Texture-based segmentation with Gabor filters, wavelet and pyramid decompositions for extracting individual surface features from areal surface topography maps

Senin, Nicola, Leach, Richard K., Pini, Stefano and Blunt, Liam A. (2015) Texture-based segmentation with Gabor filters, wavelet and pyramid decompositions for extracting individual surface features from areal surface topography maps. Measurement Science and Technology, 26 (9). 095405. ISSN 0957-0233

Metadata only available from this repository.

Abstract

Areal topography segmentation plays a fundamental role in those surface metrology applications concerned with the characterisation of individual topography features. Typical scenarios include the dimensional inspection and verification of micro-structured surface features, and the identification and characterisation of localised defects and other random singularities. While morphological segmentation into hills or dales is the only partitioning operation currently endorsed by the ISO specification standards on surface texture metrology, many other approaches are possible, in particular adapted from the literature on digital image segmentation. In this work an original segmentation approach is introduced and discussed, where topography partitioning is driven by information collected through the application of texture characterisation transforms popular in digital image processing. Gabor filters, wavelets and pyramid decompositions are investigated and applied to a selected set of test cases. The behaviour, performance and limitations of the proposed approach are discussed from the viewpoint of the identification and extraction of individual surface topography features.

Item Type:	Article
Subjects:	Q Science > Q Science (General) T Technology > T Technology (General)
Schools:	School of Computing and Engineering School of Computing and Engineering > Centre for Precision Technologies School of Computing and Engineering > Centre for Precision Technologies > Surface Metrology Group
Related URLs:	Author
Depositing User:	Sharon Beastall
Date Deposited:	11 Nov 2015 10:30
Last Modified:	11 Nov 2015 10:30
URI:	http://eprints.hud.ac.uk/id/eprint/26464

Downloads

Repository Staff Only: item control page

View Item