高光谱遥感数据最佳波段选择方法试验研究

分析了多光谱遥感数据最佳波段选择的信息量诸方法的内在联系,说明了信息量方法用于高光谱遥感数据最佳波段选择的局限性,提出了基于类间可分性的最佳波段选择原则和方法。通过试验,说明了各种处理方法的有效性、局限性和计算复杂度。     

高光谱遥感数据最佳波段选择方法试验研究

刘建平,赵英时,孙淑玲(中国科技大学研究生院　北京　100039)摘要:分析了多光谱遥感数据最佳波段选择的联合熵、行列式值及最佳指数等信息量计算方法的内在联系,说明了信息量方法用于高光谱遥感数据最佳波段选择的局限性,提出了基于类间可分性的最佳波段选择原则和方法。通过试验,说明了各种处理方法的有效性、局限性和计算复杂度。关　键　词:高光谱遥感数据;最佳波段选择;信息量;可分性中图分类号:TP 751.1/TP 79     

基于多目标遗传算法的超光谱波段选择

超光谱波段图像的信息冗余量很大,可以通过波段选择的方式简化分析.以白血病血液样本的光谱数据为例,以最佳指数和协方差为选择依据,通过多目标遗传算法实现了超光谱的波段提取.为了验证该法的有效性,对多目标算法与单目标遗传算法进行了比较实验,结果表明NSGA的选择结果总体更优于单目标算法,从一定程度上说明了该算法在光谱提取上的...     

高光谱影像波段选择算法研究*

基于高光谱影像数据的特点,分析了高光谱数据的降维方法。着重探讨了波段选择的若干算法:熵及联合熵、最佳指数因子、自动子空间划分、自适应波段选择、波段指数和最优波段指数等算法。分析了各种算法的有效性、局限性和计算复杂度,并针对波段指数的不足,设计了最优波段指数(OBI)波段选择新算法。最后通过具体的试验,验证了各种算法的性能。     

高光谱数据的光谱信息特点及面向对象的特征参数选择——以北京顺义区为例

波段宽度为纳米级的高光谱数据,具有几十乃至几百个光谱通道,它们各有不同的特点。如何根据具体的应用目的,在这众多的波段中选择出最佳波段和特征参数,对于有效地进行高光谱数据的处理、分析及信息提取至关重要。以北京顺义区高光谱数据为例,首先分析了通道间的相关性,根据通道的相关性大小和分组块状结构特点,将其分为若干组;然后全面分析了高光谱数据的光谱信息特征,在综合考虑各波段的信息含量、波段间的相关性以及地物光谱的吸收特性和可分性等因素
的基础上,提出了面向对象的分层多次选择高光谱数据最佳波段和提取特征参数的基本思路和方法;最后用其它地区的成像光谱数据对此方法进行了验证。     

高光谱遥感图像波段选择研究进展综述

高光谱成像遥感技术可获取地物的光谱、辐射和空间信息,在国民经济的各个领域得到广泛的应用.但其狭窄的波段间距带来丰富光谱信息的同时,也带来了信息冗余,增加了数据处理的难度.因此,高光谱遥感数据在进行实际应用前,需要进行波段选择并提取光谱特征,降低数据维数.对高光谱遥感图像的波段选择研究进展进行了综述,在分析、归纳波段选择...     

土壤重金属Cu含量遥感反演的波段选择与最佳光谱分辨率研究

高光谱数据以其高光谱分辨率和多而连续的光谱波段为预测土壤重金属污染提供了有力工具,但波段选择方法与光谱分辨率的影响不容忽视。利用实验室测定的181个土壤光谱样本数据,利用逐步回归法进行土壤Cu含量反演的波段选择,进而利用偏最小二乘方回归PLSR方法建模,分析了波段数对Cu含量反演的影响;此外,采用高斯响应函数重采样方法,探讨了光谱分辨率降低对反演精度的影响。实验表明,预测重金属元素Cu含量的最佳波段数为10个,模型可决系数R2=0.7523,拟合均方根误差RMSE=0.4699;预测Cu含量的最佳光谱采样间隔为32 nm,R2=0.7028,RMSE=0.5147。该结果可能为将来设计低廉实用的高光谱卫星传感器提供指标论证,为模拟卫星传感器波段预测土壤重金属含量提供理论依据。     

