基于譜線形狀與信息量差異的高光譜解混NMF初始化方法
發(fā)布時間：2019-06-18 09:31
【摘要】：在高光譜像元解混應(yīng)用中,好的端元光譜矩陣初始化方法對于提高盲信號分解精度具有重要意義。針對空間分辨率較高的高光譜數(shù)據(jù),提出了一種新的面向非負(fù)矩陣分解(non-negative matrix factorization,NMF)的初始化方法。該方法通過計算像元在譜線形狀和信息量差異等方面的參數(shù),利用像元譜線峭度、KL散度和光譜角等參量,從眾多混合像元中識別出純像元;并分辨出不同類型純像元(或類純像元)之間的差別,從中選擇最適合代表每一類型端元的純像元(或類純像元)作為算法的初值像元,完成端元矩陣的初始化。將此方法分別用于模擬數(shù)據(jù)和真實數(shù)據(jù)的實驗結(jié)果表明,該方法能夠明顯提高高光譜混合數(shù)據(jù)的NMF精度,相比其他常用初始化方法具有更好的效果。
[Abstract]:In the application of high-spectral image element, a good end-element spectral matrix initialization method is of great significance to improve the accuracy of blind signal decomposition. A new method for initializing non-negative matrix factorization (NMF) is proposed for high spectral data with higher spatial resolution. The method comprises the following steps of: calculating a parameter of an image element in a spectral line shape and an information quantity difference and the like, identifying a pure image element from a plurality of mixed image elements by using a parameter such as a spectral line similarity, a KL divergence angle and a spectral angle, and distinguishing the difference between different types of pure image elements (or quasi-pure image elements), A pure image element (or quasi-image element), which is most suitable for representing each type of end element, is selected as the initial value image element of the algorithm, and the initialization of the end element matrix is completed. The experimental results show that the method can obviously improve the NMF accuracy of the high-spectral mixed data, and has a better effect than the other common initialization methods.
【作者單位】： 南陽理工學(xué)院數(shù)學(xué)與統(tǒng)計學(xué)院;南陽理工學(xué)院經(jīng)濟(jì)管理學(xué)院;
【基金】：河南省高等學(xué)校重點(diǎn)科研項目“Smith正規(guī)型在有限域上有理點(diǎn)個數(shù)中的應(yīng)用”(編號:17A110010)資助
【分類號】：TP391.41
本文編號：2501370
[Abstract]:In the application of high-spectral image element, a good end-element spectral matrix initialization method is of great significance to improve the accuracy of blind signal decomposition. A new method for initializing non-negative matrix factorization (NMF) is proposed for high spectral data with higher spatial resolution. The method comprises the following steps of: calculating a parameter of an image element in a spectral line shape and an information quantity difference and the like, identifying a pure image element from a plurality of mixed image elements by using a parameter such as a spectral line similarity, a KL divergence angle and a spectral angle, and distinguishing the difference between different types of pure image elements (or quasi-pure image elements), A pure image element (or quasi-image element), which is most suitable for representing each type of end element, is selected as the initial value image element of the algorithm, and the initialization of the end element matrix is completed. The experimental results show that the method can obviously improve the NMF accuracy of the high-spectral mixed data, and has a better effect than the other common initialization methods.
【作者單位】： 南陽理工學(xué)院數(shù)學(xué)與統(tǒng)計學(xué)院;南陽理工學(xué)院經(jīng)濟(jì)管理學(xué)院;
【基金】：河南省高等學(xué)校重點(diǎn)科研項目“Smith正規(guī)型在有限域上有理點(diǎn)個數(shù)中的應(yīng)用”(編號:17A110010)資助
【分類號】：TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前5條
1 鄭金彬;歐陽利軍;;基于密度等高線的原型初始化方法[J];南華大學(xué)學(xué)報(自然科學(xué)版);2007年01期
2 汪西莉;劉芳;焦李成;;一種小波域HMT模型參數(shù)初始化方法[J];計算機(jī)科學(xué);2003年01期
3 曹彩鳳;面向?qū)ο蟮腃++程序設(shè)計中各類對象初始化方法[J];計算機(jī)應(yīng)用與軟件;2003年03期
4 楊圣云;袁德輝;賴國明;;一種新的聚類初始化方法[J];計算機(jī)應(yīng)用與軟件;2007年08期
5 孫澤強(qiáng);謝紅薇;郝曉燕;;NMF模型在挖掘基因功能關(guān)系中的研究與應(yīng)用[J];計算機(jī)工程與設(shè)計;2014年04期
相關(guān)碩士學(xué)位論文 前1條
1 李靖;基于小波變換和NMF的數(shù)字圖像水印算法[D];湖北工業(yè)大學(xué);2011年
，本文編號：2501370
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2501370.html
最近更新
教材專著
