosystems Engineering , Volume 180, April 2019, Pages 1-15.
[19]Chunbo Xiu, Yunfei Ma, "Image Stitching Based on Improved Gradual Fusion Algorithm", 2019 Chinese Control And Decision Conference (CCDC).
[20]Murtadha Alomran, Douglas Chai, "Feature-based panoramic image stitching", 2016 14th International Conference on Control, Automation, Robotics and Vision (ICARCV).
[21]Ruirui Ma, Chunlin Zhang, Qing Guo, Fangyi Wan, "Research on Aerial Image Stitching Technology", 2019 International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC).
[22]Weibin Li, Bin Hu, Chao Song, Fan Zhao, Honglin Ma, Yuejun Wang, "An image stitching method for blades of wind turbine based on background removal preprocessing", 2020 5th International Conference on Communication, Image and Signal Processing (CCISP).
[23]Jingfan Hang, Xianqiang Yang, Xinghu Yu, "Sanitary Ceramic Surface Defect Detection Method Based on Neighborhood Pixel Gray Information", IECON 2020 The 46th Annual Conference of the IEEE Industrial Electronics Society.
[24]Zhichao Liu,Baida Qu, "Machine vision based online detection of PCB defect", Microprocessors and Microsystems, Volume 82, April 2021, 103807.
[25]Zhike Zhao, "Review of non-destructive testing methods for defect detection of ceramics", Ceramics International , Volume 47, Issue 4, 15 February 2021, Pages 4389-4397.
[26]張軒維，”應用線掃描相機於陶瓷基板表面瑕疵高低解析度檢測系統開發”，碩士論文，自動化工程系研究所，台灣，2020。
[27]劉士銓，”使用面掃描相機於陶瓷基板表面瑕疵檢測系統開發”，碩士論文，自動化工程系研究所，台灣，2020。
[28]Tongjia Zhang, Chengrui Zhang, Yanjie Wang, Xiaofu Zou, Tianliang Hu, "A vision-based fusion method for defect detection of milling cutter spiral cutting edge", Measurement, Volume 177, June 2021, 109248.
[29]Sung-Hyok Ro, Se-HunKim, "An image stitching algorithm for the mineralogical analysis", Minerals Engineering , Volume 169, 1 August 2021, 106968.
[30]SenWang, Chang Liu, Yinhui Zhang, "Fully convolution network architecture for steel-beam crack detection in fast-stitching images", Mechanical Systems and Signal Processing, Volume 165, 15 February 2022, 108377.
[31]Guanping Dong, Shanwei Sun, Nanshou Wu, Xiangyang Chen, Pingnan Huang, Zixi Wang, "A rapid detection method for the surface defects of mosaic ceramic tiles", Ceramics International, Volume 48, Issue 11, 1 June 2022, Pages 15462-15469.
[32]研華科技, https://www.advantech.tw
[33]Yaskawa, https://www.yaskawa.com
[34]JM Vistec Sysem Pte Ltd, https://www.jm-vistec.com/
[35]碁仕科技, https://www.g4.com.tw
[36]Basler, https://www.baslerweb.com
[37]ES Technology Co, http://www.esstartek.com/yellowpage/hot_cg17313.html
[38]Myutron, https://www.myutron.com/en/lens/
[39]宸軒科技, https://www.chtco.com.tw
[40]Keysight, https://www.keysight.com
[41]Halcon, https://www. mvtec.com
[42]鍾國亮，2015，影像處理與電腦視覺第六版，第5章 門檻值決定與區域的分割，東華書局，台灣。
[43]鍾國亮，2015，影像處理與電腦視覺第六版，第3章第7節 頻率域濾波器設計，東華書局，台灣，。
[44]鍾國亮，2015，影像處理與電腦視覺第六版，第2章第2節 形態學，東華書局，台灣。