Enhanced Fingerprint Recognition System Using Minutiae-Based Analysis and FFT-Based Feature Extraction

Fingerprint recognition is a widely adopted biometric authentication technique due to the uniqueness and permanence of fingerprint patterns. This paper presents a fingerprint recognition system developed using MATLAB, incorporating an interactive graphical user interface (GUI) and advanced image processing techniques. The system employs histogram equalization for contrast enhancement, Fast Fourier Transform (FFT)-based enhancement for ridge structure clarity, and minutiae-based feature extraction for precise fingerprint identification. The preprocessing steps, including region of interest (ROI) extraction, thinning, and noise removal (H-breaks and spikes), optimize the feature representation for improved matching accuracy. The minutiae-based matching algorithm detects ridge endings and bifurcations, ensuring robust authentication. The experimental results demonstrate the system's effectiveness in enhancing fingerprint image quality and feature extraction accuracy, leading to high matching precision. The system achieves reliable performance across varying fingerprint quality levels, confirming its suitability for biometric authentication applications. Future work aims to integrate deep learning models and multimodal biometric approaches to further enhance recognition accuracy and system adaptability.