English  |  正體中文  |  简体中文  |  Items with full text/Total items : 2737/2828
Visitors : 269191      Online Users : 5
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ir.lib.stu.edu.tw:80/ir/handle/310903100/1267

Title: 可以支援公共安全應用之通用型隨插即用技術-以車牌辨識為例
An Approach to Public Security based on UPnP Technology, An Example of License Plate Recognition
Authors: 杜宛玲
Du, Wan-Ling
Contributors: Horng, Mong-Fong
資訊工程學系
Keywords: 通用型隨插即用;車牌辨識;公共安全
UPnP;Linense Plate Recognition;Public Safty
Date: 2008
Issue Date: 2011-05-24 15:12:03 (UTC+8)
Publisher: 高雄市:[樹德科技大學資訊工程學系]
Abstract: 本論文提出一可支援公共安全應用之通用型隨插即用技術與應用,並以車牌辨識為例來驗證所提出的方法。透過所建置的平台,攝影機可以擷取車牌影像進行辨識,並將辨識結果即時通報使用者的終端裝置,使用者可透過UPnP網路取得更多的警報資訊。所探討的主要技術包含:影像辨識與自動通報兩大部份。在影像辨識部份,本論文採用類神經網路技術,利用類神經網路自我學習與高速運算的能力,來進行基礎影像的學習與實際影像辨識。其次,利用UPnP技術,使用者毋需設定網路組態,便可在不同無線網路中,快速地建立連線,即時取得違規車輛警報的資訊。此外,使用者也可以透過PDA或手提電腦遙控攝影機的動作模式。所提方法可以應用於可疑車輛的自動化查詢與通報系統。依據實驗結果顯示,當系統發現違規車輛時,能夠即時將警報訊息通報使用者,且使用者可藉由UPnP技術而便利地查詢更多的警報資訊。即時且便利的服務可初步提昇監控品質,加強公共安全。
In this paper, UPnP-based license plate recognition (ULPR) is developed for public safety. To overcome the problems encountered in traditional license plate recognition (LPR) using statistics approach, the proposed system employees intelligent computing technology, such neural networks, to offer a flexible and robust pattern recognition. The developed system ULPR is composed of tow main components called recognition module (RM) and the alerting module (AM). RM is in charge of (1) image capturing (2) plate positioning (3) image segmentation (4) plate recognition (5) recognition learning. AM is in charge of (1) discovering new-attached devices (2) identifying the users and (3) message alerting. Thus the camera on server can capture images to identify plates and send users with results through the UPnP Network. Besides, a Back-propagation neural network is designed to be trained and to learn the diverse plate images. Through a three-layer BPN, all numeric and alphabetic images captured from real vehicles are fed to train. The training data set is composed of lots of irregular patterns, including screwing; tilting and shrinking patterns. The experimental results show the developed system achieves an accuracy of 93% for insides testing. According to the experimental results, the system demonstrates a prototype for detecting suspicious vehicles to support public safety.
Appears in Collections:[資訊工程系(所) ] 博碩士論文

Files in This Item:

File Description SizeFormat
可以支援公共安全應用之通用型隨插即用技術-以車牌辨識為例__臺灣博碩士論文知識加值系統.htm國圖105KbHTML336View/Open
可以支援公共安全應用之通用型隨插即用技術-以車牌辨識為例.pdf3417KbAdobe PDF1024View/Open


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

 


無標題文件

著作權政策宣告:

1.

本網站之數位內容為樹德科技大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
 
2. 本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本校護人員(clairhsu@stu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
 
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback