Related Experiment Video
Updated: Mar 21, 2026
09:30
Super-resolution Imaging of Neuronal Dense-core Vesicles
Published on: July 2, 2014
10.2K
A 3D Feature Descriptor Recovered from a Single 2D Palmprint Image.
Summary
A novel palmprint matching descriptor, DoN (Descriptor of Normals), captures 3D information from 2D images. This method offers stable, efficient, and accurate palmprint identification and verification, achieving state-of-the-art results.
Area of Science:
- Computer Vision
- Biometrics
- Pattern Recognition
Background:
- Efficient and accurate feature descriptors are crucial for computer vision.
- Palmprint matching is a key biometric identification method.
- Illumination variations pose challenges in contactless imaging.
Purpose of the Study:
- To propose a new feature descriptor, DoN, for 2D palmprint matching.
- To leverage 3D information from 2D palmprints for improved stability.
- To enhance the efficiency and accuracy of palmprint recognition systems.
Main Methods:
- Developed the Descriptor of Normals (DoN) based on ordinal measures of neighboring point normal vectors.
- Extracted 3D information from single 2D palmprint images.
- Evaluated DoN using four publicly available 2D palmprint databases for identification and verification.
Main Results:
- DoN descriptor captures 3D information, showing stability under illumination variations.
- DoN is computationally simple, efficient for storage, and easy to match (1-bit size per point).
- Achieved state-of-the-art performance in palmprint identification and verification across all tested databases.
Conclusions:
- The proposed DoN descriptor effectively extracts 3D information from 2D palmprints.
- DoN offers a robust and efficient solution for palmprint matching, outperforming existing methods.
- This approach advances the field of biometrics through improved feature descriptor design.
