节点文献

磨损磨粒铁谱分析中图像预处理技术研究

Study of image preprocessing in the ferrographic diagnosis of wear debris

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 詹松郑寿森胡献国

【Author】 ZHAN Song,ZHENG Shou-sen,HU Xian-guo (School of Machinery and Automobile Engineering, Hefei University of Technology, Hefei 230009, China)

【机构】 合肥工业大学机械与汽车工程学院合肥工业大学机械与汽车工程学院 安徽合肥 230009安徽合肥 230009安徽合肥 230009

【摘要】 利用计算机图像处理技术实现铁谱诊断自动化是铁谱技术发展的目标,故铁谱原始图像需要进行预处理,如平滑、滤波、二值化及颗粒筛选等,能够有效减少噪声污染和简化数据处理。文章探讨了铁谱图像的预处理技术应用,采用空白窗卷积方法,对实际铁谱图像进行处理,从而验证这些方法的有效性。

【Abstract】 How to realize the automatization of ferrographic diagnosis by PC image processing is one of the research targets in the development of ferrographic technique,and in order to reduce noise and simplify data processing, the preprocessing of original images is the necessary step,which directly affects the following processing. In this paper,some methods of processing original ferrographic images such as smoothing,filtering,and threshold method are discussed,and a new method-debris selection is developed,which is based on the iterative convolution in the whole map with a blank, size variable window. These methods are applied to the processing of a real ferrographic image and some satisfactory results are obtained,which verifies the validity of the methods.

【基金】 安徽省自然科学基金资助项目(00046103)
  • 【文献出处】 合肥工业大学学报(自然科学版) ,Journal of Hefei University of Technology(Natural Science) , 编辑部邮箱 ,2004年01期
  • 【分类号】TH117.1
  • 【被引频次】16
  • 【下载频次】217
节点文献中: 

本文链接的文献网络图示:

本文的引文网络