Warinthorn Kiadtikornthaweeyot Evans, Pattaraporn Somapamit

Abstract: Fingerprint recognition remains one of the most broadly adopted biometric methods for personal identity verification. Though, its effectiveness can be compromised by degraded fingerprint image quality, often caused by incomplete image data. Such degradation may arise from occupational factors involving intensive manual labour, physical injuries, or adverse environmental conditions, such as cold temperatures leading to dry skin or high humidity causing excessive perspiration. These challenges consequence in poor-quality images that hinder the performance of authentication systems. This study proposes a different fingerprint image enhancement approach integrating Adaptive Histogram Equalization, Gabor filtering, and Contrast Limited Adaptive Histogram Equalization to improve ridge sharpness and overall image quality. Experimental evaluation using the NIST Fingerprint Image Quality2 algorithm demonstrates important improvements. Employing only Adaptive Histogram Equalization and Gabor filtering, image quality increased by 19.55%, whereas the addition of Contrast Limited Adaptive Histogram Equalization further improved performance to 21.73%. In challenging cases such as Image_01 and Image_04, enhancements reached 150.0% and 133.3%, respectively. On average, the proposed method increased the NFIQ2 score from 45.60 for original images to 53.85 for enhanced images, while reducing the standard deviation from 20.72 to 12.00, indicating more constant and superior quality. These results approve the effectiveness of the proposed method in addressing low-quality fingerprint image challenges.

Keywords: Fingerprints improvement, Contrast Limited Adaptive Histogram Equalization, Image Quality, Adaptive histogram, Gabor Filter

Date Published: November 19, 2025
DOI: 10.11159/jmids.2025.006

View Article