目標物點狀散射點特徵在合成孔徑雷達目標物影像辨識上非常重要，因此找尋合適的萃取方法就十分關鍵，目標物點狀特徵的萃取原則在於維持主波瓣的強度，並降低旁波瓣的影響，並且能有效反應真實目標物散射點位置。藉由這些資訊我們可以有效利用在目標物辨識、分類上。其中本研究以增顯目標物點狀散射點為主要目標。 合成孔徑雷達影像依取像方式可分成兩類，焦點式和掃描式。本研究主要是由M. Çetin 所提出針對焦點式合成孔徑雷達增顯地表反射係數的演算邏輯為基礎，其設計理念是以回波訊號方程式為基礎，納入強烈假設的相加性雜訊在演算的模型上，並以雷達回波訊號的指數項，建構成為演算法之轉換核心矩陣，再針對這個模型作最佳化的運算。為了將原有限定在焦點式合成孔徑雷達的地表參數運算擴張到掃描式合成孔徑雷達的資料上，本研究以掃描式的訊號接收數學式重新建構出一個新的演算法核心矩陣，並將散射點增顯後的結果簡化成目標物點狀散射點相對位置的圖形，方便未來研究上的應用。 本研究使用調控演算法(Regularization Algorithm)增顯散射點特徵，其好處在於可以視使用者的需要來增加代價方程式(Cost function)的二次或非二次項，由於合成孔徑雷達的資料為複數資料，求解過程以複數共軛梯度為主要概念並配合準牛頓演算法(Quasi-Newton iteration)和柯列斯基分解(Cholesky decomposition)處理最佳化的問題。由於使用兩種資料形式的不同，本研究試著將掃描式的處理以切割的方式來完成，並先尋找背景的標準差為區分背景區域和不標物區域的準則，防止區域極大值的產生，並增加演算的速度，因此配合數個地真調查的目標物在數個參數上重新定義，本研究使用RADARSAT高解析度掃描式合成孔徑雷達影像為測試資料，並配合地真調查的結果來佐證經過演算法增顯出的目標物點狀散射點特徵，並也使用NASA/JPL AIRSAR全偏極資料為第二組例子。 其後會與傳統方法MV(Minimum variance)和MUSIC (Multiple Signal Classification)的效能作比較，展示散射點增強的結果並使用評鑑指標如目標物對背景混亂物比、主波瓣3dB寬和實際運算時間，評鑑效能且列表表示。結果顯示在經過規則化適當調整演算法參數之後，無論是在評鑑指標或是在統計機率密度分佈圖上，目標點旁波瓣的抑制和目標物與背景分離上皆有明顯的效果。 Target identification and recognition of SAR images require good feature selection and enhancement. Due to the coherent process, it is difficult to discriminate the SAR target feature properties simply using the shape, shadow, tone, color and texture, to name a few. The scattering center is one of the important properties for extracting the SAR feature. This involves working on raw data (amplitude and phase) as part of image formation. In this paper, we modified an algorithm based a conjugate gradient (CG) optimum method originally proposed for spot mode SAR images, in order to enhance the Stripmap mode SAR targets. The SAR image reconstruction method for spotlight mode uses the range profile data that displays on spatial frequency domain in polar form. By the conventional method it is resampled to be in rectangular form, followed by using inverse Fourier transform with properly weighted window. In papers the recompiling method is done by introducing a projection operator kernel. Directly applying CG method, one has to change the data format to range profile of polar form but suffer from information loss in this recompiling processing. To avoid this information degradation, it has been suggested to replace SAR projection operator kernel by Fourier transform kernel, and let it have suitability for Stripmap mode data. To validate the effectiveness and efficiency of the modified method, a series of RADARSAT SAR images at fine mode were tested with ground truth available overpass the image acquisition. In addition to, we import the fully polarization SAR data from NASA/JPL AIRSAR acquired from south Taiwan when Sep. 27, 2000. We also compared the performance with MV (Minimum Variance) and MUSIC (Multiple Signal Classification) methods. Performance indices include target to clutter ratio, 3 dB mainlobe width and CPU time. From the logarithmic probability density distribution and column ordering plot of enhanced image and original image, it was demonstrated that modified method provides the best performance among the three methods besides CPU time.
