Component based feature space partition and combination in multiple colour spaces for texture classification
Component based feature space partition and combination in multiple colour spaces for texture classification
- Author(s): S. Chindaro ; K. Sirlantzis ; M.C. Fairhurst
- DOI: 10.1049/cp:20050092
For access to this article, please select a purchase option:
Buy conference paper PDF
Buy Knowledge Pack
IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.
IEE International Conference on Visual Information Engineering (VIE 2005) — Recommend this title to your library
Thank you
Your recommendation has been sent to your librarian.
- Author(s): S. Chindaro ; K. Sirlantzis ; M.C. Fairhurst Source: IEE International Conference on Visual Information Engineering (VIE 2005), 2005 p. 211 – 218
- Conference: IEE International Conference on Visual Information Engineering (VIE 2005)
- DOI: 10.1049/cp:20050092
- ISBN: 0 86341 507 5
- Location: Glasgow, UK
- Conference date: 4-6 April 2005
- Format: PDF
The use of colour and texture for visual processing applications has been traditionally carried out in different colour spaces or transformations of colour spaces. This has resulted in various proposals advocating the use of different colour spaces for different applications, without consensus on which colour space is the best overall. In this paper we propose a different approach which takes advantage of the wide range of colour spaces and use two feature space analysis techniques to perform classification in partitions of a fused multiple-colour space. We systematically partition this enlarged colour feature space using correlation reduction and independence seeking techniques, and use these in ensemble systems. We compare the performance of the ensembles and individual partitions so-formed with those of the partitions which are based on the colour space models. The results have shown that in the majority of cases, there are more significant gains obtained by using the subsets of features obtained using proposed partition methodologies over those obtained from colour-space based partitions.
Inspec keywords: independent component analysis; feature extraction; image colour analysis; image segmentation; image classification; principal component analysis; image texture
Subjects: Other topics in statistics; Optical, image and video signal processing; Other topics in statistics; Computer vision and image processing techniques
Related content
content/conferences/10.1049/cp_20050092
pub_keyword,iet_inspecKeyword,pub_concept
6
6