Title :
Research on the Key Technique of Image Preprocessing in the Fingerprint Identification
Author_Institution :
Dept. of Comput. Sci., Zaozhuang Univ., Zaozhuang, China
Abstract :
The theory and algorithms of fingerprint identification were firstly introduced. By comparing kinds of methods and analysing the characteristics of fingerprint image, the new preprocessing techniques based on histogram equalization of wavelet frequency division and multi-scale product of wavelet transform based on direction pattern were proposed. The experiment results showed that this algorithm formed single pixel edge of the fingerprint and achieved the effects of edge enhancement, noise cancellation, smooth, binarization and edge thinning as well as time complexity reduction.
Keywords :
computational complexity; fingerprint identification; image enhancement; image thinning; statistical analysis; wavelet transforms; direction pattern; edge enhancement; edge thinning; fingerprint identification; histogram equalization; image preprocessing; multiscale product; time complexity; wavelet frequency division; wavelet transform; Feature extraction; Fingerprint recognition; Frequency conversion; Frequency domain analysis; Histograms; Image matching; Noise cancellation; Wavelet analysis; Wavelet domain; Wavelet transforms; fingerprint identification; histogram equalization; multi-scale product; wavelet transform;
Conference_Titel :
Software Engineering, 2009. WCSE '09. WRI World Congress on
Conference_Location :
Xiamen
Print_ISBN :
978-0-7695-3570-8
DOI :
10.1109/WCSE.2009.93