Monitoring Surface Roughness of Turning by Using Image Processing Technology


Article Preview

This paper is pointed to the turning surface. A surface roughness prediction model is constructed by polynomial network and image processing technology. After threshold treatment, we can get the gray scale central line arithmetic mean value, Gra and the distance between black and white color, Gf. The prediction model is constructed by Gra value, Gf value and cutting condition. By the experimental results, it can be found that the surface roughness prediction model constructed by this study is very accurately. It can be used to on-line surface roughness monitoring for turning.



Materials Science Forum (Volumes 532-533)

Edited by:

Chengyu Jiang, Geng Liu, Dinghua Zhang and Xipeng Xu




W. S. Lin et al., "Monitoring Surface Roughness of Turning by Using Image Processing Technology", Materials Science Forum, Vols. 532-533, pp. 1164-1167, 2006

Online since:

December 2006




[1] A.R. Barron: Predicted Square Error: A Criterion for Automatic Model Selection, Self-organizing Methods in Modeling: GMDH Type Algorithms (Marcel-Dekker, Inc., New York 1984).

[2] G.J. Montgomery and K.C. Drake: Neurocomputing, Vol. 2 (1991), pp.97-104.

[3] A. Mital and M. Mehta: Int. J. Prod. Res., Vol. 26 (1988) No. 12, pp.1861-1876.