English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27533/39387
Visitors : 2542425      Online Users : 32
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/987654321/16043

Title: 應用衡量保持切割法於SAR影像油污染之偵測
Detection of Oil Slicks in SAR Images by Moment-Preserving Segmentation
Authors: Tsai-Hsing Tang
唐再興
Contributors: NTOU:Department of Merchant Marine
國立臺灣海洋大學:商船學系所
Keywords: SAR衛星影像;衡量保持法;影像處理;油污偵測;合成孔徑雷達
SAR image;Moment-Preserving Principle;image process;oil detection;Synthetic Aperture Radar
Date: 2003
Issue Date: 2011-06-30T08:32:30Z
Abstract: 摘 要 SAR合成孔徑雷達衛星影像具有涵蓋面積廣闊、解析度高、不受雲雨干擾影響,且可定時長期監測等優點,目前已被廣泛地應用在土地資源利用、海洋環境監測等各個不同領域中。隨著科技不斷地進步,人類污染海洋的情況日益嚴重,如何利用SAR衛星影像進行油污染偵測變得十分迫切。文獻中對於油污染偵測,多以利用影像處理技術進行油污邊緣偵測,對於油污資料模式建置相關論述卻很少。因此,本論文乃先針對油污與海洋建置其模式,並依此提出一以影像區間為基礎之油污偵測法。 本研究首先以衡量保持法(Moment-Preserving Principle)將影像進行切割分類,再利用大數法則考量空間特性以進行影像區間合併。利用影像切割結果決定油污區域,並建立油污與海洋之資料模式。 接著,利用建立之油污及海洋資料模式,結合偵測理論之Generalized Likelihood Ration Test準則,提出一以影像區間為基礎之油污偵測法。在固定錯誤警報機率的條件下,可自動決定決策準則之臨界值。 電腦模擬結果驗証此法不僅對訓練過的影像有極佳的偵測結果,對未曾訓練的影像,其偵測效果亦很好,故藉此偵測技術可達海洋油污自動判別之目的。 關鍵字:SAR衛星影像、合成孔徑雷達、影像處理、衡量保持法、油污偵測
Abstract Synthetic Aperture Radar images have become an important tool in ocean oil slick monitoring due to its high resolution and wide coverage under all weather condition, day and night. Now it make widely use of land utilization of resources, marine environmental monitoring, etc. With the constantly progress of science and technology, the situation that the human pollutes the ocean is serious day by day. How to utilize SAR satellite image to detect oil pollution becomes very urgent. In the literature about oil pollution detection, most of it use image process technology to detect oil edge, but about oil-data model is very less. In the study, we propose a region-based oil detection method which is based on the data model of oil polluted and sea image area. We segment the image into proper regions by using Moment-Preserving Principle at first. Then, concerning the image spacial characteristics, we use Law of Large Number to remerge the image region obtained from the Moment-Preserving Principle. Applying image segmentation results, we build the data model for the oil polluted and sea image area, respectively. According to the oil/sea data model and the Generalized Likelihood Ration Test (GLRT) decision rule, we propose a region-based oil detection method. Under the condition of Constant False Alarm Ration (CFAR), we may determine the threshold automatically. Simulations validate the efficiency of the proposed method both for trained and untrained images. Keyword: SAR image, Synthetic Aperture Radar, image process, Moment-Preserving Principle, oil detection
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M91710003
http://ntour.ntou.edu.tw/ir/handle/987654321/16043
Appears in Collections:[商船學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML224View/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