English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27533/39387
Visitors : 2542514      Online Users : 28
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://ntour.ntou.edu.tw:8080/ir/handle/987654321/18377

Title: 以平行處理進行高頻譜影像目標偵測之研究
Parallel Approaches for Hyperspectral Image Target Detection
Authors: Chih-Chien Chan
詹志謙
Contributors: NTOU:Department of Communications Navigation and Control Engineering
國立臺灣海洋大學:通訊與導航工程系
Keywords: 高頻譜影像;目標偵測;影像分類;電腦叢集;平行處理;部份濾波法
Hyperspectral image;Target detection;Image classification;PC cluster;Partial filter approach;Parallel processing
Date: 2009
Issue Date: 2011-07-04
Abstract: 高頻譜影像提供大量豐富的頻譜資訊,對於目標偵測與影像分類有較好的辨識能力。但是高頻譜影像的頻帶數量很多,相對的目標偵測運算也需要花大量的時間。為了提昇高頻譜影像目標偵測和影像分類的效能,論文中依據高頻譜影像頻帶具有高相關的特性,提出部份濾波法來減少目標偵測和影像分類的計算複雜度。此法,利用聚類頻帶分群法將具相關性高的頻帶群聚為一群組,並且搭配部份濾波器的設計,以減少頻帶數量達到降低原來濾波器設計的計算量。 此外,為了進一步提昇高頻譜影像處理的效能,我們建構電腦叢集的平行處理系統來實現部份濾波法。由於部份濾波法的結構很適合平行處理的方式,即將一個複雜龐大的工作,切割成數個簡單的工作交給多個處理器負責,以降低工作時間。於是,我們使用四台相同硬體配備的個人電腦與一台伺服器電腦透過路由器和網路線連接建構成電腦叢集,並且利用Matlab提供的Parallel Computing Toolbox和Matlab Distributed Computing Engine兩套軟體,使用分散式運算進行部份濾波法的平行處理。 藉由建置的電腦叢集快速的驗證部份濾波法的效能。經模擬結果顯示,論文所提部份濾波法,除了可減少原來濾波器法於高頻譜影像處理的計算量,也會提高目標偵測和影像分類的效能。此外,也驗證當大量的高頻譜影像資料作運算時,以電腦叢集進行平行處理的速度提昇值有明顯的上昇。
The hyperspectral images provide extensive spectral information and offer a better identification in target detecting and image classifying than traditional methods. However, higher spectral resolution increases the computation complexity in the target detection processing. In order to improve the efficiency in target detection and image classification, we propose a partial filter approach which exploits high degree correlations in spectral domain. First, bands with highly correlated features are grouped together by using a clustering algorithm. Then, the matching partial filter is designed for each group and the designed partial filters are combined to be a target detector. The partial filter approach can improve the accuracy of the target detection with less computation complexity. Since the structure of the partial filter is very suitable for the parallel processing, we then divide the design of a huge target detector into a few simple partial filter design works. Thus, to improve the efficiency of the proposed method, we build a PC clustered parallel computing system to realize the partial filter approach. The PC cluster is constructed by four same personal computers and one server computer through the router and the network cable. And we use the Parallel Computing Toolbox provided by Matlab and the Matlab Distributed Computing Engine to execute the parallel process of the partial filter through the Distributed Computing. The PC clustered parallel computing system fast validates the efficiency of the partial filter. Simulation results show that the proposed partial filter can improve the efficiency of the target detection with less computation complexity than that of the conventional target detection schemes. In addition, the processing rate of the proposed partial filter approach can be speeded up by the PC clustered parallel computing system for hyperspectral images.
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M96670017
http://ntour.ntou.edu.tw/ir/handle/987654321/18377
Appears in Collections:[通訊與導航工程學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML195View/Open


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

 


著作權政策宣告: 本網站之內容為國立臺灣海洋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,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback