基于核密度估計的光譜數(shù)據(jù)分類與回歸方法研究
發(fā)布時間：2019-01-01 13:05
【摘要】：本文針對遙感和傅立葉紅外透射兩種不同形式的光譜進(jìn)行了基于核密度估計的高光譜遙感數(shù)據(jù)分類和傅立葉紅外透射光譜回歸分析研究。高光譜遙感數(shù)據(jù)分類面臨的難點(diǎn)就是如何有效處理譜線波段強(qiáng)相關(guān)問題，而傅里葉紅外透射光譜回歸分析的難點(diǎn)在于如何準(zhǔn)確識別譜線中的隱含峰。已有的基于譜線匹配和統(tǒng)計特征（例如支持向量機(jī)）的高光譜分類方法往往忽略了譜線波段間的相關(guān)性，從而限制了這些分類方法的表現(xiàn)。同時，由于傳統(tǒng)的基于最小二乘法的擬合方法存在正規(guī)方程組病態(tài)化的缺陷，使其往往無法識別譜線中的隱含峰，從而導(dǎo)致不精確的光譜數(shù)據(jù)分析結(jié)果。為了有效地解決上述光譜數(shù)據(jù)分類和回歸分析中面臨的難點(diǎn)和問題，本文從分析光譜數(shù)據(jù)的概率密度函數(shù)分布入手，主要開展了以下三部分的研究工作，提出了基于等效概率的靈活樸素貝葉斯分類器FNBEPNSK、基于聯(lián)合概率密度函數(shù)估計的非樸素貝葉斯分類器NNBC以及基于模糊積分的核回歸集成算法KREFI。 1）研究了Gaussian，Uniform，Triangular，Epanechnikov，Biweight，Triweight和Cosine七種不同的核函數(shù)對基于邊緣概率密度函數(shù)估計的樸素貝葉斯分類器表現(xiàn)的影響，其中Gaussian核為光滑核函數(shù)，后六種為非光滑核函數(shù)。通過分析這七種不同核函數(shù)在概率密度函數(shù)估計中的效能，給出了非光滑核函數(shù)的作用條件，并針對非光滑核函數(shù)在概率密度函數(shù)估計中的缺點(diǎn)，提出了基于等效概率的靈活樸素貝葉斯分類器FNBEPNSK。在標(biāo)準(zhǔn)的UCI數(shù)據(jù)集和真實(shí)高光譜數(shù)據(jù)集上的驗證結(jié)果表明，等效概率的應(yīng)用顯著改善了樸素貝葉斯分類器的分類表現(xiàn)。 2）為了有效處理樣本條件屬性之間的相關(guān)性，我們將聯(lián)合概率密度估計的思想引入到了高光譜遙感數(shù)據(jù)分類器的設(shè)計中，提出了基于聯(lián)合概率密度估計的非樸素貝葉斯分類器NNBC。為了確定聯(lián)合概率密度估計中的帶寬參數(shù)，本文設(shè)計了基于積分均方誤差最小化的參數(shù)選擇標(biāo)準(zhǔn)，保證了最佳帶寬參數(shù)的選取，并通過與基于積分方差最小化的參數(shù)選擇標(biāo)準(zhǔn)比較證實(shí)其有效性。同時，理論證明了當(dāng)屬性之間存在強(qiáng)相關(guān)時聯(lián)合概率密度函數(shù)估計的最優(yōu)性。最后，，在UCI數(shù)據(jù)集以及真實(shí)高光譜數(shù)據(jù)集上的實(shí)驗結(jié)果表明非樸素貝葉斯分類器在獲得較高概率密度函數(shù)估計質(zhì)量的同時，顯著地提升了樸素貝葉斯分類器的分類精度。 3）針對硅基薄膜傅里葉紅外透射光譜曲線波動頻率大、隱含峰難識別的特點(diǎn)設(shè)計了一款能夠充分考慮光譜數(shù)據(jù)概率分布信息、具有高穩(wěn)定性和高準(zhǔn)確度的核回歸集成方法。通過在6個標(biāo)準(zhǔn)的測試函數(shù)上的實(shí)驗比較，證實(shí)了基于交叉驗證帶寬選擇策略Priestley-Chao核回歸器PCKE1和PCKE2的高方差的特性。之后，設(shè)計了基于模糊積分的核回歸集成模型KREFI對四種不同的Priestley-Chao核回歸器進(jìn)行了融合以提高核回歸算法的穩(wěn)定性，其中模糊積分中的模糊測度使用了三種不同的粒子群優(yōu)化算法進(jìn)行確定。最后，在標(biāo)準(zhǔn)的測試函數(shù)以及28條不同形式的硅基薄膜傅里葉紅外透射光譜上對核回歸集成算法KREFI的表現(xiàn)進(jìn)行了驗證，結(jié)果表明KREFI獲得了良好的回歸表現(xiàn)，在一定程度上解決了譜線隱含峰無法識別的問題。
[Abstract]:In this paper, a high-spectral remote sensing data classification and a Fourier infrared transmission spectrum regression analysis based on nuclear density estimation are carried out for two different forms of spectrum of remote sensing and Fourier infrared transmission. The difficulty of the high-spectral remote sensing data classification is how to effectively deal with the strong correlation of the spectral line, and the difficulty of the Fourier infrared transmission spectrum regression analysis is how to accurately identify the hidden peaks in the spectral line. the existing high-spectral classification methods based on spectral line matching and statistical features, such as support vector machines, tend to ignore the correlation between the spectral line bands, thereby limiting the performance of these classification methods. At the same time, because the traditional method of fitting least square method has the defect that the normal set of equations is ill, it is often unable to identify the hidden peaks in the spectral line, resulting in inaccurate spectral data analysis results. In order to effectively solve the difficulties and problems of the above-mentioned spectral data classification and regression analysis, this paper starts with the analysis of the probability density function distribution of the spectral data, mainly carries out the research work of the following three parts, and puts forward the flexible and simple Bayesian classifier FNBEPNSK based on the equivalent probability. The non-naive Bayesian classifier NNBC based on the joint probability density function estimation and the kernel regression integration algorithm KREFI based on the fuzzy integral. 