A fast clonal selection algorithm for feature selection in hyperspectral imagery

Clonal selection feature selection algorithm (CSFS) based on clonal selection algorithm (CSA), a new computational intelligence approach, has been proposed to perform the task of dimensionality reduction in high-dimensional images, and has better performance than traditional feature selection algorithms with more computational costs. In this paper, a fast clonal selection feature selection algorithm (FCSFS) for hyperspectral imagery is proposed to improve the convergence rate by using Cauchy mutation instead of non-uniform mutation as the primary immune operator. Two experiments are performed to evaluate the performance of the proposed algorithm in comparison with CSFS using hyperspectral remote sensing imagery acquired by the pushbroom hyperspectral imager (PHI) and the airborne visible/infrared imaging spectrometer (AVIRIS), respectively. Experimental results demonstrate that the FCSFS converges faster than CSFS, hence providing an effective new option for dimensionality reduction of hyperspectral remote sensing imagery.     

台风"杜鹃"影响期间福建大风天气的特点及成因

台风“杜鹃”影响期间，福建大风天气出现的比较早，而且范围比较大．本文通过对天气尺度的环境场、具有代表性的单站资料的对比分析，寻求大风出现的原因．还深入利用MM5模式对台风的气压场、风场、流场等进行数值模拟，分析结果表明台湾地形倒槽的形成对此次大风天气出现有着重要的作用，此外从流场分型的角度进一步分析台湾地形倒槽的形成原因．     

风廓线雷达资料在一次强降水天气过程中的分析应用

针对风廓线雷达资料分辨率高但目前业务应用不广泛的问题,通过对2006年6月29日一次因飑线天气引发的强降水的观测,分析水平风场、垂直气流和大气虚温的变化,以及高低空急流出现的时机并结合地面观测实况,探讨风廓线雷达资料在此类天气过程中的应用.结果表明,水平风场可以清晰地显示高空急流和低空急流出现的时刻和高度,高空急流区和低空急流区不断发展并相互靠近可以预示阵风锋的到来;垂直速度正值突然增大(气流向上为正,向下为负)并迅速变成负值是判断阵风锋过境的一个依据,同时也预示降水即将开始;另外,高空温度递减率的突然加大对降雨的出现也有一定的预示作用.     

五矩二流太阳风等离子体特性的数值研究

本文数值求解了各向同性二流太阳风的五矩方程组,得出了1 R_s-2AU区域内太阳风密度、速度、电子和质子温度、它们的热流通量密度q以及非麦克斯韦分布尾部粒子过剩量ξ随日心距离的变化关系.文中比较了二流太阳风五矩模型、四矩模型（ξ=0）和低阶矩模型（不包括q和ξ二个矩方程）的等离子体特性,着重讨论了量ξ对质子温度及其热流通量的影响.结果表明,包括言的五矩方程可改善T_e/T_p和q_e/q_p的计算值与观测值的符合程度.     

Comparison among four kinds of data of sea surface wind stress in the South China Sea

By using remote sensing (ERS) data, FSU data, COADS data and Hellerman & Rosen-stein objective analysis data to analyze the sea surface wind stress in the South China Sea, it is found that the remote sensing data have higher resolution and more reasonable values. Therefore we suggest that remote sensing data be chosen in the study of climatological features of sea surface wind stress and its seasonal variability in the South China Sea, especially in the study of small and middle scale eddies.     

从遥感影像提取道路特征的方法综述与展望

从遥感影像中提取道路特征已有许多研究,但仍存在许多问题有待解决。本文探讨了道路特征提取的基本思路,对道路特征提取的基本方法和现状进行了较全面的综述,介绍了具有代表性的特征提取方法,并对各个方法的特点进行了分析,最后对道路特征提取的研究前景进行了展望。     

Research on wind-induced responses of a large-scale membrane structure

The wind-induced responses of a large-scale membrane structure, Expo Boulevard, are evaluated in this study. To obtain the wind pressure distribution on the roof surface, a wind tunnel test is performed. A brief analysis of wind pressure on the membrane roof is conducted first and then an analysis of the wind-induced responses of the structure is carried out using a numerical integral method in the time domain. In the process of calculation, the geometrical nonlinearity is taken into account. Results indicate that mean, RSM and peak values of the structure responses increase nonlinearly while the approaching flow velocity increases. Strong nonlinear characteristics are observed in the displacement responses, whereas the responses of nodal stress and cable axial force show minimal nonlinear properties when the membrane structure is subjected to wind loads. Different values of the damping ratio only have a minimal impact on the RSM response of the structure because the background component is a dominant part of the total dynamic response and the resonant component is too small. As the damping ratio increases from 0.02 to 0.05, the RMS responses of vertical displacement, nodal stress and cable axial force decrease by 8.1%, 6.7% and 17.9%, respectively. Since the mean component plays a significant role in the wind-induced response, the values of the gust response factor are not high for Expo Boulevard.     

0917号台风“芭玛”折向东南向移动原因分析

利用NCEP全球数据同化系统(GDAS)1°×1°分析资料,对0917号台风"芭玛"折向东南向移动原因进行了分析。结果表明,中高纬度环流调整是"芭玛"折向东南向移动的根本原因;0918号台风"茉莉"通过改变外围环境场的强度、形状对"芭玛"台风产生间接影响,而两台风之间逆时针互旋以及台风"茉莉"外围强大的环流对台风"芭玛"的直接作用是台风"芭玛"折向东南向移动的关键。对台风"芭玛"经纬向UV最大风速变化诊断分析表明,"芭玛"经纬向UV最大风速中心的转移对"芭玛"折向东南向移动有重要影响,经纬向UV最大风速差的变化对"芭玛"转向具有预示作用,经纬向最大风速差的合成风方向与台风中心未来移动方向有一定的关系。     

自组织双重空间聚类算法的城市扩张结构分析应用

双重空间聚类是能顾及空间连续性和属性相似性的空间数据分析,而常规空间聚类算法难以同时顾及2方面的约束条件。本文采用自组织双重空间聚类算法,对城市扩张结构分析进行了研究。通过改造自组织特征映射的最佳匹配神经元搜索的算法机制,在空间域和属性域进行迭代聚类搜索,实现了自组织双重空间聚类。以武汉市扩张斑块的位置信息和扩张程度指数为输入数据,使用自组织双重空间聚类算法,实现了城市扩张动态结构的识别。自组织双重空间聚类算法使得聚类结果,既在空间域上连续,又在属性域上相近,算法过程具有自组织性,减少了人为影响。     

不变矩特征在医学CT图像中的应用分析

本文以医学CT图像为研究对象,探讨不变矩特征在医学CT图像中的应用。首先实验验证不变矩特征所具有的旋转、平移、缩放等不变特性,其后以基于内容的医学图像检索为例,揭示不变矩特征在基于内容的医学图像检索中的应用原理,最后分析图像不变矩特征的应用场合。实验成果对不变矩特征在医学图像识别、图像匹配、形状分析、图像分割以及基于内容的图像检索等领域的应用具有推动作用。