English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27454/39300
Visitors : 2533995      Online Users : 34
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/52375

Title: 軟性顯示器摺疊測試的光學檢測之類神經網路視覺自動對位研究
A Study of Neural Network Vision Alignment for Optical Inspection of Flexible Displays by Folding Test
Authors: Hung, Pin-Hsuan
洪品軒
Contributors: NTOU:Department of Mechanical and Mechatronic Engineering
國立臺灣海洋大學:機械與機電工程學系
Keywords: 軟性顯示器;自動光學檢測系統;類神經網路;影像回饋;對位控制
flexible display;automatic optical detection system;neural network;image feedback;alignment control
Date: 2018
Issue Date: 2019-11-18T07:49:05Z
Abstract: 為了生產及應用軟性顯示器,不可或缺的便是對這種顯示器的撓曲與摺疊能力進行測試,以達到可多次重複使用的目的。而對於軟性顯示器的檢測,除了最普遍的電性檢測外,對經過測試後的軟性顯示器之擷取影像進行光學分析也是一種檢測的方法。為了以光學方法檢測軟性顯示器的耐用程度,本研究利用類神經網路作為控制法則建立一個自動光學檢測系統。此系統包含一XYZ平台,結合一附有轉軸平台的CCD攝影機於一四自由度的可動影像擷取平台。由於摺疊測試軟性顯示器時,軟性顯示器的尺寸不同會決定軟性顯示器夾具的旋轉角度,本研究利用此影像擷取平台對一可反覆摺疊軟性顯示器的全面性摺疊機台上不同角度的軟性顯示器夾具進行影像擷取。根據影像中軟性顯示器夾具上設置的定位點以影像回饋方式取得點座標作為類神經網路的輸入值,並將相對應角度下四軸平台正確對位時所需的位移量紀錄作為輸出,以建立一個對位控制的光學檢測之監督式類神經網路模型。因此將類神經網路套用於影像擷取平台的控制程式內,可以針對不同角度的軟性顯示器夾具進行自動正交定位的動作。最後本研究利用訓練完成的類神經網路控制影像擷取平台,而定位時誤差最大百分比落在1.77%。另外,根據膽固醇液晶軟性顯示器經過摺疊測試後其影像的灰階值變化結果可以發現摺疊次數遞增時軟性顯示器的光學衰減特性也會增加。除此之外,在四個量測區域中的灰階值最終差異百分比最大可達15.1%。
For the production and application of flexible displays, the deflection and folding capabilities test of the display is indispensable so as to achieve repeated use. For flexible displays testing, in addition to the most public electrical tests, optical analysis of the captured images of the tested flexible display is also an important detection method. To optically detect the durability of a flexible display, this study creates an automatic optical detection system, comprises a neural network as control algorithm, XYZ platform and a CCD camera with a rotating stage on a four degree-of-freedom image capture platform. During the folding test of a flexible display, the different sizes of the flexible display require corresponding rotation angles of the flexible display fixture. This study captures images of flexible displays at different angles on the whole-folding folding platform that can repeatedly fold flexible displays. According to the coordinates of the positioning points in the image which are set on the flexible display fixture as the input values by image feedback and the displacement values of the four degree-of-freedom image capture platform as output values, a Supervised neural network model is set up for the alignment control of the optical inspection. Therefore, the neural network is used in the control program of the image capture platform to orthogonally and automatically position for different fixture angles. Finally, this study uses the neural network to control the four degree-of-freedom image capture platform, and the maximum error percentage is 1.77%. According to the results of the change of the grayscale value of the flexible cholesteric liquid crystal display after the folding test, it can be found that the decay optical characteristics increase as folding cycles. In addition, the maximum difference percentage of the grayscale value in the four regions of interest measurement in the study is 15.1 %.
URI: http://ethesys.lib.ntou.edu.tw/cgi-bin/gs32/gsweb.cgi?o=dstdcdr&s=G0010472007.id
http://ntour.ntou.edu.tw:8080/ir/handle/987654321/52375
Appears in Collections:[機械與機電工程學系] 博碩士論文

Files in This Item:

File SizeFormat
index.html0KbHTML1View/Open


All items in NTOUR are protected by copyright, with all rights reserved.

 


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback