Comparative study on projected clustering methods for hyperspectral imagery classification

In this study, projected clustering is introduced to hyperspectral imagery for unsupervised classification. The main advantage of projected clustering lies in its ability to simultaneously perform feature selection and clustering. This framework also allows selection of different sets of dimensions (features/bands) for different clusters. This framework provides an effective way to address the issues associated with the high dimensionality of the data. Experiments are conducted on both synthetic and real hyperspectral imagery. For this purpose, projected clustering algorithms are implemented and compared with k-means and k-means preceded by principal component analysis. Preliminary analyses of studied algorithms on synthetic hyperspectral imagery demonstrate good results. For real hyperspectral imagery, only ORCLUS is able to produce acceptable results as compared to other unsupervised methods. The main concern lies with identification of right parameter settings. More experiments are required in this direction.     

综合聚类和上下文特征的高光谱影像分类

常规高光谱影像逐像素分类往往没有考虑空间相关性,分类结果未体现地物的空间关联和分布特征。为了在分类中充分利用空间特征,利用聚类信息并结合隐马尔可夫随机场模型讨论了高光谱遥感影像光谱-空间分类方法。首先,在不同特征提取方法（最小噪声分离、独立成分分析和主成分分析）下,使用不同聚类方法（k-均值、迭代自组织分析算法和模糊c-均值算法）借助隐马尔可夫随机场获取优化的分割图;然后,采用4连通区域标记法对分割区域标记生成图像对象,并根据支持向量机的逐像素分类结果采用多数投票法对图像对象进行分类;最后,借助凹槽窗口邻域滤波技术改进分类结果,削弱“椒盐”现象。该方法综合了监督分类和非监督分类的优势,通过聚类引入地物空间相关性信息,通过隐马尔可夫随机场引入上下文特征,较好地弥补了单纯基于光谱信息分类的不足。     

An ensemble pansharpening approach for finer-scale mapping of sugarcane with Landsat 8 imagery

We tested the effects of three fast pansharpening methods – Intensity-Hue-Saturation (IHS), Brovey Transform (BT), and Additive Wavelet Transform (AWT) – on sugarcane classification in a Landsat 8 image (bands 1–7), and proposed two ensemble pansharpening approaches (band stacking and band averaging) which combine the pixel-level information of multiple pansharpened images for classification. To test the proposed ensemble pansharpening approaches, we classified “sugarcane” and “other” land cover in the unsharpened Landsat multispectral image, the individual pansharpened images, and the band-stacked and band-averaged ensemble images using Support Vector Machines (SVM), and assessed the classification accuracy of each image. Of the individual pansharpened images, the AWT image achieved higher classification accuracy than the unsharpened image, while the IHS and BT images did not. The band-stacked ensemble images achieved higher classification accuracies than the unsharpened and individual pansharpened images, with the IHS-BT-AWT band-stacked image producing the most accurate classification result, followed by the IHS-BT band-stacked image. The ensemble images containing averaged pixel values from multiple pansharpened images achieved lower classification accuracies than the band-stacked ensemble images, but most still had higher accuracies than the unsharpened and individual pansharpened results. Our results indicate that ensemble pansharpening approaches have the potential to increase classification accuracy, at least for relatively simple classification tasks. Based on the results of the study, we recommend further investigation of ensemble pansharpening for image analysis (e.g. classification and regression tasks) in agricultural and non-agricultural environments.     

一种高光谱遥感图像快速谱聚类算法

为获得分类效果更优良的遥感图像分类方式并解决高光谱遥感图像分类运算速度缓慢的问题,集成Lanczos算法与谱聚类算法,探讨了高光谱遥感图像谱聚类算法应用于遥感图像分类的可行性,提出了一种面向高光谱遥感图像的快速谱聚类算法;通过对比美国圣地亚哥机场高光谱遥感图像K-均值算法与谱聚类算法的分类结果,发现面向高光谱遥感图像的谱聚类算法易于识别线性地物,且分类的速度能得到较大提升。     

Improving classification accuracy of spectrally similar land covers in the rangeland and plateau areas with a combination of WorldView-2 and UAV images

