一种基于最小二乘支持向量机法的常见废弃塑料分类方法  

Classification method for the common waste plastics based on the least squares support vector machines

作　　者：武影影[1] 邓文怡[1] 李晓英[1] 刘桂礼[1] 王晓飞[1] 

机构地区：[1]北京信息科技大学现代测控技术教育部重点实验室,北京100192

出　　处：《北京信息科技大学学报（自然科学版）》2011年第1期89-92,共4页Journal of Beijing Information Science and Technology University

基　　金：现代测控技术教育部重点实验室资助项目

摘　　要：将最小二乘支持向量机法(LS_SVM)应用于中红外光谱分析,建立一种新的对常见废弃塑料进行分类的方法。依据不同类别的塑料在红外波段具有不同的特征吸收峰,采用LS_SVM方法对塑料的中红外光谱数据进行处理,并与全局相关法、系统比较法进行比较。实验结果表明,LS_SVM进行分类的正确率为0.92。与全局相关法和系统比较法相比,LS_SVM分类正确率较高,其解决小样本问题效果显著,可应用于常见废弃塑料的分类。The least squares support vector machines method （LS_SVM） is used in the mid-infrared spectroscopy analysis of plastics to establish a new classification method of common waste plastics. According to different absorption peak positions of different plastic types in the mid-infrared bands, the infrared spectral data are processed by LS_SVM method. The result is compared with the results of the overall correlation method and the system comparison method. Experimental results show that classification accuracy rate of LS_SVM method is 0.92,which is higher than those of the other two metboces. And the effect of LS_SVM method solving the problem of small samples is remarkable. So it can be applied into the classification of common waste plastics.

关 键 词：最小二乘支持向量机 塑料分类 中红外光谱分析 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]