English  |  正體中文  |  简体中文  |  Items with full text/Total items : 51948/87093 (60%)
Visitors : 8507438      Online Users : 266
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/987654321/104569

    Title: Vanishing-Point Detection Based on a Fuzzy Clustering Algorithm and New Clustering Validity Measure
    Authors: Chou, Chien-Hsing;Zhao, Yu-Xiang;Tai, Hsien-Pang
    Date: 2015-06-15
    Issue Date: 2016-01-06 11:03:17 (UTC+8)
    Abstract: This paper proposes an image-preprocessing method combined with a fuzzy clustering algorithm and new validity measure to detect the vanishing point in an image. The proposed method segments the object in an image by using the clustering algorithm and then extracts critical vanishing lines. By examining the intersection of the vanishing lines, the vanishing point is located. To locate the vanishing point accurately, the initial cluster number of the fuzzy clustering algorithm should be provided correctly. Therefore, the study proposes a new clustering validity measure, the area measure, to estimate the initial cluster number according to the information of cluster areas. Experimental results on 29 images show that the proposed preprocessing method and validity measure can accurately identify the location of the vanishing point and vanishing lines. In addition, compared with several validity measures, the new validity measure achieves satisfactory experimental results and outperforms six other validity measures.
    Relation: Journal of Applied Science and Engineering 18(2), pp.105-116
    DOI: 10.6180/jase.2015.18.2.02
    Appears in Collections:[電機工程學系暨研究所] 期刊論文

    Files in This Item:

    File Description 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