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.pdfVIP

26 1 Vo.l 26 o. 1 2011 2 JournalofBeijing Information Science and Technology University Feb. 2011 : 1674- 6864( 2011)01- 0089- 04 武影影, 邓文怡, 李晓英, 刘桂礼, 王晓飞 ( , 100192) : 将最小二乘支持向量机法( LS_SVM )应用于中红外光 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 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 : 最小二乘支持向量机; 塑料分类; 中红外光谱分析 : TP 391 : A C lassif ication m ethod for the comm on waste p lastics based on the least squares support vector m ach ines WU Yingying, DE G W eny,i LIXiaoying, LIU Guil,iWA G Xiaofei ( Key Laboratory ofModern Measurement ControlTechnology, M inistry ofEducation, Beijing Information Science and Technology University, Beijing 100192, China) Ab stract: The least squares supportvectormachinesmethod (LS_SVM ) is used in them idinfra red spectroscopy analysis ofplastics to establish a new classification method of common waste plastics. According to different absorption peak positions of different plastic types in them idinfrared bands, the infrared spectraldata are processed by LS_SVM method. The result is compared w ith the results of the o verall correlation method and the system comparison method. Expermi entalresults show that classification accuracy rate of LS_SVM m ethod is 0. 92, which is higher than those of the other wt o metboces. And the effect of LS_SVM m ethod solving the problem of small samples is remarkable. So it can be applied into the classification of common waste plastics. Key w ords: least squares supportvectormachines; classificationmethod of commonwasteplastics; them idinfrare