2^n元数神经网络模型：存贮容量分析

The 2￣n-element Number Neural Network Model: the Analysis of the Storage Capacity

在神经网络中引入了量子力学的Ｄｉｒａｃ符号表示，采用算符形式统一描述了２￣ｎ元数神经网络模型，从理论和数值模拟比较了各模型的存贮容量，该模型可应用于灰度图象或彩色图象的识别．

The Dirac symbol is introduced to express the Hopfield neural network mod-el and the 2￣n-element number neural network model is expressed unifiedly. The storage capacityof the model is analyzed by sigal-to-noise theory and the numerical simulation。The present modelcan be applied to recognize the multi-state grey or color patterns.