Authors:
Dana E. Ilea
1
;
Paul F. Whelan
2
and
Ovidiu Ghita
1
Affiliations:
1
Dublin City University, Ireland
;
2
Vision Systems Group, Dublin City University, Ireland
Keyword(s):
Texture Segmentation, Multi-resolution Integration, Image Orientation, Texture Distribution.
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Feature Extraction
;
Features Extraction
;
Image and Video Analysis
;
Informatics in Control, Automation and Robotics
;
Segmentation and Grouping
;
Signal Processing, Sensors, Systems Modeling and Control
Abstract:
The major aim of this paper consists of a comprehensive quantitative evaluation of adaptive texture descriptors when integrated into an unsupervised image segmentation framework. The techniques involved in this evaluation are: the standard and rotation invariant Local Binary Pattern (LBP) operators, multi-channel texture decomposition based on Gabor filters and a recently proposed technique that analyses the distribution of dominant image orientations at both micro and macro levels. The motivation to investigate these texture analysis approaches is twofold: (a) they evaluate the texture information at micro-level in small neighborhoods and (b) the distributions of the local features calculated from texture units describe the texture at macro-level. This adaptive scenario facilitates the integration of the texture descriptors into an unsupervised clustering based segmentation scheme that embeds a multi-resolution approach. The conducted experiments evaluate the performance of these te
chniques and also analyse the influence of important parameters (such as scale, frequency and orientation) upon the segmentation results.
(More)