A Composite Semisupervised SVM for Classification of Hyperspectral Images

This letter presents a novel composite semisupervised support vector machine (SVM) for the spectral-spatial classification of hyperspectral images. In particular, the proposed technique exploits the following: 1) unlabeled data for increasing the reliability of the training phase when few training samples are available and 2) composite kernel functions for simultaneously taking into account spectral and spatial information included in the considered image. Experiments carried out on a hyperspectral image pointed out the effectiveness of the presented technique, which resulted in a significant increase of the classification accuracy with respect to both supervised SVMs and progressive semisupervised SVMs with single kernels, as well as supervised SVMs with composite kernels.     

Ensemble Stacked Auto-encoder Classification on LIDAR Remote Sensing Images

Light detection and ranging system (LIDAR) can obtain diverse remote sensing datasets which contains different land cover information. The datasets offer vital and significant features for land cover classification. As a new and effective deep learning algorithm, stacked auto-encoders (SAE) consists of multiple auto-encoders in which the code of each auto-encoder is the input of the successive one. The classification precision is closely related to hidden layers, and the number of samples in fine-tuning step also affects classification results. In this paper we study the classifiers based on different number of samples and hidden layers. According to appropriate parameters, we promote SAE with adaptive boosting ensemble strategy to build new classification method. Two tests which are based on LIDAR datasets are implemented. The experiment results prove that the fusion of deep learning and ensemble learning is effective to LIDAR remote sensing images. The proposed method is robust to similar scenes classification. The overall accuracy increases 6% compared with bagging method on test 1.     

基于自动搜索和光谱匹配技术的训练样本纯化算法

提出了一种基于局部自动搜索和光谱匹配技术的监督分类训练样本的纯化方法。该方法首先利用遥感影像中像元的灰度信息在图像上局部范围内自动搜索和选择最佳样区位置，然后利用光谱匹配的思想对寻找到的最佳样区在光谱空间上进一步纯化。实验结果证明，通过手工选择样区的辅助，该算法能够自动有效地搜寻到最佳样区的位置，并对最佳样区进行纯化处理。原始遥感图像经过本文的样区纯化算法处理后，无论是目视判读效果，还是分类后混淆矩阵的统计及分类精度，均优于纯化处理前的分类结果，具有一定的实用价值。     

一种切空间协同表示的高光谱遥感影像分类方法

协同表示分类（collaborative representation classification,CRC）算法近年来成为高光谱遥感分类的研究热点。地物类别间区分性不高会严重影响现有CRC算法的性能。流形结构可有效地解决非线性问题,并解决高光谱遥感影像因数据冗余导致的类别间区分性低的问题。提出了一种基于切空间的高光谱遥感影像协同表示分类算法（tangent space collaborative representation classification,TCRC）和一种基于欧氏距离的自适应加权的切空间协同表示分类算法（weighted tangent space collaborative representation classification,WTCRC）。TCRC算法利用测试样本的切平面来估计区域流形,在测试样本的切空间中使用协同表示算法,寻找测试样本在各类训练样本中的最优线性表示估计,并用其最小误差来对测试样本进行分类。在此基础上,利用测试样本邻域像元、训练样本与测试样本的欧氏距离作为权矩阵来自适应调整各样本对测试样本的影响。实验采用ROSIS（reflective optics system image spectro-meter）和AVIRIS（airbone visible infrared imaging spectrometer）高光谱遥感影像对所提出算法的性能进行了评价,结果表明TCRC和WTCRC在分类效果上比CRC有明显的提升,WTCRC相较于TCRC具有更好的分类效果,具有更强鲁棒性。     

