基于多特征融合的衛(wèi)星遙感圖像分類研究
發(fā)布時間：2018-11-23 13:23
【摘要】：隨著遙感事業(yè)的蓬勃發(fā)展,衛(wèi)星遙感圖像受到人們越來越多的關(guān)注。高光譜圖像作為衛(wèi)星遙感圖像的一個重要分支,其本身具有的高維數(shù)據(jù)蘊(yùn)含了豐富的信息待我們深入挖掘。高光譜圖像分類問題是現(xiàn)階段遙感圖像研究領(lǐng)域的一個熱門問題,該問題涉及計(jì)算機(jī)圖像學(xué)、數(shù)理統(tǒng)計(jì)學(xué)、矩陣論等多個學(xué)科理論。在高光譜圖像分類領(lǐng)域,目前較為流行的分類方法是利用基于統(tǒng)計(jì)學(xué)習(xí)的機(jī)器學(xué)習(xí)分類算法(監(jiān)督學(xué)習(xí)方法,無監(jiān)督學(xué)習(xí)方法),通過建立分類模型來預(yù)測預(yù)測圖像中每個像素點(diǎn)的類別。當(dāng)使用有監(jiān)督分類算法時,由于高光譜圖像每個像素點(diǎn)對應(yīng)的特征向量(光譜特征)維度較高,并且一般可供使用的訓(xùn)練樣本數(shù)量稀少,會導(dǎo)致很嚴(yán)重的“Hughes”現(xiàn)象,這種特性導(dǎo)致了高光譜數(shù)據(jù)對分類算法的選擇頗為敏感。根據(jù)前人的大量實(shí)驗(yàn)可知,支持向量機(jī)分類算法(SVM)比較適合高光譜圖像的分類問題,此算法能夠很好地克服訓(xùn)練集少、特征維度高等問題。在高光譜圖像分類時,現(xiàn)有方法一般僅依賴于光譜特征,從而忽略了高光譜圖像所蘊(yùn)含的空間特征(即空間地理信息)。因此,為了提高圖像分類精度,在分類過程中如何有效地融合這兩種特征所提供的信息成為亟待解決的問題。本文中使用屬性特征方法(Attribute Profiles,APs)提取空間特征,該方法利用屬性過濾器(Attribute Filters,AFs)掃描高光譜圖像的每個通道,計(jì)算AF覆蓋區(qū)域的屬性值,與一系列閾值比較后,得到對應(yīng)像素點(diǎn)的空間特征;將得到的空間特征和光譜特征加權(quán)相加從而實(shí)現(xiàn)多特征融合,利用融合后的特征建立SVM分類模型。在特征融合之前,需要對數(shù)據(jù)中夾雜的噪聲進(jìn)行處理。近些年來,基于字典學(xué)習(xí)的矩陣稀疏表達(dá)方法越來越多地被應(yīng)用于數(shù)據(jù)特征處理。在圖像領(lǐng)域比較有代表性的稀疏表達(dá)方法是矩陣低秩分解,并結(jié)合字典學(xué)習(xí)后轉(zhuǎn)換為矩陣低秩表達(dá)算法(LRR)。該方法通過矩陣分解還原矩陣的低秩性,使得噪聲數(shù)據(jù)與原來的數(shù)據(jù)分解開來,從而得到高質(zhì)量的特征。結(jié)合衛(wèi)星遙感圖像的特點(diǎn),根據(jù)相鄰元素具有相同類別的假設(shè),對整幅圖像分塊進(jìn)行低秩表達(dá),本文提出使用基于區(qū)域劃分的LRR對高光譜圖像的光譜特征和空間特征進(jìn)行去噪處理,對得到的新特征進(jìn)行特征融合,最后再利用SVM分類算法建立分類模型;通過實(shí)驗(yàn)證明,本文提出的方法對于高光譜圖像分類問題可以得到較高的分類精度,另與基于核函數(shù)的特征融合方法對比具有明顯的優(yōu)勢。
[Abstract]:With the rapid development of remote sensing, people pay more and more attention to satellite remote sensing image. As an important branch of satellite remote sensing images, hyperspectral images contain abundant information for us to mine deeply. The problem of hyperspectral image classification is a hot issue in the field of remote sensing image research at present. This problem involves many disciplines such as computer graphics mathematical statistics matrix theory and so on. In the field of hyperspectral image classification, the most popular classification methods are machine learning (supervised learning, unsupervised learning), which is based on statistical learning. The classification model is established to predict the classification of each pixel in the image. When the supervised classification algorithm is used, because of the high dimension of the feature vector (spectral feature) corresponding to each pixel of hyperspectral image and the small number of training samples generally available for use, the phenomenon of "Hughes" will be very serious. This property leads to hyperspectral data being sensitive to the selection of classification algorithms. According to a large number of previous experiments, support vector machine (SVM) classification algorithm (SVM) is more suitable for hyperspectral image classification. This algorithm can overcome the problems of less training set and higher feature dimension. In the classification of hyperspectral images, the existing methods generally only depend on spectral features, thus ignoring the spatial features (i.e. spatial geographic information) contained in hyperspectral images. Therefore, in order to improve the accuracy of image classification, how to effectively fuse the information provided by these two features in the process of classification has become an urgent problem. In this paper, attribute feature method (Attribute Profiles,APs) is used to extract spatial features. In this method, attribute filter (Attribute Filters,AFs) is used to scan each channel of hyperspectral image, and the attribute value of AF coverage area is calculated, which is compared with a series of thresholds. The spatial features of the corresponding pixels are obtained. The spatial features and spectral features are weighted together to achieve multi-feature fusion, and the SVM classification model is established by using the fused features. It is necessary to deal with the noise in the data before feature fusion. In recent years, the sparse representation of matrix based on dictionary learning has been applied to data feature processing more and more. The sparse representation method in image field is matrix low rank decomposition, which is converted to matrix low rank representation algorithm (LRR). After dictionary learning. By decomposing the low rank of the matrix, the noise data is decomposed from the original data, and the high quality characteristic is obtained. Considering the characteristics of satellite remote sensing images and the assumption that adjacent elements have the same category, the block representation of the whole image is carried out in low rank. In this paper, LRR based on region division is used to de-noise the spectral and spatial features of hyperspectral images, and the new features are fused. Finally, the classification model is established by using the SVM classification algorithm. The experimental results show that the proposed method can achieve high classification accuracy for hyperspectral image classification and has obvious advantages compared with the kernel-based feature fusion method.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP751