1) The effect of seven different kernel functions of Gaussian, Uniform, Triangular, Epantechikov, Biweght, Triweight and Cosine on the performance of the naive Bayesian classifier based on the estimation of the edge probability density function is studied. By analyzing the performance of the seven different kernel functions in the estimation of the probability density function, the working condition of the non-smooth kernel function is given, and the shortcoming of the non-smooth kernel function in the estimation of the probability density function is given, and the flexible and simple Bayesian classifier FNBEPNS based on the equivalent probability is proposed. K. The results of the verification on the standard UCI data set and the real high spectrum data set show that the application of the equivalent probability significantly improves the classification table of the Naive Bayes classifier (2) In order to effectively deal with the correlation between sample condition attributes, we introduce the idea of joint probability density estimation into the design of high-spectral remote sensing data classifier, and put forward the non-naive Bayesian classifier N based on joint probability density estimation. In order to determine the bandwidth parameters in the joint probability density estimation, this paper designs a parameter selection criterion based on the minimization of the integral mean square error, guarantees the selection of the best bandwidth parameter, and compares it with the parameter selection criteria based on the minimization of the integral variance. validity. At the same time, the theory proves that the joint probability density function is estimated when there is a strong correlation between the attributes Finally, on the UCI data set and the real high spectrum data set, the experimental results show that the non-naive Bayesian classifier significantly improves the score of the Naive Bayes classifier while the higher probability density function estimation quality is obtained. Based on the characteristics of the large fluctuation frequency of the Fourier infrared transmission spectrum of the silicon-based thin-film and the difficult recognition of the implicit peak, a kind of kernel back-back with high stability and high accuracy can be considered in full consideration of the characteristic of the spectral data probability distribution information. Based on the experimental comparison of the six standard test functions, the high performance of the Priestley-Chao kernel regression PCKE1 and PCKE2 based on the cross-verification bandwidth selection strategy is proved. Then, a kernel regression integration model KREFI based on fuzzy integral is designed to fuse four different Priestley-Chao nucleators to improve the stability of the kernel regression algorithm, in which the fuzzy measure in the fuzzy integral uses three different particle swarm optimization. Finally, the performance of the kernel regression integration algorithm KREFI is verified by the standard test function and 28 different forms of the silicon-based thin film Fourier infrared transmission spectrum. The result shows that the KREFI has a good regression performance, and the hidden peak of the spectral line cannot be solved to a certain extent.
