本文選題：極化SAR分類 + Wishart分布�。� 參考：《西安科技大學(xué)》2013年碩士論文

【摘要】：極化SAR影像分類是極化SAR應(yīng)用中一項(xiàng)非常重要的研究問題。全極化數(shù)據(jù)多視極化相干矩陣或協(xié)方差矩陣所服從的復(fù)Wishart分布是目前極化SAR影像分類領(lǐng)域應(yīng)用最廣泛、最著名的統(tǒng)計(jì)模型。但應(yīng)用該模型需要假定目標(biāo)的散射分量服從復(fù)高斯分布，這使得復(fù)Wishart模型與異質(zhì)性區(qū)域數(shù)據(jù)的匹配效果較差。為解決這一問題，本文研究基于K-Wishart分布模型的極化SAR影像分類。K-Wishart模型基于乘積模型，由Wishart相干斑模型和Gamma紋理分布模型推導(dǎo)而得，該模型將極化信息與紋理信息進(jìn)行融合，，適應(yīng)了不同條件下的場(chǎng)景描述。 本研究以山東泰安地區(qū)徂徠山一帶作為實(shí)驗(yàn)區(qū)，獲取了覆蓋實(shí)驗(yàn)區(qū)的一景ALOSPALSAR全極化數(shù)據(jù)，一景TM數(shù)據(jù)和土地類型覆蓋數(shù)據(jù)。 首先，通過引入紋理參量，本文發(fā)展了一種具有非高斯性質(zhì)的統(tǒng)計(jì)模型，即K-Wishart統(tǒng)計(jì)模型。將基于有限混合K-Wishart模型和基于有限混合Wishart模型的極化SAR影像進(jìn)行分類對(duì)比，結(jié)果發(fā)現(xiàn)，分類精度由74.1935%提高到了88.9276%，且在一定程度上抑制了地形起伏的影響。 其次，本文還發(fā)展了一種新的參數(shù)估計(jì)方法，即基于Mellin變換的矩陣對(duì)數(shù)累積量參數(shù)估計(jì)法，對(duì)極化SAR影像的K-Wishart分布紋理參數(shù)進(jìn)行估計(jì)，不僅簡(jiǎn)化了參數(shù)估計(jì)過程，而且提高了估計(jì)性能。 此外，本文還提出一種基于有限混合Gamma模型的初始化方法。該方法不僅保證了Wishart和K-Wishart兩種分類器進(jìn)行對(duì)比所需要的同等初始條件，而且還降低了時(shí)間復(fù)雜度。
[Abstract]:The classification of polarized SAR images is a very important research problem in the application of polarized SAR. The complex Wishart distribution followed by multi-polarimetric coherent matrix or covariance matrix of fully polarized data is the most widely used and most famous statistical model in the field of polarimetric SAR image classification. However, it is necessary to assume that the scattering component of the target is distributed from the complex Gao Si, which makes the matching effect between the complex Wishart model and the heterogeneity region data poor. In order to solve this problem, this paper studies the classification of polarized SAR images based on K-Wishart distribution model. K-Wishart model is derived from Wishart speckle model and Gamma texture distribution model based on product model. It adapts to the scene description under different conditions. Taking Culai Mountain area in Tai'an area as the experimental area, a scene ALOSPALSAR full polarization data, a scene TM data and land type coverage data were obtained in this study. Firstly, by introducing texture parameters, we develop a non-Gao Si statistical model, that is, K-Wishart statistical model. The polarimetric SAR images based on finite mixed K-Wishart model and finite mixed Wishart model are compared. The results show that the classification accuracy is increased from 74.1935% to 88.9276, and the effect of terrain fluctuation is restrained to some extent. Secondly, this paper also develops a new parameter estimation method, which is matrix logarithmic cumulant parameter estimation method based on Mellin transform, which not only simplifies the process of parameter estimation, but also estimates the texture parameters of K-Wishart distribution in polarized SAR images. Moreover, the estimation performance is improved. In addition, an initialization method based on finite hybrid Gamma model is proposed. This method not only guarantees the same initial conditions required for the comparison of Wishart and K-Wishart classifiers, but also reduces the time complexity.
【學(xué)位授予單位】：西安科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TN957.52;P225.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 吳永輝;計(jì)科峰;李禹;郁文賢;;基于Wishart分布和MRF的多視全極化SAR圖像分割[J];電子學(xué)報(bào);2007年12期

2 曹芳;洪文;吳一戎;;基于Cloude-Pottier目標(biāo)分解和聚合的層次聚類算法的全極化SAR數(shù)據(jù)的非監(jiān)督分類算法研究[J];電子學(xué)報(bào);2008年03期

3 談璐璐;楊汝良;商建;;利用Shannon熵參數(shù)的極化干涉SAR圖像非監(jiān)督分類[J];電子學(xué)報(bào);2010年10期

4 陳強(qiáng);蔣詠梅;陸軍;匡綱要;;基于目標(biāo)散射相似性的POLSAR圖像無(wú)監(jiān)督地物散射分類新方案[J];電子學(xué)報(bào);2010年12期

5 楊震,楊汝良,劉秀清;SAR圖像的極化干涉非監(jiān)督Wishart分類方法和實(shí)驗(yàn)研究[J];電子與信息學(xué)報(bào);2004年05期

6 吳永輝;計(jì)科峰;郁文賢;;基于H-α和改進(jìn)C-均值的全極化SAR圖像非監(jiān)督分類[J];電子與信息學(xué)報(bào);2007年01期

7 譚衢霖,邵蕓;雷達(dá)遙感圖像分類新技術(shù)發(fā)展研究[J];國(guó)土資源遙感;2001年03期

8 高旅端,陳志,王家潤(rùn);一種加速EM算法收斂的方法[J];數(shù)理統(tǒng)計(jì)與應(yīng)用概率;1998年04期

9 鄒同元;楊文;代登信;孫洪;;一種新的極化SAR圖像非監(jiān)督分類算法研究[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2009年08期

10 高貴;;SAR圖像統(tǒng)計(jì)建模研究綜述[J];信號(hào)處理;2009年08期

本文編號(hào)：1883562