Palmprint identification using Log Transformation of Transform Domain Features

Author

Shashikala, K.P. ; Raja, K.B.

Author_Institution

Dept. of ECE, Dayananda Sagar Coll. of Eng., Bangalore, India

fYear

2014

fDate

13-14 Feb. 2014

Firstpage

1

Lastpage

5

Abstract

The Palmprint is an efficient physiological biometric trait to identify a person. In this paper we propose Palmprint Identification using Log Transformation of Transform Domain Features. The Region of Interest (ROI) of palmprint image is extracted using preprocessing. The KWT and DWT are applied on preprocessed image to generate features. The KWT and DWT features of test image and database images are compared using Euclidian distance to compute EER and TSR values. The EER and TSR values of KWT and DWT are fused using Log Transformation to get better performance parameters. It is observed that the values of performance parameters are better in the case of proposed algorithm compared to existing algorithms.

Keywords

discrete wavelet transforms; feature extraction; palmprint recognition; DWT; EER value; Euclidian distance; KWT; Kekre wavelet transform; ROI; TSR value; discrete wavelet transforms; log transformation; palmprint identification; physiological biometric trait; region of interest; transform domain features; Databases; Discrete wavelet transforms; Feature extraction; Support vector machines; DWT; ED; EER; KWT; Palmprint; TSR;

fLanguage

English

Publisher

ieee

Conference_Titel

Electronics and Communication Systems (ICECS), 2014 International Conference on

Conference_Location

Coimbatore

Print_ISBN

978-1-4799-2321-2

Type

conf

DOI

10.1109/ECS.2014.6892668

Filename

6892668