本文選題：高光譜圖像 + 亞像元　； 參考：《南京理工大學(xué)》2014年碩士論文

【摘要】：高光譜圖像光譜分辨率高,具有圖譜合一的特性,能夠提供區(qū)分不同物質(zhì)的診斷性光譜信息,結(jié)合該光譜信息可提高對(duì)目標(biāo)和背景進(jìn)行定量分析的能力,因此高光譜目標(biāo)檢測(cè)技術(shù)在目標(biāo)檢測(cè)領(lǐng)域具有獨(dú)特的優(yōu)勢(shì)。由于地物分布情況復(fù)雜和成像光譜儀空間分辨率的較低等原因,待檢測(cè)的目標(biāo)通常與其他地物共同組成混合像元,此時(shí)目標(biāo)以亞像元形式存在。高光譜亞像元目標(biāo)檢測(cè)是目標(biāo)檢測(cè)研究的前沿和難點(diǎn),本文著眼于如何利用高光譜數(shù)據(jù)的稀疏性提高檢測(cè)效果,對(duì)高光譜圖像亞像元目標(biāo)檢測(cè)技術(shù)進(jìn)行了研究,主要工作和成果如下： 1.研究了高光譜遙感圖像的光譜混合模型,詳細(xì)的介紹了多元信號(hào)估計(jì)和信號(hào)檢測(cè)理論,推導(dǎo)了四種經(jīng)典的亞像元目標(biāo)檢測(cè)算法——約束能量最小化方法(CEM)、基于加權(quán)樣本自相關(guān)矩陣的CEM、正交子空間投影算法(OSP)、適應(yīng)匹配子空間檢測(cè)算法(AMSD),并在第四章用實(shí)驗(yàn)證明了這些算法的可行性。 2.給出了基于稀疏約束的線性混合光譜分解模型,推導(dǎo)了四種經(jīng)典的稀疏性解混算法——正交匹配追蹤算法(OMP)、迭代光譜混合分析算法(ISMA)、變量分離的增廣拉格朗日算法(SUNSAL)、基于加權(quán)L1正則化的SUNSAL算法,給出了這四種算法的具體實(shí)現(xiàn)步驟,并提出了基于L1/2正則化的稀疏性解混方法。實(shí)驗(yàn)證明基于L1/2正則化的稀疏性解混方法在圖像信噪比較小的情況下性能比其他四種算法好而且更加穩(wěn)定。 3.利用高光譜數(shù)據(jù)的稀疏性進(jìn)行目標(biāo)檢測(cè)。本文將稀疏性分解算法和自適應(yīng)匹配子空間檢測(cè)算法相結(jié)合提出了SU-AMSD算法,并介紹了該算法的基本流程與具體實(shí)現(xiàn)。為了提高目標(biāo)檢測(cè)的效率與精度,本文對(duì)該算法進(jìn)行了改進(jìn),提出了一種基于Lib-IEA的SU-AMSD算法。實(shí)驗(yàn)證明,與上面介紹的四種經(jīng)典亞像元目標(biāo)檢測(cè)算法相比基于光譜庫(kù)和IEA算法的SU-AMSD算法的效果更好,而基于Lib-IEA的SU-AMSD算法則沒(méi)有達(dá)到預(yù)期的目標(biāo)。
[Abstract]:The spectral resolution of hyperspectral image is high, and it has the characteristic of unifying the spectrum, which can provide diagnostic spectral information that distinguishes different substances. Combined with this spectral information, the ability of quantitative analysis of target and background can be improved. Therefore, hyperspectral target detection technology has a unique advantage in the field of target detection. Due to the complex distribution of ground objects and the low spatial resolution of the imaging spectrometer, the target to be detected usually forms a mixed pixel with other ground objects, and the target exists in the form of sub-pixel. Hyperspectral sub-pixel target detection is the frontier and difficulty of target detection. This paper focuses on how to use the sparsity of hyperspectral data to improve the detection effect, and studies the sub-pixel target detection technology of hyperspectral image. The main work and results are as follows: 1. The spectral mixing model of hyperspectral remote sensing image is studied, and the theory of multivariate signal estimation and signal detection is introduced in detail. In this paper, four classical subpixel target detection algorithms, constrained energy minimization method, CEM based on weighted sample autocorrelation matrix, orthogonal subspace projection algorithm, adaptive matching subspace detection algorithm are derived. In chapter 4, we use the Experiments show that these algorithms are feasible. 2. A linear mixed spectral decomposition model based on sparse constraints is presented. Four classical sparse demultiplexing algorithms, orthogonal matching tracking algorithm, iterative spectral mixed analysis algorithm, extended Lagrangian algorithm with variable separation, SUNSAL algorithm based on weighted L1 regularization, are derived. The implementation steps of the four algorithms are given, and a sparse demultiplexing method based on L 1 / 2 regularization is proposed. Experiments show that the sparse demultiplexing method based on L1 / 2 regularization is better and more stable than the other four algorithms in the case of low SNR. 3. Target detection is carried out by using the sparsity of hyperspectral data. In this paper, SU-AMSD algorithm is proposed by combining sparse decomposition algorithm with adaptive matching subspace detection algorithm, and its basic flow and implementation are introduced. In order to improve the efficiency and accuracy of target detection, this paper improves the algorithm and proposes a SU-AMSD algorithm based on Lib-IEA. Experimental results show that the SU-AMSD algorithm based on spectral library and IEA algorithm is more effective than the four classical subpixel target detection algorithms mentioned above, but the SU-AMSD algorithm based on Lib-IEA does not achieve the desired goal.
【學(xué)位授予單位】：南京理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 孫林;鮑金河;劉一超;;高光譜圖像目標(biāo)檢測(cè)算法分析[J];測(cè)繪科學(xué);2012年01期

