English  |  正體中文  |  简体中文  |  Items with full text/Total items : 70585/70585 (100%)
Visitors : 23188410      Online Users : 618
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version


    Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/48126


    Title: 類神經網路應用於LED模組的色相與亮度研究;The application of Neural Networks to the luminance and color uniformity for the LED modules
    Authors: 徐志凱;Zhi-kai Xu
    Contributors: 光電科學研究所
    Keywords: 發光二極體;類神經網路;倒傳遞法;LED;neural networks;back-propagation
    Date: 2011-07-11
    Issue Date: 2012-01-05 14:34:37 (UTC+8)
    Abstract: 目前在市場上檢驗LED (Light Emitting Diode) 模組這方面的技術許多尚停留在由肉眼視覺方式檢驗,此方法由於視覺殘留容易造成人為的誤差,並且可能會對人眼造成不良的傷害。在不同的檢驗者、時間、地點與環境都可能會造成不同的結果,這樣對於LED模組產品的品質將會造成相當的問題。 基於上述原因,本研究使用類神經網路的概念,利用類神經網路倒傳遞法(Back Propagation)建立一套LED模組自動檢測系統,不僅可檢驗出亮度的不均勻性,同時對色彩的偏離亦能同時量測出來。此系統藉由CCD攝影機經由光學質心法求出LED模組中合格的紅、綠、藍三色的灰階值,做為學習的輸入與目標值,給予類神經網路做適當的學習,再利用學習完成的類神經網路架構對待測的LED做檢測,得到電腦自動檢測的目的。 經本研究結果來測試待測樣本,得到的結果多落在理想值的90%至100%之間,對於此測試結果尚稱合理。對於極少數落在70%~90%之間的模糊地帶,則是因訓練時樣本本身品質的落差與其他環境因素造成,除了將這些樣本加入學習檔重複學習,未來亦可考慮以模糊邏輯(Fuzzy logic)來解決這些問題,亦將是未來延續本研究的方向與目標。 關鍵字:發光二極體、類神經網路、倒傳遞法 The technologies used for the brightness and color testing of the LED (Light Emitting Diode) modules on the market, are mostly remain in the manual ways, manual inspection could induce the human errors and cause the damage of human eyes. Due to the different operators, timing and environment, the results of the inspection might be different. The study is utilizing the concept of the neural networks, using the back-propagation neural network method, to construct a LED module automatic inspection system. The uniformity of the brightness and the color deviation of the module were tested. The CCD camera captures the red, green, and blue grayscale values from the samples under test, the captured data are divided into 3 different bins, one for the learning, one for the testing and retraining, and one for the final accuracy conformation purpose. The optical centroid qualified method is used for the LED RGB gray level images. The results of the study are fairly reasonable, some deviations are caused by the environmental factors, adding these samples to the learning bin, making some repeated learning, and the acceptable results are obtained. The fuzzy logic methodology is considered to be the future study for the advanced research, and more stable results would be expected. Keywords: LED, neural networks, back-propagation
    Appears in Collections:[光電科學研究所] 博碩士論文

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML424View/Open


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

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback  - 隱私權政策聲明