【摘要】：高光譜遙感影像具有所含光譜信息量大、相關(guān)性強(qiáng)的大數(shù)據(jù)量等特點(diǎn)，若用傳統(tǒng)分類(lèi)算法對(duì)其進(jìn)行分類(lèi)易產(chǎn)生“維數(shù)災(zāi)難”，因此對(duì)高維數(shù)據(jù)進(jìn)行降維處理則顯得尤為重要。在諸多降維算法中，如主成分分析(PCA)算法、線性判別分析(LDA)算法等，它們或是不能有效利用數(shù)據(jù)中的類(lèi)別信息，或是對(duì)數(shù)據(jù)的類(lèi)別信息要求嚴(yán)格。針對(duì)這些問(wèn)題，論文提出一種半監(jiān)督局部保持投影(SSLPP)算法。 論文首先對(duì)高光譜圖像及其自身特點(diǎn)作簡(jiǎn)單介紹，并結(jié)合監(jiān)督學(xué)習(xí)與非監(jiān)督學(xué)習(xí)對(duì)高維數(shù)據(jù)的特征提取方法進(jìn)行總結(jié)和分析，提出SSLPP算法；其次從SSLPP算法的原理、算法流程等方面，對(duì)算法進(jìn)行詳細(xì)介紹；為驗(yàn)證SSLPP算法的有效性，與目前幾種主流特征提取算法進(jìn)行對(duì)比性實(shí)驗(yàn)，如主成分分析(PCA)算法、局部保持投影(LPP)算法、監(jiān)督局部保持投(SLPP)算法。實(shí)驗(yàn)中對(duì)兩種實(shí)際情況下的高光譜遙感圖像數(shù)據(jù)進(jìn)行分類(lèi)實(shí)驗(yàn)，，首先用各種算法對(duì)原數(shù)據(jù)集進(jìn)行降維處理，然后使用K近鄰分類(lèi)器對(duì)低維數(shù)據(jù)進(jìn)行判類(lèi)識(shí)別，計(jì)算出各個(gè)算法的總體分類(lèi)精度，由此實(shí)驗(yàn)結(jié)果對(duì)SSLPP算法的有效性進(jìn)行驗(yàn)證；最后為了探究SSLPP算法與各種分類(lèi)器的融合性，分別用三種分類(lèi)器與之結(jié)合對(duì)四種遙感圖像進(jìn)行地物分類(lèi)實(shí)驗(yàn)，結(jié)果表明在該算法下各分類(lèi)器均獲得較高識(shí)別率，由此驗(yàn)證SSLPP算法具有較好的融合性。 經(jīng)過(guò)實(shí)驗(yàn)分析，論文所提SSLPP算法相比較于其他特征提取算法具有以下幾點(diǎn)優(yōu)勢(shì)：①SSLPP算法相對(duì)于非監(jiān)督降維算法，它充分利用了數(shù)據(jù)中的類(lèi)別信息，使高維數(shù)據(jù)經(jīng)過(guò)低維映射后具有較好的可分性；②SSLPP算法相對(duì)于監(jiān)督降維算法，其不僅利用了數(shù)據(jù)中的標(biāo)記樣本并同時(shí)充分利用大量的未標(biāo)記樣本，使得在進(jìn)行低維投影時(shí)更好的把握原始數(shù)據(jù)的整體性；③在對(duì)高光譜數(shù)據(jù)進(jìn)行分類(lèi)處理，SSLPP保證較高分類(lèi)精度的同時(shí)，又避免了對(duì)原始數(shù)據(jù)的全類(lèi)別標(biāo)定工作，從而很好的提高數(shù)據(jù)計(jì)算處理效率。 綜上所述，論文主要研究了高光譜遙感圖像基于半監(jiān)督學(xué)習(xí)的特征提取與分類(lèi)方法，提出一種半監(jiān)督數(shù)據(jù)特征提取算法，通過(guò)對(duì)幾種實(shí)際高光譜遙感圖像的分類(lèi)識(shí)別實(shí)驗(yàn)證明了論文算法的有效性。
[Abstract]:Hyperspectral remote sensing images are characterized by large amount of spectral information and large amount of data with strong correlation. If the traditional classification algorithm is used to classify hyperspectral remote sensing images, it is easy to produce "dimensionality disaster", so it is very important to reduce the dimension of high-dimensional data. In many dimensionality reduction algorithms, such as principal component analysis (PCA) algorithm, linear discriminant analysis (LDA) algorithm and so on, they either can not effectively use the category information in the data or require strictly the data category information. In order to solve these problems, this paper presents a semi-supervision department preserving projection (SSLPP) algorithm. Firstly, the hyperspectral image and its own characteristics are briefly introduced, and the feature extraction methods of high-dimensional data are summarized and analyzed by combining supervised learning and unsupervised learning, and then the SSLPP algorithm is proposed, and then the principle of SSLPP algorithm is introduced. In order to verify the effectiveness of the SSLPP algorithm, a comparative experiment is carried out with several popular feature extraction algorithms, such as principal component analysis (PCA) (PCA) algorithm, local preserving projection (LPP) algorithm, and so on. The Supervisory Department maintains the (SLPP) algorithm. In the experiment, two kinds of hyperspectral remote sensing image data are classified. Firstly, the original data set is reduced by various algorithms, and then the low-dimensional data is identified by K-nearest neighbor classifier. The overall classification accuracy of each algorithm is calculated, and the validity of SSLPP algorithm is verified by the experimental results. Finally, in order to explore the fusion of SSLPP algorithm with various classifiers, Three classifiers are used to classify the ground objects of four remote sensing images respectively. The results show that each classifier has a high recognition rate under this algorithm, which verifies that the SSLPP algorithm has a better fusion performance. After experimental analysis, compared with other feature extraction algorithms, the proposed SSLPP algorithm has the following advantages over the unsupervised dimensionality reduction algorithm, which makes full use of the class information in the data. Compared with the supervised dimensionality reduction algorithm, the high-dimensional data has good separability after low-dimensional mapping. It not only makes use of the labeled samples in the data, but also makes full use of a large number of unlabeled samples at the same time. In order to better grasp the integrity of the original data in the low-dimensional projection, the classification of the hyperspectral data can be processed by SSLPP to ensure a higher classification accuracy, and at the same time, the whole classification of the original data can be avoided. In order to improve the efficiency of data calculation and processing. To sum up, this paper mainly studies the feature extraction and classification method of hyperspectral remote sensing image based on semi-supervised learning, and proposes a semi-supervised data feature extraction algorithm. The effectiveness of the algorithm is proved by the classification and recognition experiments of several real hyperspectral remote sensing images.
【學(xué)位授予單位】：重慶大學(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7條