2 賀霖;潘泉;趙永強(qiáng);;量測(cè)重構(gòu)線性混合模型高光譜圖像目標(biāo)檢測(cè)[J];電子學(xué)報(bào);2007年01期

3 胡秀清;劉京晶;邱康睦;范天錫;張玉香;戎志國(guó);張立軍;;神舟3號(hào)飛船中分辨率成像光譜儀場(chǎng)地替代定標(biāo)新方法研究[J];光譜學(xué)與光譜分析;2009年05期

4 賀霖,潘泉,趙永強(qiáng),鄭紀(jì)偉,魏坤;基于波段子集特征融合的高光譜圖像異常檢測(cè)[J];光子學(xué)報(bào);2005年11期

5 賀霖;潘泉;趙永強(qiáng);鄭紀(jì)偉;;基于子空間投影的未知背景航拍高光譜圖像恒虛警目標(biāo)檢測(cè)[J];航空學(xué)報(bào);2006年04期

6 董廣軍;張永生;范永弘;;PHI高光譜數(shù)據(jù)和高空間分辨率遙感圖像融合技術(shù)研究[J];紅外與毫米波學(xué)報(bào);2006年02期

7 李慶利;薛永祺;王建宇;白智全;;PHI高光譜圖像的大氣校正算法[J];紅外與毫米波學(xué)報(bào);2006年04期

8 賀霖;潘泉;邸椺;;自適應(yīng)結(jié)構(gòu)化背景和形狀特征子空間高光譜圖像多類(lèi)目標(biāo)檢測(cè)[J];紅外與毫米波學(xué)報(bào);2007年05期

9 耿修瑞;趙永超;;高光譜遙感圖像小目標(biāo)探測(cè)的基本原理[J];中國(guó)科學(xué)(D輯:地球科學(xué));2007年08期

10 吳澤彬;韋志輝;孫樂(lè);劉建軍;;基于迭代加權(quán)L1正則化的高光譜混合像元分解[J];南京理工大學(xué)學(xué)報(bào);2011年04期

相關(guān)博士學(xué)位論文 前1條

1 杜博;高光譜遙感影像亞像元小目標(biāo)探測(cè)研究[D];武漢大學(xué);2010年

，

本文編號(hào)：1922918