This study aims to increase the accuracy of the object based classification approach to differentiate the spectrally similar land cover types to create thematic maps depicting the current land use status in rangeland. Firstly, the multispectral and panchromatic bands of a WorldView-2 MS and Pan images are fused. The fused WV-2 image is then classified with object based approach using Support Vector Machines (SVMs) classifier (Method 1). The overall classification accuracy for Method 1 is found to be 88.6%. Secondly, UAV ortho-image is utilised for segmentation process, which is required for the object based SVM classification of the WV-2 MS image (Method 2). The overall classification accuracy for Method 2 is obtained as 92.4%. It is realised that the Method 2 increases the object based classification accuracy by 4%, compared to Method 1. This result reveals that the object based classification of the UAV and WV-2 MS images makes significant contribution to the classification accuracy.     

Assessing the performance of two unsupervised dimensionality reduction techniques on hyperspectral APEX data for high resolution urban land-cover mapping

Despite the high richness of information content provided by airborne hyperspectral data, detailed urban land-cover mapping is still a challenging task. An important topic in hyperspectral remote sensing is the issue of high dimensionality, which is commonly addressed by dimensionality reduction techniques. While many studies focus on methodological developments in data reduction, less attention is paid to the assessment of the proposed methods in detailed urban hyperspectral land-cover mapping, using state-of-the-art image classification approaches. In this study we evaluate the potential of two unsupervised data reduction techniques, the Autoassociative Neural Network (AANN) and the BandClust method – the first a transformation based approach, the second a feature-selection based approach – for mapping of urban land cover at a high level of thematic detail, using an APEX 288-band hyperspectral dataset. Both methods were tested in combination with four state-of-the-art machine learning classifiers: Random Forest (RF), AdaBoost (ADB), the multiple layer perceptron (MLP), and support vector machines (SVM). When used in combination with a strong learner (MLP, SVM) BandClust produces classification accuracies similar to or higher than obtained with the full dataset, demonstrating the method’s capability of preserving critical spectral information, required for the classifier to successfully distinguish between the 22 urban land-cover classes defined in this study. In the AANN data reduction process, on the other hand, important spectral information seems to be compromised or lost, resulting in lower accuracies for three of the four classifiers tested. Detailed analysis of accuracies at class level confirms the superiority of the SVM/Bandclust combination for accurate urban land-cover mapping using a reduced hyperspectral dataset. This study also demonstrates the potential of the new APEX sensor data for detailed mapping of land cover in spatially and spectrally complex urban areas.     

面向对象的高光谱遥感影像稀疏表示分类

稀疏表示用于高光谱遥感影像分类多是基于像素层次来处理的。文中提出一种面向对象的高光谱遥感影像稀疏表示分类方法。首先从高光谱影像中提取4个波段组成标准的多波段影像,进行面向对象的影像分割;然后计算各对象在各波段上的光谱均值,并选取少量样本进行训练;最后利用基于Fisher字典学习的稀疏表示进行高光谱遥感影像的分类。实验结果表明,该方法可以利用较少的样本得到较好的分类效果,与基于像素层的稀疏分类相比较,分类精度与效率均有所提高,分类结果更接近真实地物,避免了零碎图斑。     

鲁棒多特征谱聚类的高光谱影像波段选择

传统谱聚类的高光谱影像波段选择模型中,采用的波段相似矩阵受到噪声或异常值的影响且仅能表征波段的单一相似特征,导致波段子集的选取结果受到限制.本文从波段选择的目的 出发,提出鲁棒多特征谱聚类方法,整合多个特征的波段相似矩阵来形成综合相似矩阵以解决上述问题.该方法假设4种相似性度量包括光谱信息散度、光谱角度距离、波段相关性...     

Evaluation of classifiers for processing Hyperion (EO-1) data of tropical vegetation

