单形体体积最小化的差分进化光谱解混算法

目的 光谱解混是高光谱遥感图像处理的核心技术。当图像不满足纯像元假设条件时,传统算法难以适用,基于(单形体)体积最小化方法提供了一种有效的解决途径。然而这是一个复杂的约束最优化问题,更由于图像噪声等不确定性因素的存在,导致算法容易陷入局部解。方法 引入一种群智能优化技术-差分进化算法(DE),借助其较强的全局搜索能力以及优越的处理高维度问题的能力,并通过对问题编码,提出了一种体积最小化的差分进化(VolMin-DE)光谱解混算法。结果 模拟数据和真实数据实验的结果表明,与现有算法相比,该算法在15端元时精度(光谱角距离)可提高7.8%,当端元数目少于15个时,其精度普遍可以提高15%以上,特别是10端元时精度可以提高41.3%;在20~50 dB的噪声范围内,精度变化在1.9~3.2(单位:角度)之间,传统算法在2.2~3.5之间,表明该算法具有相对较好的噪声鲁棒性。结论 本文算法适用于具有纯像元以及不存在纯像元(建议最大纯度不低于0.8)这两种情况的高光谱遥感图像,并可在原始光谱维度进行光谱解混,从而避免降维所带来的累计误差,因此具有更好的适应范围和应用前景。     

Unmixing of soil types and estimation of soil exposure with simulated hyperspectral imagery

Hyperspectral determination of soil types has the potential to become an important addition to the methods used for classification and mapping of soils. In this study laboratory measured spectra of different soils, vegetation and crop residue were combined to simulate hyperspectral remote sensing imagery. The overall aim was to examine the spectral unmixing of these materials under laboratory conditions to better understand the limits to prediction of soil types and determination of cover fractions. Two different methods were utilized to mix spectra of the soil and vegetation and substantial differences were observed in the unmixing results from the different image types, particularly in mixed pixels. Results found pure soils were easily distinguished from each other when not mixed with vegetation, while some mixes of soil and vegetation were confused as pure soil spectra. The accuracy of abundance fractions retrieved in the unmixing process also varied substantially with soil type and vegetation cover.     

Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated‐forest hyperspectral data

The spectral unmixing of mixed pixels is a key factor in remote sensing images, especially for hyperspectral imagery. A commonly used approach to spectral unmixing has been linear unmixing. However, the question of whether linear or nonlinear processes dominate spectral signatures of mixed pixels is still an unresolved matter. In this study, we put forward a new nonlinear model for inferring end‐member fractions within hyperspectral scenes. This study focuses on comparing the nonlinear model with a linear model. A detail comparative analysis of the fractions ‘sunlit crown’, ‘sunlit background’ and ‘shadow’ between the two methods was carried out through visualization, and comparing with supervised classification using a database of laboratory simulated‐forest scenes. Our results show that the nonlinear model of spectral unmixing outperforms the linear model, especially in the scenes with translucent crown on a white background. A nonlinear mixture model is needed to account for the multiple scattering between tree crowns and background.     

结合空间光谱预处理和约束非负矩阵分解的高光谱图像混合像元分解

目的 混合像元问题在高光谱遥感图像处理分析中普遍存在,非负矩阵分解的方法被引入到高光谱图像解混中。本文提出结合空间光谱预处理和约束非负矩阵分解的混合像元分解流程。方法 结合空间光谱预处理的约束非负矩阵分解,如最小体积约束、流行约束等,通过加入邻域的空间和光谱信息进行预处理获得更优的预选端元,从而对非负矩阵分解的解混结果进行优化。结果 在5组不同信噪比的模拟数据实验中,空间预处理（SPP）和空间光谱预处理（SSPP）均能够有效提高约束非负矩阵分解（最小体积约束的非负矩阵分解和图正则非负矩阵分解）的解混结果,其中SPP在不同信噪比的情况下都能优化约束非负矩阵分解的结果,而SSPP在低信噪比的情况下,预处理效果更佳。利用美国内华达州Cuprite矿区数据进行真实数据实验,SPP提高了约束非负矩阵分解的解混精度,而SSPP在复杂场景下,解混精度更佳。模拟数据和真实数据的实验均表明,空间光谱预处理能够有效地提高约束非负矩阵分解的解混精度,特别是对于信噪比较低的情况下,融合空间和光谱信息对噪声有很好的鲁棒性。结论 本文对约束非负矩阵分解的解混算法添加空间光谱预处理,利用高光谱遥感数据的空间和光谱信息,优化预选端元,加入空间光谱预处理的非负矩阵解混实验流程,在复杂场景情况下,对噪声具有较好的鲁棒性。     

