Finger minutiae extraction based on the use of YOLO-V5

In this study, we propose a novel method for detecting minutiae in fingerprint images using YOLOv5, a state-of-the-art object detection algorithm. Our approach utilizes a convolutional neural network (CNN) to identify and locate minutiae points, such as ridge endings and bifurcations, within a fingerprint image. We trained the CNN on a dataset of fingerprint images and corresponding minutiae annotations, and evaluated its performance using standard metrics such as precision, recall, mAP 0.5 and mAP 0.5:0.95. Our results indicate that the proposed method is able to accurately detect minutiae in fingerprint images with high precision - 91%, recall - 82%, mAP 0.5 - 89% and mAP 0.5:0.95 - 39%. Furthermore, we demonstrate that the YOLOv5-based approach is significantly faster than traditional minutiae detection methods, making it suitable for real-time applications. In conclusion, this study presents a promising approach for the automated detection of minutiae in fingerprint images using YOLOv5.

• Autores: Trusiak, Krzysztof; Saeed, Khalid
• Tipo de recurso: Conferencia (Ponencia)
• Fecha de publicación: 2023
• Institución: Corporación Universidad de la Costa
• Repositorio: REDICUC - Repositorio CUC
• Idioma: eng
• OAI Identifier: oai:repositorio.cuc.edu.co:11323/14081
• Acceso en línea: https://hdl.handle.net/11323/14081; https://repositorio.cuc.edu.co/
• Palabra clave: YOLOv5; Fingerprint images; Convolutional neural network (CNN)
• Rights: closedAccess
• License: Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)