食品科学 ›› 0, Vol. ›› Issue (): 0-0.

• 基础研究 •    下一篇

基于主成分分析和聚类分析的猕猴桃品质指标综合评价

傅隆生1,宋思哲1,邵玉玲1,李平平2,王海峰1,崔永杰1   

  1. 1. 西北农林科技大学
    2. 青岛中科昊泰新材料科技有限公司
  • 收稿日期:2014-05-27 修回日期:2014-09-05 出版日期:2014-10-15 发布日期:2014-10-17
  • 通讯作者: 崔永杰 E-mail:cuiyongjie@nwsuaf.edu.cn
  • 基金资助:
    猕猴桃采摘机器人果实信息感知与无损采摘方法研究;猕猴桃生产中分级标准制定与自动分级装置研发;果实的自动分级研究

Comprehensive Evaluation of Kiwifruit Quality based on Principal Component and Cluster Analysis

Long-Sheng FU 2, 2, 2, 2,   

  • Received:2014-05-27 Revised:2014-09-05 Online:2014-10-15 Published:2014-10-17

摘要: 为了更准确的评价猕猴桃品质,选取了陕西省眉县的157个海沃德猕猴桃,对果实的单果重、长轴、短轴、厚度、体积、果皮颜色和糖度、酸度、硬度9个分级指标进行了描述统计和相关分析,采用主成分分析法建立综合得分数学模型,对综合得分进一步做K-means聚类分析,最后利用Fisher判别分析法对样品重新进行聚类以验证K-means聚类分析方法的可靠性。结果表明,除体积与单果重间差异不明显外,其余各分级指标之间均存在显著差异;按综合得分将样品聚为3类:优为0.1~1.39,中为-0.44~0.09,差为-1.27~-0.46;判别分析对聚类结果的正确率达到98.72%,所以两者具有较高的一致性。此结果对猕猴桃品质评价有较大指导意义。

关键词: 猕猴桃, 质量评价, 主成分分析, 聚类分析, 判别分析

Abstract: In order to evaluate kiwifruit quality more accurately, 157 kiwifruits were selected from Meixian county, Shaanxi. Nine indices, including weight, long axis, short axis, thickness, volume, color, sugar content, acidity, and firmness were measured and analyzed by descriptive statistics and correlation analysis. After that, principal component analysis was employed to build a composite score mathematical model, then the composite score was analyzed by K-means cluster. Finally, Fisher discriminate analysis was applied to re-cluster fruits for evaluating the K-means cluster analysis. Results showed that grading indices have significant differences except the weight and volume. The 157 samples were classified into three according to their composite score, whose quality and scores were as follows: the best, 0.1~1.39; moderate, -0.44~0.09; poor, -1.27~-0.46. The success rate of discriminant analysis compared to the clustering results reached 98.72%, which meant a good consistency between them. The results had significant on the assessment of kiwifruit quality.

Key words: Kiwifruits, Quality evaluation, Principal Component Analysis, Cluster Analysis, Discriminant Analysis

中图分类号: