...足小渡函数的容许性条件I 9(f)dt. J一∞ 利用文献[3]的实验数据,建立神经网络浓度模型.采用三层多分辨率小波网络(MRNN),输入层 为30个变量,输出层为1个变量,有一个隐层.对实验数据进行处理,从波长250nm到波长430nm,每 隔6nm波长墩一个光谱的强...
基于20个网页-相关网页
文章提出一种基于多分辨率学习的正交基小波神经网络结构的设计方法,网络权值的学习采用阻尼递推最小二乘算法。
A designing method of wavelet neural network structure based on multiresolution learning is put forward, and the studies of network weights adopt damped least squares.
良好的局部放大特性和多分辨率学习特性使得小波神经网络比神经网络有更强的自适应能力、更快的收敛速度和更高的预报精度。
The good local amplification and multi-resolution characteristic make the wavelet network have strong adaptive capacity, fast convergence speed and high precision of prediction.
提出一种基于二进小波变换与多层分组神经网络的自由手写体数字的多分辨率识别算法。
In this paper, a new scheme of multiresolution recognition of unconstrained handwritten numerals based on dyadic wavelet transform and multilayer cluster neural network is presented.
应用推荐