1 王立志;黃鴻;馮海亮;;基于SSMFA與kNNS算法的高光譜遙感影像分類(lèi)[J];電子學(xué)報(bào);2012年04期

2 黃鴻;李見(jiàn)為;馮海亮;;基于有監(jiān)督核局部線性嵌入的面部表情識(shí)別[J];光學(xué)精密工程;2008年08期

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

4 袁國(guó)強(qiáng);肖倩;劉強(qiáng);;帶有最小風(fēng)險(xiǎn)準(zhǔn)則的兩階段模糊運(yùn)輸模型[J];計(jì)算機(jī)工程與應(yīng)用;2011年35期

5 申中華;潘永惠;王士同;;有監(jiān)督的局部保留投影降維算法[J];模式識(shí)別與人工智能;2008年02期

6 任廣波;張杰;馬毅;鄭榮兒;;生成模型學(xué)習(xí)的遙感影像半監(jiān)督分類(lèi)[J];遙感學(xué)報(bào);2010年06期

7 趙慧潔;葛文謙;李旭東;;最小誤差準(zhǔn)則與脈沖耦合神經(jīng)網(wǎng)絡(luò)的裂縫檢測(cè)[J];儀器儀表學(xué)報(bào);2012年03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(lèi)方法研究[D];國(guó)防科學(xué)技術(shù)大學(xué);2010年

2 黃鴻;圖嵌入框架下流形學(xué)習(xí)理論及應(yīng)用研究[D];重慶大學(xué);2008年

本文編號(hào)：2162081