Band Selection and Dimension Estimation for Hyperspectral Imagery—a New Approach Based on Invasive Weed Optimization

Currently, hyperspectral images have potential applications in many scientific areas due to the high spectral resolution. Extracting suitable and adequate bands/features from high dimensional data is a crucial task to classify such data. To overcome this issue, dimension reduction techniques have direct effects to improve the efficiency of classifiers on hyperspectral images. One common approach for decreasing the dimensionality is the feature/band selection by considering the optimum dimensionality of the hyperspectral imagery. In this paper, a new method was proposed to select optimal band for classification application, based on a metaheuristic Invasive Weed Optimization (IWO) algorithm. In this regard, the K-nearest neighbour (K-NN) technique was used as the classifier. Moreover, as a by-product of our band selection method, a new method was proposed to estimate an optimum dimension of the reduced hyperspectral images for better classification. Experimental results over three real-world hyperspectral datasets clearly showed that the proposed IWO-based band selection algorithm of this study led to the significant progress in selecting suitable bands for classification applications and estimation of optimum dimensionality of these datasets. In this regard, the overall accuracy (OA) of classification of the proposed IWO-based band selection algorithm was 92.02, 93.57, and 89.72 % for each dataset, respectively. Moreover, results reveal the superiority of the proposed IWO-based band selection algorithm against the other algorithms including GA, SA, ACO, and PSO for band selection purpose.     

Visual method for spectral band selection

We present a new method for performing band selection experiments with spectral data. This method allows for the visual inspection and assessment of the experiment results, and includes a statistical significance test. The method follows a standard feature selection approach in which a multivariate distance measure is used as a figure of merit in a search-optimization procedure. For this letter, we have chosen the Jeffries-Matusita distance between each sample and its immediate background. The band selection methodology uses either an exhaustive search over all possible combinations of 1-4 bands or sequential forward selection. To analyze the band selection results, we count the number of times that each band is selected as a member of the best set by the protocol, and we plot the results as a band frequency histogram. This allows us to visually discern spectral patterns that are not evident otherwise, and thus better assess the utility of each spectral band. We can compute band frequency histograms over individual classes of samples or over groups of classes. In addition, we can compute a significance statistic that gives us the probability that a given histogram is not the result of random band selection outcomes.     

Band Selection for Hyperspectral Image Classification Using Mutual Information

Spectral band selection is a fundamental problem in hyperspectral data processing. In this letter, a new band-selection method based on mutual information (MI) is proposed. MI measures the statistical dependence between two random variables and can therefore be used to evaluate the relative utility of each band to classification. A new strategy is described to estimate the MI using a priori knowledge of the scene, reducing reliance on a "ground truth" reference map, by retaining bands with high associated MI values (subject to the so-called "complementary" conditions). Simulations of classification performance on 16 classes of vegetation from the AVIRIS 92AV3C data set show the effectiveness of the method, which outperforms an MI-based method using the associated reference map, an entropy-based method, and a correlation-based method. It is also competitive with the steepest ascent algorithm at much lower computational cost     

高光谱图像子空间的波段选择

为降低高光谱遥感数据光谱空间的冗余度,提出一种快速的波段选择方法。该方法在波段子空间下进行,依次选择各子空间中方差最大的波段作为初始波段,设定目标函数,然后逐子空间替换波段使得目标性能更加优化,直至没有替换可以使得目标更优为止。在两个公开高光谱影像数据集上对比3种常用波段选择方法(ABC、AP、ABS)来验证提出方法的有效性,实验结果表明:(1)在印第安纳数据上,本文方法与ABC、AP、ABS所选波段子集相比平均相关性分别降低22.04%、52.61%、55.71%,最佳指数分别提高0.58%、51.73%、0.95%,总体分类精度分别提高0.16%、1.39%、23.07%,在搜索效率上与同类型的ABC方法相比提高6.61%—69.02%;(2)在帕维亚大学数据上,本文方法与ABC、AP、ABS所选波段子集相比平均相关性分别降低2.38%、0.51%、32.83%,最佳指数分别提高1.34%、17.97%、12.92%,总体分类精度分别提高0.31%、0.69%、8.53%,在搜索效率上与同类型的ABC方法相比提高19.13%—86.34%。本文提出的波段选择方法能够选择合适的波段子集满足不同的应用需要,是一种有效的波段选择方法。     

