| 基于近红外光谱的汽油牌号快速识别 |
| |
| 引用本文： | 张其可,戴连奎.基于近红外光谱的汽油牌号快速识别[J].化工自动化及仪表,2005,32(4):53-57. |
| |
| 作者姓名： | 张其可 戴连奎 |
| |
| 作者单位： | 工业控制技术国家重点实验室,浙江大学,智能系统与决策研究所,浙江杭州310027 |
| |
| 基金项目： | 国家高技术研究发展计划项目(2003AA412110) |
| |
| 摘 要： | 提出基于近红外(NIR)光谱的汽油牌号快速识别算法,主要包括预处理、特征提取和分类建模几部分,比较了各种分类方法的识别能力。实验结果表明:采用主元分析(PCA)提取特征进行模式识别的性能普遍优于直接在光谱波长域的方法,通过选择合适的PCA主元可以获得满意的分类效果,可用于汽油产品的牌号快速识别。
|
| 关 键 词： | 主元分析 近红外光谱 汽油牌号 模式识别 分类 |
| 文章编号： | 1000-3932（2005）（04）-0053-05 |
| 收稿时间： | 2005-06-13 |
| 修稿时间： | 2005年6月13日 |
Fast Recognition of Gasoline Brands Based on Near-infrared Spectroscopy |
| |
| ZHANG Qi-ke,DAI Lian-kui.Fast Recognition of Gasoline Brands Based on Near-infrared Spectroscopy[J].Control and Instruments In Chemical Industry,2005,32(4):53-57. |
| |
| Authors: | ZHANG Qi-ke DAI Lian-kui |
| |
| Abstract: | The fast recognition algorithms of gasoline brands based on the near-infrared (NIR) spectroscopy is presented.It includes pretreatment,feature extraction and classification modeling.The recognizing capabilities of different kinds of classification methods are compared.Experiment results show that capabilities of recognition methods using principal component analysis(PCA) to extract features are generally better than those used directly in spectrum wavelength domain.The satisfied classification performance can be obtained through selecting proper PCA factors,so it is used in fast recognition of brands of gasoline products. |
| |
| Keywords: | principal component analysis near-infrared spectroscopy gasoline brands pattern recognition classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
