2-D Dual Tree Complex Wavelet Transform (DT-CWT) based fingerprint ridge structure quality enhancement

Nowadays fingerprint is widely used to ascertain identity of an individual. The performance of an automated fingerprint verification system depends on the quality of the fingerprint image captured by the sensor. In this paper we propose fingerprint ridge structure quality enhancement using 2-D Dual Tree Complex Wavelet Transform (DT-CWT). The DT-CWT provides multiscale description of fingerprint images with good directional selectivity. The DT-CWT transforms fingerprint images into six directionally selective subbands at multiple scales. The wavelet coefficients in all subbands are compensated adaptively based on the compensation coefficients obtained from the singular values of subbands of fingerprint image and a referred Gaussian template. The original image is then reconstructed using inverse DT-CWT. The experimental results show that the DT-CWT is more effective than the Discrete Wavelet Transform in terms of improvement in singular point detection accuracy.