差分拉曼光谱结合聚类分析检验运动鞋鞋底

为建立一种快速、准确、无损的区分检验运动鞋鞋底的方法，采用差分拉曼光谱仪结合聚类分析方法对40个不同品牌，不同系列的白色运动鞋鞋底样品进行了检验研究。结果表明，根据鞋底样品的主要成分不同可将样品分为5类；同一类样品可依据添加的主要填料成分不同进行分组；同一组样品可依据特征峰的相对峰高比不同进行区分；利用聚类分析方法处理样品数据，当并类距离为2时，样品被分为5类，分类结果较为理想。该方法操作简便快速、结果准确可靠且无损检材，可为公安机关实际办案提供帮助。

Discrimination and Classification of Sneaker Sole by Combination of Differential Raman Spectroscopy and Cluster Analysis

In order to establish a fast, accurate and non?destructive method for distinguishing the sneaker soles, a differential Raman spectrometer combined with cluster analysis was used to examine the sole samples obtained from 40 different brands and different series of white sports shoes. The experimental results indicated that these samples could be divided into five categories according to the different main components of the sole samples. The same types of samples could be grouped according to the different main filler components added in the sole materials, and the same group of samples could be distinguished based on different relative peak height ratios of characteristic peaks. The data from these samples was processed by a cluster analysis method. When the parallel distance was 2, the samples were divided into five categories with an ideal classification result. This method was found to be simple and quick for operation, the obtained results were accurate and reliable, and there was no destruction in the inspected materials. The method developed by this work can provide help for police offices to handle the criminal cases.