分段2维主成分分析的超光谱图像波段选择

目的 超光谱图像具有极高的谱间分辨率,巨大的数据量给分类识别等后续处理带来很大压力。为了有效降低图像数据维数,提出基于分段2DPCA的超光谱图像波段选择算法。方法 首先根据谱间相关性对原始图像进行波段分组,然后根据主成分反映每个光谱波段的信息比重分别对每组图像进行波段选择,从而实现超光谱图像的谱间降维。结果 该算法有效降低了超光谱图像的光谱维数,选择的波段明显反映出不同地物像元矢量的区别。结论 实验结果表明,该波段选择算法相对传统算法速度更快,并且较好地保留了原始图像的局部重要信息,对后续处理有积极意义。     

基于萤火虫算法的高光谱遥感波段选择方法

高光谱图像在遥感领域中的应用越来越广泛,但由于自身的高数据维、波段间的高冗余度等特性给图像处理带来了一定困难,针对这个问题,提出一种基于类间可分性准则的改进萤火虫仿生算法,进行高光谱遥感波段选择。在分析萤火虫算法机理的基础上,阐述了利用该算法进行高光谱波段选择的思路,并构造波段相似性矩阵,选择欧氏距离、JM距离、光谱信息散度和离散度作为可分性准则来设置目标函数,根据目标函数值的优劣选择优势波段。最后,使用HYDICE Washington DC Mall和 HyMap Purdue Campus两个高光谱遥感影像数据进行实验验证,并利用支持向量机分类器对最佳波段组合进行精度评价,证明该算法的可行性和有效性。
     

一种改进的高光谱图像波段选择方法

根据高光谱波段选择的基本准则,将子空间划分、基于矩阵模式的高光谱波段选择方法(BSMM)、波段指数(OIF)三者相结合,提出了一种新的波段选择方法ABO。该方法首先根据各波段之间的相关性进行子空间划分;然后,在全波段范围内利用基于矩阵模式的高光谱波段选择方法得到单一量化指标W,选出各子空间中量化指标W取最大值所对应的波段;其次,针对已选波段计算任意3个波段的波段指数(OIF),波段指数最大值所对应的3个波段即为所选波段;最后,利用AVIRIS真实高光谱数据进行仿真实验,对所选3个波段进行RGB合成与HSV变换以及RX异常检测,通过与以往波段选择方法进行对比验证了所提方法的有效性。     

月面地形重构系统中的并行Delaunay算法设计

三角剖分过程是影响三维重建系统实时性的瓶颈之一,为提高三角剖分速度,基于共享内存多核计算机设计并实现了并行Delaunay算法。该算法在分治三角剖分算法的基础上,通过改进子三角网归并过程及Delaunay三角网优化过程避免了并行计算中的数据竞争问题。利用月面仿真实验场真实地形数据在50万到500万不同规模的点云数据集上进行了实验,加速比最高可达6.44。除此之外,对算法复杂度、加速比以及并行效率进行了全面分析,并将算法实际应用于月面地形重构系统,实现了虚拟地形的快速构建。     

New framework for hyperspectral band selection using modified wind-driven optimization algorithm

