基于光谱复原的NMF遥感图像融合改进算法

在研究了非负矩阵分解(NMF)用于遥感图像融合技术的基础上,改进了一种基于光谱复原的NMF遥感图像融合算法,首先利用矩阵非负性分解求取图像的基向量矩阵,然后综合考虑图像融合后的光谱特性和波形变化情况,对基向量矩阵进行直方图规则化处理,最后得到融合后的图像。实验结果表明,该方法具有较好保持原始图像的光谱信息和空间信息的优点。     

水下收缩裂隙形成过程及裂缝充填模式研究

通过野外实地考察测试,对水下收缩裂隙整体形成过程、裂隙内充填物沉积模式、影响因素等进行了详细讨论,并且建立了相应的地质理论模型。水下收缩裂隙的形成共分4个阶段,依次为:泥水混合物进入低洼地带的初始混浊状态;沉积压实稳定阶段;水位线下降,盐度增大,裂隙形成阶段;沉积裂隙充填阶段。其中“异常高压作用”,对于初始裂隙的形成、沉积物孔隙水的有效排出以及主要渗流通道的发育起到了重要作用。单一泥质地层裂隙内充填物沉积模式与沙泥互层的多旋回地层略有差异,主要表现为沙层的润滑作用以及对早期裂隙内充填物类型的影响。上述两种地层裂隙内充填物沉积特征均与后期充注期次相对应,表现出很好的韵律性。裂隙纵剖面中生物发育层的出现,对裂隙内充填物的物理、化学性质将产生重要影响。     

基于交通网络空间权重的区域经济空间相关分析——以甘肃省为例

在区域经济中,交通是联系地理空间和区域经济活动的纽带,交通的发达程度决定了各地理单元空间相互作用的广度与深度.针对不同交通模式赋以不同的权重,基于最短加权交通网络,提出新的空间权重矩阵构建方法,构建了交通网络空间权重,与各种传统空间权重一起,对比研究甘肃省各县域单元之间的区域经济的空间相关性.研究结果表明,利用交通网络空间权重生成的空间权重矩阵,能更真实地反映区域间实际的空间过程;甘肃省的区域经济具有空间相关性但不显著,核心城市经济外溢现象不明显.     

基于显著性图像与纹理特征的遥感影像云检测

遥感影像云检测是遥感影像处理中非常关键的环节,准确识别影像含云区域能够提升影像的利用价值.根据遥感影像的成像特点,将阈值法和纹理特征结合实现云和下垫面的分割.首先将影像从RGB(red-green-blue)空间转化为HSI(hue-saturation-intensity)空间,进而构建影像的显著性图像,利用Otsu...     

Macroscopic criteria for Green type porous materials with spheroidal voids: application to double porous materials

Combined effects of matrix plastic compressibility and void shape are investigated for ductile porous materials. To this end, a spheroidal volume containing a confocal spheroidal (prolate or oblate) void subjected to uniform strain rate boundary conditions has been first studied. A Green type matrix is chosen as a prototype for investigating effects of plastic compressibility. This is carried out by using a kinematics limit analysis theory from which a closed‐form expression of the macroscopic criterion is established for the considered class of materials. These results are then extended to ductile porous materials made up of a green matrix containing randomly oriented spheroidal voids. In the framework of a two‐step homogenization procedure, the obtained results are implemented to describe the macroscopic behavior of double porous materials involving spherical voids at the microscale and randomly oriented and distributed spheroidal voids at the mesoscale. For validation purpose, the new derived criteria are assessed and validated by comparing their predictions to available upper bounds and numerical data from literature. Copyright © 2017 John Wiley & Sons, Ltd.     

Evaluation of a probabilistic approach to simulation of alkalinity and electrical conductivity at a river bank filtration site

In river bank filtration, impurities present in the river water travel with the bank filtrate towards the pumping well. During this passage, certain types of impurities, such as turbidity, total coliform, and so forth, may get attenuated; however, it is interesting to note that some of the instant raw river water quality parameters, such as alkalinity and electrical conductivity, increase after the passage of water through the porous medium. This occurs because water, when passing through the soil pores, absorbs many of the solutes that cause an increase in alkalinity and electrical conductivity. Measurements at a river bank filtration site for a year showed that alkalinity of 116–32 mg l^?1 in river water increased to 222.4–159.9 mg l^?1 in the river bank filtered water. Likewise, the electrical conductivity increased from 280–131 μS cm^?1 to 462–409.6 μS cm^?1. This study uses a probabilistic approach for investigating the variation of alkalinity and electrical conductivity of source water that varies with the natural logarithm of the concentration of influent water. The probabilistic approach has the potential of being used in simulating the variation of alkalinity and electrical conductivity in river bank filtrate. Copyright © 2011 John Wiley & Sons, Ltd.     

三分量地震资料中各向异性检测

各向异性研究是当令地震学研究领域中重要课题之一；三分量地震资料中各向异性的检测是各向异性理论研究成果解决地球科学实际问题的重要桥梁。本文简要描述了各向异性介质中横波分裂现象，分别介绍了现有的四种主要检测技术，即偏振图法、协方差拒阵法、质，点振动分辨率法及传输矩阵法，并对这四种技术作了一定的评述。     

用解析有限元法嵌入地下水管理模型求最优解

为了弥补当前地下水资源优化管理模型中通用的两种构模方法的缺点,即嵌入法的代数方程组过于庞大、上机困难,和响应矩阵法不保证全局最优,提出了用解析有限元法建立整体非稳定(即带有时间变量)抽水响应矩阵作为约束条件嵌入线性规划模型.这一方法由于具备了空间上表达任一节点在任一时刻的地下水运动规律的能力,克服了上述构模的两大缺点,其推广应用将有利于提高地下水资源的优化开发和管理.     

Rangeland production: use of models incorporating aggregated knowledge and fuzzy construction

This work reports on a method using fuzzy membership functions to construct an aggregated interaction matrix in which the summation of variables is scaled according to the way rainfall and soil variables affect water availability to plants and hence influence rangeland productivity. Aggregation of the variables gives a comprehensive value which can be used to predict production. The model increases the predicability of production to 81% compared to models using rainfall alone and a multiplicative parametric one which give predictibilities of 61 and 76% respectively. The results showed that (1) the importance of rainfall in determining production was most important at lower rainfalls i.e. <350 mm; (2) soil texture and particularly slope were important through out the rainfall range (149–700 mm) investigated; and (3) soil depth was only important at the higher >350 mm rainfalls. The aggregated interaction matrix gives a measure of land productive capability.