基于核密度估计的泥土物证分类方法

Classification Method of Soil Evidence Based on Kernel Density Estimation

利用核密度估计建立了泥土物证元素数据分布不符合正态分布时,估计样本总体概率密度函数的方法,为城市样本间基于似然比检验的分类与比对建立基础。在估计过程中采用积分均方误差衡量窗宽优劣,以积分均方误差最小化准则求得核密度估计的最优窗宽。利用判别分析与似然比检验为特定地域泥土物证的分类提供了可靠的分类分析方法。

The kernel density estimation is used to establish a method for estimating the probability density function of the sample population when the data distribution of the soil evidence does not conform to the normal distribution,and establish the basis for the classification and comparison based on the likelihood ratio test between urban samples.In the fitting process,the mean integral square error is used to measure the window width.Discriminant analysis and likelihood ratio test are used to provide a reliable classification analysis method for the classification of soil evidence in specific regions.