Title: New colour SIFT descriptors for image classification with applications to biometrics

Authors: Abhishek Verma, Chengjun Liu, Jiancheng Jia

Addresses: Department of Computer Science, New Jersey Institute of Technology, Newark, NJ 07102, USA. ' Department of Computer Science, New Jersey Institute of Technology, Newark, NJ 07102, USA. ' Department of Test Engineering, International Game Technology, Reno, NV 89521, USA

Abstract: This paper first presents a new oRGB-SIFT descriptor, and then integrates it with other colour SIFT features to produce the novel Colour SIFT Fusion (CSF) and the Colour Greyscale SIFT Fusion (CGSF) descriptors for image classification with special applications to biometrics. Classification is implemented using a novel EFM-KNN classifier, which combines the Enhanced Fisher Model (EFM) and the K Nearest Neighbour (KNN) decision rule. The effectiveness of the proposed descriptors and classification method are evaluated using 20 image categories from two large scale, grand challenge datasets: the Caltech 256 database and the UPOL Iris database.

Keywords: oRGB-SIFT descriptor; CSF; colour SIFT fusion; CGSF; colour greyscale SIFT fusion; EFM-KNN classifier; image classification; biometrics.

DOI: 10.1504/IJBM.2011.037714

International Journal of Biometrics, 2011 Vol.3 No.1, pp.56 - 75

Published online: 24 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article