臺灣博碩士論文加值系統

傳統半伺服口罩生成設備在生產口罩時，因調整時需要耗費大量時間，而且單純以一個動力馬達帶動,所以調整上更為困難，有時候光是調整或是維修就需要耗上大半天的時間，所以現在市面上都已改為全伺服控制的口罩機為主流，而且加上自動補正技術讓步更為流暢，解省工時之餘還可以不用讓人一直盯著機器運作，深怕一個不小心布料偏移導致幾百公尺的布料都報廢，大致上把傳統口罩機的缺點大幅度改善，但這類的口罩機還缺少了辨識口罩耳帶是否有確實的熔接，也因此有這個缺點而有本論文之研究構思。
本論文研究取材自市面上常見口罩機為構思基礎，以視覺辨識系統去判斷口罩熔接處是否有正確熔接，來免去不必要的損失，並使用辨識軟體matlab來分析，並算出良率值，以便之後機台改造之方向。
經由matlab分析之後，以成像觀察去變更超音波熔接位置及發波時間，更甚更改超音波熔接之高度和能量及電流量，更可以觀察是否是因為軌道過窄所導致口罩反摺導致無法熔接之問題，改善這些問題，達到機器之最高效能。

When traditional semi-servo mask generation equipment produces masks, it takes a lot of time to adjust and is driven by a power motor, so it is more difficult to adjust. Sometimes it takes more than half a day to adjust or repair. , So now the market has been changed to a fully servo-controlled mask machine as the mainstream, and with the automatic correction technology to make the concession smoother, save time and save time, you don’t need to keep people staring at the operation of the machine. Be careful of the fabric offset, which results in scraps of hundreds of meters of fabric. Generally speaking, the shortcomings of traditional mask machines have been greatly improved. However, this type of mask machine lacks the ability to identify whether the mask earbands are actually welded, and therefore have this shortcoming And there is the research concept of this thesis.
The research of this thesis is based on the concept of common mask machines on the market. It uses a visual recognition system to determine whether the mask is welded correctly to avoid unnecessary losses. It uses the recognition software matlab to analyze and calculate the yield value. So that the direction of machine transformation later.
After analyzing by matlab, use imaging observation to change the ultrasonic welding position and wave time, and even change the ultrasonic welding height, energy and current. It can also be observed whether it is because the track is too narrow that the mask is reflexed and cannot be welded. The problems, improve these problems, and achieve the highest performance of the machine.

誌 謝 i
摘 要 ii
Abstract iii
目錄 v
圖目錄 viii
表目錄 x
第一章、緒論 1
1.1研究背景與動機 1
1.2 研究背景 2
1.3文獻探討 3
1.4論文架構 8
第二章、口罩生成設備相關研究 9
2.1半伺服口罩機設備介紹 9
2.1.1半伺服口罩機布料架說明 10
2.1.3半伺服口罩機耳帶熔接設備說明 14
3.1視覺辨識理念介紹 18
3.2視覺辨識架構介紹 19
3.3雲台伺服馬達控制補正 20
3.5雲台架構介紹 20
3.6程式編碼 24
3.7影像分析介紹 25
3.8視覺辨識程式碼 27
3.9程式碼細部介紹 28
3.10影像前處理流程 31
3.11分段線性灰階變換 33
3.12基本線性灰階變換 35
3.13均衡處理 36
3.14大背光盒 41
3.15大背光盒材料 42
第四章、研究總結與未來方向 45
4.1研究成果 45
4.2研究貢獻 45
4.3未來研究方向 45
參考文獻 46
附錄 49
燈盒尺寸圖 49
實驗數據 50

[1]A. Taylor Stuart, CCD and CMOS Imaging Array Technology Review Technical report, 1998.
[2]Alberto, P. J. C., Karina, P. J. S., Irene, S. A. F., & Nahomee, R. Á. A. (2021). Waste reduction in printing process by implementing a video inspection system as a human machine interface. Procedia Computer Science, 180, 79-85.
[3]Bailly, M., Tulet, M., & Flamenbaum, S. (1986). CCD imaging sensor. In Automatic Control in Space 1985 (pp. 231-238). Pergamon.
[4]Cherry, T. L., James, A. G., & Murphy, J. (2021). The impact of public health messaging and personal experience on the acceptance of mask wearing during the COVID-19 pandemic. Journal of economic behavior & organization.
[5]Gong, Y., Song, Z., Ning, H., Hu, N., Peng, X., Wu, X., ... & Liu, Q. (2020). A comprehensive review of characterization and simulation methods for thermo-stamping of 2D woven fabric reinforced thermoplastics. Composites Part B: Engineering, 108462.
[6]Heijmans, H. J. (2020). Morphology for grey-scale images. Advances in Imaging and Electron Physics.
[7]Huang, Z., Wang, Z., Zhang, J., Li, Q., & Shi, Y. (2021). Image enhancement with the preservation of brightness and structures by employing contrast limited dynamic quadri-histogram equalization. Optik, 226, 165877.
[8]Ibhadode, O., Zhang, Z., Bonakdar, A., & Toyserkani, E. (2021). IbIPP for topology optimization—An Image-based Initialization and Post-Processing code written in MATLAB. SoftwareX, 14, 100701.

[9]Ito, K., Fujisawa, T., & Iwata, T. (2007). Novel fast-arrayed CCD X-ray detector using interline transfer-type CCD for time-resolved X-ray diffraction measurements. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 582(2), 673-682.

[10]Litos, L., Gray, D., Johnston, B., Morgan, D., & Evans, S. (2017). A maturity-based improvement method for eco-efficiency in manufacturing systems. Procedia Manufacturing, 8, 160-167.

[11]Liu, Y., Zhang, F., Zhang, Y., Li, X., & Zhang, C. (2021). Image smoothing based on histogram equalized content-aware patches and direction-constrained sparse gradients. Signal Processing, 183, 108037.
[12]Luo, T. D., Li, B. K., Yang, S. H., Guo, M. A., & Yan, M. (2016). Novel driver method to improve ordinary CCD frame rate for high-speed imaging diagnosis. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 822, 43-47
[13]May 2020, [online] Available: https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
[14]N. Norouzi and M.E. Moghaddam, "Motion blur identification using image derivative", Signal Processing and Information Techonology. ISSPIT 2008, pp. 380-384, 2008.
[15]Rawle, R. J., Boxer, S. G., & Kasson, P. M. (2019). Detecting and Controlling Dye and Illumination Effects in Single-Virus Fusion Experiments. Biophysical Journal, 116(3), 181a.
[16]Van de Straete, H. J., Degezelle, P., De Schutter, J., & Belmans, R. (1998). Optimal selection procedure for induction motor based servo drives. IFAC Proceedings Volumes, 31(27), 327-333.
[17]Wang, F., & Peng, G. (2021). Graph-based saliency detection using a learning joint affinity matrix. Neurocomputing.
[18]Zhang, X., Wang, L., & Su, Y. (2021). Visual place recognition: A survey from deep learning perspective. Pattern Recognition, 113, 107760.
[19]Zhang, Z., Cheng, D., Shi, Y., & Zhang, J. (2017). The terminal vibration of laser spot tail in dual channel type linear CCD. Optics & Laser Technology, 96, 238-242.