A novel convolutional neural network architecture of multispectral remote sensing images for automatic material classification

For real-world simulation, terrain models must combine various types of information on material and texture in terrain reconstruction for the three-dimensional numerical simulation of terrain. However, the construction of such models using the conventional method often involves high costs in both manpower and time. Therefore, this study used a convolutional neural network (CNN) architecture to classify material in multispectral remote sensing images to simplify the construction of future models. Visible light (i.e., RGB), near infrared (NIR), normalized difference vegetation index (NDVI), and digital surface model (DSM) images were examined.This paper proposes the use of the robust U-Net (RUNet) model, which integrates multiple CNN architectures, for material classification. This model, which is based on an improved U-Net architecture combined with the shortcut connections in the ResNet model, preserves the features of shallow network extraction. The architecture is divided into an encoding layer and a decoding layer. The encoding layer comprises 10 convolutional layers and 4 pooling layers. The decoding layer contains four upsampling layers, eight convolutional layers, and one classification convolutional layer. The material classification process in this study involved the training and testing of the RUNet model. Because of the large size of remote sensing images, the training process randomly cuts subimages of the same size from the training set and then inputs them into the RUNet model for training. To consider the spatial information of the material, the test process cuts multiple test subimages from the test set through mirror padding and overlapping cropping; RUNet then classifies the subimages. Finally, it merges the subimage classification results back into the original test image.The aerial image labeling dataset of the National Institute for Research in Digital Science and Technology (Inria, abbreviated from the French Institut national de recherche en sciences et technologies du numérique) was used as well as its configured dataset (called Inria-2) and a dataset from the International Society for Photogrammetry and Remote Sensing (ISPRS). Material classification was performed with RUNet. Moreover, the effects of the mirror padding and overlapping cropping were analyzed, as were the impacts of subimage size on classification performance. The Inria dataset achieved the optimal results; after the morphological optimization of RUNet, the overall intersection over union (IoU) and classification accuracy reached 70.82% and 95.66%, respectively. Regarding the Inria-2 dataset, the IoU and accuracy were 75.5% and 95.71%, respectively, after classification refinement. Although the overall IoU and accuracy were 0.46% and 0.04% lower than those of the improved fully convolutional network, the training time of the RUNet model was approximately 10.6 h shorter. In the ISPRS dataset experiment, the overall accuracy of the combined multispectral, NDVI, and DSM images reached 89.71%, surpassing that of the RGB images. NIR and DSM provide more information on material features, reducing the likelihood of misclassification caused by similar features (e.g., in color, shape, or texture) in RGB images. Overall, RUNet outperformed the other models in the material classification of remote sensing images. The present findings indicate that it has potential for application in land use monitoring and disaster assessment as well as in model construction for simulation systems.     

Classification of multispectral images based on a fuzzy-possibilistic neural network

In this paper, a new Hopfield-model net based on fuzzy possibilistic reasoning is proposed for the classification of multispectral images. The main purpose is to modify the Hopfield network embedded with fuzzy possibilistic C-means (FPCM) method to construct a classification system named fuzzy-possibilistic Hopfield net (FPHN). The classification system is a paradigm for the implementation of fuzzy logic systems in neural network architecture. Instead of one state in a neuron for the conventional Hopfield nets, each neuron occupies 2 states called membership state and typicality state in the proposed FPHN. The proposed network not only solves the noise sensitivity fault of Fuzzy C-means (FCM) but also overcomes the simultaneous clustering problem of possibilistic C-means (PCM) strategy. In addition to the same characteristics as the FPCM algorithm, the simple features of this network are clear potential in optimal problem. The experimental results show that the proposed FPHN can obtain better solutions in the classification of multispectral images.     

Multiset canonical correlations analysis and multispectral, trulymultitemporal remote sensing data

This paper describes two- and multiset canonical correlations analysis (CCA) for data fusion, multisource, multiset, or multitemporal exploratory data analysis. These techniques transform multivariate multiset data into new orthogonal variables called canonical variates (CVs) which, when applied in remote sensing, exhibit ever-decreasing similarity (as expressed by correlation measures) over sets consisting of (1) spectral variables at fixed points in time (R-mode analysis), or (2) temporal variables with fixed wavelengths (T-mode analysis). The CVs are invariant to linear and affine transformations of the original variables within sets which means, for example, that the R-mode CVs are insensitive to changes over time in offset and gain in a measuring device. In a case study, CVs are calculated from Landsat Thematic Mapper (TM) data with six spectral bands over six consecutive years. Both Rand T-mode CVs clearly exhibit the desired characteristic: they show maximum similarity for the low-order canonical variates and minimum similarity for the high-order canonical variates. These characteristics are seen both visually and in objective measures. The results from the multiset CCA R- and T-mode analyses are very different. This difference is ascribed to the noise structure in the data. The CCA methods are related to partial least squares (PLS) methods. This paper very briefly describes multiset CCA-based multiset PLS. Also, the CCA methods can be applied as multivariate extensions to empirical orthogonal functions (EOF) techniques. Multiset CCA is well-suited for inclusion in geographical information systems (GIS).     

