摘要
分析了磨粒的自动识别研究在铁谱技术发展和应用中的重要性。指出铁谱磨粒图像分割是磨粒自动识别的重要环节,铁谱图像所包含的彩色信息对磨粒识别和磨损形式分析非常重要。将最小二乘支持向量机LS-SVM用于彩色磨粒图像的分割,利用模糊c均值选取训练样本,把磨粒图像的R,G,B分量,同一像素点各分量的方差以及像素点邻域像素均值作为特征,对彩色磨粒图像分割,取得了很好的效果,为磨粒自动识别提供了更有效的信息。
The importance of wear debris intelligent recognition for the development and application of ferrography was analyzed. It was pointed out that the segmentation method of ferrograph debris image was important for wear debris intelligent recognition because the color information contained in ferrograph image was important to particle identification and wear form analysis. The least square support vector machine (LS-SVM) was used in the segmentation of color image. The training data was acquired with the help of fuzzy c-means (FCM) algorithm, and then the R, G, B components, the component variance of same pixel point and the average of neighborhood pixels were taken as the identification characteristics. Finally the color image of ferrography was segmented with a good effect, which brought useful information for the intelligent recognition of wear debris.
出处
《车用发动机》
北大核心
2009年第1期81-83,共3页
Vehicle Engine
基金
国家自然科学基金(50705097)
中国人民解放军军械工程学院科研基金资助项目(YJJXM08009)
关键词
铁谱
磨粒
图像分割
最小二乘支持向量机
模糊C均值
ferrography
wear debris
image segmentation
least square support vector machine(LS-SVM)
fuzzy c-means (FCM)