There is an urgent necessity to monitor changes in the natural surface features of earth. Compared to broadband multispectral data, hyperspectral data provides a better option with high spectral resolution. Classification of vegetation with the use of hyperspectral remote sensing generates a classical problem of high dimensional inputs. Complexity gets compounded as we move from airborne hyperspectral to Spaceborne technology. It is unclear how different classification algorithms will perform on a complex scene of tropical forests collected by spaceborne hyperspectral sensor. The present study was carried out to evaluate the performance of three different classifiers (Artificial Neural Network, Spectral Angle Mapper, Support Vector Machine) over highly diverse tropical forest vegetation utilizing hyperspectral (EO-1) data. Appropriate band selection was done by Stepwise Discriminant Analysis. The Stepwise Discriminant Analysis resulted in identifying 22 best bands to discriminate the eight identified tropical vegetation classes. Maximum numbers of bands came from SWIR region. ANN classifier gave highest OAA values of 81% with the help of 22 selected bands from SDA. The image classified with the help SVM showed OAA of 71%, whereas the SAM showed the lowest OAA of 66%. All the three classifiers were also tested to check their efficiency in classifying spectra coming from 165 processed bands. SVM showed highest OAA of 80%. Classified subset images coming from ANN (from 22 bands) and SVM (from 165 bands) are quite similar in showing the distribution of eight vegetation classes. Both the images appeared close to the actual distribution of vegetation seen in the study area. OAA levels obtained in this study by ANN and SVM classifiers identify the suitability of these classifiers for tropical vegetation discrimination.     

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.     

Improving Spectral–spatial Classification of Hyperspectral Imagery Using Spectral Dimensionality Reduction Based on Weighted Genetic Algorithm

This paper aims to improve the accuracy and the efficiency of high resolution land cover mapping in urban area. To this end, an improved approach for classification of hyperspectral imagery is proposed and evaluated. This approach benefits from both inherent spectral and spatial information of an image. The weighted genetic (WG) algorithm is first used to obtain the subspace of hyperspectral data. The obtained features are then fed into the enhanced marker-based minimum spanning forest (EMSF) classification algorithm. In this algorithm, the markers are extracted from the classification maps obtained by both support vector machine and watershed segmentation algorithm classifiers. For this purpose, the class’s pixels with the largest population in the classification map are kept for each region of the segmentation map. Then, the most reliable classified pixels are chosen from among the exiting pixels as markers. To evaluate the efficiency of the proposed approach, three hyperspectral data sets acquired by ROSIS-03, Hymap and Hyper-Cam LWIR are used. Experimental results showed that the proposed WG–EMSF approach achieves approximately 9, 8 and 6% better overall accuracy than the original MSF-based algorithm for these data sets respectively.     

小样本的高光谱图像降噪与分类

在样本数目稀少情况下实现高光谱图像精细分类是个挑战性的问题。高光谱图像信噪比提高比较困难,噪声大小对分类结果有最直接的影响。利用高光谱图像相邻波段之间的相关性和相邻像素之间的相关性,提出多级降噪滤波的高光谱图像分类方法,通过改进的两阶段稀疏与低秩矩阵分解方法,去除高光谱图像中能量较高的噪声,利用主成分分析方法去除高光谱图像中能量较低的噪声,引导滤波方法去除分类结果图中的"椒盐噪声"。选取两幅真实高光谱图像进行实验,结果表明,两阶段稀疏与低秩矩阵分解法和主成分分析法两种降噪方法具有较强的互补性;引导滤波方法使得分类图更加平滑且分类精度更高。与其他光谱空间分类方法相比,本文方法分类精度更高,且在样本极少时能获得很高的分类精度。     

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.     

遥感图像分割的低维纹理特征算子与双变异蝴蝶优化算法

针对光学遥感影像受天气影响导致部分地物间形成弱边缘问题,提出一种低维纹理特征算子与双变异蝴蝶优化算法。首先提出一种适用于遥感影像的低维完备局部三值模式的纹理特征提取算子,并将其引入简单线性迭代聚类算法,对遥感影像进行初始分割,减小了噪音影响,同时增强算法对弱边缘的敏感度和分割准确性;然后采用双变异蝴蝶优化的支持向量机合并同质超像素块,以简单线性迭代聚类算法和低维纹理特征算子得到的综合特征作为输入,得到最终分割图像。利用2组高分辨率遥感影像进行分割实验,并与当下流行的卷积神经网络进行对比,实验结果表明,所提算法相较于传统算法对弱边缘有更好的分割效果,数据一的边界回归率(boundary recall,BR)值较对比算法平均提高了1.9%,Kappa系数平均提高了0.036;数据二的BR值较对比算法平均提高了2.33%,Kappa系数平均提高了0.027。对比实验证明了所提算法相较于卷积神经网络有更好的泛化性。     

Investigating Issues in Map Accuracy When Using an Object-Based Approach to Map Benthic Habitats

In a recent study, benthic habitat maps were created of the Texas Gulf Coast from digital aerial imagery. The images were classified using an object-based image analysis (OBIA) approach and a classification and regression tree (CART) technique. The map was manually edited, changing 26% of the polygons' labels. Accuracy assessments of the unedited map and the edited map revealed the two were not significantly different. The research in this paper evaluates why these maps may have similar accuracies. Our analyses indicate that the small segmentation scale parameter used over-segmented the imagery, reducing the effectiveness of the CART technique and editing.     

