DocumentCode
1803101
Title
Identification of critical bands in DCT domain representation for fingerprint recognition
Author
Shah, Yasir Ali ; Ahmad, Nasir ; Naeem, Muhammad
Author_Institution
Dept. of Comput. Syst. Eng., Univ. of Eng. & Technol., Peshawar, Pakistan
fYear
2012
fDate
7-8 Sept. 2012
Firstpage
1
Lastpage
4
Abstract
The low frequency sub bands of Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) are mostly used for recognition purposes. In this paper, we propose that frequency sub bands other than low frequency sub bands contain discriminative information and can be more helpful in recognition purposes. As a result, the hypothesis was checked on the fingerprint images as fingerprint images have finer and sharper details and higher frequency sub bands contain more information which should be more helpful in recognition than the low frequency sub bands. Our proposed method has accomplished promising performance. Medium and higher frequency sub bands outperform low frequency sub bands outperforms low frequency sub bands in terms of recognition rates.
Keywords
discrete cosine transforms; discrete wavelet transforms; fingerprint identification; DCT domain representation; DWT; critical bands identification; discrete cosine transform; discrete wavelet transform; fingerprint images; fingerprint recognition; low frequency sub bands; Databases; Discrete cosine transforms; Feature extraction; Fingerprint recognition; Frequency conversion; Image matching; Frequency bands; discrete cosine transform; feature extraction; fingerprint recognition;
fLanguage
English
Publisher
ieee
Conference_Titel
Automation and Computing (ICAC), 2012 18th International Conference on
Conference_Location
Loughborough
Print_ISBN
978-1-4673-1722-1
Type
conf
Filename
6330528
Link To Document