A dynamic learning neural network for remote sensing applications

The neural network learning process is to adjust the network weights to adapt the selected training data. Based on the polynomial basis function (PBF) modeled neural network that is a modified multilayer perceptrons (MLP) network, a dynamic learning algorithm (DL) is proposed. The presented learning algorithm makes use of the Kalman filtering technique to update the network weights, in the sense that the stochastic characteristics of incoming data sets are implicitly incorporated into the network. The Kalman gains which represent the learning rates of the network weights updating are calculated by using the U-D factorization. By concatenating all of the network weights at each layer to form a long vector such that it can be updated without propagating back, the proposed algorithm improves the performance of convergence to which the backpropagation (BP) learning algorithm often suffers. Numerical illustrations are carried out using two categories of problems: multispectral imagery classification and surface parameters inversion. Results indicates the use of Kalman filtering algorithm not only substantially increases the convergence rate in the learning stage, but also enhances the separability for highly nonlinear boundaries problems, as compared to BP algorithm, suggesting that the proposed DL neural network provides a practical and potential tool for remote sensing applications     

多光谱可见光遥感图像压缩系统设计

为了实现多光谱可见光遥感图像高质量压缩的要求,提出以JPEG2000压缩标准为理论,将FPGA与专用压缩芯片ADV212相结合的空间遥感图像压缩方法.该系统设计采用ADV212,通过小波变换及熵编码实现对大数据量的空间遥感图像进行高质量实时压缩,并且采用FPGA完成图像数据输入、压缩码流输出、图像预处理以及对ADV212的工作模式进行控制.实验结果表明,该系统设计功耗低、成本低、调试简单合理,具有较好的压缩效果,可满足多光谱可见光遥感图像对高质量压缩比的要求.     

The least-squares mixing models to generate fraction images derivedfrom remote sensing multispectral data

Constrained-least-squares (CLS) and weighted-least-squares (WLS) mixing models for generating fraction images derived from remote sensing multispectral data are presented. An experiment considering three components within the pixels-eucalyptus, soil (understory), and shade-was performed. The generated fraction images for shade (shade image) derived from these two methods were compared by considering the performance and computer time. The derived shade images are related to the observed variation in forest structure, i.e. the fraction of inferred shade in the pixel is related to different eucalyptus ages     

基于卷积神经网络的卫星遥感图像区域识别

为了提高卫星遥感图像的识别与分类效果,提出一种基于卷积神经网络的卫星遥感图像识别与分类方法。该方法通过导向滤波去雾和旋转图像数据提高了模型的泛化能力,同时采用了双全连接层网络结构增强了模型数据表达能力。实验证明,该方法在卫星遥感图像的识别与分类上优于传统图像识别方法和一般卷积神经网络模型。     

遥感图像分类中的遗传算法LVQ神经网络运用

对于遥感图像分类过程中的问题,提出遗传算法LVQ神经网络来实现遥感图像的分类。将LVQ神经网络结合遗传算法,使用遗传算法最优阈值与权值实现网络训练,使分类精度得到提高。之后融合相似灰度值创建分类图像特征矢量,使特征矢量在神经网络中输入实现训练。学习矢量量化神经算法对初值非常敏感,对遥感图像分类精度具有一定影响。最后,为了对性能进行测试,在实验过程中对比本文分类方法和SVM决策树分类方法,通过实验结果表示,文中提出的分类方法的遥感图像分类精度为95.82%,与其他分类方法相比,分类精度得到进一步提高。     

基于遥感影像和神经网络的城市用地功能分类

城市用地功能分类的准确识别对精准把握城市现状、优化城市空间结构有重要意义。基于此,利用高分辨力遥感影像,提出一种针对中国城市用地功能分类的模型。设计一种多分辨力特征融合的卷积神经网络识别遥感影像中的特定功能区;针对中国城市功能区分布的特点,建立一个用于城市用地功能分类的新数据集。实验显示,本文算法在6种用地功能类型上的分类精确度达88%,表明算法对城市用地功能分类识别具有较高的准确性。最后,通过对北京部分主要城区的案例研究,验证了所提出的模型在城市规划相关领域提供数据支持的价值和有效性。     

多偏移遥感图像的BP神经网络亚像元定位

提出了一种借助多偏移遥感图像来改进基于BP神经网络(BPNN)的亚像元定位新方法.不同于原BPNN方法使用单幅低空间分辨率观测图像,新方法利用多幅带有亚像元偏移的低空间分辨图像来确定亚像元属于各类的概率,然后根据概率值和地物覆盖比例确定亚像元类别,以降低BPNN定位模型中的不确定性和误差.实验表明,提出方法在视觉和定量评价上,均能获得更高精度的亚像元定位结果,验证了提出方法的有效性.     

