基于KNN算法的机器嗅觉及白酒分类研究
时间:2018-06-08 21:15 来源:毕业论文 作者:毕业论文 点击:次
| 摘要： 本课题的主要讲述了使用搭载了KNN算法的机器嗅觉来识别未知白酒并分类的问题。由于在课题研究的过程中缺少器材和实验样品，因此，本次研究主要是通过仿真，模拟研究白酒的分类。为了能够更加方便地对白酒进行分类，即不需要借助人的帮助，单纯地由机器完成一系列分类的工作，才有了本课题的出现，本次课题研究，主要通过模式识别和机器嗅觉来完成。本课题的研究意义，在于通过搭载了KNN算法的机器嗅觉——电子鼻的投入使用，从而达到代替认为感官识别白酒的目的。本课题所研究及探讨的正是基于机器嗅觉的检测、KNN算法的分析和识别的能力，以达到给酒类饮品更准确地分类。本次研究以就白酒分类为主体，其内容通过电子鼻以及KNN算法来对白酒芳香的特征值进行数据分析和比较。23959 毕业论文关键词： 机器嗅觉；白酒分类；KNN算法；模式识别；算法分析 Liquor classification and machine olfaction based on KNN algorithm Abstract: This topic focuses on the Identification and classification of unknown Liquor through the machine olfaction based on KNN algorithm. Because of the lack of equipment and samples in the process of study, the study is continued by the simulation study of the sample data of wines. In order to be more convenient to categorize liquor, namely do not need with the help of people, a series of classification work, simply done by machines has the appearance of this topic, this topic research, mainly done through pattern recognition and machine the sense of smell. This topic research significance lies in using the machine smell of KNN algorithm - electronic nose put into use, so as to achieve the aim of sensory recognition liquor are instead. This topic research and discussion is based on machine sense of smell detection, analysis and recognition ability of KNN algorithm, for alcoholic drinks in order to achieve more accurate classification. This study with liquor is classified as the main body, the content of liquor flavoring by electronic nose and KNN algorithm of eigenvalue for data analysis and comparison. Keywords: machine olfaction; liquor classification; KNN algorithm; Pattern recognition; algorithm analysis 目录 摘要 i Abstract ii 目录 iii 1 引言 1 1.1 全文内容简介 2 2 技术概念及开发工具介绍 4 2.1 技术概念介绍 4 2.1.1 模式识别 4 2.1.2 机器嗅觉 6 2.1.3 KNN算法 7 2.1.4 电子鼻 8 2.2 开发工具介绍 8 2.2.1 Java语言 8 2.2.2 NetBeans IDE 8.0 10 2.2.3 ACCESS数据库 10 3 白酒分类简介 12 3.1 白酒分类研究的概况 12 3.2 白酒香型 13 3.3 白酒分类设计 14 4 程序详细介绍 16 4.1 总体流程框架图 16 4.2 各模块介绍 18 4.2.1 数据预处理模块 18 4.2.2 输入模块 19 4.2.3 KNN算法模块 21 4.2.4 输出模块 23 (责任编辑：qin) |