• DocumentCode
    3562617
  • Title

    Prediction of surface roughness in end milling process by machine vision using neuro fuzzy network

  • Author

    Palani, S. ; Kesavanarayana, Y.

  • Author_Institution
    Dept. of Mech. Eng., Vel Tech Multi Tech Tech Dr. Rangarajan Dr. Sakunthala Eng. Coll., Avadi, India
  • fYear
    2014
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    The roughness of the machined surface is a main concern because product fitness depends on surface roughness. The monitoring of roughness on the workpiece in end milling process by applying machine vision method is presented in this research work. The captured machined image of the work piece is extorted by image processing method. A neuro fuzzy model is used to relate the actual and predicted roughness at various cutting parameters in end milling operation. Measurement of milled surface is monitored with less error when the extracted milled image and milling parameters are fed into the model. The values of the surface roughness predicted by neuro fuzzy model are then verified with experiments and are compared. The prediction accuracy motivating that computer vision technique could be used to various on-line automated manufacturing sectors.
  • Keywords
    computer vision; cutting; feature extraction; fuzzy neural nets; milling; production engineering computing; surface roughness; computer vision technique; cutting parameters; end milling process; image processing method; machine vision method; milled image extraction; milled surface measurement; neuro fuzzy model; neuro fuzzy network; online automated manufacturing sectors; product fitness; roughness monitoring; surface roughness prediction; Computational modeling; Machine vision; Milling; Monitoring; Rough surfaces; Surface roughness; Surface treatment; Computer vision; Milling operation; Neural fuzzy Network; Non-contact inspection; Surface roughness;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Science Engineering and Management Research (ICSEMR), 2014 International Conference on
  • Print_ISBN
    978-1-4799-7614-0
  • Type

    conf

  • DOI
    10.1109/ICSEMR.2014.7043574
  • Filename
    7043574