Learning with transductive SVM for semisupervised pixel classification of remote sensing imagery

Land cover classification using remotely sensed data requires robust classification methods for the accurate mapping of complex land cover area of different categories. In this regard, support vector machines (SVMs) have recently received increasing attention. However, small number of training samples remains a bottleneck to design suitable supervised classifiers. On the other hand, adequate number of unlabeled data is available in remote sensing images which can be employed as additional source of information about margins. To fully leverage all of the precious unlabeled data, integration of filtering in a transductive SVM is proposed.Using two labeled image datasets of small size and two large unlabeled image datasets, the effectiveness of the proposed method is explored. Experimental results show that the proposed technique achieves average overall accuracies of around 4.5–7.8%, 0.8–2.6% and 0.9–2.2% more than the standard inductive SVM (ISVM), progressive transductive SVM (PTSVM) and low density separation (LDS) classifiers, respectively on larger domains in case of labeled datasets. Using image datasets, visual interpretation from the classified images as well as the segmentation quality reveal that the proposed method can efficiently filter informative data from the unlabeled samples.     

Multi-Feature Classification Approach for High Spatial Resolution Hyperspectral Images

High spatial resolution hyperspectral images not only contain abundant radiant and spectral information, but also display rich spatial information. In this paper, we propose a multi-feature high spatial resolution hyperspectral image classification approach based on the combination of spectral information and spatial information. Three features are derived from the original high spatial resolution hyperspectral image: the spectral features that are acquired from the auto subspace partition technique and the band index technique; the texture features that are obtained from GLCM analysis of the first principal component after principal component analysis is performed on the original image; and the spatial autocorrelation features that contain spatial band X and spatial band Y, with the grey level of spatial band X changing along columns and the grey level of spatial band Y changing along rows. The three features are subsequently combined together in Support Vector Machine to classify the high spatial resolution hyperspectral image. The experiments with a high spatial resolution hyperspectral image prove that the proposed multi-feature classification approach significantly increases classification accuracies.     

结合mRMR选择和IFCM聚类的遥感影像分类算法

为解决高分影像特征间相关性大冗余度高、FCM聚类稳健性差带来的分类精度不佳问题,提出一种基于mRMR选择和改进FCM聚类的影像分类算法。首先基于对象置信度指标（OC）进行影像分割,然后利用mRMR算法实现特征选择,解决特征冗余问题,最后将提取的特征输入分类器通过IFCM聚类,得到最终分类结果。试验结果表明,本文算法能减少特征间相关性,降低冗余,并有效提高影像分类精度。     

像素级和对象级高分辨率遥感图像变化检测分析

高分辨率遥感图像具有丰富的纹理信息,而像素级变化检测方法主要分析图像的光谱信息,导致将像素级变化检测方法用于高分辨率遥感图像具有一定的局限性。因此,本文提出了一种像素级与对象级相结合的高分辨率遥感图像变化检测方法,解决了像素级与对象级变化检测方法中存在的椒盐现象、误检等问题。首先,结合高分辨率遥感图像的多维特征,构建遥感图像变化检测模型;其次,利用随机森林分类器对图像进行分类,得到像素级变化检测结果;最后,将像素级变化检测结果与图像对象分割结果进行融合,得到图像变化区域和不变区域。试验结果表明,该算法具有较高的准确率和检测精度。     

利用SVM-CRF进行高光谱遥感数据分类

提出了一种改进的随机场模型SVM-CRF,它以支持向量机作为条件随机场的一阶势能项,结合了支持向量机和条件随机场的优点。采用AVIRIS高光谱遥感数据进行实验,对SVM-CRF模型进行了分析,结果表明,在分类精度上SVM-CRF优于支持向量机和传统条件随机场模型。