【學(xué)位授予單位】：河北大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：O438.2;TP79
本文編號：2397607
[Abstract]:In this paper, a high-spectral remote sensing data classification and a Fourier infrared transmission spectrum regression analysis based on nuclear density estimation are carried out for two different forms of spectrum of remote sensing and Fourier infrared transmission. The difficulty of the high-spectral remote sensing data classification is how to effectively deal with the strong correlation of the spectral line, and the difficulty of the Fourier infrared transmission spectrum regression analysis is how to accurately identify the hidden peaks in the spectral line. the existing high-spectral classification methods based on spectral line matching and statistical features, such as support vector machines, tend to ignore the correlation between the spectral line bands, thereby limiting the performance of these classification methods. At the same time, because the traditional method of fitting least square method has the defect that the normal set of equations is ill, it is often unable to identify the hidden peaks in the spectral line, resulting in inaccurate spectral data analysis results. In order to effectively solve the difficulties and problems of the above-mentioned spectral data classification and regression analysis, this paper starts with the analysis of the probability density function distribution of the spectral data, mainly carries out the research work of the following three parts, and puts forward the flexible and simple Bayesian classifier FNBEPNSK based on the equivalent probability. The non-naive Bayesian classifier NNBC based on the joint probability density function estimation and the kernel regression integration algorithm KREFI based on the fuzzy integral. 1) The effect of seven different kernel functions of Gaussian, Uniform, Triangular, Epantechikov, Biweght, Triweight and Cosine on the performance of the naive Bayesian classifier based on the estimation of the edge probability density function is studied. By analyzing the performance of the seven different kernel functions in the estimation of the probability density function, the working condition of the non-smooth kernel function is given, and the shortcoming of the non-smooth kernel function in the estimation of the probability density function is given, and the flexible and simple Bayesian classifier FNBEPNS based on the equivalent probability is proposed. K. The results of the verification on the standard UCI data set and the real high spectrum data set show that the application of the equivalent probability significantly improves the classification table of the Naive Bayes classifier (2) In order to effectively deal with the correlation between sample condition attributes, we introduce the idea of joint probability density estimation into the design of high-spectral remote sensing data classifier, and put forward the non-naive Bayesian classifier N based on joint probability density estimation. In order to determine the bandwidth parameters in the joint probability density estimation, this paper designs a parameter selection criterion based on the minimization of the integral mean square error, guarantees the selection of the best bandwidth parameter, and compares it with the parameter selection criteria based on the minimization of the integral variance. validity. At the same time, the theory proves that the joint probability density function is estimated when there is a strong correlation between the attributes Finally, on the UCI data set and the real high spectrum data set, the experimental results show that the non-naive Bayesian classifier significantly improves the score of the Naive Bayes classifier while the higher probability density function estimation quality is obtained. Based on the characteristics of the large fluctuation frequency of the Fourier infrared transmission spectrum of the silicon-based thin-film and the difficult recognition of the implicit peak, a kind of kernel back-back with high stability and high accuracy can be considered in full consideration of the characteristic of the spectral data probability distribution information. Based on the experimental comparison of the six standard test functions, the high performance of the Priestley-Chao kernel regression PCKE1 and PCKE2 based on the cross-verification bandwidth selection strategy is proved. Then, a kernel regression integration model KREFI based on fuzzy integral is designed to fuse four different Priestley-Chao nucleators to improve the stability of the kernel regression algorithm, in which the fuzzy measure in the fuzzy integral uses three different particle swarm optimization. Finally, the performance of the kernel regression integration algorithm KREFI is verified by the standard test function and 28 different forms of the silicon-based thin film Fourier infrared transmission spectrum. The result shows that the KREFI has a good regression performance, and the hidden peak of the spectral line cannot be solved to a certain extent.
【學(xué)位授予單位】：河北大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：O438.2;TP79
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 馮長根,趙軍,劉霞;多金屬氧酸鹽的紅外光譜[J];光譜學(xué)與光譜分析;2005年01期
2 琚宜文,姜波,侯泉林,王桂梁;構(gòu)造煤結(jié)構(gòu)成分應(yīng)力效應(yīng)的傅里葉變換紅外光譜研究[J];光譜學(xué)與光譜分析;2005年08期
3 張琳,張黎明,李燕,劉丙萍,王曉斐,王俊德;偏最小二乘法在傅里葉變換紅外光譜中的應(yīng)用及進(jìn)展[J];光譜學(xué)與光譜分析;2005年10期
4 吳景貴,席時權(quán),姜巖;紅外光譜在土壤有機(jī)質(zhì)研究中的應(yīng)用[J];光譜學(xué)與光譜分析;1998年01期
5 胡全一;祝民強(qiáng);;火星高光譜遙感大氣校正方法[J];東華理工大學(xué)學(xué)報(自然科學(xué)版);2008年03期
6 婁全勝;陳蕾;王平;張晨;謝健;;高光譜遙感技術(shù)在海洋研究的應(yīng)用及展望[J];海洋湖沼通報;2008年03期
7 梁堯欽;曾輝;;高光譜遙感在植被特征識別研究中的應(yīng)用[J];世界林業(yè)研究;2009年01期
8 王桂清,劉敏娜;紅外光譜技術(shù)的近代進(jìn)展及其應(yīng)用[J];現(xiàn)代儀器;2002年02期
9 唐延林,黃敬峰;農(nóng)業(yè)高光譜遙感研究的現(xiàn)狀與發(fā)展趨勢[J];遙感技術(shù)與應(yīng)用;2001年04期
10 李志忠;楊日紅;黨福星;張顯峰;譚炳香;趙慧潔;;高光譜遙感衛(wèi)星技術(shù)及其地質(zhì)應(yīng)用[J];地質(zhì)通報;2009年Z1期
相關(guān)博士學(xué)位論文 前2條
1 陳進(jìn);高光譜圖像分類方法研究[D];國防科學(xué)技術(shù)大學(xué);2010年
2 周陽;透明ZnO薄膜與鐵電薄膜的集成研究[D];河北大學(xué);2011年
本文編號：2397607
本文鏈接：http://sikaile.net/guanlilunwen/gongchengguanli/2397607.html
最近更新
教材專著
