基于模拟退火算法的SAR图像复原

根据吉布斯马尔可夫随机场模型和SAR图像斑点噪声的伽玛分布统计特性，应用模拟退火算法．实现雷达截面的全局优化伪似然估计。本文将新解4选1的方法从极化SAR图像推广到一般SAR图像；提出一个实用有效的图像特征检测法；说明了一阶模型和零阶模型之间的联系．并给出满意的处理结果。     

Results from the study of an M7.6 flare and its associated CME

An M7.6 flare was well observed on October 24, 2003 in active region 10486 by a few instruments and satellites, including GOES, TRACE, SOHO, RHESSI and NoRH. Multi-wavelength study shows that this flare underwent two episodes. During the first episode, only a looptop source of <40 keV was observed in reconstructed RHESSI images, which showed shrinkage with a velocity of 12–14 km s⁻¹ in a period of about 12 min. During the second process, in addition to the looptop source, two footpoint sources were observed in energy channel of as high as ∼200 keV. One of them showed fast propagation along one of the two TRACE 1600 Å flare ribbons and the 195 Å loop footpoints, which could be explained by successive magnetic reconnection. The associated CME showed a mass pickup process with decreasing center-of-mass velocity. The decrease of the CME kinetic energy and the increase of its potential energy lead to an almost constant total energy during the CME propagation. Our results reveal that the flare and its associated CME have comparable energy content, and the flare is of non-thermal property.     

微孔光学阵列长宽比对X射线聚焦性能的影响

微孔光学阵列聚焦镜头是实现X射线探测系统高探测效率、轻小型化的有效手段,近年来得到了深入研究。X射线探测系统通过测量脉冲星辐射的X射线光子的到达时间,重构脉冲轮廓来实现航天器导航。为了获得更精确的脉冲星信号,必须尽可能多地收集X射线光子,因此探测系统的聚焦性能是X射线探测系统的核心指标,也是微孔光学阵列主要的设计目标。文章根据脉冲星能谱的分布特点,分析了微孔光学阵列通道长宽比对X射线聚焦性能的影响,提出了针对不同脉冲星选取特定长宽比的方法,实现了不同的脉冲星能谱分布下最优的聚焦效率,为后续微孔光学阵列的研制和工程搭载提供了参考。基于这种方法,文章对Crab脉冲星的能谱分布进行了微孔光学阵列设计,设计结果显示,其聚焦性能得到明显提高。     

一种基于石墨烯的脉冲星导航X射线探测器

针对脉冲星导航要求X射线探测器具有高能量分辨率和高时间分辨率的需求,研究基于石墨烯电场效应来实现X射线探测的可行性。通过蒙特卡罗方法模拟不同能量X射线和不同厚度探测器基底的相互作用过程,得到不同厚度探测器对X射线的探测效率。利用有限元方法分析X射线在探测器内电离产生载流子的输运过程,验证了X射线入射对石墨烯电学特性的影响。理论分析计算探测器的时间分辨率和能量分辨率等指标参数,结果表明：基于石墨烯电场效应的X射线探测器具有较高的能量分辨率和时间分辨率,是一种潜力巨大的X射线探测方式,可满足未来脉冲星导航的工程应用。     

基于时间误差补偿的HJ-1C卫星几何定位改进

利用距离多普勒算法对 HJ-1C（“环境一号”C）卫星图像进行定位,其初始定位精度为1100~1400m,不能很好地满足实际应用的需要。进一步分析后发现,HJ-1C卫星图像几何定位误差主要分布在方位向,通过对影响HJ-1C卫星几何定位精度的因素进行分析,得出合成孔径雷达（Synthetic Aperture Radar, SAR）时间误差是影响几何定位的关键因素,因此文章提出通过时间误差补偿提高HJ-1C卫星几何定位精度。该方法首先计算SAR时间误差,然后对SAR载荷时间进行补偿,最后利用距离多普勒算法进行定位,并对HJ-1C卫星图像进行了验证。结果表明,HJ-1C卫星定位精度提高到300m左右,得到了有效的改进。     

基于X射线脉冲星和星间链路的导航星座自主导航仿真研究

主要对基于X射线脉冲星和对基于星间链路的导航星座自主导航方法进行了仿真研究.同时提出了基于2种导航方法相融合的新方法,仿真研究证明两者融合的导航方法有效提高了位置估计精度,为导航星座自主导航研究提供了一种新的思路和实现途径.     

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。     

码流映射的JPEG2000遥感图像无损信息隐藏

遥感图像经过飞行器JPEG2000压缩编码传输到地面,在传输过程中有时需要搭载其他一些遥测信息,将附加信息嵌入到压缩文件中。对JEPG2000的数据结构进行了研究,将压缩文件的数据部分变换成图像并将其分块,统计256个灰度值,找到出现次数最多的灰度值以及次数为0的灰度值。将这些灰度值所对应的字节进行映射,不同的灰度值代表不同的二进制秘密信息。接收端通过映射将秘密信息提取后,再将载体更改后的字节恢复。隐藏容量平均有了500bit的提升,载密图像的PSNR大于30dB,同时载体数据仍然可以无损恢复。试验结果说明,该算法可以无损地实现JPEG2000图像的信息搭载,同时文件没有膨胀。     

A synchrotron-based facility for the in-situ location,chemical and mineralogical characterization of ∼10 μm particles captured in aerogel

NASA’s Stardust mission collected dust from the coma of Comet Wild-2 on January 2nd, 2004, by direct capture into aerogel cells that flew through the dust coma at ∼6 km/s. Stardust collected several hundred comet particles >10 μm in size. These comet samples were delivered to Earth on January 15th, 2006. We developed a facility at the National Synchrotron Light Source at Brookhaven National Laboratory (Upton, NY, USA) for the in-situ characterization of ∼10 μm particles collected in aerogel. These analytical instruments allow us to perform extensive chemical, mineralogical, and size-frequency characterization of particles captured in aerogel. These analyses are conducted without any invasive extraction, minimizing the possibility of contamination or particle loss during preparation. This facility was used to determine the chemical composition, the oxidation state, the mineralogy and to provide an indication of the grain size of the Wild-2 particles before they were removed from the aerogel. This information provides a catalog of particle types, allowing a more reasoned allocation of the particles to subsequent investigators based on a relatively detailed knowledge of the chemical composition and mineralogy of each particle. These measurements allowed a comparison of the chemical and mineralogical properties of the Wild-2 particles with other types of extraterrestrial materials, including interplanetary dust particles and meteorites. The success of in-situ analysis for Wild 2 particles demonstrates that synchrotron-based facilities will be important for the analysis of particles collected in aerogel on future earth-orbiting satellites and spacecraft.