基于磨粒颜色特征的识别方法研究

英文篇名：Research on Wear Debris Color Recognition with Factor Analysis
中文刊名：润滑与密封
英文刊名：Lubrication Engineering
作者：殷勇辉
英文作者：Yin Yonghui(School of Mechanical and Power Engineering ; East China University of Science and Technology ; Shanghai 200237 ; China)
中文关键词：因子分析法 ; 磨粒识别 ; 多维尺度分析
英文关键词：factor analysis method ; wear debris recognition ; multi-scale analysis
出版日期：2006-06-15
机构：华东理工大学机械与动力工程学院上海200237
年：2006
期：06
出版单位：润滑与密封

摘要

磨粒监测是机械设备故障诊断的有效方法,其中磨粒的识别是目前研究的重点和难点,磨粒的颜色参数是识别磨粒的重要特征。基于对因子分析方法的介绍,将彩色磨粒图像提取的颜色变量减少为5个主要特征,简化了描述参数。结果表明,选择适合的HSV获得不同磨粒的空间分布,设置分割线,就可用于不同颜色的磨粒分类和识别。
Wear debris recognition is important and difficulty in wear monitoring of mechanical facult diagnosis.A new method based on color paramete analysis was presented,the color variable was reduced to five main parameters through factor analysis method,so the different wear debris was classed and recognized by setting partition line in HSV color space.

引文

[1]杨维权,等.多元统计分析[M].北京:高等教育出版社,1989.
    [2]王碧泉,陈祖阴.模式识别理论方法和应用[M].北京:地震出版社,1989.
    [3]曾黄鳞,虞阙邦,曾谦.基于主成分分析的特征简化[J].四川轻化工学院学报,1999,12(1).Zen Huanglin,Yu Quebang,Zen Q ian.System Feature Re-duction on Principal Component Analysis[J].Journal of Si-chuan Institute of Light Industry and Chem ical Technology,1999,12(1).
    [4]Myshkin N K,et al.The use of color in wear debris[J].W ear,2001,251:1218-1226.
    [5]T D Pham,H Yan.Color image segmentation using fuzzy inte-gral and mountain clustering[J].Fuzzy Sets and Systems,1999,107:121-130.