Enhanced Vein Detection Using NIR and RGB-D Imaging

Abstract

In this study, a robotic-based system for the venipuncture process was proposed. The system was composed of an RGB-D camera and an NIR camera mounted on the TCP end-effector of an industrial robot manipulator. By using the NIR camera and an IR light array, blood vessels near the skin surface, outside the visible light spectrum, were identified to determine an appropriate point for venipuncture. To enable the robot to utilize the vessel entry point, depth information from the RGB-D camera was aligned with the NIR image, providing depth data with an accuracy of up to 0.1 mm. The calibration procedures and vein detection methodology used in this study were thoroughly examined. The proposed system demonstrated that the vessel entry point could be identified with high precision using two cameras. © 2025 IEEE.