Core Concepts
The sweet platform is a modular and extensible system designed to explore various sensors and technologies for improving the state-of-the-art in contactless reflective vascular hand biometrics.
Abstract
The sweet platform is designed to capture images of fingers, palm and wrist using several modalities, including multi-spectral Near-Infrared (NIR), RGB-color, Stereo Vision (SV), and Photometric Stereo (PS). This allows for the collection of high-quality data on vascular and surface features of the hand.
The key highlights of the sweet platform include:
Contactless acquisition of hand vascular and surface features
Support for multiple modalities: multi-spectral NIR, RGB, SV, and PS
Ability to capture images of multiple fingers simultaneously
Development of a powerful 3D pipeline for pre-processing the acquired data
Experiments on Finger-Vein Recognition (FVR) demonstrating state-of-the-art performance
The authors have collected a new dataset, Candy v3 - Part 1, using the sweet platform, and have made it publicly available for research purposes. They have also released the acquisition software, parts of the hardware design, and the source code for their experiments.
Stats
The sweet platform can capture 20 usable image-frames per camera, under a variety of illumination conditions, for each biometric sample.
The Candy v3 - Part 1 dataset contains data from 120 subjects, with 5 samples per hand.
Quotes
"Current finger-vein or palm-vein recognition systems usually require direct contact of the subject with the apparatus. This can be problematic in environments where hygiene is of primary importance."
"Besides introducing the hardware prototype, we also present FVR experiments to study the efficacy of our approach."
"The FVR experiments discussed here are based on a new dataset collected from 120 subjects using the sweet platform."