Independent Base Vector Representation to Address Endmember Variability in Hyperspectral Unmixing

The normal compositional model (NCM) is a well-known and powerful model in hyperspectral unmixing which represents endmembers as independent Gaussian vectors to capture endmember variability. However, the assumption of independent endmembers diminishes the model accuracy because the high degree of correlation between endmembers of a scene and identical sources of variability demonstrate that the endmembers are dependent. This paper proposes a new hyperspectral unmixing algorithm which represents endmembers using dependent Gaussian vectors to estimate abundance fractions. To overcome the higher complexity caused by dependence assumption, this algorithm introduces new independent Gaussian vectors named Base Vectors to represent different endmembers by a weighted linear combination. Also, the proposed unmixing algorithm uses maximum likelihood method to estimate weight coefficients of Base Vectors which are used to represent mixed pixel. Finally, abundance estimation can be done using the new representation for endmembers and mixed pixel. The proposed algorithm is evaluated and compared with other state-of-the-art unmixing algorithms using simulated and real hyperspectral images. Experimental results demonstrate that the proposed unmixing algorithm can unmix pixels composed of correlated endmembers in hyperspectral images in the presence of spectral variability more accurately than previous methods.     

采用目标端元修正的高光谱混合像元盲分解

针对非负矩阵盲信号分离(NMF)用于混合像元分解易陷入局部极小值的不足,将非监督端元提取与盲分解方法相结合,构建了一种基于目标端元修正的混合像元盲分解模型(ATGP-NMF)。ATGP-NMF模型利用非监督正交子空间投影算法(ATGP)和非负最小二乘法(NNLS)获取NMF盲分离的初始值,然后将获得初始目标端元光谱与丰度输入NMF模型,通过迭代运算不断逼近优化目标而得到最终的端元光谱和端元丰度。为了检验模型对于各类数据的有效性和适用性,将ATGP-NMF与传统NMF分别应用于模拟仿真数据、室内控制数据和真实遥感影像3类实验数据进行分析验证。结果表明,ATGP-NMF模型具有较好的适用性,在没有先验信息、先验信息很少,以及纯像元假设不存在情况下都能较好地分解混合像元,且能够更好克服局部极小问题,提高混合像元分解的精度。     

FAN模型的双种群差分进化光谱解混算法

双线性混合模型是近年来非线性光谱解混的研究重点之一,其克服了线性混合模型无法描述地物多重散射作用的缺陷,能够更精确地还原真实的地物光谱混合过程。然而,限于模型的复杂性,目前在缺乏准确的端元先验知识的条件下进行双线性光谱解混仍是一项具有挑战性的任务。差分进化算法(DE)是一种具有良好全局搜索能力的群智能优化算法,其优化求解过程无需进行复杂的数学推导,为双线性光谱解混问题提供了一种有效的解决途径。为此,本文以FAN双线性混合模型为例,提出了一种双种群机制的差分进化算法(记为DEFAN),实现非监督双线性光谱解混。DE-FAN算法通过建立端元与丰度两个种群的交替进化机制寻找最优解,同时在迭代中引入自适应重构策略增强种群多样性,降低算法陷入局部最优解的风险,最终实现端元与丰度的同时估计。通过模拟图像及真实图像的解混实验进行算法检验,证明DE-FAN算法较之传统非线性解混算法具有更高的解混精度及解混效率。     

高光谱数据截断加权核范数稀疏解混

受仪器和观测条件限制,高光谱数据易受噪声污染,给数据解译带来挑战。针对传统稀疏解混模型抗噪性能差的问题,本文提出一种截断加权核范数稀疏解混方法,利用高光谱图像像元之间的相关性减轻噪声对丰度估计的干扰。该方法借助低秩表示在挖掘数据内在低维结构方面的优势,在稀疏解混中加入基于截断加权核范数的低秩约束,并结合加权稀疏技术,在稀疏正则项中引入空间邻域权重。截断加权核范数对丰度矩阵的奇异值向量分段处理,可以更好地实现丰度矩阵的低秩逼近,使丰度图像保持空间一致性并保留更多细节信息,空间加权策略则增强了丰度图像的空间连续性。模拟高光谱数据、Cuprite矿区真实数据和红树林高光谱数据实验表明,与其他先进的稀疏解混方法相比,所提方法具有更好的抗噪性,能够提高解混精度。     