The potential role of thermal infrared multispectral scanners in geological remote sensing

Thermal infrared scanners with multiple channels in the 8-14-µm wavelength region are important for geological remote sensing because this spectral region contains important compositional information about silicate rocks and minerals that cannot be duplicated by remote sensors operating elsewhere in the electromagnetic spectrum. Emittance minima in this spectral region, caused by interatomic oscillations, occur at different wavelengths depending on silicate rock type. It has been demonstrated that an image constructed from a signal that is proportional to a ratio of radiances in two thermal scanner channels can be used to map compositional (chemical and mineral) variations in silicates, while suppressing temperature variations across a scanned scene. Theoretical studies indicate that future infrared scanners with eight to twelve channels in the 8-14-µm region might be used to produce an image that could be simply level-sliced (divided into discrete gray levels) to map silicate rocks according to traditional rock classification charts. This is a field in which sensor technology is still the limiting factor. However, improvements in extrinsic sensor properties, especially an increase in the number of spatially coregistered detector elements in a single dewar, are more important than improvements in intrinsic properties, such as detectivity.     

Classification of multispectral image data by the binary diamondneural network and by nonparametric, pixel-by-pixel methods

The study deals with the application of nonparametric pixel-by-pixel classification methods in the classification of pixels, based on their multispectral data. A neural network, the binary diamond, is introduced, and its performance is compared with a nearest neighbor algorithm and a back-propagation network. The binary diamond is a multilayer, feedforward neural network, which learns from examples in unsupervised one-shot mode. It recruits its neurons according to the actual training set, as it learns. The comparisons of the algorithms were done using a realistic database, consisting of approximately 90000 Landsat 4 Thematic Mapper pixels. The binary diamond and the nearest neighbor performances were close, with some advantages to the binary diamond. The performance of the back-propagation network lagged behind. An efficient nearest neighbor algorithm, the binned nearest neighbor, is described. Ways for improving the performances, such as merging categories and analyzing nonboundary pixels, are addressed and evaluated     

应用相位一致性评价多光谱遥感图像条带噪声

条带噪声是影响多光谱遥感图像质量的重要因素之一,严重影响遥感数据的解译和信息提取。提出一种应用相位一致性进行多光谱遥感图像条带噪声质量评价的方法。该方法不受图像亮度或对比度变化的影响,以频域中相位一致性探测图像特征点,综合考虑图像中条带噪声的数量、长度、宽度和强度建立评价因子,客观评价遥感图像中存在的条带噪声对图像质量的影响。实验证明该方法评价结果符合人眼视觉感官评价。     

基于ISOMAP的高光谱遥感数据的降维与分类

为挖掘高光谱遥感数据内在的非线性结构特性,采用全局化流形学习算法等距特征映射（ISOMAP）对高光谱遥感数据进行非线性降维,并取得了优于常用的最小噪声分离（MNF）变换方法的结果,具有更好的数据压缩性能。将光谱角相似性度量方法用于ISOMAP 算法,取得良好的降维效果。通过把ISOMAP 降维算法和k-最邻近分类器相结合对降维后子空间特征进行分类,实验表明:ISOMAP 利用较少的特征维数获得比MNF 更高的分类精度,并达到较高稳定的分类精度,尤其对难以区分、光谱相似的两类别问题,ISOMAP 的特征维数能够有效的提高两类别的可分性。     

基于非线性核空间映射与人工免疫网络的高光谱遥感图像分类

提出了一种基于非线性核空间映射人工免疫网络的高光谱遥感图像分类算法.根据生物免疫网络基本原理构建了人工免疫网络模型,利用非线性核函数将高光谱训练样本映射到高维空间,完善了人工免疫网络中目标样本核空间相似性分选方法,降低了人工免疫网络识别样本所需的抗体数量,提升了算法的分类精度和运算效率.为了验证算法的有效性,利用两组高光谱遥感数据将多种高光谱分类方法进行了对比实验.实验表明该算法分类精度和算法运算时间上都有较大改善,是一种分类精度更高、运算速度更快的改进型基于人工免疫网络的高光谱遥感图像分类新方法.     

