Share Email Print

Optical Engineering

Surface roughness extraction based on Markov random field model in wavelet feature domain
Author(s): Lei Yang; Li-qiao Lei
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Based on the computer texture analysis method, a new noncontact surface roughness measurement technique is proposed. The method is inspired by the nonredundant directional selectivity and highly discriminative nature of the wavelet representation and the capability of the Markov random field (MRF) model to capture statistical regularities. Surface roughness information contained in the texture features may be extracted based on an MRF stochastic model of textures in the wavelet feature domain. The model captures significant intrascale and interscale statistical dependencies between wavelet coefficients. To investigate the relationship between the texture features and surface roughness Ra, a simple research setup, which consists of a charge-coupled diode camera without a lens and a diode laser, was established, and the laser speckle texture patterns are acquired from some standard grinding surfaces. The research results have illustrated that surface roughness Ra has a good monotonic relationship with the texture features of the laser speckle pattern. If this measuring system is calibrated with the surface standard samples roughness beforehand, the surface roughness actual value Ra can be deduced in the case of the same material surfaces ground at the same manufacture conditions.

Paper Details

Date Published: 9 September 2014
PDF: 7 pages
Opt. Eng. 53(12) 122414 doi: 10.1117/1.OE.53.12.122414
Published in: Optical Engineering Volume 53, Issue 12
Show Author Affiliations
Lei Yang, Hefei Univ. of Technology (China)
Li-qiao Lei, Hefei Univ. of Technology (China)

© SPIE. Terms of Use
Back to Top