神经网络

The gray images filtering by these skinlike regions are then been fed into a welltrained neural network to verify if they are faces.

最后把经过肤色区域滤波的灰度图像作为特征送入训练好的神经网络，由神经网络判别该图像是否一张人脸。

Then, according to phase space reconstruct technique from one-dimensional and multi-dimensional time series, the quantitative criterion and rule of system chaos which combine neural network, analyses, computations and sort are conducted on EEG signals of five kinds of human consciousness activities. The author finds that statistic results of the determinism computation in time series indicate that chaos characteristic may lie in human consciousness activities, and central tendency measure is consistent with phase graph, so it may become one division way of attractor; approximate entropy of different subjects exists discrepancy; correlation dimension and Lyapunov exponent of different consciousness activities indicate that attractors of human activities are all fractional; nonlinear quantitative criterion and rule, which unites neural network, can distinguish different styles of consciousness activity in principle, and the result of sortation takes on that consciousness activity type of arithmetic is distinguished more easily than that of abstract.

采用一维与多维时间序列相空间重构技术和系统混沌的定量判据准则，结合神经网络，对五种思维作业方式的脑电信号进行了分析、计算与分类，发现确定性计算的统计结果表明人类的意识活动中可能蕴含混沌特性，中心趋向测量(Central Tendency Measure，CTM)与相图吻合较好，可作为EEG吸引子的区分方法之一；功率谱分析反映出单个受试者的各种意识形态的差异很小，但是不同意识活动谱中的活跃频段还是略有差异的；个体之间的近似熵存在差异，同种状态下近似熵大的人可能具有更好的创新性；关联维数和Lyapunov指数的计算结果表明大脑的运动落在具有分维的奇怪吸引子上；非线性量化方法与神经网络相结合，可对人脑思维活动进行较好的分类，并且从分类结果来看数理计算类思维活动较抽象类思维活动有着更好的区分度。

The other is to use a linear neural network for feature extraction, which is simultaneously trained with the subneural networks dealing with different situations.

另一种方法是应用线性神经网络进行特征抽取，并与各子神经网络进行统一训练。

This article introduced the fuzzy logic and the artificial neural networks elementary theory, then unifies both carries on the superiority supplementarily, has constructed the fault diagnosis method which one kind of artificial neural networks and the fuzzy reasoning technology unify.

本文介绍了模糊逻辑和人工神经网络的基本理论，然后将两者结合起来进行优势互补，构建了一种人工神经网络和模糊推理技术相结合的故障诊断方法。

The performance of synergetic neural network is discussed.

摘 要：本文对一类全新的神经网络——协同式神经网络的性能进行了研究。

Up to now, the whole operational speed of existent CNN chips has reached Tera magnitude, which can great improve the speed and real time of video processing.

细胞神经网络在视频运动对象分割中的应用研究研究细胞神经网络用于视频运动对象分割的原因在于其并行高速的特点和适于图象处理的网络结构。

There have some difficulties in the applications of Neural Networks, the convergence for large scale problem is very slowly, the maturity of samples often can not be guaranteed, the calculating results usually trustless because the bad capability of extrapolation of Neural Networks.

人工神经网络模型在实际应用中存在一些困难，包括对于较大规模的问题学习过程缓慢，样本的完备性难以保证，由于人工神经网络的外推计算能力差而导致计算结果的可靠性降低等。

Because the traditional PID exists the shortcomings of unadjustment of parameters and inferior real time nature,the Genetic BP neural networks PID control algolithm is proposed in the dissertation. The Genetic algorithm and BP neural networks are introduced to optimize three control parameters.

由于传统PID控制算法存在控制参数不可调，控制系统实时性差等缺点，本文提出遗传BP神经网络PID控制算法，用遗传算法和BP神经网络对PID控制参数进行在线优化。

A merge-split algorithm based on study property of neural net and sum aggregate is presented. Its merit is that its merge is disconnected a little and freedom satisfies the adaptability, and it is realized during learning rather than stipulating in advance. Furthermore, the method of fitting edge of uncontinuous image with neural net is given. The method can carry on the fit at self-learning, using its learning function fully under the circumstances without the mathematical function form of edge.

提出了一种基于神经网络学习特性和集合论的边缘分段算法(Merge-Split)，优点在于它的分段间断点、自由度满足自适应性，能在学习过程中实现而不必事先规定；进一步提出神经网络拟合不连续的图像边缘的方法，在无边缘数学函数形式的情况下，充分利用其学习函数的任意逼近和自学习性进行拟合，实践证明该方法可以克服以往边缘拟合方法的不足。

So during the collection of fault sample data and checking data, the paper sets up the mathematics model of control system first. We cause all kinds of fault in the control system, then collects a variety of fault data. After unitarily dealing, the fault data server as the sample data of training neural network. At the same time, according to the collection of checking data, it examines whether the trained neural network have the function of fault diagnosis.

在故障样本数据及检验数据的采集过程中，首先建立了控制系统的数学模型，人为地让控制系统发生各种故障，从而采集到各种故障数据，经过归一化处理后，作为训练神经网络的样本数据；同时，也采集检验数据，用以检验训练出的神经网络是否能够起到故障诊断的作用。

We have no common name for a mime of Sophron or Xenarchus and a Socratic Conversation; and we should still be without one even if the imitation in the two instances were in trimeters or elegiacs or some other kind of verse--though it is the way with people to tack on 'poet' to the name of a metre, and talk of elegiac-poets and epic-poets, thinking that they call them poets not by reason of the imitative nature of their work, but indiscriminately by reason of the metre they write in.

索夫农 、森那库斯和苏格拉底式的对话采用的模仿没有一个公共的名称；三音步诗、挽歌体或其他类型的诗的模仿也没有——人们把&诗人&这一名词和格律名称结合到一起，称之为挽歌体诗人或者史诗诗人，他们被称为诗人，似乎只是因为遵守格律写作，而非他们作品的模仿本质。

The relationship between communicative competence and grammar teaching should be that of the ends and the means.

交际能力和语法的关系应该是目标与途径的关系。