基于Hessian图正则稀疏NMF的高光谱解混

基于非负矩阵分解(Nonnegative Matrix Factorization, NMF)的高光谱解混(Hyperspectral Unmixing,HU)方法引起了大家的关注,因为可以将一个非负高光谱图像(Hyperspectral Imagery, HSI)数据矩阵分解为两个非负矩阵的乘积,分别对应于端元矩阵和丰度系数矩阵。目前,图约束的NMF算法已经被证明对高光谱解混是有效的,因为它们可以捕获HSI的几何特性。为了挖掘数据在混合过程中的几何结构和稀疏性,提出了一种稀疏的Hessian图正则化NMF(SHGNMF)算法。SHGNMF算法是将丰度矩阵的L1/2正则化器和Hessian图正则化项都添加到每个NMF模型中,同时采用乘法更新规则。最后用模拟数据和真实数据进行实验,验证了所提出的SHGNMF算法相对于其他NMF算法的优越性。     

可变神经网络结构下的遥感影像光谱分解方法

多层感知神经网络(MLP)是主流的非线性分解方法,但是目前缺乏有效方法处理MLP分解结果中的丰度负值问题。为此,提出一种可变神经网络结构的方法,逐步去除负值丰度对应的端元,并调整相应的网络结构使之针对剩余的端元进行分解。通过武汉地区模拟TM遥感影像实验可以发现,该方法与传统MLP方法以及线性光谱分解方法的平均误差分别为0.077 7、0.081 9、0.094 3,说明该方法的分解精度高于其他2种分解方法,能克服丰度负值问题。     

基于混合像元分解的喀斯特石漠化地物丰度估测

我国西南喀斯特地区长期存在以石漠化为特征的土地退化问题,是我国三大生态问题之一。喀斯特地区地表复杂度高,具有高度时空异质性,像元混合现象严重,植被、裸岩和裸土为喀斯特地区典型地物,使得评价喀斯特石漠化的关键指标(如裸岩率、植被覆盖度)获取比较困难,高光谱遥感在混合像元分解方面有独特优势,可以获取地物端元的丰度。通过地面试验表明光谱指数能够表征地物覆盖度,进而以Hyperion高光谱影像为数据源,利用连续最大角凸锥方法从影像中提取这3类地物的端元,运用半约束和全约束线性光谱分解方法估算其丰度。研究表明:半约束线性分解得到的丰度优于全约束分解结果,其反演的植被、裸土和裸岩的丰度与相应的光谱指数间具有显著线性相关性,确定系数R²分别为0.92、0.66与0.84,表明地物丰度能够表征其覆盖度。因此,通过混合像元分解算法反演地物丰度来提取喀斯特石漠化因子具有一定的可行性,这为高光谱遥感在喀斯特石漠化中的评价和监测奠定了理论和算法基础。     

Remote sensing image subpixel mapping based on adaptive differential evolution

In this paper, a novel subpixel mapping algorithm based on an adaptive differential evolution (DE) algorithm, namely, adaptive-DE subpixel mapping (ADESM), is developed to perform the subpixel mapping task for remote sensing images. Subpixel mapping may provide a fine-resolution map of class labels from coarser spectral unmixing fraction images, with the assumption of spatial dependence. In ADESM, to utilize DE, the subpixel mapping problem is transformed into an optimization problem by maximizing the spatial dependence index. The traditional DE algorithm is an efficient and powerful population-based stochastic global optimizer in continuous optimization problems, but it cannot be applied to the subpixel mapping problem in a discrete search space. In addition, it is not an easy task to properly set control parameters in DE. To avoid these problems, this paper utilizes an adaptive strategy without user-defined parameters, and a reversible-conversion strategy between continuous space and discrete space, to improve the classical DE algorithm. During the process of evolution, they are further improved by enhanced evolution operators, e.g., mutation, crossover, repair, exchange, insertion, and an effective local search to generate new candidate solutions. Experimental results using different types of remote images show that the ADESM algorithm consistently outperforms the previous subpixel mapping algorithms in all the experiments. Based on sensitivity analysis, ADESM, with its self-adaptive control parameter setting, is better than, or at least comparable to, the standard DE algorithm, when considering the accuracy of subpixel mapping, and hence provides an effective new approach to subpixel mapping for remote sensing imagery.     

