本文關(guān)鍵詞： 高光譜圖像 拉普拉斯支持向量機(jī) 半監(jiān)督學(xué)習(xí)　出處：《廈門大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：基于高光譜圖像的地物分類是遙感領(lǐng)域的研究熱點(diǎn)。高光譜遙感數(shù)據(jù)最主要的特點(diǎn)是：將傳統(tǒng)的圖象維與光譜維信息融合為一體,在獲取地表空間圖象的同時(shí),得到每個(gè)地物的連續(xù)光譜信息,從而實(shí)現(xiàn)依據(jù)地物光譜特征的地物成份信息反演與地物識(shí)別。高光譜內(nèi)部存在著流形結(jié)構(gòu)和內(nèi)部數(shù)據(jù)空間信息等特點(diǎn)。高光譜標(biāo)記樣本少而昂貴,高光譜的維度高且數(shù)據(jù)量很大,使得對(duì)高光譜的數(shù)據(jù)處理出現(xiàn)問題。本文從高光譜數(shù)據(jù)特點(diǎn)入手,對(duì)高光譜地物分類進(jìn)行分析。主要研究成果如下： 首先,引入拉普拉斯支持向量機(jī)對(duì)高光譜數(shù)據(jù)進(jìn)行地物分類。普拉斯算子能夠?qū)⒏呔S的高光譜數(shù)據(jù)映射到低維空間中進(jìn)行非線性降維,發(fā)現(xiàn)內(nèi)在的流形結(jié)構(gòu)。因此,將拉普拉斯應(yīng)用于高光譜分類能有效的利用高光譜的特點(diǎn)使地物分類達(dá)到更高的準(zhǔn)確度。在解決方案中,初始預(yù)共軛梯度優(yōu)化方法和對(duì)偶方法是兩個(gè)不同的解法。預(yù)共軛梯度解法的拉普拉斯支持向量機(jī)解法在原始方法的對(duì)偶框架上加入了基于預(yù)測(cè)穩(wěn)定性的梯度下降早期停止條件。本文將拉普拉斯預(yù)共軛梯度引入到高光譜圖像的地物分類當(dāng)中,提出了利用預(yù)共軛梯度解法來(lái)對(duì)高光譜數(shù)據(jù)進(jìn)行地物分類,極大地減少了訓(xùn)練和分類時(shí)間以及復(fù)雜度。、 其次,針對(duì)分類器參數(shù)優(yōu)化方面,本文提出了模擬退火算法對(duì)各類分類器進(jìn)行參數(shù)的優(yōu)化。在拉普拉斯的三個(gè)參數(shù)即核參數(shù)、懲罰因子、拉普拉斯因子的設(shè)置方面,模擬退火算法對(duì)參數(shù)進(jìn)行優(yōu)化,與傳統(tǒng)的網(wǎng)格搜索優(yōu)化算法相比節(jié)省了大量的時(shí)間,特別是存在多個(gè)參數(shù)的情況下,較少了計(jì)算復(fù)雜度。 本文通過利用Indian Pine數(shù)據(jù)集中的六類地物類別進(jìn)行分類試驗(yàn),同時(shí)對(duì)比了其他不同種類的分類器,顯示了本論文所提出方法的優(yōu)越性和分類的精確性。
[Abstract]:The classification of ground objects based on hyperspectral images is a hotspot in the field of remote sensing. The most important feature of hyperspectral remote sensing data is that the traditional image dimension and spectral dimension information are integrated to obtain the surface spatial image at the same time. Get the continuous spectral information for each object. In order to realize the retrieval and recognition of the composition information of the ground objects according to the spectral characteristics of the ground objects, the hyperspectral interior has the characteristics of manifold structure and internal data spatial information, and the hyperspectral labeling samples are small and expensive. The hyperspectral dimension is high and the amount of data is very large, which makes the processing of hyperspectral data appear problems. This paper starts with the characteristics of hyperspectral data, and analyzes the classification of hyperspectral features. The main research results are as follows: Firstly, Laplace support vector machine is introduced to classify hyperspectral data. Plath operator can map high-dimensional hyperspectral data to low-dimensional space for nonlinear dimensionality reduction. Therefore, the application of Laplace to hyperspectral classification can effectively make use of the characteristics of hyperspectral to achieve higher accuracy in the classification of ground objects. The initial preconjugate gradient optimization method and the dual method are two different solutions. The Laplace support vector machine method for the preconjugate gradient method adds gradient descent based on predictive stability to the dual frame of the original method. In this paper, Laplace preconjugate gradient is introduced into the classification of ground objects in hyperspectral images. A preconjugate gradient method is proposed to classify the ground objects of hyperspectral data, which greatly reduces the training and classification time and complexity. Secondly, for the optimization of classifier parameters, this paper proposes simulated annealing algorithm to optimize the parameters of all kinds of classifiers. In Laplacian, three parameters, kernel parameters, penalty factor. In the setting of Laplace factor, the simulated annealing algorithm optimizes the parameters, which saves a lot of time compared with the traditional mesh search optimization algorithm, especially in the case of multiple parameters. Less computational complexity. In this paper, by using the Indian Pine data set of the six categories of ground objects classification experiments, at the same time compared with other different types of classifiers. The advantages of the proposed method and the accuracy of classification are demonstrated.
【學(xué)位授予單位】：廈門大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP751;TP181

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 杜培軍;林卉;孫敦新;;基于支持向量機(jī)的高光譜遙感分類進(jìn)展[J];測(cè)繪通報(bào);2006年12期

2 楊國(guó)鵬;余旭初;劉偉;陳偉;;面向高光譜遙感影像的分類方法研究[J];測(cè)繪通報(bào);2007年10期

3 堯德中;劉鐵軍;雷旭;楊平;徐鵬;張楊松;;基于腦電的腦-機(jī)接口:關(guān)鍵技術(shù)和應(yīng)用前景[J];電子科技大學(xué)學(xué)報(bào);2009年05期

4 趙春霞,錢樂祥;遙感影像監(jiān)督分類與非監(jiān)督分類的比較[J];河南大學(xué)學(xué)報(bào)(自然科學(xué)版);2004年03期

5 譚琨;杜培軍;;基于支持向量機(jī)的高光譜遙感圖像分類[J];紅外與毫米波學(xué)報(bào);2008年02期

6 孟海東;郝永寬;宋飛燕;格日勒?qǐng)D;;遙感圖像非監(jiān)督計(jì)算機(jī)分類方法的研究[J];計(jì)算機(jī)與現(xiàn)代化;2008年07期

7 陳毅松,汪國(guó)平,董士海;基于支持向量機(jī)的漸進(jìn)直推式分類學(xué)習(xí)算法[J];軟件學(xué)報(bào);2003年03期

8 麻永平;張煒;劉東旭;;高光譜偵察技術(shù)特點(diǎn)及其對(duì)地面軍事目標(biāo)威脅分析[J];上海航天;2012年01期

9 包芳;潘永惠;;神經(jīng)網(wǎng)絡(luò)泛化性能優(yōu)化算法[J];計(jì)算機(jī)工程與設(shè)計(jì);2009年22期

10 蘇紅軍;杜培軍;;高光譜數(shù)據(jù)特征選擇與特征提取研究[J];遙感技術(shù)與應(yīng)用;2006年04期

相關(guān)博士學(xué)位論文 前1條

1 高恒振;高光譜遙感圖像分類技術(shù)研究[D];國(guó)防科學(xué)技術(shù)大學(xué);2011年

，

本文編號(hào)：1478550