Mammogram classification using curvelet GLCM texture features and GIST features

This paper presents a feature fusion technique that can be used for classification of ROIs in breast cancer into normal and abnormal classes. The texture features are extracted using geometric invariant shift transform and statistical features from the curvelet grey level co-occurrence matrices. Fir...

Full description

Main Authors: Gardezi, S.J.S., Faye, I., Adjed, F., Kamel, N., Eltoukhy, M.M.
Format: Article
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-eprints.20333 /
Published: Springer Verlag 2017
Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-84994508050&doi=10.1007%2f978-3-319-48308-5_67&partnerID=40&md5=be837cf29475b0685f78e75292854546
http://eprints.utp.edu.my/20333/
Tags: Add Tag
No Tags, Be the first to tag this record!
id utp-eprints.20333
recordtype eprints
spelling utp-eprints.203332018-04-23T01:04:49Z Mammogram classification using curvelet GLCM texture features and GIST features Gardezi, S.J.S. Faye, I. Adjed, F. Kamel, N. Eltoukhy, M.M. This paper presents a feature fusion technique that can be used for classification of ROIs in breast cancer into normal and abnormal classes. The texture features are extracted using geometric invariant shift transform and statistical features from the curvelet grey level co-occurrence matrices. First classification accuracy of both methods were evaluated independently. Later, feature fusion is done to improve the classification performance. Support vector machine classifier with polynomial kernel was implemented using 2 × 5 folds cross validation. Fusion of features produces better results with accuracy of 92.39 as compared to 77.97 and 91 for GIST and CGLCM respectively. © Springer International Publishing AG 2017. Springer Verlag 2017 Article NonPeerReviewed https://www.scopus.com/inward/record.uri?eid=2-s2.0-84994508050&doi=10.1007%2f978-3-319-48308-5_67&partnerID=40&md5=be837cf29475b0685f78e75292854546 Gardezi, S.J.S. and Faye, I. and Adjed, F. and Kamel, N. and Eltoukhy, M.M. (2017) Mammogram classification using curvelet GLCM texture features and GIST features. Advances in Intelligent Systems and Computing, 533 . pp. 705-713. http://eprints.utp.edu.my/20333/
institution Universiti Teknologi Petronas
collection UTP Institutional Repository
description This paper presents a feature fusion technique that can be used for classification of ROIs in breast cancer into normal and abnormal classes. The texture features are extracted using geometric invariant shift transform and statistical features from the curvelet grey level co-occurrence matrices. First classification accuracy of both methods were evaluated independently. Later, feature fusion is done to improve the classification performance. Support vector machine classifier with polynomial kernel was implemented using 2 × 5 folds cross validation. Fusion of features produces better results with accuracy of 92.39 as compared to 77.97 and 91 for GIST and CGLCM respectively. © Springer International Publishing AG 2017.
format Article
author Gardezi, S.J.S.
Faye, I.
Adjed, F.
Kamel, N.
Eltoukhy, M.M.
spellingShingle Gardezi, S.J.S.
Faye, I.
Adjed, F.
Kamel, N.
Eltoukhy, M.M.
Mammogram classification using curvelet GLCM texture features and GIST features
author_sort Gardezi, S.J.S.
title Mammogram classification using curvelet GLCM texture features and GIST features
title_short Mammogram classification using curvelet GLCM texture features and GIST features
title_full Mammogram classification using curvelet GLCM texture features and GIST features
title_fullStr Mammogram classification using curvelet GLCM texture features and GIST features
title_full_unstemmed Mammogram classification using curvelet GLCM texture features and GIST features
title_sort mammogram classification using curvelet glcm texture features and gist features
publisher Springer Verlag
publishDate 2017
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-84994508050&doi=10.1007%2f978-3-319-48308-5_67&partnerID=40&md5=be837cf29475b0685f78e75292854546
http://eprints.utp.edu.my/20333/
_version_ 1741196354860875776
score 11.62408