本实验的目的是学习Parzen窗估计和k最近邻估计方法。在之前的模式识别研究中，我们假设概率密度函数的参数形式已知，即判别函数J(.)的参数是已知的。本节使用非参数化的方法来处理任意形式的概率分布而不必事先考虑概率密度的参数形式。在模式识别中有躲在令人感兴趣的非参数化方法，Parzen窗估计和k最近邻估计就是两种经典的估计法。(The purpose of this experiment is to study the Parzen window estimation and the k nearest neighbor estimation method. In previous pattern recognition studies, we assume that the parameter form of the probability density function is known, that is, the parameter of the discriminant function J (...) is known. This section uses nonparametric methods to handle any form of probability distribution without having to consider the parameter form of probability density. In pattern recognition, there are hidden nonparametric methods that are interesting. Parzen window estimation and K nearest neighbor estimation are two classical estimation methods.)