利用統計方法改善永久散射體干涉雷達技術於地表變形偵測之應用;Statistical Methods for Improving the Application of a PSInSAR Technique to Surface Displacement Detection

NCUIR > Engineering College > Graduate Institute of Civil Engineering > Electronic Thesis & Dissertation > Item 987654321/59109

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/59109

Title:	利用統計方法改善永久散射體干涉雷達技術於地表變形偵測之應用;Statistical Methods for Improving the Application of a PSInSAR Technique to Surface Displacement Detection
Authors:	黃郁棠;Huang,Yu-tang
Contributors:	土木工程學系
Keywords:	地表變形速率;差分干涉;永久散射體;雷達影像;Differential InSAR;Persistent Scatterers;surface deformation rate;Synthetic Aperture Radar
Date:	2013-01-25
Issue Date:	2013-03-25 15:46:44 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著近年來遙測技術的發展，光學影像在土地資訊的獲得不只擁有一套完整的流程，且隨著太空科技之發展，光學衛星已不勝枚舉；雖然合成孔徑雷達影像還未開發至極，但雷達影像除可穿透雲層與日夜皆可拍攝外，更有利用干涉雷達製作數位地形模型與全極化影像分類等技術可與光學相輔相成。本研究主要目的為，利用合成孔徑雷達影像的干涉技術量測地形；干涉雷達技術主要係藉由兩個不同位置的接收器，所收到的回波訊號之相位差，來獲取地表三維資訊。當擁有多時期干涉影像即可計算出地表變形速率。更進一步的利用永久散射體技術可以提升量測的精度，並忽略雜訊較多的區域，但台灣地形地貌變化迅速，使得永久散射體在山區、叢林地區的密度相較於都市區較低，而造成永久散射體干涉合成孔徑雷達技術的效果不彰。論文成果使用統計方法，幫助適應性的濾波處理，不只使雜訊降低，同時保留永久散射體的回波正確性，並成功的在永久散射體密度較低的區域，找到大量的分布散射體。最後成功得結合永久散射體與分部散射體的訊號計算出地表變形率。In recent development of remote sensing, there are complete procedures to acquire land information by optical images. And optical satellites have been launched too numerous to mention. Although the Synthetic Aperture Radar (SAR) technology has not yet developed extremely, the advantages of radar images can complement the optical images.The purpose of this research is using Interferometric SAR (InSAR) algorithm to detect the topography. Interferograms were produced from the phase difference between two receivers. When there are interferograms with time series, then the surface deformation rate can be calculated. Further, using the Persistent Scatterers InSAR (PSInSAR) algorithm can improve the accuracy of measurement and ignore the noisy area, when the image quality is not so good. However, there is a limitation on the PS algorithm, the density of measurement points will reduce in mountain area or forest. It is because in those areas the coherence values are typically low.Results of this research come from the statistical method by using the space adaptive filtering. It not only reduces the noise ratio but also keep the accuracy of PS point returns. Finding the Distributed Scatterers (DS) is successful in the area where the density of measurement points is low. Final, the PS and DS points can be jointly processed and then the surface deformation rate is estimated.
Appears in Collections:	[Graduate Institute of Civil Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	737	View/Open

社群 sharing

Loading...