全偏極合成孔徑雷達非監督式目標分類與極化方位角偏移效應估算之研究; Unsupervised Target Classification and Estimation of The Orientation Angle Shift Using Fully Polarimetric SAR

NCUIR > College of Earth Sciences > Graduate Institute of Space Science > Department of Earth Sciences > Item 987654321/5176

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/5176

Title:	全偏極合成孔徑雷達非監督式目標分類與極化方位角偏移效應估算之研究;Unsupervised Target Classification and Estimation of The Orientation Angle Shift Using Fully Polarimetric SAR
Authors:	王邦達;Pang-Ta Wang
Contributors:	太空科學研究所
Keywords:	非監督式分類;極化方位角;全偏極合成孔徑雷達;Orientation Angle Shift;Fully Polarimetric SAR;Unsupervised Target Classification
Date:	2004-05-31
Issue Date:	2009-09-22 09:46:29 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	摘要由於全偏極合成孔徑雷達(Fully Polarimetric SAR) 所獲得的資訊比一般單偏極雷達或多偏極雷達較豐富，因此我們希望利用全偏極合成孔徑雷達觀測影像，以利我們做更完整的資料分析。全偏極雷達資料的運用是雷達遙測領域一個重大的進展。所有被探測目標的資訊皆可以由散射矩陣得到。散射矩陣描述了目標物與電磁波之間複雜的交互作用。矩陣內每一個元素都與電磁波的頻率、極化狀態、物體的形狀與排列方向等因素有關，可知散射矩陣包含了雷達目標物最完整的資訊。本研究中分為兩大部分，一為使用全偏極合成孔徑雷達影像於非監督性分類上，此研究使用目標散射矩陣分解（Target Decomposition）方法，並運用於分類與辨識上，我們對全偏極雷達影像利用目標矩陣分解估算出三個參數Entropy（H）、Alpha Angle（）和Anisotropy（A），最後結合聚類法形成的非監督性分類法，經由嘉義鰲鼓農場的實地調查和比較實驗結果之後，發現分類後的地物類別和實際地物相吻合，加入Anisotropy此參數後，影像中分類類別較多並更接近實際地表覆蓋情況，其中很多原本無法看出的區塊也都一一呈現出來。另一為研究為使用全偏極合成孔徑雷達影像於雷達極化方位角的偏移效應(Polarization Orientation Angle Shift)，當發射機發射訊號觀測目標物時，會因目標物所在的地形坡度而使雷達觀測的回波極化方位角產生偏移，故本次研究利用全偏極雷達影像的特性，並配合圓形極化演算法(Circular Polarization Algorithm)，此演算法可藉由圓形極化的方式計算出回波極化方位角的偏移，再由此雷達收到回波極化方位角的偏移了解地形原有的紋理。此外，並利用數位高層模型(Digital Elevation Model)模擬此地形坡度，以利比較回波極化方位角偏移的估算結果。而在實驗之後發現台灣地區全偏極合成孔徑雷達影像資料的估算結果並不易察覺，我們推測可能影響的因素包括地形坡度與觀測角之影響、植被覆蓋與雷達頻率的影響和影像資料校正，對於估算雷達極化方位角之偏移效應時都具有決定性的影響。 Abstract This study included two parts of using fully Polarimetric Synthetic Aperture Radar (POLSAR) data. First, classification of Earth terrain components using a POLSAR image which is one of the important applications of Radar Polarimetric; second, we used POLSAR to measure azimuth slopes that are related to shifts in polarization orientation angle. Polarization orientation angle is one of the parameters among the wealth of polarimetric information when analyzing POLSAR data. At first, we used a method for unsupervised classification of terrain types and man-made objects by POLSAR data. This technique is a combination of the unsupervised classification based on Polarimetric target decomposition and the maximum likelihood classifier based on the complex Wishart distribution for the Polarimetric covariance matrix. This unsupervised classification based on the use of two-dimensional Entropy (H) / Alpha angle (α) classification plane, where all random scattering mechanisms can be represented. Then, we appended the Anisotropy (A) information to the unsupervised classification combined Wishart classifier. After comparing the Au-Gu farm ground truth map with our classification result, the experiment results agreed with ground truth map. We introduced the Anisotropy (A) information, and found it allowed the improvement of the capability to distinguish between different classes whose cluster centers end in the same Entropy (H) – Alpha (α) zone. Second, we used POLSAR to measure azimuth slopes. When SAR images a rugged terrain area, surface slopes have two main effects on the SAR image response. The first is the change of radar cross section per unit image area, and the second is that polarization states are also affected, because azimuth slopes induce polarization orientation changes. We used the polarization characteristic, and the circular Polarization algorithm to estimate polarization orientation angle shifts. For comparison, the orientation angles from a digital elevation model (DEM) generated using C-band interferometry. However, the experiment results in TAIWAN POLSAR data sets were very noisy. At last, we listed and discussed some possible sources that affected the experiment results.
Appears in Collections:	[Graduate Institute of Space Science] Department of Earth Sciences

Files in This Item:

File	Size	Format

社群 sharing

Loading...