提高中巴卫星IR MSS图像空间分辨能力的光谱保真融合方法

介绍一种提高中巴资源卫星IRMSS图像空间分辨能力的光谱保真融合方法。通过计算低分辨率图像上每一个像元对应的高分辨率图像上一组子像元的平均亮度值及二者之差,将该差值与高分辨率图像上相应子像元亮度求和,形成新的图像。该图像具有高分辨率图像的空间细节,又具有低分辨率图像的光谱信息,从而实现融合图像信息保真。试验表明,光谱保真融合方法可以在不改变光谱信息的前提下提高IRMSS图像的空间分辨能力,是一种新的简单实用的数据处理方法。     

基于小波分量特征值匹配的高光谱影像分类

提出了一种基于小波分量特征值的高光谱影像分类算法。针对每个像素构建一个能反映该分量特征的函数，得到其特征值。再利用这些特征值与参考光谱的特征值进行匹配，从而对整幅影像实现分类。实验证明，该方法比传统的光谱角制图法和交叉相关系数法的分类精度有较大提高。     

结合空间信息选取最优端元组合的混合像元分解

传统的混合像元分解算法认为每个像元都包含图像中所能提取的全部端元组分,但这并不符合实际情况。实际上图像中大多数混合像元仅由少部分端元混合而成。由于端元提取精度及噪声的影响,采用全部端元对混合像元进行分解,会使得混合像元中实际并不存在的端元的丰度估计值不为零,分解结果存在较大误差。由于混合像元大多存在于不同地物的交界处,基于此,本文提出了一种结合图像的空间信息选取混合像元最优端元子集的方法。利用一个空间结构元素,从混合像元的附近邻域开始搜索,将搜索到的纯净像元光谱与所提取的图像端元光谱进行对比,并确定混合像元的端元子集进行分解。根据RMSE大小和变化情况,逐步扩大结构元素的大小,不断调整搜索范围,直至得到最优端元组合。模拟数据和真实数据的试验结果表明,该方法相比传统的全端元光谱分解方法,在总体上获得了更好的分解效果。     

密集卷积残差网络的遥感图像融合

针对传统的遥感图像融合方法通常会引起光谱失真的问题和大多数基于深度学习的融合方法忽略充分利用每个卷积层信息的不足,本文结合密集连接卷积网络和残差网络的特性,提出了一个新的融合网络.该网络通过建立多个密集卷积块来充分利用卷积层的分级特征,同时块与块之间通过过渡层加快信息流动,从而最大程度地对特征进行极致利用并提取到丰富的...     

融合形状和光谱的高空间分辨率遥感影像分类

提出了一种像元形状指数及基于形状和光谱特征融合的高（空间）分辨率遥感影像分类方法。形状和光谱是遥感影像纹理的具体表现形式,尤其在高分辨率影像中地物细节得到充分表达,相邻像元的关系及其共同表征的形状特性成为分类的重要因素。本文用像元及其邻域的关系来描述其空间结构,同时为了更全面地利用影像特征,提出了基于支持向量机的形状和光谱融合分类方法。实验证明,该方法计算简便且能有效表达高分辨率影像的地物特征,提高分类精度。     

开封市Spot5全色波段与多光谱影像融合评价

以Spot5全色波段与多光谱波段为例,用PCA、Brovey、IHS等三种像素级融合方法对其进行融合研究。分别运用灰度均值、相关系数、平均梯度等三个传统的指标在影像亮度、光谱保真度以及高频信息融入度三个方面对其评价。最后综合这些指标在信息量以及融合指数方面对融合影像进行综合评价,从而建立一套完整的影像评价体系。     

A new pan-sharpening method using multiobjective particle swarm optimization and the shiftable contourlet transform

In this paper, a novel approach based on multiobjective particle swarm optimization (MOPSO) is presented for panchromatic (Pan) sharpening of a multispectral (MS) image. This new method could transfer spatial details of the pan image into a high-resolution version of the MS image, while color information from the low-resolution MS image is well preserved. The pan and MS images are locally different because of different resolutions, and therefore we cannot directly combine them in the spatial domain. For this reason, we generate two initial results, which are more appropriate for a weighted combination. First, the pan and the MS images are histogram matched. Then we use the shiftable contourlet transform (SCT) to decompose the histogram-matched pan and MS images. The SCT is a new shiftable and modified version of the contourlet transform. In this step, an algorithm based on the SCT is used to generate two initial results of the high-resolution MS images. Our objective is to produce two modified high-resolution MS images, in which one has high spatial similarity to the pan image and the other one has high radiometric quality in each band. Therefore, we have used two different fusion rules to integrate the high-frequency contourlet coefficients of the pan and MS images to generate two initial results of high-resolution MS image or the pan-sharpened (PS) image. Finally, we can find the optimal PS image by applying the MOPSO algorithm and using the two initial PS results. Specifically, the PS image is obtained via a weighted combination of the two initial results, in which the weights are locally estimated via a multiobjective particle swarm optimization algorithm to generate a PS image with high spatial and radiometric qualities. Based on experimental results obtained, the produced pan-sharpened image also has good spectral quality. The efficiency of the proposed method is tested by performing pan-sharpening of high-resolution (Quickbird and Wordview2) and medium-resolution (Landsat-7 ETM +) datasets. Extensive comparisons with the state-of-the-art pan-sharpening algorithms indicate that our new method provides improved subjective and objective results.     

