【摘要】：高光譜數(shù)據(jù)波段多、波段之間相關(guān)性強(qiáng),導(dǎo)致信息冗余嚴(yán)重,增加了數(shù)據(jù)處理的工作量,有效準(zhǔn)確地在眾多波段中選擇具有代表性的波段尤為重要。首先用Wilks'Lambda(WL),隨機(jī)森林(random forest,RF)與自適應(yīng)波段選擇(adaptive band selection,ABS)這3種方法對(duì)高光譜數(shù)據(jù)進(jìn)行降維處理。然后提出了基于曲線誤差指數(shù)的評(píng)價(jià)方法,用此指數(shù)的趨勢(shì)來確定每種降維方法所要選擇的合適波段數(shù)量,同時(shí)用指數(shù)的大小評(píng)價(jià)不同降維方法的優(yōu)劣,并用分類方法對(duì)評(píng)價(jià)結(jié)果加以驗(yàn)證。結(jié)果顯示:Wilks'Lambda最終選擇的波段數(shù)為10個(gè),α6-α平穩(wěn)值(選擇6個(gè)波段時(shí)的曲線誤差值與曲線誤差平穩(wěn)值之間的差值)為0.05;隨機(jī)森林最終選擇的波段數(shù)為13個(gè),α6-α平穩(wěn)值為0.06;自適應(yīng)波段選擇方法最終選擇的波段數(shù)為20個(gè),α6-α平穩(wěn)為0.14。Wilks'Lambda的總體分類精度為80.56%,Kappa系數(shù)為0.77;隨機(jī)森林的總體分類精度為79.11%,Kappa系數(shù)為0.76;自適應(yīng)波段選擇方法的總體分類精度為49.94%,Kappa系數(shù)為0.41。得出以下結(jié)論:(1)基于曲線誤差指數(shù)的方法得出Wilks'Lambda有最小的α6-α平穩(wěn)值,是最佳的波段降維方法 ;分類結(jié)果顯示:Wilks'Lambda有最大的總體分類精度與Kappa系數(shù),是最佳的波段降維方法。(2)基于曲線誤差指數(shù)的評(píng)價(jià)方法與基于分類結(jié)果的誤差一致,說明此方法具有可行性。
[Abstract]:There are many bands of hyperspectral data and strong correlation between bands, which leads to serious redundancy of information, and increases the workload of data processing. It is particularly important to select representative bands in many bands effectively and accurately. Firstly, the hyperspectral data are reduced by using Wilks'Lambda (WL), random forest (random forest,RF) and adaptive band selection (adaptive band selection,ABS). Then an evaluation method based on curve error index is proposed. The trend of this index is used to determine the appropriate number of bands to be selected for each dimensionality reduction method. At the same time, the advantages and disadvantages of different dimensionality reduction methods are evaluated by the magnitude of exponent. The evaluation results are verified by the classification method. The results show that the final number of bands chosen by 10 bands is 10, the 偽 6- 偽 stationary value (the difference between the curve error value and the curve error stationary value) is 0.05, and the number of bands selected by random forest is 13, and the 偽 6- 偽 stationary value is 0. 05. 0.06; the number of bands selected by adaptive band selection method is 20; the overall classification accuracy of 偽 6- 偽 stationary 0.14.Wilks'Lambda is 80.56 kappa coefficient is 0.77; the total classification accuracy of random forest is 79.11kappa coefficient 0.76; the overall classification accuracy of adaptive band selection method is 0.76. The precision is 49.94 and the Kappa coefficient is 0.41. The following conclusions are obtained: (1) based on the curve error index method, Wilks'Lambda has the smallest 偽 6- 偽 stationary value, which is the best band dimension reduction method, and the classification results show that the Wilks'Lambda has the largest overall classification accuracy and Kappa coefficient. (2) the evaluation method based on curve error index is consistent with the error based on classification results, which shows that this method is feasible.
【作者單位】： 浙江農(nóng)林大學(xué)浙江省森林生態(tài)系統(tǒng)碳循環(huán)與固碳減排重點(diǎn)實(shí)驗(yàn)室;浙江農(nóng)林大學(xué)省部共建亞熱帶森林培育國家重點(diǎn)實(shí)驗(yàn)室;中國林業(yè)科學(xué)研究院資源信息研究所;
【基金】：國家自然科學(xué)基金青年基金資助項(xiàng)目(41201365) 浙江農(nóng)林大學(xué)科研發(fā)展基金資助項(xiàng)目(2013FR052,2014FR004);浙江農(nóng)林大學(xué)林學(xué)重中之重一級(jí)學(xué)科學(xué)生創(chuàng)新基金資助項(xiàng)目(201513)
【分類號(hào)】：TP751

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 唐貴華;基于密度排序聚類和超像素分割的高光譜遙感影像降維方法研究[D];深圳大學(xué);2016年

，

本文編號(hào)：2234512