一种基于支撑向量机的遥感影像不完全监督分类新方法

不完全监督分类是研究在只有目标类训练样本的情况下如何准确地将目标类从数据集中提取出来。在许多遥感应用问题中,往往只需要从遥感影像中提取某一类地物。如果分类过程中只要选取目标类训练样本,将节省在训练样本选取过程中的大量人力物力。因此不完全监督分类是一个值得研究的遥感分类问题。提出了一种基于加权无标识样本支撑向量机(WUS-SVM),并在其基础发展出一种不完全监督分类方法。该方法分3个步骤:(1)在影像中随机选取一定量的无标识样本,将它们作为具有不同权重的非目标类训练样本;(2)用目标类的训练样本和加权无标识训练样本一起训练WUS-SVM,得到初步的分类器;(3)利用初步的分类器确定无标识样本的类别,并与原目标类训练样本一起再次训练SVM得到最终的分类器。通过对模拟数据和遥感影像的分类试验初步证明了该分类方法的有效性。     

基于小波与遗传算法的特征提取与特征选择

高维遥感数据的分类与识别与传统的多光谱遥感分类技术具有明显的区别。本文提出了一种基于遗传算法和小波／小波包分析相结合的特征提取方法用于高维遥感数据降维与分类。该方法综合了遗传算法的全局优化和小波／小波包分析的多尺度、多分辨率的特点。首先,通过离散的小波变换(DWT)或小波包变换(WP)将高光谱信号变换到特征域进行光谱分解。由于DWT变换是一种线性变换,不同尺度的DWT系数可作为线性光谱特征。然后,对这些线性光谱特征利用遗传算法结合训练样本计算类内／类间距离搜索最优分类子集,其具体染色体编码取可能的特征号,适应度函数基于样本平均Jeffries-Matusita距离计算。所用的分类器采用最大似然分类器。试验结果表明该方法与常规特征提取算法如主成分变换(PCA)、判别分析特征提取(DAFE)、决策边界特征提取(DBFE)相比,能提高分类精度约1.1％-6.5％。     

SVM-Based Data Editing for Enhanced One-Class Classification of Remotely Sensed Imagery

This paper studies a specific one-class classification problem where the training data are corrupted by significant outliers. Specifically, we are interested in the one-class support vector machine (OCSVM) approach that normally requires good training data. However, perfect training data are usually hard to obtain in most real-world applications due to the inherent data variability and uncertainty. To address this issue, we propose an OCSVM-based data editing and classification method that can iteratively purify the training data and learn an appropriate classifier from the trimmed training set. The proposed method is compared with a general OCSVM approach trained from two types of bootstrap samples, and applied to the mapping and compliance monitoring tasks for the U.S. Department of Agriculture's Conservation Reserve Program using remotely sensed imagery. Experimental results show that the proposed method outperforms the general OCSVM using bootstrap samples at a lower computational load.     

Hyperspectral Image Classification Using Relevance Vector Machines

This letter presents a hyperspectral image classification method based on relevance vector machines (RVMs). Support vector machine (SVM)-based approaches have been recently proposed for hyperspectral image classification and have raised important interest. In this letter, it is genuinely proposed to use an RVM-based approach for the classification of hyperspectral images. It is shown that approximately the same classification accuracy is obtained using RVM-based classification, with a significantly smaller relevance vector rate and, therefore, much faster testing time, compared with SVM-based classification. This feature makes the RVM-based hyperspectral classification approach more suitable for applications that require low complexity and, possibly, real-time classification.     

Feature fusion to improve road network extraction in high-resolution SAR images

This letter aims at the extraction of roads and road networks from high-resolution synthetic aperture radar data. Classical methods based on line detection do not use all the information available; indeed, in high-resolution data, roads are large enough to be considered as regions and can be characterized also by their statistics. This property can be used in a classification scheme. Therefore, this letter presents a road extraction method which is based on the fusion of classification (statistical information) and line detection (structural information). This fusion is done at the feature level, which helps to improve both the level of likelihood and the number of the extracted roads. The proposed approach is tested with two classification methods and one line extractor. Results on two different datasets are discussed.     

