铁谱磨粒图像色彩自动识别的一种新方法  

A new method of automated recognition on image's colors of ferrographic wear particles

作　　者：姜雪丽[1] 裴峻峰[1] 朱恩全[2] 

机构地区：[1]江苏工业学院机械系 [2]江苏油田分公司

出　　处：《石油机械》2008年第1期45-48,84,共4页China Petroleum Machinery

基　　金：江苏省油气井口装备工程中心建设项目(BM2004007);江苏省科技成果转化资金项目“基于油样分析技术的往复泵状态监测系统研究”(BA2005041)

摘　　要：磨粒色彩识别的准确率极大地影响着诊断的准确率,而铁谱图像自动识别方法是用Matlab图像处理工具箱,将灰色理论与色度学相结合,综合考虑磨粒特征之后得出的新方法。图像处理工具箱可完成图像组合、空间变换、图像分析、图像增强等工作。应用实例表明,采用RGB色度理论对铁谱磨粒色彩自动识别能筒化色彩识别过程,减少经验知识和人为操作的影响,RGB值的界定是影响色彩识别准确率的关键因素。The development of ferrography technology makes identifying for colors of ferrographic wear particles more and more important in wear debris recognition in ferrography technology. The automatic and accurate identifying for colors of images of ferrographic wear particles will influence the application of ferrography technology. Based on colorimetry theory, this paper applies the chromaticity description method in RGB tricolor theory to automatic identifying for colors of images of ferrographic wear particles. And MATLAB's image processing toolbox is used to compile program to process image automatically and recognize colors. It has realized the intelligent analysis for colors of wear particles. This method avoids the disadvantage factor which exists in the artificial analysis, and improves the accurate rate of automatic identifying of wear particles.

关 键 词：铁谱磨粒 灰色理论 色彩识别 图像处理 色度学 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TS212.7[自动化与计算机技术—计算机科学与技术]