The low power chips were used to insure that the whole system power is under 1.5w . a feature extraction and b-p neural networks algorithm embedded in the single chip processor were presented in order to avoid large amount of computation 在硬件设计中,着重于低功耗电路的设计,采用低功率器件和省电管理模式,使整个系统的工作功率低于1.5w。
The chip shapes in the course of cutting processing was carefully analyzed and studied and put forward feature expanded and radbas function neural network algorithms for recognition chip shape, more than experiments program was developed 对于切削加工中的切屑形态进行了分析研究,提出了特征拓展和径向基函数神经网络识别切屑形态的算法,并开发了相应的程序软件。
Anew neural network algorithm for decision level fusion is also presented in this dissertation . the architecture of this network is novel . it is the thresholds, not the conjunction weights, which are modified, when the network is being trained 本文还提出了一种新的神经网络算法用于决策层融合目标识别,该网络结构新颖,网络训练时修改的是门限而不是连接权值。
At the same time attempting and groping for the combination of fuzzy logic control and neural network algorithm, putting forward the fuzzy neural network structure and control algorithm that is in series and validating it in practice 在模糊逻辑控制与神经网络计算相结合方面作了尝试和探索。提出了网络学习型的模糊神经网络控制器以及相应的学习算法,并在实践中进行了验证。
The actuality of the bp neural network algorithm, the wavelet transforms theory and gabor wavelet neural network algorithm and their application of target recognition in the world are introduced . the problems are analyzed 论文包括以下诸方面内容:对目前国际、国内bp神经网络算法、小波变换理论、gabor小波神经网络算法以及在目标识别中的应用现状进行了综述,对存在的问题进行了剖析。
The actuality of the bp neural network algorithm, the wavelet transforms theory and gabor wavelet neural network algorithm and their application of target recognition in the world are introduced . the problems are analyzed 论文包括以下诸方面内容:对目前国际、国内bp神经网络算法、小波变换理论、gabor小波神经网络算法以及在目标识别中的应用现状进行了综述,对存在的问题进行了剖析。
The microsoft neural network algorithm creates classification and regression mining models by constructing a multilayer perceptron network of neurons, providing support for nonlinear models that are too complex to derive by using other algorithms microsoft神经网络算法通过构造神经元的多层感知器网络来创建分类和回归挖掘模型,同时为过于复杂而无法使用其他算法派生的非线性模型提供支持。
And it firstly analyzes some character recognition algorithms used much at present, such as template matching algorithm, characters features algorithm and artificial neural network algorithm . and then it puts forward thoughtway of combination of character recognition algorithm 同样,先对目前常用的字符识别算法(模板匹配法、字符特征法、人工神经网络)进行分析,然后提出组合识别法字符识别思想。
Secondly, considering the complexity of cement rotary kiln system, we build up the models of this system with the integration techniques of artificial neural network and expert system . the improved ebp neural network algorithm is used in this system 第二,根据水泥回转窑系统具有多变量、多回路、非线性、分布复杂和经典数学模型难以建立的特点,运用神经网络与专家系统集成技术对预分解窑系统进行了建模与研究。