空间与谱间相关性分析的NMF高光谱解混

非负矩阵分解(NMF)技术是高光谱像元解混领域的研究热点。为了充分利用高光谱图像中丰富的空间与光谱相关性特征,改善基于NMF的高光谱解混算法性能,提出一种结合了空间与谱间相关性分析的NMF解混算法。算法针对NMF的通用性和局部极小问题,引入并结合高光谱图像两种典型的相关性特征,具体包括:基于马尔可夫随机场(MRF)模型,建立描述相邻像元空间相关特征的约束;通过复杂度映射技术,建立描述相邻波段谱间相关(光谱分段平滑)特征的约束;并将上述两种约束同时引入NMF解混目标函数中。实验结果表明,对于一般自然地物场景或人造地物场景,相对于分段平滑和稀疏约束的非负矩阵分解(PSNMFSC)、交互投影子梯度的非负矩阵分解(APSNMF)和最小体积约束的非负矩阵分解(MVCNMF)这3种代表性NMF解混参考算法,该算法可进一步提高高光谱解混精度;对于空间相关或谱间相关特征中某一种不显著的特殊场景,也具有更好的适应能力。通过将空间相关和谱间相关特征相结合,较全面地反映了高光谱数据与解混相关的重要特征,能够对绝大多数真实高光谱数据进行高精度解混,对高光谱解混及后续应用领域相关研究均具有参考价值。     

A method to compute the n-dimensional solid spectral angle between vectors and its use for band selection in hyperspectral data

This study presents the calculation of spectral angle beyond two endmember vectors to the n-dimensional solid spectral angle (NSSA). The calculation of the NSSA is used to characterize the local spectral shape difference among a set of endmembers, leading to a methodology for band selection based on spectral shape variations of more than two spectra. Equidistributed sequences used in the quasi-Monte Carlo method (ESMC) for numerical simulations are shown to expedite the calculation of the NSSA. We develop a band selection method using the computation of NSSA(ϑ_n) in the context of a sliding window. By sliding the window over all bands available for varying band intervals, the calculated solid spectral angle values can capture the similarity of the endmembers over all spectral regions available and for spectral features of varying widths. By selecting a subset of spectral bands with largest solid spectral angles, a methodology can be developed to capture the most important spectral information for the separation or mapping of endmembers. We provide an example of the merits of the NSSA-ESMC method for band selection as applied to linear spectral unmixing. Specifically, we examine the endmember abundance errors resulting from the NSSA band selection method as opposed to using the full spectral dimensionality available.     

Wheat crop inventory using high spectral resolution irs-p3 mos-b spectrometer data

Modular Optoelectronic Scanner (MOS-B) spectrometer data over parts of Northern India was evaluated for wheat crop monitoring involving (a) sub pixel wheat fractional area estimation using spectral unmixing approach and (b) growth assessment by red edge shift at different phenological stages. Red shift of 10 nm was observed between crown root initiation stage to flowering stage. Wheat fraction estimates using linear spectral unmixing on Feb. 13, 1999 acquisition of MOS-B data had high correlation (0.82) with estimates from Wide Field Sensor (WiFS) data acquired on same date by IRS-P3 platform. It was observed that five bands (4,5,8,12,13 MOS-B bands) are sufficient for signature separability of major land cover classes viz. wheat, urban, wasteland, and water based on purely spectral separability criterion using Transformed Divergence (T.D.) approach. Higher number of bands saturated the T.D. values. In contrast, performance of sub pixel fractional area estimation using unmixing decreased drastically for eight bands (4,5,6,7,8,9,12,13 MOS-B bands) chosen from optimal band selection criteria in comparison to full set of 13 bands. The relative deviation between area estimated from Wifs and MOS-B increased from 1.72 percent when all thirteen bands were used in unmixing to 26.10 percent for the above eight bands.     

