最小化

When seismic wavelet is in non-minimum phase , we can make it have smaller phase by doing exponential weighting . As for nonwhite noise problem of reflection coefficient , we may employ logarithmic power spectrum to estimate the autocorrelation function of seismic wavelet , then obtain seismic wavelet and deconvolution opera .

当地震子波为非最小相位时，可利用指数加权方法使之小相位化；对于反射系数非白噪声问题，本文提出利用对数功率谱来估算地震子波的自相关函数进而求取地震子波以及反褶积算子的方法；对于理想的期望子波，本文采用"宽带雷克子波"作为期望输出子波。

In Chapter three, by constructing a special partitioned matrix and studying the fast triangular factorization of its inverse matrix, we respectively give fast algorithms for the minimal norm least squares solutions of linear equations which coefficient matrices are n×m Toeplitz type matrices and Vandermonde-Loewner type matrices with full column rank.

第三章通过构造特殊分块矩阵并研究其逆矩阵的快速三角分解，分别给出了求以秩为m的n×m阶Toeplitz型矩阵和Vandermonde-Loewner型矩阵为系数阵的线性方程组极小范数最小二乘解的快速算法，并通过算例将新算法与已有的法方程组的方法和正交化方法作了比较。

Fast algorithm of minimal norm least square solution for Cauchy linear system;2. For the m×n Cauchy-type matrix with full column rank,the fast algorithm of the minimal norm least square solution to the linear system Cx=b was obtained by construction of a special block matrix and study of the triangular decomposition.

对于秩为n的m×n阶Cauchy型矩阵C，通过构造特殊分块矩阵并研究其三角分解，进而得到了线性方程组C x=b的极小范数最小二乘解的快速算法，所需运算量为O+O(n2)，而通常构造法方程组的方法所需运算量为O(m n2)+O(n3)，用正交化法虽然避免了构造法方程组，但所需的运算量更大些。

The said process can regenerate catalyst continuously or intermittently to maintain the catalytic activity of the alkylating catalyst for long period and has lowest loss of produce in amount and octane number while eliminating the thiofuran sulfur impurity in material.

采用该方法可对催化剂连续或间歇再生，克服烷基化催化剂一般寿命短的缺陷，并且可以长时间维持烷基化催化剂的活性；在脱除原料中噻吩类杂质硫的同时使得产品量损失最小及产品辛烷值损失最少。

By rotating the X-axis of the measurement coordinates, the straightness data of each generatrix are transformed into a unified 3D coordinate system whose Z-axis superposes the least square mean axis calculated with the data of the multiple generatrices. The computer simulations and experiments prove that the construction by rotating the abscissas can effectively eliminate the distortion of the constructed profile.

通过测量坐标系中X轴的旋转，将各条母线直线度偏差数据统一到同一个空间坐标系中，采用两端面圆的最小二乘轴心的连线作为零件轴线，消除了直接构造所带来的表面变形，并能进一步计算出被测内孔的直径、圆度和轴线直线度误差，使测量系统集成化和多功能化。

This article comminutes the symbols of the formula and records their position by the outline rectangles of symbols, and then normalized these symbol. Then designed Artificial Neural Network that recognizes these symbol.

本文通过计算公式中字符的最小外接矩形，将字符依次分割开并记录其位置信息，然后进行归一化处理，利用本文设计的适合本文中字符识别的识别模块，即二层感知器神经网络识别模块，将归一化后的字符识别出来。

This indicates that these HA have similar compositions, but different structures and properties.Results from analysing E465/E665 are the same as those obtained by analysing molecule structure using colour index coefficientΔ~ K.Compassion study of HA in NaOH and NaHCO_3 solutions show that the absorption peaks shift towards the direction of longer wavelength.

这几个吸收带的变化证明了以紫色潮土HA、石灰性黄壤HA、中性紫色HA、酸性紫色HA、腐殖土HA为序的腐殖酸的芳构化度降低、缩合程度下降的观象，说明腐殖土HA其分子在苯环C骨架的聚合程度、芳构化度和分子复杂程度等上均最小，与紫外光谱分析结果一致。

By spatially and temporally analyzing of the NSD which describes the normative difference of predicted SAD and minimum SAD after motion estimation, as well as the assumption that residual signal is Laplacian distribution, a modulation factor is given to realize early termination, thus a tradeoff between search speed and reconstructed video quality can be achieved.

本文定义了一个新的变量NSD－归一化SAD差值来描述预测所得的SAD值与实际搜索到的最小SAD之间的归一化关系。

To reduce or remove size effect, the size classification elimination method, volume average method and normalization method based on volume were presented and the calibration and prediction results indicate that these three methods could reduce or remove the size effect effectively. The normalization method is more suit to in-line application and obtained better calibration and prediction results with the r of 0.789, 0.858 and 0.947, the RMSEC of 0.596, 0.499 and 0.313, and the RMSEP of 0.606, 0.686 and 0.413 than the results without normalization with the r of 0.716, 0.793 and 0.848, the RMSEC of 0.678, 0.592 and 0.515 and the RMSEP of 0.825, 0.764 and 0.714 for the original spectra, the first derivative spectra and the second derivative spectra with soluble solids content using partial least square method , respectively. 4. Owing to much thick skin of watermelon, contrast experiment was conducted in skin-peeled watermelon and intact watermelon to compare the influence of skin on spectra.

针对西瓜果型大且差异显著的特点，研究果型差异对可见／近红外光谱以及建模与预测结果的影响，并提出三种方法消除或减小这一影响因素，分别是分级消除法、体积平均法和基于体积的正则化方法，其中基于体积的正则化方法，更适合在线生产应用，该方法使果型大小差异明显样品的原始光谱、一阶微分光谱、二阶微分光谱在对可溶性固形物含量检测时，采用偏最小二乘法的建模与预测结果由原来的校正相关系数r分别为0.716、0.793、0.848提高到r分别为0.789、0.858、0.947，均方根校正标准偏差RMSEC由原来的0.678、0.592、0.515分别降为0.596、0.499、0.313，均方根预测标准偏差RMSEP由原来的0.825、0.764、0.714分别降为0.606、0.686、0.413； 4。

The energy which drives the evolution of a curve is constructed based on local image information,the alignment term enhances the speed of curve evolution,and the improved data fitting term which is based on the competition between of the shifted Heaviside functions,makes the curve converge to a better global stationary minimum.

该算法基于图像局部化信息创建驱动曲线演化的能量，引入的配准项提高了曲线的演化速度，基于曲线演化竞争的数据拟合项，使得曲线能更稳定地收敛到一个全局静态最小值，且算法对水平集函数初始化位置不敏感。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。