
Research Article
Palmprint Recognition Using Learning Discriminant Line Direction Descriptors
@INPROCEEDINGS{10.1007/978-3-031-28790-9_10, author={Hoang Van Thien and Thong Dinh Duy Phan and Thai Hoang Le}, title={Palmprint Recognition Using Learning Discriminant Line Direction Descriptors}, proceedings={Nature of Computation and Communication. 8th EAI International Conference, ICTCC 2022, Vinh Long, Vietnam, October 27-28, 2022, Proceedings}, proceedings_a={ICTCC}, year={2023}, month={3}, keywords={Palmprint recognition Dominant direction number Multi direction pattern Biometrics}, doi={10.1007/978-3-031-28790-9_10} }
- Hoang Van Thien
Thong Dinh Duy Phan
Thai Hoang Le
Year: 2023
Palmprint Recognition Using Learning Discriminant Line Direction Descriptors
ICTCC
Springer
DOI: 10.1007/978-3-031-28790-9_10
Abstract
Palmprint-based biometrics has received a lot of attention for personal identification. The paper proposes a novel learning discriminant feature technique for palmprint recognition, called the Learning Discriminant Line Direction Descriptor (LDLDD), that learns separately all three kind of directional pattern code. The dominant direction number (DDN) map is calculated first in this method. Then, this technique computes direction pattern maps with three multi-direction encoding methods based on the DDN map, where pixels with the same DDN values will use the same encoding strategy and belong to the same feature map. Finally, (2D)2LDA is used to train new feature subspaces that project these maps from a high-dimensional space to a discriminant space with lower dimensions. Experiments on Hong Kong Polytechnic University’s (PolyU and IITD) public databases show that the proposed method outperforms existing techniques in terms of accuracy.