利用稳健非负矩阵分解实现无监督高光谱解混

目的 基于非负矩阵分解的高光谱图像无监督解混算法普遍存在着目标函数对噪声敏感、在低信噪比条件下端元提取和丰度估计性能不佳的缺点。因此,提出一种基于稳健非负矩阵分解的高光谱图像混合像元分解算法。方法 首先在传统基于非负矩阵分解的解混算法基础上,对目标函数加以改进,用更加稳健的L₁范数作为重建误差项,提高算法对噪声的适应能力,得到新的无监督解混目标函数。针对新目标函数的非凸特性,利用梯度下降法对端元矩阵和丰度矩阵交替迭代求解,进而完成优化求解,得到端元和丰度估计值。结果 分别利用模拟和真实高光谱数据,对算法性能进行定性和定量分析。在模拟数据集中,将本文算法与具有代表性的5种无监督解混算法进行比较,相比于对比算法中最优者,本文算法在典型信噪比20 dB下,光谱角距离（spectral angle distance,SAD）增大了10.5%,信号重构误差（signal to reconstruction error,SRE）减小了9.3%;在真实数据集中,利用光谱库中的地物光谱特征验证本文算法端元提取质量,并利用真实地物分布定性分析丰度估计结果。结论 提出的基于稳健非负矩阵分解的高光谱无监督解混算法,在低信噪比条件下,能够获得较好的端元提取和丰度估计精度,解混效果更好。     

基于ANN端元估计的高光谱图像解混算法

针对高光谱图像解混问题进行研究,发现传统解混算法在保持端元数目不变的情况下,得到的解混精度不高。为此,基于人工神经网络（ANN）提出一种估计单像素点中端元数目和类别的解混算法。首先利用人工神经网络对遥感图像中各个像素的端元数目和类别进行估计;之后依据估计结果确定解混算法的目标函数,并引入改进的差分搜索算法对目标函数进行优化求解;最终获取地物丰度和待求参数,实现高光谱图像的解混。仿真数据和真实遥感数据实验表明,与现有的解混算法相比,所提解混算法具有更高的解混性能,更加符合实际场景的情况。     

基于目标优化的高光谱图像亚像元定位

目的 高光谱图像混合像元的普遍存在使得传统的分类技术难以准确确定地物空间分布,亚像元定位技术是解决该问题的有效手段。针对连通区域存在孤立点或孤立两点等特例时,通过链码长度求周长最小无法保证最优结果及优化过程计算量大的问题,提出了一种改进的高光谱图像亚像元定位方法。方法 以光谱解混结合二进制粒子群优化构建算法框架,根据光谱解混结果近似估计每个像元对应的亚像元组成,通过分析连通区域存在特例时基于链码长度求周长最小无法保证结果最优的原因,提出修改孤立区域的周长并考虑连通区域个数构造代价函数,最后利用二进制粒子群优化实现亚像元定位。为了减少算法的时间复杂度,根据地物空间分布特点,采用局部分析代替全局分析,提出了新的迭代优化策略。结果 相比直接基于链码长度求周长最小的优化结果,基于改进的目标函数优化后,大部分区域边界更明显,并且没有孤立1点和孤立两点的区域,识别率可以提高2%以上,Kappa系数增加0.05以上,新的优化策略可以使算法运算时间减少近一半。结论 实验结果表明,本文方法能有效提高亚像元定位精度,同时降低时间复杂度。因为高光谱图像中均匀混合区域不同地物的分布空间相关性不强,因此本文方法适用于非均匀混合的高光谱图像的亚像元定位。     

Citrus Recognition Methods of Hyperspectral Remote Sensing Image based on Spectral Unmixing Model

In order to monitor the citrus planting information timely and accurately,We take Huichang County of Jiangxi Province as the research area,using EO\|1 Hypersion hyperspectral remote sensing (HRS)image as a datasource to build a citrus recognition methods of hyperspectral remote sensing image based on spectral unmixing.First of all,the EO\|1 Hyperion hyperspectral remote sensing image has 242 bands,and it has a wide spectrum rang.It can extract the spectral curve of typical objects in the study area,which is based on the image pre\|processing including the band selection,the atmospheric correction and so on.Then,we use the fully constrained linear spectral mixture model of spectral unmixing to decompose the mixed pixels of the image,and then extract the abundance value of citrus.Finally,we construct the relationship between citrus abundance and the actual cultivation of citrus based on the high resolution remote sensing image.The results indicated that the unavoidable error in the extraction of the typical objects and the differences of the citrus canopy coverage can lead to the corresponding relationship between the citrus plant accurate identification and the citrus abundance threshold value.Under the condition of repeated experiments,the study area of citrus abundance thresholds in the range of 0.30~0.45,the overall accuracy can reach more than 90%,and it can meet the requirements of identification of citrus.     

