摘要
依据生物学结论,从结构上对一种在模式识别过程中具有选择注意力(Selective Attention)功能的视觉神经网络模型作了适当的改进,既保留了原模型的识别、联想记忆及选择注意力等功能,又在结构原理上与生物神经系统更吻合,而且还减少了计算复杂性.
A neural network model of the mechanism of selective attention in visual pattern recognition is improved on the basis of neurobiological theory. Its complexity of computation is reduced. In addition to retaining the properities of recognition, associative memory and selective attention, the improved model matches the animal's nervous system much better.
出处
《复旦学报（自然科学版）》
CAS
CSCD
北大核心
1992年第3期263-269,共7页
Journal of Fudan University：Natural Science
关键词
视觉神经网络
模式识别
神经感知机
postsynaptic potential
neocognitron
analog thresliold element
