值分布

We first denoise the SAR image by Lee filter.Next,the maximum likelihood classifier based on different distribution was used for SAR and SPOT images(Based on Wishart distribution and multivariate Gaussian distribution respectively),to extract the conditional probability of each pixel for each class.

首先使用李式滤波器滤除全偏极雷达数据噪声，接下来同样是使用采用最大似然法来分类融合影像，(不同的在于全偏极雷达影像使用Wishart几率分布，在光学影像采用multivariate Gaussian几率分布)将每个类别中每个像元属于某个类别的几率值计算出来，再利用Dempster-Shafer evidence theory来结合这些类别的机率值。

Ten contamination features are extracted from the contaminated insulator infrared image,viz.the highest , the lowest , the mean temperature of the background, the temperature variance of the background ,the highest ,the lowest ,the mean temperature of the contaminated insulator discal surface , the temperature variance of the insulator discal surface ,the maximum and the mean temperature increase of the discal surface contrasting to the mean temperature of the background .

本文研究从分割后的污秽绝缘子盘面图像中提取了10维原始污秽特征，即：绝缘子红外热像背景的最高温度、最低温度、平均温度、背景温度分布的方差值、盘面图像的最高温度、最低温度、平均温度、盘面温度分布的方差值、盘面温度的最大温升、盘面温度的平均温升。

Monorchis belonging to Orchidaceae had a narrower one, indicating that rare and endangered Orchidaceae plants had a less distribution in Beijing mountainous region, and only distributed in several types of forest vegetation communities. Due to the lack of suitable habits, these herbage species were in dire need of prior conservation.

结果表明：野大豆的生态位宽度值较大，为0.297；大花杓兰、沼兰和角盘兰同属于兰科植物，生态位宽度值较小，说明兰科植物在北京地区分布较少，并且仅在几种森林群落类型中有分布，适宜生存的生境较少，急需优先保护。

Among tropical elements, the genera of Pantropical distribution contribute from 13.64% to 46% of the total genera in regional floras with the highest proportion in dry and hot valleys in western and southeastern Yunnan, and the genera of Tropical Asian distribution contribute from 3.33% to 42.26% with the highest proportion in southern Yunnan.

在热带成分中，全热带分布属占13.64%~46%，居热带成分之首，其最大值出现在云南怒江和元江的干热河谷；热带亚洲分布属占3.33%~42.26%，最大值出现在云南南部。

Combination with the practice of domestic aluminium production industry, the 3—D current distribution, magnetic flux density, electro—magnetic force, melt flow velocity field of three diffrent cells, that is, 80KA Soderberg cell with vertical stub, 155KA prebake cell, 160KA prebake cell were computed and the related computer—simulated pictures were drawn by use of the softwares we made.

本文利用电流分布、磁场、电磁力场和流场的计算软件，结合我国生产实际计算了80KA上插自焙槽，155KA预焙槽，160KA预焙槽三种不同槽子的三维电流分布，槽内磁感应强度，熔体中电磁力场，熔体流速场，绘制了相应的计算机仿真图，为了增加计算软件的可靠性、用铁棒法测定了80KA上插自焙槽，155KA预焙槽的铝液流速，测定值与计算值符合较好。

The average percent value of G C (49.3%) was less than that of A T (50.7%), there was a bias of the content of G, T in the third codon; The number of transition G-A and T-C were 22, 15 respectively, which were higher than that of transition A-G and C-T, the number of tranversion C-A, T-G is 2 respectively, and other tranversion patterns didnt occur; the probability of transition was higher than that of tranversion , Ts/Tv = 9.5-19, there was the highest number of transition in the third codon. The gamma parameter a of the lst,nd and 3rd codon positions were 0.00572,0.01237 and 1.05239 respectively, they showed that there was adifferent substitution rate at different codon position. Frequences of synonymous codon usage were relatively biased. The average rates of synonymous and nonsynonymous substitution were 0.0787, 0.0011 respectively, there was a significant difference between dS and dN (Z = 4.713, p.01), and the low ratio( CD = 0.0284 .3) of dN/dS ratio impled that there were selective constraints against the nonsynonymous sites in cyt-b gene, The distribution of nonsynonymous codon substitution pattern related to Grantharm distance indicated that the purified selection at 2nd codon positions was more intensive than that at 1st codon positions. The phylogenetic trees supported the view of the double origin of Chinese goose, which means that domestic Chinese goose was derived from Anser cygnoides and domestic European goose, and Yili was derived from Anser anser.

