Logo des Repositoriums

Auflistung nach Schlagwort "and Vein"

1 - 2 von 2
  • Konferenzbeitrag
    Contactless Palmprint Recognition for Children
    (BIOSIG 2023, 2023) Akash M Godbole, Steven A Grosz
    Effective distribution of nutritional and healthcare aid for children, particularly infants and toddlers, in the world’s least developed and most impoverished countries, is a major problem due to lack of reliable identification documents. We present a mobile based contactless palmprint recognition system, Child Palm-ID, which meets the requirements of usability, cost, and accuracy for child recognition. On a contactless child palmprint database, Child-PalmDB1, with 1,020 unique palms (age range of 6 mos. to 48 mos.), Child Palm-ID achieves a TAR=94.8% at FAR=0.1%. Child Palm-ID is also able to recognize adults, achieving a TAR=99.5% on the CASIA contactless palmprint database and a TAR=100% on the COEP contactless adult palmprint database, both at FAR=0.1%. For child palmprint images captured at an interval of five months with differences in standoff distance, illumination and motion blur, the TAR drops to 80.5% at FAR=0.1%. This indicates that more research remains in contactless child palmprint recognition.
  • Konferenzbeitrag
    Exploring the Untapped Potential of Unsupervised Representation Learning for Training Set Agnostic Finger Vein Recognition
    (BIOSIG 2023, 2023) Tugce Arican, Raymond Veldhuis
    Finger vein patterns are a promising biometric trait because of their higher privacy and security features compared to face and finger prints. Finger vein recognition methods have been researched extensively, especially deep learning based methods such as Convolutional Neural Networks. These methods show promising recognition performance, but their low degree of generalization and adaptability results in much lower and inconsistent recognition performance in cross database scenarios. Despite these drawbacks, much less research has gone into the generalization and adaptability of these deep learning methods. This study addresses these issues and proposes an unsupervised learning approach, namely a patch-based Convolutional Auto-encoder for learning finger vein representations. Our proposed approach outperforms traditional baseline finger recognition methods on the UTFVP, SDUMLA-HMT, and PKU datasets, and achieves state-of-the-art performance on the UTFVP dataset with 0.24\% EER. It also indicates a noticeably higher generalization of finger vein features across different datasets compared to a supervised method. The findings of this work offer promising advancements in achieving robust finger vein recognition in real-life scenarios, due to the enhanced generalization and adaptability of our proposed method.