Contactless Fingerprint Recognition Using SIFT Enhanced Minutiae Matching

This study suggests a framework for contactless fingerprint detection that uses mobile cameras to get around the drawbacks of conventional contact-based systems, including distortions, sensor wear, and hygienic issues. Verification and enrolment are the two stages of the system's operation. High-resolution fingerprint images are taken under carefully monitored circumstances during enrolment, and they are then improved using Gabor filtering, normalisation, CLAHE, and YCbCr segmentation. Techniques for feature extraction, such as minutiae identification, thinning, and binarization, produce reference templates. Newly taken fingerprints are processed the same way during verification and compared using a minutiae-based algorithm that accounts for differences in scale and orientation. A review of current techniques emphasises combining deep learning and traditional image processing for increased accuracy. This method improves scalability, security, and user convenience for applications such as remote identity verification and mobile authentication.