基于免疫粒子群优化算法的影像纹理分类

粒子群优化算法是基于群智能的随机全局优化方法,它源于对鸟群简化社会系统的模拟。为了提高标准粒子群优化算法的收敛性能,将生物免疫系统的记忆能力和多样性引入标准粒子群优化算法,提出一种免疫粒子群优化算法。在提取纹理样本Laws纹理能量模板特征、小波特征等纹理特征的基础上,提出针对分类问题的粒子表达方法和群体寻优策略,实现了基于免疫粒子群算法的纹理分类。实验结果表明,与标准粒子群优化算法相比,免疫粒子群优化算法在获取训练样本类别中心时具有较好的收敛性能,并且基于该算法的影像纹理分类具有较高的分类精度。     

Genetic SVM Approach to Semisupervised Multitemporal Classification

The updating of classification maps, as new image acquisitions are obtained, raises the problem of ground-truth information (training samples) updating. In this context, semisupervised multitemporal classification represents an interesting though still not well consolidated approach to tackle this issue. In this letter, we propose a novel methodological solution based on this approach. Its underlying idea is to update the ground-truth information through an automatic estimation process, which exploits archived ground-truth information as well as basic indications from the user about allowed/forbidden class transitions from an acquisition date to another. This updating problem is formulated by means of the support vector machine classification approach and a constrained multiobjective optimization genetic algorithm. Experimental results on a multitemporal data set consisting of two multisensor (Landsat-5 Thematic Mapper and European Remote Sensing satellite synthetic aperture radar) images are reported and discussed.     

融合光谱-空间信息的高光谱遥感影像增量分类算法

提出了一种融合光谱和空间结构信息的高光谱遥感影像增量分类算法INC_SPEC_MPext。通过主成分分析(PCA)提取高光谱影像的若干主成分,利用数学形态学提取各主分量影像对应的形态学剖面(MP),再将所有主分量影像的形态学剖面归并联结,组成扩展的形态学剖面(MPext)。将MPext与光谱信息相结合以增加知识,最大限度地挖掘未标记样本的有用信息,优化分类器的学习能力。不断从分类器对未标记样本的预测结果中甄选置信度高的样本加入训练集,并迭代地利用扩大的训练集进行分类器构建和样本预测。以不同地表覆盖类型的AVIRIS Indian Pines和Hyperion EO-1Botswana作为测试数据,分别与基于光谱、MPext、光谱和MPext融合的分类方法进行比对。试验结果表明,在训练样本数量有限情况下,INC_SPEC_MPext算法在降低分类成本的同时,分类精度和Kappa系数都有不同程度的提高。     

灰度直方图与K-最近邻的影像分割算法

针对灰度直方图在影像分类时需要考虑阈值和K-最近邻分类算法分类效率低等问题,提出了一种基于灰度直方图与KNN相结合的影像分割算法:首先对待分割的影像进行灰度直方图统计;其次利用灰度直方图对影像进行硬阈值的划分,得到已知类别样本和未标记样本;然后选择一定数量已知类别的样本对KNN分类器进行训练;最后利用KNN分类器对未标记样本进行类别划分,得到最终分割图像。实验结果表明,该算法结合了灰度直方图高效性和KNN高精度的优势,避免了直方图分割最佳阈值的选取;与传统的KNN算法相比,本文算法大大提高了分类效率,且精度相当,满足实际生产应用的需求。     

结合nDSM的高分辨率遥感影像深度学习分类方法

针对高分辨率遥感影像因其地物类内差异大、光谱信息相对欠缺导致现有影像分类方法存在错分现象较多、地物边界残缺不完整等问题，本文提出了一种归一化数字表面模型（nDSM）约束的高分辨率遥感影像深度学习分类方法。首先，将nDSM数据作为附加波段叠加在遥感影像上并获取训练样本；然后，利用优化的U-Net网络进行模型训练得到最优模型；最后，利用最优模型对附加了nDSM波段的遥感影像进行地物分类。试验结果表明，本文方法引入nDSM数据用于U-Net模型训练和分类，可有效提高影像分类精度，得到更加真实可靠的分类结果。     

