Abstract: Through analyzing noise types of fingerprint image and the shortcomings of traditional image processing algorithms,this paper presents a kind of fingerprint image preprocessing method combining wavelet packet denoising and image quality assessment.The simulation results show that this method can effectively remove fingerprints of random noise,and reduce the effects of dry fingerprints,wet fingerprints and incomplete fingerprints to the rate of fingerprint recognition.Also it is a simple algorithm having fast processing speed to meet real-time requirements.

Key words: wavelet packet transformation, image quality assessment, median filtering, feature extraction, feature matching, fingerprint recognition