均方值

A mean of squared differences image correlation/Kalman filtering motion estimator is analyzed deeply, proving its efficiency and unlbias, and equivalent to maximum likelihood estimator.

对实际中一种常用的均方差图像相关/Kalman滤波位移估值算法进行了深入分析，证明了该算法的有效性和无偏性，并等价于极大似然估值。

In the proposed algorithm the traditional fixed step size iteration is substituted by time-varying step size iteration to improve the accuracy of directional estimation while the minimum mean square deviation is solved to achieve the purpose of increasing the convergence speed of the proposed algorithm.

以一种时变步长迭代方法取代传统的定步长迭代法，旨在提高算法在求取均方差最小值的过程中方向估值的精度，从而达到提高算法收敛速度的目的。

With the background learning process, the mean square error of grey value of each pixel was estimated by sequence images.

通过背景学习过程，利用若干帧连续图像中每点像素灰度值样本，估算对应像素点的灰度值均方差。

You can customize the parameters of the training process, namely: maximum number of cycles, a momentum value, a learning rate, a minimum value of mean square error (in other words "target error").

您可以自定义的培训工艺参数，即：周期最多，声势值，学习率，均方误差的最小值。

The non-reference-signal's adaptive digital filter algorithm of multi-sensor data fusion in the course of adaptive filter can be done without reference signals. It can adjust convergence constant automatically, it can also be recurred with real gradient value. Thus the field to apply the adaptive filter is widened, the convergence speed of the adaptive filter is sped up, the weighted disadjustmend is decreased, so that the last error of mean square approaches its minimum value more.

这章研究的基于多传感器数据融合无参考信号的自适应数字滤波算法其在自适应滤波过程中具有不用参考信号与自动调整收敛因子及使用实际梯度值进行递推的特点，这样不仅可扩大自适应滤波的使用范围，而且还使自适应滤波的收敛速度加快有效地减小权失调，从而使最后的均方误差更接近于其最小值。

The ordinary edge detection methods such as Sobel and Canny operators are immune to noise, which result in cophasal axes discontinuity due to lack of edge information caused by noise.

由于边缘检测中常用的Sobel算子和Canny算子检测同相轴时对噪声免疫差，常因噪声因素引起边缘信息缺失导致同相轴提取不连续，本文采用中值滤波并对Sobel算子进行改进，对地震剖面图进行预处理，结合均方差自动确定阈值并进行异常检测，提取完整同相轴。

The ordinary edge detection methods such as Sobel and Canny operators are immune to noise, which result in cophasal axes discontinuity due to lack of edge information caused by noise. In order to extract cophasal axes more continuous and immune to noise, a novel approach is developed based on edge detection. Firstly, the median filter and improved Sobel method are used to preprocess seismic section images. Secondly, the threshold value is determined automatically based on variance to detect abnormal information. Finally, the complete cophasal axes are extracted after a post-processing.

由于边缘检测中常用的Sobel算子和Canny算子检测同相轴时对噪声免疫差，常因噪声因素引起边缘信息缺失导致同相轴提取不连续，本文采用中值滤波并对Sobel算子进行改进，对地震剖面图进行预处理，结合均方差自动确定阈值并进行异常检测，提取完整同相轴。

The eminent performance of this method is proved by its higher SNR(signal- to- noise rate), smaller MSE(Minimizes the mean squared error) The discal surface of the contaminated insulators is the interesting area of the research.

数值试验证明，与固定阈值法和极大极小阈值法比较，运用该方法去噪后的图像具有更高的信噪比和更小的最小均方误差。

Result 1-3-β-D glucosan was found exceeding the diagnostic criteria (10pg/ml) of the kit in 18 cases of fungal rhinosinusitis; and it was found no more than 10pg/ml in all 5 cases of purely nasal septum deviation patients. One of the 18 cases of fungal rhinosinusitis patients is chronic invasive fungal rhinosinusitis, and its level of 1-3-β-D glucosan reaches to 21.82 pg/ml. We compared the result of the study group to the control groups, and found that there is significant deviation beteen the two groups.

结果 经过实验检测20例真菌性鼻窦炎(19例为真菌球型，1例为慢性侵袭性真菌性鼻窦炎)及5例对照组患者血浆中1-3-β-D葡聚糖的含量，可以发现其中18例真菌性鼻窦炎患者血浆中1-3-β-D葡聚糖的含量大于试剂盒的诊断标准10pg/ml，其中一例慢性侵袭性真菌性鼻窦炎患者血浆中1-3-β-D葡聚糖的含量达到21.82pg/ml，2例真菌性鼻窦炎患者血浆中1-3-β-D葡聚糖的含量小于10pg/ml，而对照组患者血浆中1-3-β-D葡聚糖的含量均小于10pg/ml，病例组与对照组间采用卡方检验测得卡方值为16.071，(P＜0.05)两组间有差异，有统计学意义。

To demodulate the OFDM signals accurately, channel estimation algorithms are adopted. This paper examines the slow-fading estimation algorithms based on block pilot arrangement: LS and LMMSE algorithms, as well as the fast-fading channel estimation algorithms based on comb-type pilot arrangement: linear and transform domain interpolation algorithms.

本文分析了导频分布的两种方案：1块状导频分布，对基于块状导频分布方案的慢衰落信道估计算法：最小平方和线性最小均方误差算法进行了计算机仿真，比较了二者的估计性能，以及各自的优缺点；2梳状导频分布，对基于这种分布方案的快衰落信道估计算法进行了研究，对两类经典算法：线性插值法和基于变换域的插值法分别进行了理论分析和计算机仿真。

But we don't care about Battlegrounds.

但我们并不在乎沙场中的显露。

Ah! don't mention it, the butcher's shop is a horror.

啊！不用提了。提到肉，真是糟透了。