The Enhancement Method for Low-Quality Fingerprint Images Based on Fast Discrete Curvelet Transform

The performance of a fingerprint recognition system relies on the accuracy of minutiae extraction. Fingerprint image quality may degrade due to the acquisition environment and the condition of the finger (dry or moist), which negatively affects minutiae extraction. Therefore, it is necessary to enhance low-quality fingerprint images. In this paper, a low-quality fingerprint image enhancement method based on Fast Discrete Curvelet Transform is proposed. First, multi-scale analysis of the low-quality fingerprint image is performed using Fast Discrete Curvelet Transform. Then, a soft threshold function and Gabor filtering are applied to remove noise and enhance ridge continuity. To ensure accurate ridge orientation extraction, the grayscale intensity difference method is used to estimate the fingerprint ridge direction. Finally, the image is reconstructed. Experimental results show that the proposed method restores the structure of broken ridges and achieves higher accuracy in minutiae extraction.