[Abstract]:With the rapid development of remote sensing, people pay more and more attention to satellite remote sensing image. As an important branch of satellite remote sensing images, hyperspectral images contain abundant information for us to mine deeply. The problem of hyperspectral image classification is a hot issue in the field of remote sensing image research at present. This problem involves many disciplines such as computer graphics mathematical statistics matrix theory and so on. In the field of hyperspectral image classification, the most popular classification methods are machine learning (supervised learning, unsupervised learning), which is based on statistical learning. The classification model is established to predict the classification of each pixel in the image. When the supervised classification algorithm is used, because of the high dimension of the feature vector (spectral feature) corresponding to each pixel of hyperspectral image and the small number of training samples generally available for use, the phenomenon of "Hughes" will be very serious. This property leads to hyperspectral data being sensitive to the selection of classification algorithms. According to a large number of previous experiments, support vector machine (SVM) classification algorithm (SVM) is more suitable for hyperspectral image classification. This algorithm can overcome the problems of less training set and higher feature dimension. In the classification of hyperspectral images, the existing methods generally only depend on spectral features, thus ignoring the spatial features (i.e. spatial geographic information) contained in hyperspectral images. Therefore, in order to improve the accuracy of image classification, how to effectively fuse the information provided by these two features in the process of classification has become an urgent problem. In this paper, attribute feature method (Attribute Profiles,APs) is used to extract spatial features. In this method, attribute filter (Attribute Filters,AFs) is used to scan each channel of hyperspectral image, and the attribute value of AF coverage area is calculated, which is compared with a series of thresholds. The spatial features of the corresponding pixels are obtained. The spatial features and spectral features are weighted together to achieve multi-feature fusion, and the SVM classification model is established by using the fused features. It is necessary to deal with the noise in the data before feature fusion. In recent years, the sparse representation of matrix based on dictionary learning has been applied to data feature processing more and more. The sparse representation method in image field is matrix low rank decomposition, which is converted to matrix low rank representation algorithm (LRR). After dictionary learning. By decomposing the low rank of the matrix, the noise data is decomposed from the original data, and the high quality characteristic is obtained. Considering the characteristics of satellite remote sensing images and the assumption that adjacent elements have the same category, the block representation of the whole image is carried out in low rank. In this paper, LRR based on region division is used to de-noise the spectral and spatial features of hyperspectral images, and the new features are fused. Finally, the classification model is established by using the SVM classification algorithm. The experimental results show that the proposed method can achieve high classification accuracy for hyperspectral image classification and has obvious advantages compared with the kernel-based feature fusion method.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP751
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 張帆;陳俊杰;郭浩;;基于腦功能超網(wǎng)絡(luò)的多特征融合分類方法[J];計(jì)算機(jī)工程與應(yīng)用;年期
2 劉杰;;一種基于多特征融合的自適應(yīng)目標(biāo)跟蹤策略[J];大眾科技;2016年11期
3 劉紅;王曄;雷長海;;基于多特征融合的中醫(yī)舌像檢索研究[J];計(jì)算機(jī)應(yīng)用研究;2010年02期
4 王益堯;;基于多特征融合問句理解問答系統(tǒng)研究[J];數(shù)據(jù)通信;2018年02期
5 李志欣;李艷紅;張燦龍;;一種多特征融合的場景分類方法[J];小型微型計(jì)算機(jī)系統(tǒng);2018年05期
6 霍富功;王詩琴;;基于多特征融合的轉(zhuǎn)移粒子濾波紅外小目標(biāo)跟蹤算法研究[J];信息與電腦(理論版);2018年12期
7 蔣曉玲;章慧;;基于多特征融合的激光醫(yī)學(xué)圖像拼接[J];激光雜志;2015年12期
8 孫汝星;范榮雙;;基于支持向量機(jī)的多特征融合影像云檢測[J];測繪與空間地理信息;2018年06期
9 喬瑞萍;孫賀;董員臣;王方;;基于多特征融合的井蓋檢測系統(tǒng)實(shí)現(xiàn)[J];電子技術(shù)應(yīng)用;2018年06期
10 顧鑫;李U，
本文編號：2351756
本文鏈接：http://sikaile.net/guanlilunwen/gongchengguanli/2351756.html
最近更新
教材專著