基于差分搜索的高光谱图像解混算法

针对高光谱图像解混问题进行研究,发现高光谱图像中各个端元的分布不完全独立,不能将盲源分离方法直接应用于高光谱图像解混。为此,提出了一种基于差分搜索的高光谱图像解混算法。该算法根据高光谱图像丰度非负和丰度和为一特性构造相应的约束项,与互信息相结合作为目标函数,利用差分搜索算法对该目标函数进行优化求解来实现高光谱图像解混。仿真数据和实际数据实验表明,该算法能够有效解决高光谱图像解混问题,与已有其它算法相比,提高了图像解混的精度,并且针对不含纯像元的高光谱图像具有很好的解混效果。     

地球观测1号高光谱与全色图像融合的最佳方法

受制于成像原理及制造技术等因素,航天高光谱遥感图像的空间分辨率相对较低,为此提出将高光谱图像与高空间分辨率图像进行融合处理,设计最佳的增强高光谱遥感图像空间分辨率的融合算法。针对地球观测1号(EO-1)Hyperion高光谱图像和高级陆地成像仪（ALI）全色波段图像的特点,从9种具体遥感图像融合算法中选用4种融合算法开展山区与城市的数据融合实验,即Gram-Schmidt光谱锐化融合法、平滑调节滤波(SFIM)变换融合法、加权平均法（WAM）融合法和小波变换（WT）融合法,并分别从定性、定量和分类精度三方面对这些方法的融合效果进行综合评价与对比分析,从而确定适合EO-1高光谱与全色图像融合的最佳方法。实验结果显示：从图像融合效果看,在所采用的4种融合方法中,Gram-Schmidt光谱锐化融合法的效果最好;从图像分类效果看,基于融合图像的分类效果要优于基于源图像的分类效果。理论分析与实验结果均表明：Gram-Schmidt光谱锐化融合法是一种较为理想的高光谱与高空间分辨率遥感图像的融合算法,为提高高光谱遥感图像的清晰度、可靠性及图像的地物识别和分类的准确性提供有力的支持。     

高光谱图像非线性解混方法的研究进展

由于空间分辨率的限制,高光谱遥感图像中存在大量混合像元,对混合像元的解混是实现地物精确分类和识别的前提。与传统的线性解混方法相比,非线性解混方法在寻找组成混合像元的端元以及每个端元的丰度时具有较高的精度。分析了光谱非线性混合的原理,总结了近年来提出的非线性解混算法,重点对双线性模型、神经网络、基于核函数的非线性解混算法以及基于流形学习的非线性解混算法进行了介绍和分析。最后总结了混合像元非线性解混未来发展的趋势。     

基于互信息和随机优化的超光谱遥感图像配准

精确的谱间配准是从超光谱遥感图像中获取光谱信息的基本前提之一。而谱间配准的主要困难在于宽的成像光谱范围使波长间隔远的图像缺乏相似性, 超光谱图像本身海量的数据也限制了配准算法的复杂性。提出了一种结合互信息和随机优化技术的多分辨率配准方法。该方法采用互信息作为相似性测度, 能很好的适应超光谱图像光谱特征的变化; 二阶同步试探随机逼近(2SPSA )算法的应用, 解决了互信息的多变量优化问题; 通过一种具有平移和旋转不变性的小波分解实现算法的多分辨率形式, 能明显加快算法的收敛速度并保证搜索结果的全局最优性。实验结果表明该算法适用于配准波长范围很宽的超光谱图像, 并能达到子像素的配准精度。     

Linear spectral unmixing using endmember coexistence rules and spatial correlation

