ANALYSIS OF DIGITAL IMAGE PROCESSING ALGORITHMS FOR FINGERPRINT IMAGES IN BIOMETRIC IDENTIFICATION

Abstract

Biometric identification technologies are becoming increasingly important in modern information security systems. Among various biometric methods, fingerprint recognition systems are widely used due to their high reliability, uniqueness, and stability throughout a person’s lifetime. This paper analyzes digital image processing algorithms used in fingerprint-based biometric identification systems. The study examines the main stages of fingerprint image processing, including preprocessing, segmentation, binarization, ridge thinning, minutiae extraction, and matching algorithms. In addition, the effectiveness of Gabor filters, Wavelet transform methods, correlation-based techniques, and Convolutional Neural Network (CNN) models in fingerprint recognition systems is investigated. The research shows that image enhancement and accurate minutiae extraction significantly improve the accuracy, reliability, and security of biometric authentication systems. Furthermore, the paper discusses modern challenges such as fingerprint spoofing and low-quality image processing, while highlighting the advantages of artificial intelligence–based approaches in fingerprint recognition technologies.