基于混合Contourlet变换和主成分变换的高光谱图像融合

针对高光谱图像波段多、存在噪声干扰等特点,提出了一种混合Contourlet和主成 分 分析（principal component analysis, 简称PCA）变换的高光谱图像融合方法。首先将多 个波段的高光谱图像进行Contourlet变换,得到系列多尺度、方向各异的子带系数,然后利 用PCA变换对各子带系数分别进行自适应融合处理。实验结果表明该算法可以有效地进行高 光谱图像融合,消除噪声干扰,获得比直接应用Contourlet变换和PCA变换更好的融合效果 。〖JP〗     

基于小波和支持向量回归的高光谱遥感图像压缩

高光谱遥感成像技术能够获取探测对象丰富的空间信息和光谱信息,但所得图像海量数据的存储与传输制约了其应用。为此,提出了一种基于三维整数小波变换和小波支持向量回归的高光谱遥感图像压缩方法。首先采用三维整数小波变换把高光谱遥感图像分解成不同尺度的多个子带。然后对低频子带直接进行DPCM编码,而对高频子带则利用小波支持向量回归学习其小波系数之间的相关性,并采用小部分训练样本即支持向量来稀疏表示小波系数,以此达到压缩高频小波系数的目的。最后对支持向量及其相应的权重进行熵编码。文中给出了实验结果,并与基于3D SPIHT和JPEG2000的高光谱遥感图像压缩方法进行了比较,结果表明：所提出的方法在相同比特率下能够获得更高的峰值信噪比。     

Sub-pixel mineral mapping of a porphyry copper belt using EO-1 Hyperion data

The main aim of the present study was to examine the feasibility of the EO-1 Hyperion data in discriminating and mapping diagnostic alteration minerals around porphyry copper deposits (PCDs), verified by field surveys and laboratory analyses. A partial sub-pixel method, mixture tuned matched filtering (MTMF), was implemented on a pre-processed and calibrated Hyperion dataset. The tested area is situated at the Central Iranian Volcano-Sedimentary Complex, where abundant porphyry copper deposits like Sarcheshmeh, Darrehzar, and Sereidun are located. The characteristic alteration minerals identified by Hyperion data included biotite, muscovite, illite, kaolinite, goethite, hematite, jarosite, pyrophyllite, and chlorite. Discrimination of these minerals especially biotite and iron oxide (hematite and goethite) can provide valuable evidences for PCD exploration projects. Results revealed that Hyperion data prove to be powerful in discriminating and mapping various types of alteration zones while the data were subjected to adequate pre-processing.     

高光谱图像融合最佳波段选择方法

针对高光谱图像高数据维给图像处理带来的困难和影响,本文构造了高光谱图像融合的最佳波段选择新模型-联合偏度-峰度指数(Joint Skewness-Kurtosis figure,JSKF)模型,利用JSKF指数进行自适应子空间的分解和波段选择,降低高光谱数据维数;并将选择出的最佳波段组合进行了融合,实验结果表明,该方法所选择的波段信息差异较大、互补特征明显,融合后图像包含的信息量丰富,效果优于传统的自适应波段选择方法和主成分分析累计贡献率方法.     

基于激光测振仪分离装置测量方法研究

针对目前采用的高速摄像测量方法存在误差较大与操作困难问题,提出了一种基于激光测振仪的新型测量方法,可以有效提高分离装置试验测量精度,为后续分离装置机构精确设计奠定基础。     

一种最大压缩误差可控的高光谱图像压缩算法

提出了一种可以实时控制压缩误差的高光谱图像有损压缩算法。该算法对高光谱 图像矩阵进行奇异值分解得到奇异值矩阵和奇异向量矩阵;用部分奇异值及其对应的奇异向 量重构图像;根据测量系统精度要求确定量化因子并对重构图像与原始图像间的光谱误差进 行量化;采用预测编码和算术编码对用于图像重构的奇异值及其对应奇异向量进行无损压缩 ;设计了非零值编码算法完成对重构误差的无损压缩。对Luna Lake和Low Altitude图像的 仿真结果为：最大相对误差分别控制在0.44％和0.33％时,压缩比为41.5:1和24.6:1, 信噪比为 50.4 dB 和47.8 dB。
     