Mixed pixels are often formed when surface materials are smaller than the spatial resolution of a sensor, or two or more ground features fall within a pixel. Spectral unmixing, decomposing a mixed pixel into a set of endmembers and their corresponding abundance fractions, is an important method for extracting the underlying spectral and spatial information from remote sensing images. Recent studies have shown that it is difficult to increase the accuracy of unmixing using single pixel processing. Here, we suggest combining information on the fundamental interrelations of ground components and a priori knowledge on how ground components co-exist or exclude each other according to general geographic and geomorphic relations with spectral information may allow improved unmixing. Therefore, we propose a novel spectral unmixing method to estimate endmember abundances based on linear spectral mixing model with endmember coexistence rules and spatial correlation (LSMM-R&C). This method was implemented by incorporating endmember coexistence rules along with spatial correlation into a weighted least square method. Experiments with both synthetic and real satellite images were carried out to verify the proposed method, and its performance was also evaluated in comparison to the commonly used LSMM (linear spectral mixture method), LAU (local adaptive unmixing), ISU (iterative spectral unmixing) and ISMA (iterative spectral mixture analysis) methods. LSMM-R&C showed the smallest error, and was more effective at revealing the detailed spatial distribution of endmembers’ abundance, showing high potential for solving the problem of spatial heterogeneity among neighbouring pixels.     

Postfire soil burn severity mapping with hyperspectral image unmixing

Burn severity is mapped after wildfires to evaluate immediate and long-term fire effects on the landscape. Remotely sensed hyperspectral imagery has the potential to provide important information about fine-scale ground cover components that are indicative of burn severity after large wildland fires. Airborne hyperspectral imagery and ground data were collected after the 2002 Hayman Fire in Colorado to assess the application of high resolution imagery for burn severity mapping and to compare it to standard burn severity mapping methods. Mixture Tuned Matched Filtering (MTMF), a partial spectral unmixing algorithm, was used to identify the spectral abundance of ash, soil, and scorched and green vegetation in the burned area. The overall performance of the MTMF for predicting the ground cover components was satisfactory (r² = 0.21 to 0.48) based on a comparison to fractional ash, soil, and vegetation cover measured on ground validation plots. The relationship between Landsat-derived differenced Normalized Burn Ratio (dNBR) values and the ground data was also evaluated (r² = 0.20 to 0.58) and found to be comparable to the MTMF. However, the quantitative information provided by the fine-scale hyperspectral imagery makes it possible to more accurately assess the effects of the fire on the soil surface by identifying discrete ground cover characteristics. These surface effects, especially soil and ash cover and the lack of any remaining vegetative cover, directly relate to potential postfire watershed response processes.     

高光谱遥感图像的单形体分析方法

将n个波段的高光谱图像像元与n维空间里的散点联系起来，结合凸体几何中单形体概念研究高光谱遥感图像纯净像元提取方法，实现图像的地物精确分类识别及像元波谱分解。寻找高光谱遥感图像n维空间里的单形体并认知分析单形体是该研究方法的重要环节。通过MNF(minimum noise fraction)变换和PPI(pixel purity index)计算技术寻找到单形体，基于单形体进行像元分解分析单形体，并结合应用实例和SAM(spectral angle mapper)分类技术完成高光谱图像地物精确分类制图，验证了该研究方法的可操作性。该研究方法的优点在于不需要用户提供地物波谱信息，用于制图和波谱分解的终端单元可由图像本身得到，并由用户控制分类制图和波谱分解的详细程度。     

Spatial-Spectral Preprocessing based on Nonnegative Matrix Factorization to Unmix Hyperspectral Data

Non-negative Matrix Factorization (NMF)method of blind spectral unmixing can obtain the spectrum and abundance of the endmember by synchronous optimization,without supervising the selection of endmember.Therefore,NMF has been developed rapidly in the application of hyperspectral unmixing.However,traditional blind spectral unmixing NMF method tends to fall into the local optimum and it is difficult to obtain a stable optimal solution.In this paper,we propose an improved Non-negative Matrix Factorization (NMF)method based on Spatial\|Spectal Preprocessing for spectral unmixing of hyperspectral data (SSPP-NMF).First,the SSPP algorithm is used to combine spatial and spectral information to select reasonable and effective dataset.Then,the NMF algorithm is used to unmix this dataset to obtain the final optimized endmember spectrum.Finally,the Non\|Negative Least Squares (NNLS)method is used to obtain the final abundance of the whole study area.The validity and applicability of the proposed method were analyzed based on a set of synthetic hyperspectral data and real hyperspectral images;and then the results were compared with that from three algorithms including the existing NMF algorithm,MVC\|NMF algorithm and ATGP-NMF algorithm.Results show that compared with ATGP-NMF and MVC-NMF,the SSPP algorithm can effectively suppress the influence of noise,significantly improve the performance of the NMF method of blind spectral unmixing algorithm.