论文部分内容阅读
模糊技术和神经网络在模式识别领域中已有了广泛应用，二者有着各自的优势。针对神经网络模式识别中所遇到的问题，为了进一步提高分类器在样本分布不清晰情况下的识别能力，本文提出了两种模糊将机制引入神经网络的方法—输入模糊化方法、隐层模糊化方法，并在此基础上分别构造了模糊神经网络。实验结果表明，模糊神经网络较好地结合了神经网络和模糊技术的优点，取得了比传统神经网络更好的识别结果。
Fuzzy techniques and neural networks have been widely used in the field of pattern recognition, both of which have their own advantages. In order to solve the problems encountered in pattern recognition of neural networks, in order to further improve the recognition ability of classifiers when the sample distribution is not clear, two methods of introducing mechanism into neural network are proposed in this paper-input blurring method, On the basis of this, a fuzzy neural network is constructed respectively. The experimental results show that the fuzzy neural network better combines the advantages of neural networks and fuzzy techniques, and achieves better recognition results than the traditional neural networks.