文化大學機構典藏 CCUR:Item 987654321/22030
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 47249/51115 (92%)
造访人次 : 14540084      在线人数 : 1461
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜寻范围 查询小技巧:
  • 您可在西文检索词汇前后加上"双引号",以获取较精准的检索结果
  • 若欲以作者姓名搜寻,建议至进阶搜寻限定作者字段,可获得较完整数据
  • 进阶搜寻
    主页登入上传说明关于CCUR管理 到手机版


    Please use this identifier to cite or link to this item: https://irlib.pccu.edu.tw/handle/987654321/22030


    Title: Automatic Detection of a License Plate by Using Wavelet Transform
    Other Titles: 使用小波轉換偵測車牌位置
    Authors: 逄霖生
    陳俞州
    陳柏詮
    Contributors: 工學院
    Keywords: 對比
    車牌偵測
    數位影像處理
    灰度
    小波轉換
    Contrast
    License plate detection
    Digital image processing
    Graylevel wavelet transform
    Date: 2007-06-01
    Issue Date: 2012-04-18 13:36:15 (UTC+8)
    Abstract: 自動車牌辨識是項有用的數位影像處理的應用。本研究探索使用小波轉換(wavelet Transform)來偵測車牌在一部靜止的車輛中的位置,我們利用車牌高對應比的特性,並用小波轉換將這些高對比的訊號分解為高頻的信號,這些資料的頂點代了對比的位置與差別,這些特性可偵測車牌。再水平與垂直的相鄰的映像點間,對資料加以計算與比較,高對比的區域應此可以被察覺,若符合條件,車牌的位置就可以找出,反之非車牌的區域則被剔除。

    License plate recognition is a wide demanding image processing application. We have explored the possibility by using wavelet transform to detect the location of a license plate for non-moving vehicle. The license plate is designed to have high contrast differences which can be easily seen by human eyes. The decomposed high frequency component of wavelet transform extracts high contrast part of license plate. The large magnitude peaks of high frequency component indicate the edges of the foreground character and white background. We search these edges on the horizontal and vertical directions separately. By judging the similarity of neighboring pixels of edge points, we successfully detect the location of license plate.
    Relation: 華岡工程學報 21期 p.87 -92
    Appears in Collections:[工學院] 學報-華岡工程學報

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML773View/Open


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


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©  2006-2025  - Feedback