Publication: Hand shape identification using palmprint alignment based on intrinsic local affine-invariant fiducial points
Issued Date
2015-01-01
Resource Type
Other identifier(s)
2-s2.0-84923017877
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
BMEiCON 2014 - 7th Biomedical Engineering International Conference. (2015)
Suggested Citation
Choopol Phromsuthirak, Supakorn Suwan, Arthorn Sanpanich, Chuchart Pintavirooj Hand shape identification using palmprint alignment based on intrinsic local affine-invariant fiducial points. BMEiCON 2014 - 7th Biomedical Engineering International Conference. (2015). doi:10.1109/BMEiCON.2014.7017384 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/35953
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Hand shape identification using palmprint alignment based on intrinsic local affine-invariant fiducial points
Other Contributor(s)
Abstract
© 2014 IEEE. Palmprint is the mostly popular biometrics used in security system. However, it is difficult to acquire the palmprint features with the common problems of pose, lighting, orientation, gesture etc. of palmprint image. So, these problems have the effect to reduce the level of confidence in personal authentication. In this paper, we proposed a new hand shape identification using palmprint alignment without guidance pegs algorithm for improving the level of confidence in palmprint identification system. The palmprint alignment based on a set of fiducial points which are intrinsic, local and preserved under affine transformation. The fiducial points are relative affine invariant to affine transformations, they allow for alignment where position of the palm relative to camera orientation can be arbitrary set. Moreover, before palmprint alignment process, the web camera which was used to capture the palmprint image was calibrated by Camera Calibration Toolbox developed by Jean-Yves Bouguet. The performance of the identification algorithm was tested in 2 types: intra-class identification and inter-class identification. The intra-class identification has the most of distance map error was started from 1.4 pixels to 4.5 pixels and the inter-class identification has 18 percent equal error rate.