基于对应分析的训练样本的选择

本文提出一种基于对应分析的训练样本的选择方法。它从训练样本中自动地选择有代表性的典型训练样本,使得在自动分类中充分利用所采集的样本信息,以便得到满意的分类结果。通过实验与分析证明,该方法是可行的,它明显优于人工随机选择训练样本的方式。与基于Q型因子分析的训练样本选择方法相比,可以更快地得到较少的典型样本,满意的分类精度。     

A Context-Sensitive Clustering Technique Based on Graph-Cut Initialization and Expectation-Maximization Algorithm

This letter presents a multistage clustering technique for unsupervised classification that is based on the following: 1) a graph-cut procedure to produce initial segments that are made up of pixels with similar spatial and spectral properties; 2) a fuzzy c-means algorithm to group these segments into a fixed number of classes; 3) a proper implementation of the expectation-maximization (EM) algorithm to estimate the statistical parameters of classes on the basis of the initial seeds that are achieved at convergence by the fuzzy c-means algorithm; and 4) the Bayes rule for minimum error to perform the final classification on the basis of the distributions that are estimated with the EM algorithm. Experimental results confirm the effectiveness of the proposed technique.     

基于对应分析的训练样本的选择

本文提出一种基于对应分析的训练样本的选择方法。它从训练样本中自动地选择有代表性的典型训练样本,使得在自动分类中充分利用所采集的样本信息,以便得到满意的分类结果。通过实验与分析证明,该方法是可行的,它明显优于人工随机选择训练样本的方式。与基于Q型因子分析的训练样本选择方法相比,可以更快地得到较少的典型样本,满意的分类精度。     

A Neural Network Landuse Classifier for SAR Images Using Textural and Fractal Information

Abstract

Artificial neural networks (ANN) have recently been popularly used in image classification. Input features to most ANNs are extracted based on a one class per pixel basis. This requires a large number of training samples and thus a slow training rate. In this paper, we describe the use of a windowing technique to extract textural features such as average intensity, second moment of intensity histogram and fractal surface dimension from an image. This method of image characterization reduces the number of training samples efficiently, yet retains a reasonable overall classification accuracy. The ANN is trained based on the back‐error propagation algorithm. The method is applied for landuse classification of Synthetic Aperture Radar (SAR) images. An example is given for a site in Kedah State, Malaysia. The SAR images (HH,HV,VV) were taken by the Canadian Centre for Remote Sensing (CCRS) CV‐580 airborne C‐band SAR system in November 1993 during their GlobeSAR mission in Malaysia. These multi‐polarization SAR images are co‐registered with a Landsat Thematic Mapper (TM) channel 5 image from same area. An overall classification accuracy of about 86.95% is achieved using windowing technique, as compared to 68.22% based on one class per pixel approach. This shows that through fractal and textural information, the windowing technique when applied in an ANN classifier has a great potential in remote sensing applications.     

Performance analysis of different predictive models for crop classification across an aridic to ustic area of Indian states

The purpose of this study is to present comparative performance analysis of different machine learning algorithms for large area crop classification. Ten Indian districts with significant rabi crops viz. wheat, mustard, gram, red lentils (masoor) have been selected for the study. Most popular classical ensemble models – bagging/ARCing, random forest (RF), gradient boosting and Importance Sampled Learning Ensemble (ISLE) with traditional single model (decision tree) have been selected for comparative analysis. To incorporate dependency of large area crop in different variables viz. parent material and soil, phenology, texture, topography, soil moisture, vegetation, climate etc., 35 digital layers are prepared using different satellite data (ALOS DEM, Landsat-8, MODIS NDVI, RISAT-1, Sentinental-1A) and climatic data (precipitation, temperature). In rabi season, field survey about crop type is carried out to prepare training data. Performance is evaluated on the basis of marginal rates, F-measure and Jaccard’s coefficient of community, Classification Success Index and Agreement Coefficients. Score is calculated to rank the algorithm. RF is best performer followed by gradient boosting for crop classification. Other ensemble methods ARCing, bagging and ISLE are in decreasing order of performance. Traditional non-ensemble method decision tree scored higher than ISLE.     

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。