ART neural networks for remote sensing: vegetation classificationfrom Landsat TM and terrain data

A new methodology for automatic mapping from Landsat thematic mapper (TM) and terrain data, based on the fuzzy ARTMAP neural network, is developed. System capabilities are tested on a challenging remote sensing classification problem, using spectral and terrain features for vegetation classification in the Cleveland National Forest. After training at the pixel level, system performance is tested at the stand level, using sites not seen during training. Results are compared to those of maximum likelihood classifiers, as well as back propagation neural networks and K nearest neighbor algorithms. ARTMAP dynamics are fast, stable, and scalable, overcoming common limitations of back propagation. Best results are obtained using a hybrid system based on a convex combination of fuzzy ARTMAP and maximum likelihood predictions. A prototype remote sensing example introduces each aspect of data processing and fuzzy ARTMAP classification. The example shows how the network automatically constructs a minimal number of recognition categories to meet accuracy criteria. A voting strategy improves prediction and assigns confidence estimates by training the system several times on different orderings of an input set     

应用监督近邻重构分析的高光谱遥感数据特征提取

针对高光谱遥感数据特征提取方法的研究,提出了一种新的监督近邻重构分析（Supervised Neighbor Reconstruction Analysis,SNRA）算法。该方法首先利用同一类别的近邻数据点对各数据点进行重构;然后在低维嵌入空间中保持该重构关系不变,尽可能地分离开非同类数据点,并利用总体散度矩阵来约束数据间的相关性;最后求解得到一个最佳投影矩阵,进而提取出鉴别特征。SNRA算法不仅保持了同类数据的局部结构而且增强了非同类数据的可分性,同时减少了数据的冗余信息。在Indian Pine和KSC高光谱遥感数据集上的实验结果表明:提出的方法能更好地揭示出高光谱遥感数据的内在特性,提取出更有效的鉴别特征,改善分类效果。     

Classification of multifrequency polarimetric SAR imagery using adynamic learning neural network

A practical method for extracting microwave backscatter for terrain-cover classification is presented. The test data are multifrequency (P, L, C bands) polarimetric SAR data acquired by JPL over an agricultural area called “Flevoland”. The terrain covers include forest, water, bare soil, grass, and eight other types of crops. The radar response of crop types to frequency and polarization states were analyzed for classification based on three configurations: 1) multifrequency and single-polarization images; 2) single-frequency and multipolarization images; and 3) multifrequency and multipolarization images. A recently developed dynamic learning neural network was adopted as the classifier. Results show that using partial information, P-band multipolarization images and multiband hh polarization images have better classification accuracy, while with a full configuration, namely, multiband and multipolarization, gives the best discrimination capability. The overall accuracy using the proposed method can be as high as 95% with a total of thirteen cover types classified. Further reduction of the data volume by means of correlation analysis was conducted to single out the minimum data channels required. It was found that this method efficiently reduces the data volume while retaining highly acceptable classification accuracy     

Hydrometeorological data extension for the Sudan-Sahel zone of WestAfrica using satellite remote sensing

Conventional techniques for assessing and investigating water resources in the Sudan and Sahel zones of West Africa, which are subject to frequent and prolonged droughts, are often difficult to implement. Satellite remote sensing coupled with available ground truth data from a limited number of rain gauge locations provides an alternative means of hydrometric data extension within the zone. Monthly and seasonal rainfall estimates, based on a database derived from Meteosat infrared data, are determined. Rain gauge data are regressed on five-day maximum surface temperatures and cold cloud occurrences. A linear model is used for dry month estimates. Values of the coefficient of determination vary from 0.55 to 0.69 for the monthly model, and the value is 0.83 for the seasonal model     

Classification of sport videos using edge-based features and autoassociative neural network models

In this paper, we propose a method for classification of sport videos using edge-based features, namely edge direction histogram and edge intensity histogram. We demonstrate that these features provide discriminative information useful for classification of sport videos, by considering five sports categories, namely, cricket, football, tennis, basketball and volleyball. The ability of autoassociative neural network (AANN) models to capture the distribution of feature vectors is exploited, to develop class-specific models using edge-based features. We show that combining evidence from complementary edge features results in improved classification performance. Also, combination of evidence from different classifiers like AANN, hidden Markov model (HMM) and support vector machine (SVM) helps improve the classification performance. Finally, the performance of the classification system is examined for test videos which do not belong to any of the above five categories. A low rate of misclassification error for these test videos validates the effectiveness of edge-based features and AANN models for video classification.