基于红外光谱分析的塑料识别方法  

Identification of plastic based on infrared spectrum

作　　者：李晓英[1] 邓文怡[1] 武影影[1] 何巍[1] 

机构地区：[1]北京信息科技大学光电信息与通信工程学院,北京100101

出　　处：《北京信息科技大学学报（自然科学版）》2011年第3期49-52,共4页Journal of Beijing Information Science and Technology University

基　　金：北京信息科技大学校基金项目(1125007)

摘　　要：针对废旧塑料分拣问题。利用相关法对塑料产生的红外吸收光谱进行处理,实现对其分类和识别。根据谱图中峰值的位置和谱图形状的不同,将待测塑料红外谱图中峰值位置提取与标准的谱图进行相关计算比较,可识别出待测塑料的种类。实验对94个样品,5种常见塑料做了识别,正确率可达到94.6%。实验结果表明了该方法的可行性和实用性。Aiming at the classification and identification of waste plastic, correlation algorithm is proposed to process the date of infrared absorption spectrum of the waste plastic in the paper, by which the waste plastic can be identified. Because the curve shape and peak value position in infrared spectrogram is different corresponding to the different plastic, the characteristic peak value position must be extracted from the infrared spectrogram. Then the correlation algorithm is performed between the characteristic peak value position of unknown waste plastic and the same position in spectrogram of the reference plastic. The waste plastic can be identified by the correlation coefficient. 94 specimens of 5 kinds of common plastics are selected in experiment, in which 94. 6% specimens are identified correctly. Experiment results show that this method is practical and reasonable.

关 键 词：塑料识别 红外光谱 相关法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]