IKONOS图像的线性回归波段拟合融合方法

讨论基于线性回归波段拟合的空间细节信息提取方法的可行性。首先通过全色与多光谱图像构造线性回归方程,根据全色图像的高频成分设置最小二乘求解的权系数,然后利用回归系数构造低分辨率的全色图像,提取空间细节信息,最后将空间细节注入多光谱图像中进行融合。通过IKONOS全色和多光谱图像的融合实验,比较了本文方法与基于光谱响应函数的方法,结果表明:采用本文方法提取的空间细节信息进行融合,能达到甚至超过基于光谱响应函数方法的融合质量;相对于与FastIHS融合方法,本文方法的融合质量也有较大的提高。     

结合空间上下文与慢特征分析的多光谱影像变化检测

为了提高多光谱影像变化检测的精度,本文提出了一种结合空间上下文与慢特征分析的方法。首先采用自适应空间上下文提取算法围绕像素构建自适应区域,探索像素周围的上下文信息;然后通过迭代慢特征分析,由相应像素周围的成对自适应区域定量计算成对像素之间的变化强度,增强变化区域与未变区域的可分性;最后生成变化强度图像,采用大津阈值法作二值分类,将变化强度图划分为二值变化检测图。利用Landsat 7卫星ETM+传感器的图像,与4种基于代数的方法及基于变换的方法进行对比试验,结果表明,本文方法在降低漏检方面有所改善,提高了召回率。     

结合Gram-Schmidt变换的高光谱影像谐波分析融合算法

针对高光谱影像谐波分析融合(HAF)算法在影像融合时不顾及地物光谱曲线整体反射率这一缺陷,提出了结合Gram-Schmidt变换的高光谱影像谐波分析融合(GSHAF)改进算法。GSHAF算法可在完全保留融合前后像元光谱曲线波形形态的基础上,将高光谱影像融合简化为各像元光谱曲线的谐波余相组成的二维影像与高空间分辨率影像之间的融合。它是在原始高光谱影像光谱曲线被谐波分解为谐波余项、振幅和相位后,首先将其谐波余项与高空间分辨率影像进行GS变换融合,这样便可有效地修正融合后像元光谱曲线的反射率特征,随后再利用该融合影像与谐波振幅、相位进行谐波逆变换,完成高光谱影像谐波融合。本文最后利用Hyperion高光谱遥感影像与ALI高空间分辨率影像对GSHAF算法进行可行性分析,再以HJ-1A等卫星数据对其进行普适性验证,试验结果表明,GSHAF算法不仅可以完全地保留光谱曲线波形形态,而且融合后影像的地物光谱曲线反射率更接近真实地物。     

黄河口遥感图像光谱混合分解

探讨了用逻辑斯蒂法进行了光谱混合分解的新技术，采用黄河口LM图像进行了分析。结果表明，它不仅能给出分类结果图像，而且能产生组成像元各地类的丰度图像，说明分类图像是在某种置信度下的结果。     

改进IHS变换的图像融合方法研究

经典的IHS变换融合以其快速、简单的特点得到广泛应用,但该方法在显著提高图像空间分辨率的同时引起了严重的光谱失真.本文分析了IHS变换引起光谱失真的原因,并总结了已有的针对光谱失真的IHS变换改进算法,提出基于小波的IHS变换融合,利用小波分解尽可能地保留了原I分量中的光谱信息,在提高空间分辨率的同时减少了光谱失真.实...     

