English  |  正體中文  |  简体中文  |  Items with full text/Total items : 52052/87180 (60%)
Visitors : 8897058      Online Users : 249
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU 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://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/87836

    Title: 以BIM與影像辨識為基礎之室內空間辨識雛型系統
    Other Titles: BIM-vision-based indoor localization prototype
    Authors: 邱政銘;Chiu, Cheng-Ming
    Contributors: 淡江大學土木工程學系碩士班
    Keywords: 室內定位空間辨識;工業基礎類別;影像辨識;Indoor LocalizationIndoor Localization;IFC;Image recognition;BIM
    Date: 2012
    Issue Date: 2013-04-13 11:47:36 (UTC+8)
    Abstract: 本研究宗旨在使用室內空間特徵影像做為辨識標籤並結合建築資訊模型(Building Inform-ation Modeling)內所包含的空間資訊,目的是為了提供一個經濟的室內空間定位技術,使用戶可以透過智慧型手持裝置相機鏡頭擷取所在位置的空間特徵影像,得知目前所在位置及該空間資訊。
    其方法和目前無線頻寬(Wireless)和無線射頻(Radio Frequency IDentification)室內定位技術有所不同,本研究應用的室內空間特徵影像做為位置識別標籤技術。其架構包括三個功能模組,(1)空間特徵影像資料庫,(2)空間特徵影像管理模組,(3)空間特徵影像辨識定位模組。空間特徵影像資料庫是從建築資訊模型(Building Information Modeling)收集來自工業基礎類別標準(Industry Fo-undation Classes)相關空間資訊及相對應空間的特徵影像。空間特徵影像管理模組為拍攝空間特徵影像,與資料庫連結並建立空間特徵影像集。空間特徵影像辨識定位模組是使用擴增實境技術,運用D''fusion軟體的影像辨識技術,透過手持裝置擷取所在空間特徵影像,進行影像辨識定位後,連結到資料庫將所在空間資訊顯示在手持裝置上。
    結果與討論是以BIM與影像辨識為基礎之空間辨識雛型系統,透過智慧型手持裝置的攝影鏡頭在Android平台上運行,如智慧型手機和平板電腦。透過不斷的技術測試,以了本研究的技術的可行性。根據多次測試結果,本研究室內定位雛型系統可以識別空間位置及使用者所在的空間資訊,然而當室內空間是缺乏特徵影像而無法辨識時,如室內空間為單調的牆壁及相似度過高,未能被識別。為了克服這種情況,將以二維空間標籤的QR碼(Quick Response Code),替代空間特徵影像作為辨識標籤。此外,由於空間資訊是從現有的建築資訊模型而來的,可以保證資訊的一致性。之後並透過經濟可行性進行分析評估其成本與效益。
    Purpose In order to provide an economical indoor location detective technique, this study is aimed to use photo images as the indoor spatial identification tags associated with the spatial information of the existing building information model (BIM), so that users can identify their locations via the camera on mobile device based on the real images. Method Unlike the wireless and RFID-based indoor positioning techniques, this study applied the image recognition technique to indoor location detection. Prototype architecture includes three functional modules, namely, (1) Spatial image database, (2) spatial image management module, and (3) vision-based localization module are developed. The spatial image database is the data collection of space related data from the IFC (Industry Foundation Classes) dataset of the existing BIM and space feature image data collected form users. The spatial image management module provides users an interface to collect the spatial photo images of buildings, and bind them with the location data from the spatial image database. Then, the vision-based localization module developed based on the D’fusion studio can identify the space location by recognizing the images from the vision captured with the camera on a smart phone, and the corresponding spatial data can be retrieved from the database. Results & Discussion The BIM-Vision-Based indoor localization prototype was developed as an android platform application running on the mobile device with imbedded camera such as smart phones and tablets. Technique feasibility is continuously tested in current phase. According to the basic test results, the prototype can identify the indoor locations of decorated spaces; however, once the indoor spaces are lack of recognition features, such as the empty spaces with blank and monotony walls, the recognition function failed. To overcome this defect, the Quick Response (QR) code, the trademark for a type of two-dimensional code, is used as a substitution of the photos for this prototype. Besides, since the location data is transferred from the existing building information model, the data consistency can be ensured. In the future, the economic feasibility of this prototype would be analyzed to evaluate the cost and benefit ration.
    Appears in Collections:[土木工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat

    All items in 機構典藏 are protected by copyright, with all rights reserved.

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback