基于残差网络特征融合的高光谱图像分类

深度学习技术因其在深度挖掘地物特征方面的独特优势为高光谱图像分类提供了技术手段。但是在高光谱图像的像素级地物分类中,由于样本输入尺寸的影响导致深度学习的层数受限,不能充分挖掘高光谱图像中的深度特征,为此提出基于残差网络特征融合的高光谱图像分类方法。首先通过主成分分析(principal component analysis,PCA)方法提取原始高光谱图像中的第一主成分,利用残差网络有效提取地物空谱特征;再通过反卷积算法实现特征图的扩充,将反卷积后不同维度的特征进行多尺度特征融合,充分挖掘高光谱图像中的深度特征信息,进一步提升高光谱图像分类精度。对"珠海一号"卫星拍摄的江苏太湖和安徽巢湖两个区域进行地物分类实验,结果表明,与其他方法相比,该方法有效解决了高光谱图像分类中深度特征提取不足的问题,获得了更好的分类性能。     

模态间匹配学习的高光谱和激光雷达联合分类

如何有效地提取和融合不同模态的特征是高光谱图像和激光雷达数据联合分类的关键。近年来,得益于深度学习强大的特征学习能力,其在高光谱图像和激光雷达数据联合分类领域受到了越来越多的关注。然而,现有的深度学习模型大多基于监督学习的模式,分类性能依赖标注样本的数量和质量。为此,本文提出了一种基于模态间匹配学习的联合分类方法,充分利用未标注样本的信息,减少对标注信息的依赖性。具体而言,本文首先通过高光谱图像和激光雷达数据之间的匹配关系和KMeans聚类算法,构造模态匹配标签。然后,利用该标签训练含有多个卷积层的匹配学习网络。该网络由两个并行分支构成,每个分支负责提取单个模态的特征。最后,以该网络为基础,构造高光谱图像和激光雷达数据联合分类模型。该模型的参数由匹配学习网络进行初始化,因而只需要少量标注样本进行微调即可达到理想的分类效果。为了验证本文方法的有效性,在Houston和MUUFL两个常用的高光谱图像和激光雷达数据联合分类数据集上进行了大量的实验。实验结果表明,与已有的分类模型相比,本文方法能够获得更高的分类性能。     

一种新的空谱联合高光谱图像分类方法

针对现有高光谱图像分类方法对空间和谱间信息的利用不充分,限制了地物分类的准确度的问题,该文提出一种基于正交线性判别分析和三维离散小波变换的高光谱图像空谱联合分类算法。该算法首先利用正交线判别分析对高光谱图像进行特征提取和特征缩减;然后将提取后的特征经三维离散小波从不同尺度、频率和方向上分解,以级联的方式获得具有正交类判别信息的空谱融合特征集;最后,将空谱融合特征集作为概率支持向量机的输入特征,分类的结果再通过马尔可夫随机场利用空间上下文信息细化分类图,进一步提升分类准确度。在Indian Pines和Salinas两组数据集上的实验表明,相比其他算法,该算法能达到更高的总体分类准确度和Kappa系数,并且在大部分的地物类别上的分类准确度有着较为明显地提升。     

高光谱影像分类的深度少样例学习方法

针对高光谱影像分类面临的小样本问题,提出了一种深度少样例学习算法,该算法在训练过程中通过模拟小样本分类的情况来训练深度三维卷积神经网络提取特征,其提取得到的特征具有较小类内间距和较大的类间间距,更适合小样本分类问题,且能用于不同的高光谱数据,具有更好的泛化能力。利用训练好的模型提取目标数据集的特征,然后结合最近邻分类器和支持向量机分类器进行监督分类。利用Pavia大学、Indian Pines和Salinas 3组高光谱影像数据进行分类试验,试验结果表明,该算法能够在训练样本较少的情况下（每类地物仅选取5个标记样本作为训练样本）取得优于传统半监督分类方法的分类精度。     

面向高光谱图像分类的半监督空谱判别分析

为充分利用高光谱图像蕴藏的空间信息提升分类精度,提出了面向高光谱图像分类的半监督空谱判别分析(S3 DA)算法。考虑高光谱图像数据集的空间一致性,首先利用少量标记样本定义类内散度矩阵,保存数据集同类像元的光谱近邻结构;再利用无标记样本定义空间近邻像元散度矩阵,揭示像元间的空间近邻结构和地物的空间分布结构信息。S3 DA既保持数据集在光谱域的可分性,又保存了无标记样本蕴藏的空间域近邻结构,增强了同类像元和空间近邻像元在投影子空间的聚集性,从而提升分类性能。在PaviaU和Indian Pines数据集的试验表明,总体分类精度分别达到81.50%和71.77%。与传统的光谱方法比较,该算法能有效提升高光谱图像数据集的地物分类精度。     

高光谱影像奇异谱分析特征提取方法：综述与评价

高光谱遥感影像(hyperspectral imagery, HSI)通常包含几十至数百个连续波段,具有图谱合一、光谱连续的特点,能够实现地物的精细分类,被广泛应用农业、林业、城市以及海洋等领域。HSI特征提取是高光谱应用的前提,也是遥感领域的研究热点和前沿课题之一。近年来,奇异谱分析(singular spectrum analysis, SSA)被应用于HSI领域,在光谱特征和空间特征提取方面取得了较好效果,逐渐成为特征提取的一种有效方法。本文首先分析了HSI特征提取的研究进展和存在的问题;其次对SSA方法进行了系统的梳理,分别介绍了光谱域1D-SSA、空间域2D-SSA和光谱-空间组合域SSA 3类方法的作用、效果及优缺点,并在两个公开的HSI数据集和一个高分五号HSI数据上进行了分类效果验证;最后,对SSA特征提取进行了总结,并讨论了未来的研究方向。     

结合深度学习的高光谱与多源遥感数据融合分类

高光谱数据具有丰富的光谱特征,但是其空间分辨率相对较低。一些遥感数据具有与高光谱数据互补的优势,例如提供更精细的空间信息的高空间分辨率数据和具有高度信息的激光雷达LiDAR（Light Detection and Ranging ）数据。通过将高光谱数据与多源遥感数据进行融合,可以弥补高光谱数据空间分辨率相对较低,空间特征不够丰富的缺点。近年来,基于深度学习的方法已经在遥感数据分类研究中取得了一定的进展。然而,由于深度网络的特征提取过程是一个自主的过程,往往无法精确的获取最有利于遥感数据分类的特征;同时,深度学习方法具有复杂的网络结构和大量的参数,往往会在分类训练过程中造成参数拟合困难。以上这些因素会导致分类效果不佳。针对这些问题,本文提出了一种将卷积神经网络CNN（Convolutional Neural Network）和纹理特征相结合的多源遥感数据特征级融合分类框架。该方法共3个步骤,首先,对高光谱数据或多源遥感数据提取纹理特征;然后,构造CNN,分别将原始高光谱遥感数据、原始多源遥感数据和第一步中获得的纹理特征作为深度网络的输入进行深度特征提取;最后,将分别提取到的深度特征拼接,并利用Softmax分类器进行分类。为了验证本文提出方法的分类效果,本文在休斯顿和塞特福德矿地区公开数据集上进行实验,并将该分类框架与支持向量机分类方法、像素级融合分类方法和特征级融合分类方法进行对比。由此可以分析得出,本文提出的基于深度学习的融合分类方法可以获得较高的分类精度。     

面向对象的多尺度加权联合稀疏表示的高空间分辨率遥感影像分类

针对高空间分辨率遥感影像中的地物具有多尺度特性,以及各个尺度的对象特征对地物分类精度的影响具有较强的尺度效性,并结合面向对象影像分析方法和多尺度联合稀疏表示方法在高空间分辨率遥感影像分类中的各自优点,提出了一种面向对象的多尺度加权稀疏表示的高空间分辨率遥感影像分类算法。首先,采用多尺度分割算法获得多尺度分割结果并提取对象的多尺度特征;然后,根据影像对象的多尺度分割质量测度计算各尺度的对象权重,构建面向对象的多尺度加权联合稀疏表示模型;最后,采用2个国产GF-2高空间分辨率遥感数据集和1个高光谱-高空间分辨率航空遥感数据集（WashingtonD.C.数据）验证该算法的有效性。试验结果表明,与SVM、像素级稀疏表示、单尺度和多尺度对象级稀疏表示和深度学习等算法相比较,本文算法获得了较高的OA和Kappa分类精度,提高了各个尺度地物的分类精度,有效抑止了地物分类结果中的椒盐噪声现象,同时保持大尺度地物的区域性和小尺度地物的细节信息。     

基于流形对齐的高光谱遥感图像降维和分类算法

多时相的高光谱遥感图像数据处理中会出现地物光谱特征漂移的现象。为了提高源域数据已有知识对目标域数据分类的精度,采用了基于流形对齐的分类算法。先用一个标准的线性或非线性的降维方法将2个高光谱遥感数据集映射到低维(流形)空间中,再用Procrustes分析方法将其低维嵌入之间的平移、旋转和缩放因子剔除,得到数据集间的最优对齐,最后用最近邻算法进行分类。对多个不同时相高光谱遥感图像进行实验,并对比了已有的流形对齐算法,结果表明本算法具体较好的迁移能力和分类效果。     

基于空间特征与纹理信息的高光谱图像半监督分类

传统高光谱图像分类方法主要使用图像的光谱特征信息,没有充分利用高光谱图像的空间特性及样本的其他信息。本文提出了一种基于空间特征与纹理信息的高光谱图像半监督分类方法。首先,将高光谱图像每一像素的光谱特征与其邻域范围内的光谱特征进行结合,得到了这一像素的空-谱特征;然后用灰度共生矩阵提取了高光谱图像的纹理特征,并与空-谱特征进行了融合;最后,用基于图的半监督分类算法进行了分类。通过在Indian Pines数据集和PaviaU数据集上进行试验,结果表明本文提出的方法能取得较高的分类结果。     

利用胶囊网络实现高光谱影像空谱联合分类

卷积神经网络等深度学习模型已经在高光谱影像分类任务中取得了理想的结果.然而,由于传统神经元只能进行标量计算,现有的深度学习模型无法对高光谱影像特征的实例化参数进行建模,因此无法在邻域范围受限的条件下获得令人满意的分类效果.通过引入胶囊网络结构设计了一种新型网络模型,该模型利用胶囊神经元进行向量计算,并利用权重矩阵编码特...     

Dimensionality reduction of hyperspectral data using spectral fractal feature

A new approach for dimensionality reduction of hyperspectral data has been proposed in this article. The method is based on extraction of fractal-based features from the hyperspectral data. The features have been generated using spectral fractal dimension of the spectral response curves (SRCs) after smoothing, interpolating and segmenting the curves. The new features so generated have then been used to classify hyperspectral data. Comparing the post classification accuracies with some other conventional dimensionality reduction methods, it has been found that the proposed method, with less computational complexity than the conventional methods, is able to provide classification accuracy statistically equivalent to those from conventional methods.     

基于非线性判别分析的高光谱影像特征提取

高光谱遥感影像丰富的光谱信息有利于深入挖掘目标的理化特性,精细识别不同目标间的细微差异。为了提高影像分类识别的精度与速度需要对光谱信息进行特征提取。基于核函数的判别分析能够在数据中提取非线性特征,本文将其应用到高光谱影像分类的特征提取中,并进行了最小距离分类实验取得理想结果。     

Comparative performance of convolutional neural network,weighted and conventional support vector machine and random forest for classifying tree species using hyperspectral and photogrammetric data

ABSTRACT

The classification of tree species can significantly benefit from high spatial and spectral information acquired by unmanned aerial vehicles (UAVs) associated with advanced classification methods. This study investigated the following topics concerning the classification of 16 tree species in two subtropical forest fragments of Southern Brazil: i) the potential integration of UAV-borne hyperspectral images with 3D information derived from their photogrammetric point cloud (PPC); ii) the performance of two machine learning methods (support vector machine – SVM and random forest – RF) when employing different datasets at a pixel and individual tree crown (ITC) levels; iii) the potential of two methods for dealing with the imbalanced sample set problem: a new weighted SVM (wSVM) approach, which attributes different weights to each sample and class, and a deep learning classifier (convolutional neural network – CNN), associated with a previous step to balance the sample set; and finally, iv) the potential of this last classifier for tree species classification as compared to the above mentioned machine learning methods. Results showed that the inclusion of the PPC features to the hyperspectral data provided a great accuracy increase in tree species classification results when conventional machine learning methods were applied, between 13 and 17% depending on the classifier and the study area characteristics. When using the PPC features and the canopy height model (CHM), associated with the majority vote (MV) rule, the SVM, wSVM and RF classifiers reached accuracies similar to the CNN, which outperformed these classifiers for both areas when considering the pixel-based classifications (overall accuracy of 84.4% in Area 1, and 74.95% in Area 2). The CNN was between 22% and 26% more accurate than the SVM and RF when only the hyperspectral bands were employed. The wSVM provided a slight increase in accuracy not only for some lesser represented classes, but also some major classes in Area 2. While conventional machine learning methods are faster, they demonstrated to be less stable to changes in datasets, depending on prior segmentation and hand-engineered features to reach similar accuracies to those attained by the CNN. To date, CNNs have been barely explored for the classification of tree species, and CNN-based classifications in the literature have not dealt with hyperspectral data specifically focusing on tropical environments. This paper thus presents innovative strategies for classifying tree species in subtropical forest areas at a refined legend level, integrating UAV-borne 2D hyperspectral and 3D photogrammetric data and relying on both deep and conventional machine learning approaches.     

Logistic Regression for Feature Selection and Soft Classification of Remote Sensing Data

Feature selection is a key task in remote sensing data processing, particularly in case of classification from hyperspectral images. A logistic regression (LR) model may be used to predict the probabilities of the classes on the basis of the input features, after ranking them according to their relative importance. In this letter, the LR model is applied for both the feature selection and the classification of remotely sensed images, where more informative soft classifications are produced naturally. The results indicate that, with fewer restrictive assumptions, the LR model is able to reduce the features substantially without any significant decrease in the classification accuracy of both the soft and hard classifications     

Deep Fusion of Localized Spectral Features and Multi-scale Spatial Features for Effective Classification of Hyperspectral Images

This study presents a deep extraction of localized spectral features and multi-scale spatial features convolution (LSMSC) framework for spectral-spatial fusion based classification of hyperspectral images (HSIs). First, adjacent spectral bands are grouped based on their similarity measurements, where the whole hypercube is partitioned into several sub-cubes, each corresponding to one band group. Then, the proposed localized spectral features extraction (LSF) strategy is used to extract localized spectral features, which are extracted from each band group using the 1D convolutional neural network (CNN). Meanwhile, the proposed HiASPP strategy is employed to extract the multi-scale features from the first several principal components of each sub-cube. Finally, the extracted spectral and spatial features are concatenated for spectral-spatial fusion based classification of HSI. Experiments conducted on three publicly available datasets have demonstrated that the proposed architecture outperforms several state-of-the-art approaches.     

Segmentation-based classification of hyperspectral imagery using projected and correlation clustering techniques

A partitional clustering-based segmentation is used to carry out supervised classification for hyperspectral images. The main contribution of this study lies in the use of projected and correlation partitional clustering techniques to perform image segmentation. These types of clustering techniques have the capability to concurrently perform clustering and feature/band reduction, and are also able to identify different sets of relevant features for different clusters. Using these clustering techniques segmentation map is obtained, which is combined with the pixel-level support vector machines (SVM) classification result, using majority voting. Experiments are conducted over two hyperspectral images. Combination of pixel-level classification result with the segmentation maps leads to significant improvement of accuracies in both the images. Additionally, it is also observed that, classified maps obtained using SVM combined with projected and correlation clustering techniques results in higher accuracies as compared to classified maps obtained from SVM combined with other partitional clustering techniques.     

面向小样本高光谱影像分类的轻量化关系网络

近年来,基于深度学习的高光谱影像分类取得重要进展,针对高光谱影像分类训练样本稀缺的情况,提出一种结合注意力机制的轻量化关系网络（lightweight attention depth-wise relation network, LWAD-RN）, 以解决高光谱影像小样本分类问题。该网络由嵌入层和关联层组成,在嵌入层采用结合注意力机制的轻量化卷积神经网络提取像元特征,同时引入稠密网络结构;在关联层计算关联值进行分类,并采用基于任务的模式训练网络。利用3组公开的高光谱影像数据进行对比实验,结果表明,LWAD-RN能够有效提升小样本条件下（每类5个训练样本）的分类精度,同时提高了模型训练和分类效率。     

利用天宫一号高光谱数据识别植被类型的评估研究

天宫一号高光谱数据尚未得到普遍应用,其数据的质量和应用潜力仍在进一步实践求证和挖掘.See5.0数据挖掘工具是一种能够找出训练样本中模式类隐含特征,并可以自动建立决策规则的分类算法,可避免人为建立分类规则的主观性.本文首先通过光谱曲线分析,选择地物光谱分离性最好的波段组合,然后利用See5.0工具生成规则集,再利用规则集对同一幅天宫一号高光谱数据在不同分类级别上进行分类,并利用相同的验证样本进行精度验证.经过光谱分析发现分类不同森林类型的最佳谱段中心波长分别为:655 nm、673 nm、802 nm、866 nm、984 nm,See5.0分类结果表明在同一树种不同生长期及不同亚种的分类级别上,分类精度在45%以下,表现出了一定局限性,但在树种分类级别上,天宫一号数据表现出了高光谱的优越性,分类精度皆在80%以上,植被类型分类级别,分类精度可达到90%以上.