Enhancing endmember selection in multiple endmember spectral mixture analysis (MESMA) for urban impervious surface area mapping using spectral angle and spectral distance parameters

Successful retrieval of urban impervious surface area is achieved with remote sensing data using the multiple endmember spectral mixture analysis (MESMA). MESMA is well suited for studying the urban impervious surface area because it allows the number and types of the endmembers to vary on a per-pixel basis, thereby, allowing the control of the large spectral variability. However, MESMA must calculate all potential endmember combinations of each pixel to determine the best-fit one. Therefore, it is a time-consuming and inefficient unmixing technology, especially for hyperspectral images because these images have more complicated endmember categories. Hence, in this paper, we design an improved MESMA (SASD-MESMA: spectral angle and spectral distance MESMA) to enhance the computational efficiency of conventional MESMA, and we validate this new method by analyzing the Hyperion image (Jan-2011) and the field-spectra data of Guangzhou (China). In SASD-MESMA, the parameters of spectral angle (SA) and spectral distance (SD) are used to evaluate the similarity degree between library spectra and image spectra in order to identify the most representative endmember combination for each pixel. Results demonstrate that the SA and SD parameters are useful to reduce misjudgment in selecting candidate endmembers and effective for determining the appropriate endmembers in one pixel. Meanwhile, this research indicates that the proposed SASD-MESMA performs very well in retrieving impervious surface area, forest, grass and soil distributions on the sub-pixel level (the overall root mean square error (RMSE) is 0.15 and the correlation coefficient of determination (R²) is 0.68).     

A GIHS-based spectral preservation fusion method for remote sensing images using edge restored spectral modulation

High spatial resolution and spectral fidelity are basic standards for evaluating an image fusion algorithm. Numerous fusion methods for remote sensing images have been developed. Some of these methods are based on the intensity–hue–saturation (IHS) transform and the generalized IHS (GIHS), which may cause serious spectral distortion. Spectral distortion in the GIHS is proven to result from changes in saturation during fusion. Therefore, reducing such changes can achieve high spectral fidelity. A GIHS-based spectral preservation fusion method that can theoretically reduce spectral distortion is proposed in this study. The proposed algorithm consists of two steps. The first step is spectral modulation (SM), which uses the Gaussian function to extract spatial details and conduct SM of multispectral (MS) images. This method yields a desirable visual effect without requiring histogram matching between the panchromatic image and the intensity of the MS image. The second step uses the Gaussian convolution function to restore lost edge details during SM. The proposed method is proven effective and shown to provide better results compared with other GIHS-based methods.     

Cross-sensor calibration between Ikonos and Landsat ETM+ for spectral mixture analysis

Spectral mixture analysis is an algorithm that is developed to overcome the weakness in traditional land-use/land-cover (LULC) classification where each picture element (pixel) from remote sensing is assigned to one and only one LULC type. In reality, a remotely sensed signal from a pixel is often a spectral mixture from several LULC types. Spectral mixture analysis can derive subpixel proportions for the endmembers from remotely sensed data. However, one frequently faces the problem in determining the spectral signatures for the endmembers. This study provides a cross-sensor calibration algorithm that enables us to obtain the endmember signatures from an Ikonos multispectral image for spectral mixture analysis using Landsat ETM+ images. The calibration algorithm first converts the raw digital numbers from both sensors into at-satellite reflectance. Then, the Ikonos at-satellite reflectance image is degraded to match the spatial resolution of the Landsat ETM+ image. The histograms at the same spatial resolution from the two images are matched, and the signatures from the pure pixels in the Ikonos image are used as the endmember signatures. Validation of the spectral mixture analysis indicates that the simple algorithm works effectively. The algorithm is not limited to Ikonos and Landsat sensors. It is, in general, applicable to spectral mixture analysis where a high spatial resolution sensor and a low spatial resolution sensor with similar spectral resolutions are available as long as images collected by the two sensors are close in time over the same place.     

面向高光谱图像分类的半监督空谱判别分析

为充分利用高光谱图像蕴藏的空间信息提升分类精度,提出了面向高光谱图像分类的半监督空谱判别分析(S3 DA)算法。考虑高光谱图像数据集的空间一致性,首先利用少量标记样本定义类内散度矩阵,保存数据集同类像元的光谱近邻结构;再利用无标记样本定义空间近邻像元散度矩阵,揭示像元间的空间近邻结构和地物的空间分布结构信息。S3 DA既保持数据集在光谱域的可分性,又保存了无标记样本蕴藏的空间域近邻结构,增强了同类像元和空间近邻像元在投影子空间的聚集性,从而提升分类性能。在PaviaU和Indian Pines数据集的试验表明,总体分类精度分别达到81.50%和71.77%。与传统的光谱方法比较,该算法能有效提升高光谱图像数据集的地物分类精度。     

