Ground truth for the STUMLA_HMT fingervein database (STUMLA-GR)

Paper

Groundtruth Database

On this page we provide the ground-truth masks for a part (634 samples) in the SDUMLA-HMT fingervein database for direct use (i.e. training segmentation-based CNNs).

Note that the SDUMLA-GR database only contains ground-truth not the original images. The source databases to which the ground truths apply can be found at: http://mla.sdu.edu.cn/info/1006/1195.htm

If you are using this ground-truth masks in your work, please cite the paper:

Bibtex

@inproceedings{ Jalilian18a,
      author = {Ehsaneddin Jalilian and Andreas Uhl},
      title = {Finger-vein Recognition using Deep Fully Convolutional Neural Semantic Segmentation Networks: The Impact of Training Data},
      booktitle = {Proceedings of the IEEE 10th International Workshop on Information Forensics and Security (WIFS 2018)},
      year = {2018},
      pages = {1-8},
      date = {December 11 - December 13},
      address = {Hong Kong}
}