Using FCMC, FVS, and PCA techniques for feature extraction of multispectral images

In this letter, a new nonlinear approach based on a combination of the fuzzy c-means clustering (FCMC), feature vector selection and principal component analysis (PCA) is proposed to extract features of multispectral images when a very large number of samples need to be processed. The main contribution of this letter is to provide a preprocessing method for classifying these images with higher accuracy compared to the single PCA and kernel PCA. Finally, some experimental results demonstrate that our proposed approach is effective and efficient in analyzing multispectral images.     

Improved IHS Algorithm for Fusing High Resolution Satellite Images of Urban Areas

The intensity-hue-saturation method is used frequently in image fusion due to its efficiency and high spatial quality. The main shortage is its spectral distortion stemmed from replacement of intensity band with higher resolution image. In this study, a new method is introduced to improve the spectral quality of the Intensity-Hue-Saturation (IHS) algorithm. The goal of this study is to produce the fused image that has a better spectral and spatial quality with respect to the original images in term of visual comparison and the classification result. In this regard, an improved statistical approach is developed to combine an intensity band from IHS algorithm and an input high resolution image such as SAR or Panchromatic image. Then the intensity image is replaced by the combined image band. Final fused images are attained using the inverse IHS algorithm. The proposed fusion algorithm is tested on two data sets of: a) panchromatic and multi spectral bands of IKONOS image with the same acquisition date, and b) multi spectral and HH bands of IKONOS and TerraSAR-X images respectively with different acquisition dates. Moreover, the obtained results are compared with other fusion methods like IHS, Gungor, Brovey and synthetic variable ratio. The results show less spectral discrepancy of the proposed method comparing to other methods. Finally, the outcome of proposed method is classified and classification overall accuracy is improved by 5.6 and 2 percentage for data set ‘a’ and ‘b’ respectively.     

Spectral conversion of surface albedo derived from meteosat first generation observations

Comparison of surface albedos derived from spaceborne radiometers with different spectral bands requires, first of all, the conversion of these quantities into common spectral intervals. This letter proposes a spectral conversion method specifically dedicated to surface albedo derived in a large-band instrument such as the solar channel onboard the Meteosat first-generation radiometer. This new method accounts for the retrieval algorithm assumptions and radiometer spectral limitations that might have an impact on the retrieved surface albedo in such a large band. It is also shown that the proposed approach has no impact when surface albedo is derived in narrow bands and confirms the results of previously published spectral conversion methods.     

稀疏差异先验信息支持的高光谱图像稀疏解混算法

基于光谱库的高光谱稀疏解混技术近年来得到了人们的关注,该技术利用光谱库中光谱样本作为端元,将解混问题转化为稀疏表示问题。然而,由于测量环境的差异,待解混图像的实际端元往往与光谱库中相应光谱信号存在差异。本文提出了一种光谱差异稀疏约束的联合稀疏回归解混算法。首先,假设光谱差异具有稀疏特性,建立了光谱库校正模型,使得在解混过程中可对光谱库进行自适应地调整;然后,将光谱库校正模型与联合稀疏回归解混模型结合,建立了考虑光谱差异的稀疏解混模型;最后,基于交替方向乘子法得到了迭代优化解决方案。分别利用仿真和真实高光谱数据进行了试验验证,结果表明,在光谱库不匹配的情形下,本文方法能够有效提高稀疏解混算法的解混性能。     

高分五号高光谱图像自编码网络非线性解混

针对高光谱非线性混合模型中的共线性问题,提出了一种非监督的增强型非线性自编码网络方法 ENAE(Enhanced Nonlinear Autoencoder)。通过结合自编码网络在挖掘数据内在结构、提取特征方面的优势,引入端元正则项减弱端元间的共线性效应,从而提高高光谱混合像元分解精度。ENAE方法的实现步骤主要包括两部分:一是网络结构初始化,二是非线性分解。网络结构初始化是确定编码器的节点数以及端元和丰度的初值;非线性分解则主要是实现损失函数的最小化。通过模拟数据、城市区域真实数据和高分五号卫星高光谱数据的实验,得到了相较于传统非线性分解方法更高的精度,证明了ENAE方法的鲁棒性。     