基于信息量失真测度的VQ及在高光谱图像无损压缩中的应用

随着成像光谱技术的迅速发展 ,如何高效无失真的压缩海量高光谱数据引起人们越来越多的关注。由于相似的地表区域具有相似的光谱曲线 ,矢量量化是对高光谱图像进行压缩的理想算法。提出一种基于信息量失真测度的矢量量化编码方法 ,并用于高光谱图像无损压缩。与常用的矢量量化失真测度———欧几里德平方误差测度相比 ,该算法在不增加运算复杂度的情况下 ,矢量量化后的误差图像的熵值能够降低 0 0 5bpp左右。     

Quaternion-Based Sparse Model for Pan-Sharpening of IRS Satellite Images

This paper considers the pan-sharpening problem of the IRS satellite images from the perspective of vector sparse representation model using quaternion matrix analysis. It selects the sparse basis in quaternion space, which uniformly transforms the color channels into an orthogonal color space. Moreover, the proposed quaternion model for pan-sharpening is more efficient than the conventional sparse model as the hyper-complex representation of color channels conserves the interrelationship among the chromatic channels. This paper also proposes a quaternion forward–backward pursuit algorithm that preserves the inherent chromatic structures in terms of spatial and spectral details during the vector reconstruction. The experimental result validates the efficacy of the proposed quaternion model and shows its potential as a powerful pan-sharpening tool for IRS data even for cloudy multispectral data.     

高分五号高光谱数据融合方法比较

数据融合是解决高光谱卫星在时空分辨率等指标上受限的有效途径,探讨不同方法在GF-5高光谱数据上的融合效果,对GF-5高光谱数据的信息挖掘与推广应用有着重要意义。本文本着算法简单易用、适于推广的原则,采用GS（Gram-Schmidt）葛兰—施密特正交变换融合算法、GSA（GS Adaptive）自适应GS融合算法、CNMF（Coupled Non-negative Matrix Factorization）耦合非负矩阵分解融合算法、CRISP-W（Color Resolution Improvement Software Package with Wavelet transform）基于小波变换和CRISP-B（Color Resolution Improvement Software Package with Butterworth）基于巴特沃斯滤波器的分辨率提升融合算法、GLP（Generalized Laplacian Pyramid）广义拉普拉斯金字塔融合算法共6种融合方法,分别对BJ-2、GF-2、GF-1、GF-1C、GF-1D国产卫星多光谱数据与GF-5高光谱数据进行融合实验。通过目视分析、指标评价（相关系数、通用图像质量指标、峰值信噪比、光谱角、全局综合误差）、分类应用、时间成本4种方式对融合结果进行综合比较分析。结果表明,相融合的一组图像系列相同、空间分辨率相差越小,融合结果越好。CRISP-B、CRISP-W、GLP在提升空间分辨率、光谱保真度方面能达到较好的平衡,空间重建方面,GLP稍优且更稳定,CRISP-B、CRISP-W则在光谱信息保持方面稳定性更强且效果更好。数据源会对融合方法产生一定的影响,在光谱特征信息提取、分析等对光谱保真度要求高的工作中,GLP更适合同源数据（如GF-5与GF-1/1C/1D/2）融合,而在多源数据间（如GF-5与BJ-2）进行融合时,则优先选择CRISP-W。CNMF存在一定程度的色彩畸变,且运行时间较长。GSA、GS融合效果最差,其中,GSA不论是光谱保持能力还是空间分辨率提升能力均较GS更稳定。在小样本高光谱图像分类应用中,CRISP-B融合结果分类效果稳定,分类精度较高。GSA融合结果空间细节丰富,虽光谱失真较为严重,但同时增大了地物光谱分离度,仍适用于准确勾勒建筑物、道路等地物。本研究为GF-5高光谱数据与其他国产卫星多光谱数据融合方法的选择提供参考,有助于高分五号高光谱数据的应用与推广。     

A Hierarchical Hybrid SVM Method for Classification of Remotely Sensed Data

The focus of this work is on developing a new hierarchical hybrid Support Vector Machine (SVM) method to address the problems of classification of multi or hyper spectral remotely sensed images and provide a working technique that increases the classification accuracy while lowering the computational cost and complexity of the process. The paper presents issues in analyzing large multi/hyper spectral image data sets for dimensionality reduction, coping with intra pixel spectral variations, and selection of a flexible classifier with robust learning process. Experiments conducted revealed that a computationally cheap algorithm that uses Hamming distance between the pixel vectors of different bands to eliminate redundant bands was quite effective in helping reduce the dimensionality. The paper also presents the concept of extended mathematical morphological profiles for segregating the input pixel vectors into pure or mixed categories which will enable further computational cost reductions. The proposed method’s overall classification accuracy is tested with IRS data sets and the Airborne Visible Infrared Imaging Spectroradiometer Indian Pines hyperspectral benchmark data set and presented.