Visualizing NIR Vein Patterns Using Supervised and Unsupervised Methods

Abstract

This article presents a self-supervised approach implementing an unsupervised clustering algorithm to analyze the intrinsic vascular pattern in near-infrared (NIR) light. The framework includes NIR intrinsic vascular image acquisition, pattern detection, ML multiscale filtering, feature extraction, recognition, identification, and matching based on a linear regression model to detect an optional variable worth dependent on a free factor. The approach uses ordinal NIR vein print portrayal, and the self-learning methodology achieved a 97.50% accuracy score for identifying intrinsic vascular patterns in unsupervised learning issues.