基于矩阵乘法的可压缩门限图像共享方案

基于矩阵乘法秘密共享思想,提出了一种新的可压缩的图像秘密共享方案.本方案利用可压缩矩阵乘法算法直接生成影子图像,有效地克服了秘密图像需进行置乱预处理的缺陷,并且方案的复杂度较低.本方案生成的影子图像仅为原秘密图像的1/t,有效地压缩了影子图像的尺寸.     

基于SWAT模型模拟乌江三岔河生态系统产流服务及其空间变异

产流是生态系统水源涵养服务的重要内容,也是保水固土等服务的关键驱动。脆弱的喀斯特生态系统叠加人类活动的影响引起生态退化,在此背景下亟需深刻理解喀斯特地区产流服务空间格局及其影响因子。选择典型的喀斯特峰丛洼地流域,基于率定校准的SWAT水文模型,对喀斯特流域产流服务（包括地表、地下以及总径流量）进行模拟,并结合空间梯度分析和局部回归模型,剖析不同服务变量的空间变异特征。结果表明：研究区产流量空间分异规律明显,总径流系数约为70.0%;地表径流整体处于较低水平,地下径流量丰富（约为地表径流量的2~3倍）。地形因子对喀斯特流域产流服务具有宏观控制作用,随海拔和坡度升高,总径流量、地下径流量显著增加;植被分布特征对地表径流有重要影响,在上游林地和下游园地分布区存在空间差异,在植被因素的影响下随坡度的增加呈现转折性的变化趋势。此外,空间叠加分析表明林地总径流及地下径流最大：一方面是由于森林生态系统更强的土壤水分涵养能力以及水分快速入渗的地质背景特征;另一方面反映了地表覆盖与地形特征的复合效应,即林地大多分布在相对海拔和坡度较大之处。本文有助于推进喀斯特生态系统服务研究领域的发展与创新,并为生态恢复建设提供科技支撑。     

海洋环境大数据系统建设研究进展

海洋环境大数据是诸多学科研究领域的基础,也是实施海洋强国战略和建设海洋强国奋斗目标的先决条件。文章从全球海洋环境观测计划及系统出发,探讨现阶段全球海洋环境大数据观测、国家数据库建设、数据质控方法以及数据应用现状,阐明了海洋环境大数据在国家发展中的重要性,客观地分析了目前海洋环境大数据发展在基础设施、数据处理及数据质控能力建设等方面面临的关键问题与挑战,以期为将来海洋环境大数据在多尺度、多学科、多行业的深入发展与深度融合提供可靠的资料与建议。     

关于端基分析法测定壳低聚糖的相对数均分子质量问题的探讨

为使端基分析法更具合理性。文中以铁氰化钾为显色剂。以盐酸氨基葡萄糖和N-乙酰基-D-葡萄糖胺为还原糖的标准品,建立了端基分析法测定低聚糖相对数均分子质量的标准曲线。实验结果证明：通过不同浓度的标准溶液绘制的标准曲线,测得壳低聚糖的相对数均分子质量差别比较大。以凝胶色谱法测定的相对数均分子质量为对照,经分析得：以盐酸氨基葡萄糖为标准品建立的标准曲线方程和以N-乙酰基-D-葡萄糖胺为标准品建立的标准曲线方程更为合理。     