A Multiple SVM System for Classification of Hyperspectral Remote Sensing Data

With recent technological advances in remote sensing sensors and systems, very high-dimensional hyperspectral data are available for a better discrimination among different complex land-cover classes. However, the large number of spectral bands, but limited availability of training samples creates the problem of Hughes phenomenon or ‘curse of dimensionality’ in hyperspectral data sets. Moreover, these high numbers of bands are usually highly correlated. Because of these complexities of hyperspectral data, traditional classification strategies have often limited performance in classification of hyperspectral imagery. Referring to the limitation of single classifier in these situations, Multiple Classifier Systems (MCS) may have better performance than single classifier. This paper presents a new method for classification of hyperspectral data based on a band clustering strategy through a multiple Support Vector Machine system. The proposed method uses the band grouping process based on a modified mutual information strategy to split data into few band groups. After the band grouping step, the proposed algorithm aims at benefiting from the capabilities of SVM as classification method. So, the proposed approach applies SVM on each band group that is produced in a previous step. Finally, Naive Bayes (NB) as a classifier fusion method combines decisions of SVM classifiers. Experimental results on two common hyperspectral data sets show that the proposed method improves the classification accuracy in comparison with the standard SVM on entire bands of data and feature selection methods.     

利用可分离非负矩阵分解实现高光谱波段选择

高光谱影像波段众多且相关性强,导致分类存在信息冗余且计算量较大。提出了可分离非负矩阵分解方法来选取高光谱影像的代表性波段子集,在保证分类精度的同时降低计算量。该方法假设高光谱影像的波段集合具有可分离特性,改进传统非负矩阵分解模型,将波段选择转换为可分离非负矩阵分解问题,采用迭代投影方法来依次选取能够非负线性表达其他波段的代表性波段。在此基础上,利用两个公开高光谱数据集对比几种主流方法,采用定量评价和分类精度指标来综合评价所提的波段选择方法的效果。实验结果表明,可分离非负矩阵分解方法的分类精度高于其他几种方法,而且计算效率排名第2,能够选取合适的波段子集以满足高光谱遥感的应用需求。     

Application of an Inexact-Newton Method Within the Second-Order Born Approximation to Buried Objects

The imaging of buried objects is a challenging topic in electromagnetic research. In this letter, a method previously proposed for imaging of scatterers in free-space is extended to the case of objects buried in a lossy half-space. The proposed approach is based on the solution of the underlying inverse-scattering problem within the second-order Born approximation by means of a two-step inexact-Newton algorithm. The capabilities and limitations of the method are evaluated by means of some numerical simulations. Furthermore, comparisons with the free-space approach are also reported     

A new method for retrieving band 6 of aqua MODIS

The Moderate Resolution Imaging Spectroradiometer (MODIS) is a key research instrument for the NASA Earth Observing System (EOS) mission. It was successfully launched onboard the Terra satellite in December 1999 and Aqua satellite in May 2002. Both MODIS instruments have been working well except that 15 of the 20 detectors in Aqua MODIS band 6 (1.628-1.652 /spl mu/m) are either nonfunctional or noisy. The striping in Aqua MODIS band 6 caused by its nonfunctional or noisy detectors has been a serious problem for MODIS snow products, which use band 6 primarily for snow detection. MODIS scientists have been using Aqua MODIS band 7 (2.105-2.155 /spl mu/m) instead of band 6 for computing the normalized difference snow index to continue detecting global snow coverage. The main objective of this letter is to retrieve Aqua MODIS band 6 using other bands based on their relationships in Terra MODIS. The band retrieval approach proposed in this letter is also very useful for the next generation of MODIS sensor-the Visible/Infrared Imager/Radiometer Suite (VIIRS) band M10 proxy data generation. Such proxy data can support the VIIRS prelaunch end-to-end testing, postlaunch calibration/validation, and data quality checking.