English  |  正體中文  |  简体中文  |  Items with full text/Total items : 66984/66984 (100%)
Visitors : 23005348      Online Users : 358
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/987654321/63569


    Title: 應用影像辨識技術於橋梁裂縫之研究
    Authors: 呂景羣;Lu,Ching-Chun
    Contributors: 營建管理研究所
    Keywords: 類神經網路;影像處理;影像辨識;SOM optimization;pattern recognition;bridge crack;image process
    Date: 2014-01-22
    Issue Date: 2014-04-02 15:53:48 (UTC+8)
    Publisher: 國立中央大學
    Abstract: 我國台灣本島位於歐亞板塊和菲律賓板塊的交接擠壓處,同時位於亞熱帶氣候區常年為我國帶來豐沛的雨量,但此種地理環境卻造成每年遭受自然災害的侵襲,使得經濟中樞的橋梁每年的維修費用龐大;而影像資訊在我們生活中越來越普及,為了解決傳統人工檢測橋梁過於主觀的缺點和減少時間成本的開銷,本論文利用多維矩形類神經網路及最佳化演算法開發訓練一套裂縫影像辨識軟體用於橋梁檢測工作。
    影響辨識技術準確率的主要因素為原始影像的取得方法及辨識前的處理步驟,本篇論文在影像取得前先統一拍攝條件限制,接著以手持數位相機前往橋梁劣化構件部份拍攝裂縫影像供此程式訓練,但一般在進行影像判識之前,需對其作影像處理才能使機器迅速有效率的使用這些影像,本論文的影像處理包含以下步驟:灰階化(Grayscale),高通濾波(high-pass),二值化,labeling去除雜點,Local Directional Pattern運算強化裂縫邊緣;最後挑選出36張適合的訓練樣本,利用類神經網路進行準確率的訓練,根據所得之訓練結果有至少81%的準確率,並以此程式用於八八風災過後受損橋梁裂縫辨識,其準確率有89%,以本論文所發展之裂縫判識程式可以有效地判斷出受損橋梁,減少傳統上人為因素的判斷誤差以及可以大幅減少判識時間和成本。; Deterioration inspection for bridges has been performed for years in Taiwan. The current inspection method that heavily depends on human recognition so called Degree, Extend, Relevancy, and Urgency (DERU) method is time-consuming and subjective. The research objective is to develop a crack recognition model that adopts Self Organizing Map Optimization (SOMO) integrating with image process technique. The image data collection is based on the bridge current status from the database of Taiwan Bridge Management System (TBMS), which provides detailed DERU status for bridges and leads the investigators to take samples from deteriorated bridges with DERU > 2. A total of randomly selected 216 samples of bridge cracks is collected. 40 out of 216 samples are utilized and set as training and testing datasets. The image process takes four major steps: grayscale, high-pass filter, labeling, and local directional pattern. The recognition results present high accuracy rates of 90.4% for crack recognition and 92.5% for non-crack recognition. The model demonstrates the feasibility and high-accurate recognition for crack inspection in bridge management.
    Appears in Collections:[營建管理研究所 ] 博碩士論文

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML528View/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