The presence of irrelevant and highly correlated spectral bands significantly reduces the classification accuracy of the hyperspectral images. Therefore, the selection of suitable bands from the set of available spectral bands plays a crucial role in improving the classification accuracy. In this paper, a novel band selection approach is proposed based on nature inspired meta-heuristic algorithm to mitigate the effect of curse of dimensionality. Wind-driven optimization (WDO), among other meta-heuristic algorithms, has proven to be more efficient in solving global optimization problems. However, WDO is prone to premature convergence when solving the global optimization problem due to loss of diversity of air particles. Therefore, a modified WDO (MWDO) is proposed for band selection, which is able to avoid the premature convergence and control the exploration–exploitation search trade-off. Finally, in order to further improve the performance of the classification, the selected bands are fed into the deep learning architecture to extract the high-level useful features. The experiments are carried on three widely used standard datasets such as Indian Pines, Pavia University, and Salinas. The experimental results show that the proposed approach selects an optimal subset of bands with good convergence characteristics and provide high classification accuracy with fewer bands in comparison with other approaches. The proposed method achieves an overall accuracy of 93.26%, 94.76%, and 95.96% for Indian Pines, Pavia University, and Salinas datasets, respectively.     

基于时差特征与随机森林的水稻种植面积提取

准确提取水稻种植面积是探讨气候变化背景下水稻生产与粮食安全的重要前提。我国南方的水稻种植区域,地块破碎且受云雨天气影响严重,如何充分利用有限时相的数据获得较高精度的水稻面积提取是亟需解决的关键问题。提出了一种利用两个时相的数据,通过构建差值特征突出水稻物候变化的特点,并与随机森林算法结合高精度提取水稻种植面积的方法。将之应用于湖南省常德市鼎城区的水稻种植面积提取,结果表明:采用本方法进行水稻提取的最终总体精度达到93.01%,Kappa系数0.91,与单时相提取结果相比,总体精度提高了近3%。为了进一步分析差值特征对其他分类器的改进效果,分别将差值特征与决策树和随机森林组合,并分析了两种组合提取水稻的精度。研究发现构建的差值特征能够有效反映植物的生长状况,增加地物的可区分性,可为对象的分割及分类提供更多有用的信息,能够有效改善水稻种植面积的提取精度。
     

Supervised band selection in hyperspectral images using single-layer neural networks

Hyperspectral images provide fine details of the scene under analysis in terms of spectral information. This is due to the presence of contiguous bands that make possible to distinguish different objects even when they have similar colour and shape. However, neighbouring bands are highly correlated, and, besides, the high dimensionality of hyperspectral images brings a heavy burden on processing and also may cause the Hughes phenomenon. It is therefore advisable to make a band selection pre-processing prior to the classification task. Thus, this paper proposes a new supervised filter-based approach for band selection based on neural networks. For each class of the data set, a binary single-layer neural network classifier performs a classification between that class and the remainder of the data. After that, the bands related to the biggest and smallest weights are selected, so, the band selection process is class-oriented. This process iterates until the previously defined number of bands is achieved. A comparison with three state-of-the-art band selection approaches shows that the proposed method yields the best results in 43.33% of the cases even with greatly reduced training data size, whereas the competitors have achieved between 13.33% and 23.33% on the Botswana, KSC and Indian Pines datasets.     

Electronic properties of rhombohedral graphite

Electronic properties of ABC-stacked graphite are studied by the first-principles method. There are linear and parabolic bands with strong anisotropic dispersions; both non-degenerate and degenerate bands are observed. The bandwidths of occupied π and σ bands are 8.41 eV and 16.65 eV, respectively. The low valence and conduction bands only have slight overlapping near the Fermi level, mainly owing to the interlayer atomic interactions. State degeneracy and energy dispersions are strongly affected by such interactions. The band-edge states, with the high density of states, are located near or at the high symmetry points. Some significant differences exist among ABC-, AB-, and AA-stacked graphites in electronic properties.     