碱基含量分析可知，序列的G C含量(49.3％)＜A T含量(50.7％)，密码子第三位点的G、T含量都有较强的偏倚性；序列间G→A和T→C的转换数(22次和15次)高于A→G和C→T的转换数(10次和9次)，C→A、T→G颠换数均为2次，其余颠换模式均未发生；转换数明显高于颠换数，Ts／Tv=9.5～19，密码子第三位点的转换数最高，呈现了相当强烈的转换偏倚性；密码子第一、二和三位点的gamma分布参数α值分别为0.00572、0.01237和1.05239，表明密码子第一位点的替换速率变异最大，第二位点次之，第三位点的替换速率变异相对较小；编码同一氨基酸的同义密码子并非随机使用，表现出一定程度的使用偏倚性；同义替换速率和非同义替换速率分别为0.0787和0.0011，dS与dN值间的差异极显著(Z=4.713，p＜0.01)，而ω=0.0284，明显小于0.3，表明雁属鹅细胞色素b基因序列经历了中度净化选择作用；单步非同义替换(Sing-step nonsymonymous codon substitution，SSNCS)分布模式与Grantharm距离之间的关系说明密码子的三个位点所受的净化选择强度不同；构建的最大简约树与邻接树拓扑结构一致，支持中国家鹅的双起源学说，即除伊犁鹅外的其它中国鹅品种起源于鸿雁，伊犁鹅和欧洲的郎德鹅、莱茵鹅起源于灰雁。

For understanding the workability and precision of these prediction equations, at first this study not only simulates the MHs of Taipei, Taichung, Kaoshung cities and Kao-ping area but also compares these results with meteorological models of CLAMET and TPAQM. And the comparisons show the high correlation (up to 0.8) between them. When the results of MH prediction equations model are compared with TPAQM, they get higher or lower values than TPAQM. The reasons are that Holzworth method can not concern about the warm and cold advections effect and the upper air data used do not get enough finer space-spread. About the results spread on the domain of my model, they show more different MH fields between the land and the sea around and get more sharply change of the land MHs on daytime than TPAQM.

为了解此预测式之可行性及准确度，本研究针对台北、台中、高雄及高屏地区作混合层高度值的模拟，并将模拟结果与实际观测值、CALMET气象模组及光化学轨迹模式输出结果作比对，结果显示均有0.80以上之高相关性；但与TPAQM计算结果间有明显的高估或低估现象，此为Holzworth method无法考量温度平流效应对混合层发展之影响及探空资料空间解析度不足所致；而在空间分布上，本模式所计算的混合层高度场相较於TPAQM在海陆分布上有较明显的分野，且更能够显示出内陆地区混合层高度的日变化。

The signals during transformer impulse test are considered to be evolving in time, also is non-stationary signals, so we can design a software algorithm for noise reduction based on discrete Gabor transform, and the basic principle is: The Gabor coefficients of Effective signals during transformer impulse test is centralized in the time frequency range, and the peak-to-peak value is big, whereas the noise Gabor coefficient distribute in the entire time frequency range equally, the peak-to-peak value is small, we can realize the desired signal and the noise separated through frequency mask function.

考虑到冲击试验中的信号都是时变信号，也就是非平稳信号，因此设计了一种基于离散GABOR变换的软件滤波算法对冲击信号进行滤波，其滤波理论依据是：有效信号的Gabor系数在时频域分布比较集中，幅值比较大，噪声的Gabor系数均匀分布于整个时频域，幅值较小，通过构造时频掩模函数来实现有效信号与噪声的分离。

It's assumed that the spatial distribution of geographic features is the result of the interaction of various factors and can be modeled as a complex non linear function.

本文认定地理要素的空间分布可以用一复杂的非线性函数模拟，该函数是由多种因素综合作用的结果，即地理要素的值是这些因素的函数，如果以各因素为输入、对应地理要素值为期望输出，对网络进行训练可对地理要素的空间分布进行模拟。

Furthermore, the weight value of traffic density is taken into account in this paper, which is introduced in traffic forecasting based on simulated annealing.

并且提出了业务密度权值的计算方法，从而在用模拟退火算法进行业务分布预测的基础上，给出了利用业务密度权值进行业务分布预测的算法。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。