Spectral-spatial target detection based on data field modeling for hyperspectral data

Target detection is always an important application in hyperspectral image processing field. In this paper, a spectral-spatial target detection algorithm for hyperspectral data is proposed. The spatial feature and spectral feature were unified based on the data filed theory and extracted by weighted manifold embedding. The novelties of the proposed method lie in two aspects. One is the way in which the spatial features and spectral features were fused as a new feature based on the data field theory, and the other is that local information was introduced to describe the decision boundary and explore the discriminative features for target detection. The extracted features based on data field modeling and manifold embedding techniques were considered for a target detection task. Three standard hyperspectral datasets were considered in the analysis. The effectiveness of the proposed target detection algorithm based on data field theory was proved by the higher detection rates with lower False Alarm Rates (FARs) with respect to those achieved by conventional hyperspectral target detectors.     

基于小波阈值降噪和经验模态分解的高光谱图像分类算法

为了充分降低高光谱图像中的噪声以获得高精度的分类结果,本文结合小波阈值降噪(WTD)和经验模态分解(EMD)的优点,提出了一种基于小波阈值降噪-经验模态分解的高精度支持向量机(SVM)高光谱图像分类算法(WTD-EMD-SVM)。首先对高光谱图像进行小波阈值降噪,除去高光谱数据中的高频噪声;然后再对高光谱图像进行EMD,获得含有高光谱数据本质特征的内固模态函数(IMF)和含有低频噪声的残差;最后采用内固模态函数重构高光谱图像,并对高光谱图像进行SVM分类。将其应用到AVIRIS数据92AV3C,仿真结果表明该算法不仅提高了高光谱图像分类精度,同时可减少支持向量数目,以提高高光谱图像分类速度。     

Exploring for natural gas using reflectance spectra of surface soils

Reflectance spectra in the visible and near-infrared wavelengths provide a rapid and inexpensive means for determining the mineralogy of samples and obtaining information on chemical composition. Hydrocarbon microseepage theory establishes a cause-and-effect relation between oil and gas reservoirs and some special surface anomalies, which mainly include surface hydrocarbon microseepage and related alterations. Therefore, we can explore for oil, gas by determining reflectance spectra of surface anomalies. This idea has been applied to the R&D project of exploring for natural gas in Qinghai province of China using NASA EO-1 satellite with the Hyperion sensor (June 2005 to June 2006). In this project, in order to improve the accuracy of exploration targets of natural gas mapped in the field studied, an integrated practical system of exploration of oil and gas was built by the analysis of not only hyperspectral remote sensing data but also data provided from field work. In this paper, our efforts were focused on the analysis of the 799 reflectance spectra provided from the field work. In order to properly define the typical form of hydrocarbon microseepage with spectroscopy and fulfill the data analysis, it was necessary to build a spectral model. In this spectral model the most important features of hydrocarbon microseepage in the surface of our study area, i.e., diagnostic spectral macroscopic features and diagnostic spectral absorption features, were proposed and extracted, respectively. The distribution of coexisting anomalies, which results from both alteration minerals and hydrocarbons, is estimated by the diagnostic macroscopic features mainly using Spectral Angle Mapper (SAM) classifier. On the other hand, the diagnostic absorption features of two main absorption bands presented abundant local information, based on deep analysis of which, we are able to map the anomalies of alteration minerals and hydrocarbons, respectively. Additionally, a general framework of analysis and key classification algorithms applied to the Hyperion data have been introduced briefly. In our work, three exploration targets of natural gas were identified from the study area which covers 2100 km². In the three exploration targets, three wildcats have been drilled by China National Petroleum Corporation (CNPC) since July 2006, and all the three wells have been proven some industrial reserves.