Unsupervised hyperspectral band selection using ranking based on a denoising error matching approach

In this paper, we focus on utilizing the image denoising method for ranking of significant bands in hyperspectral imagery. We make use of the fact that the denoising error of bands varies with the significant information content of the bands in hyperspectral imagery. The denoising error is computed for each band individually and compared using a matching parameter with the denoising error of a reference image. The reference image is selected to be the first principal component corresponding to the maximum information. Three matching parameters including mutual information (MI), correlation coefficient (r) and the structural similarity index (SSIM) were used for ranking the bands based on the match with the denoising error of the reference image. The proposed algorithm is tested using three datasets, namely, Indian Pines, Salinas and Dhundi. The Indian Pines and Salinas datasets were acquired from the Airborne Visible Infrared Imaging Spectrometer (AVIRIS) sensor and comprised rural and agricultural area. The Dhundi dataset of Hyperion comprises mostly of features corresponding to snow-covered mountainous regions. To assess the accuracy of the proposed method, a supervised classification was carried out using a random forest classifier with 20% training pixels selected randomly from the ground reference. The proposed method yielded significantly better results determined by the kappa coefficient (κ) of 0.756, 0.910 and 0.996 for the Indian Pines, Salinas and Dhundi datasets, respectively, over several other state of the art methods. The classification results of the proposed method also yielded better results than those obtained by the state-of-the-art methods for hyperspectral band selection.     

SAR imaging simulation for an inhomogeneous undulated lunar surface based on triangulated irregular network

Based on the statistics of the lunar cratered terrain, e.g., population, dimension and shape of craters, the terrain feature of cratered lunar surface is numerically generated. According to the inhomogeneous distribution of the lunar surface slope, the triangulated irregular network (TIN) is employed to make the digital elevation of lunar surface model. The Kirchhoff approximation of surface scattering is then applied to simulation of lunar surface scattering. The synthetic aperture radar (SAR) image for comprehensive cratered lunar surface is numerically generated using back projection (BP) algorithm of SAR imaging. Making use of the digital elevation and Clementine UVVIS data at Apollo 15 landing site as the ground truth, an SAR image at Apollo 15 landing site is simulated. The image simulation is verified using real SAR image and echoes statistics. Supported by the National Matural Science Foundation of China (Grant No. 40637033)     

Gray Wolf Optimizer for hyperspectral band selection

In this paper, we propose a new optimization-based framework to reduce the dimensionality of hyperspectral images. One of the most problems in hyperspectral image classification is the Hughes phenomenon caused by the irrelevant spectral bands and the high correlation between the adjacent bands. The problematic is how to find the relevant bands to classify the pixels of hyperspectral image without reducing the classification accuracy rate. We propose to reformulate the problem of band selection as a combinatorial problem by modeling an objective function based on class separability measures and the accuracy rate. We use the Gray Wolf Optimizer, which is a new meta-heuristic algorithm more efficient than Practical Swarm Optimization, Gravitational Search Algorithm, Differential Evolution, Evolutionary Programming and Evolution Strategy. The experimentations are performed on three widely used benchmark hyperspectral datasets. Comparisons with the state-of-the-art approaches are also conducted. The analysis of the results proves that the proposed approach can effectively investigate the spectral band selection problem and provides a high classification accuracy rate by using a few samples for training.     

使用深度对抗子空间聚类实现高光谱波段选择

高光谱图像（HSI）由数百个波段组成,波段之间的相关性强且具有较高的冗余度,导致出现维度灾难并且分类的复杂性很高。为此,使用深度对抗子空间聚类（DASC）网络进行高光谱的波段选择,并引入拉普拉斯正则化使网络更优,在保证分类精度的前提下降低分类的复杂度。该网络通过在编码器和解码器中引入自表达层来模仿传统子空间聚类的“自表达”属性,充分运用光谱信息和非线性特征转换得到波段之间的相互关系,解决传统波段选择方法无法同时考虑光谱和空间信息的问题。同时,引入对抗学习来监督自编码器的样本表示和子空间聚类,使得子空间聚类具有更好的自表达性能。为了使网络性能更优,加入拉普拉斯正则化来考虑反映图像几何信息的局部流形结构。实验在两个公开的高光谱数据集上进行,所提出的方法和几种主流的波段选择方法进行对比的结果表明,DASC方法在分类精度上优于对比方法,其选出的波段子集可以满足应用需求。