THE FINE CRUSTAL STRUCTURE OF THE SOUTHERN MARGIN OF NORTH CHINA BLOCK REVEALED BY DEEP SEISMIC REFLECTION PROFILE

The study area is located at the junction of the northern margin of the Qinling orogenic belt and the southern margin of the North China Block. In order to study the fine crustal structure and the deep-shallow structural features of faults in this area, we conducted deep seismic reflection profiling with the seismic profile of 100km long, directing NE-SW in Zhumadian City, Henan Province, and got clear lithospheric structure images along the profile. As regards the data acquisition, we applied the geometry of 25m group interval, 1000 recording channels and more than 60 folds. Seismic wave exploding applies the 30kg shots of dynamite source with the borehole depth of 25m. The shot interval is 200m. In data processing, we focused on improving the signal-to-noise ratio. Data processing methods mainly include first break removal, tomographic static correction, abnormal amplitude elimination, amplitude compensation, pre-stack denoising, surface consistent deconvolution, velocity analysis, several iterations of the residual static correction, dip moveout, post-stack time migration and post-stack denoising, etc. The profile with high signal-to-noise ratio was obtained. The reflection wave group characteristics is obvious in the crust, which reflects abundant information about geological structure. Along the profile, the crust is characterized by double-layer reflection structure, and the Moho surface is composed of a series of laminated arc-shaped strong reflections. The thickness of the upper crust is about 14.8~20.7km, and the total thickness of the crust is about 32.0~35.1km. The upper crust is dominated by the inclined, densely stratified or arc-shaped reflections. The lower crust is dominated by arc-shaped and inclined reflection, and there is a reflective transparent zone under the Moho surface. The reflection sequences with different directions and shapes in the upper crust constitute the nappe structure in southwest segment and the structural model of two concaves with one uplift in NE segment, which correspond to the north Qinling nappe, Zhumadian-Huaibin depression, Pingyu-Xiping uplift and a secondary depression, respectively. There are abundant arc-shaped reflection sequences in the lower crust, which may represent multi-stage magmatic activities. The deep seismic reflection profile shows that faults in the upper crust are well developed. According to the characteristics of reflected wave field in the profile, four groups of fault structure which contain ten faults with different scales are interpreted. Among them, faults F_P1, F_P2 and F_P3 constitute the thrust fault system in the northern margin of Qinling Mountains, and F_P5 and F_P7 are boundary faults of Zhumadian-Huaibin depression. These faults are all developed within the upper crust. In addition, the Fault F_PM is a large fault that cuts through the lower crust and Moho surface. The deep seismic reflection profile reveals the crustal structure and deep-shallow structural features of faults at the junction of the northern margin of the Qinling orogenic belt and the southern margin of the North China block, which provides seismological evidence for the analysis of structural differences, the deep earth's interior processes and deep-shallow structural relationships between the Qinling-Dabie orogenic belt and the southern margin of the North China block. The lower crust of the study area is divided into two parts by deep faults that dislocate the Moho surface. These two parts have distinct reflective structures, suggesting that the area has experienced intense complex tectonic movements. The faults in the upper crust control the formation of basin-mountain structure and stratigraphic deposition of this area. And deep faults in the crust that disrupt Moho surface create conditions for the upwelling and energy exchange of deep materials. All of these have regulated the composition of material and the distribution of energy in the crust. The deep faults cutting through the lower crust and Moho surface and the south-dipping arc-shaped and inclined strong reflection sequences developed in the lower crust should indicate the large-scale subduction of the southern margin of the North China block towards the south-trending Qinling orogenic belt.     

基于云模型和差异驱动原理的农用地分等研究

在农用地分等的过程中,评价因素涵盖了定量和定性的数据,定量数据准确地反映该因子的特征,而定性因素则无法用准确的数据来表示的。语言值或者概念和数学符号的最大区别,就是其中的不确定性。此外专家打分确定的权重也存在随机性,各评价因素对农用地等别影响的侧重也不能得到确切的再现。针对以上理由,本文提出采用云模型的不确定性分析和基于差异驱动原理的赋权,试图改进现行的农用地分等方法中因素作用的判读及其对分等结果的影响。     

卫星图像斜方位投影正反解变换研究

本文针对卫星动态获取的遥感图像,探讨其空间斜方位投影问题。利用空间几何理论建立其斜方位投影的数学模型,得到了该投影的精确表达式。并且利用矢量解法研究该投影模型的正解和反解变换算法以及星下点坐标计算方法,最后给出了算例。根据算例的结果分析表明,该投影的正反解变换均达到较高的精度。     

TerraSAR影像干涉测量数据的地理编码方法

InSAR技术广泛用于获取高精度的地形信息,但是经过干涉处理获得的干涉测量数据必须经过地理编码,才能得到可实用的DEM。本文针对高分辨率TerraSAR影像,提出了利用InSAR的影像和高度信息,结合卫星的位置矢量和速度矢量以及SAR距离-多普勒构像模型,进行高程图地理编码处理的方法,直接将干涉生成的DEM校正到所需的地面参考坐标系统中,以获取干涉数据的地面三维信息。通过试验,证明了该方法的可行性。     

面向智能终端的地震应急信息移动发布研究

讨论一种新的信息发布模式,实现将地震应急信息面向智能手持终端发布和浏览。该系统采用智能PUSH技术、SMS技术实现信息的发布,采用GPRS实现信息的传输。研究地震应急信息源和地震应急移动信息发布服务模式,设计地震应急移动信息发布系统的结构功能。目前,地震应急信息移动发布系统已经在国务院抗震救灾指挥部得到应用,并取得一定的成效。     

基于面向对象和影像认知的遥感影像分类方法——以都江堰向峨乡区域为例

基于面向对象思想和影像认知科学研究,探讨影像理解的新思路,将语义特征和传统的低级影像特征综合考虑进行对象识别,该方法对于提高影像分类精度和效率及经验知识的重复利用都有重要意义。在识别过程中,首先进行知识驱动的对象探测,分析语义特征;再进行数据驱动的影像多尺度分割,生成低级特征;最后应用模糊分类整合语义特征和低级特征用于对象识别和分类。将该方法应用于四川都江堰向峨乡QuickBird等数据,表明新方法比传统分类方法有更高的精度和效率。