Commit edd6c41c authored by Philip ABBET's avatar Philip ABBET

Add utfvp/3 (api change: beat.backend.python v1.4.2)

parent 44a46e2b
This diff is collapsed.
This diff is collapsed.
.. Copyright (c) 2017 Idiap Research Institute, http://www.idiap.ch/ ..
.. Contact: beat.support@idiap.ch ..
.. ..
.. This file is part of the beat.examples module of the BEAT platform. ..
.. ..
.. Commercial License Usage ..
.. Licensees holding valid commercial BEAT licenses may use this file in ..
.. accordance with the terms contained in a written agreement between you ..
.. and Idiap. For further information contact tto@idiap.ch ..
.. ..
.. Alternatively, this file may be used under the terms of the GNU Affero ..
.. Public License version 3 as published by the Free Software and appearing ..
.. in the file LICENSE.AGPL included in the packaging of this file. ..
.. The BEAT platform is distributed in the hope that it will be useful, but ..
.. WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY ..
.. or FITNESS FOR A PARTICULAR PURPOSE. ..
.. ..
.. You should have received a copy of the GNU Affero Public License along ..
.. with the BEAT platform. If not, see http://www.gnu.org/licenses/. ..
The UTFVP Database of Finger vein
---------------------------------
Changelog
=========
* **Version 3**, 02/Nov/2017:
- Port to beat.backend.python v1.4.2
* **Version 2**, 20/Jan/2016:
- Port to Bob v2
* **Version 1**, 06/Nov/2014:
- Initial release
Description
===========
The University of Twente Finger Vascular Pattern (UTFVP) database is a
realistic and challenging finger vein database to support and stimulate
research efforts in the area of developing, testing and evaluating algorithms
for vascular pattern recognition.
The collected dataset contains 1440 finger vascular pattern images in total
which have been collected from 60 volunteers at Twente University during the
2011-2012 academic year. Images were captured in two identical sessions with an
average time lapse of 15 days. For each volunteer the vascular pattern of the
index, ring and middle finger of both hands has been collected twice at each
session. This means that each individual finger has been captured four times in
total. The captured images have a resolution of 672 x 380 pixels and have a
pixel density of 126 pixels per centimetre (ppcm). The images are stored using
the lossless 8 bit grey scale Portable Network Graphics(PNG) format.
The percentage of male volunteers was 73% and the percentage of right handed
volunteers was 87%. The dataset represents a young population with 82% of the
volunteers falling in the age range of 19-30, the remaining volunteers were
older than this. The quality of the collected images varies from person to
person, but the variation in quality of the images from the same person is
small. The width of the visible blood vessels range from 4-20 pixels which
corresponds to vessel widths of approximately 0.3-1.6 mm. These vessel widths
are approximate numbers because the pixel density was determined assuming a
flat surface.
Associated with the database is the UTFVP protocol called **1vsall**, which is
based on the database reference paper [Ton+13].
The `Idiap Research Institue <http://www.idiap.ch/>`_ and `The Swiss Center for
Biometrics Research and Testing <http://www.biometrics-center.ch/>`_ define the
Normal Operation Mode (NOM) protocols which set of data to use for training,
evaluation and testing. Performing experiments according to the protocol allows
institutions to easily compare their results to others.
Description: The nom protocols divide the database on three subsets: world
(subjects 1-10), development (subjects 11-28) and test (subjects 29-60). Only
the images in world set should be used to train system components such as
world/background models, PCA/LDA subspaces, etc., or to otherwise use as
background data, for example for score normalisation, etc. The development set
only should be used to train system hyper-parameters such as the decision
threshold, number of dimensions in a subspace, feature extraction and
preprocessing hyper-parameters, coefficients for linear fusion, etc., to
minimise the chosen error rate metric. Finally, the test set should be used to
test finger vein verification accuracy. The decision threshold must be
determined by tuning on the development set, and then blindly applied to finger
vein verification scores produced on the test set.
In the **nom** protocol the different fingers of one subject are considered
different subjects. Therefore, a total number of 60 x 6 = 360 subjects will be
considered for the experiments. The remaining protocols: **nomLeftRing**,
**nomLeftMiddle**, **nomLeftIndex**, **nomRightIndex**, **nomRightMiddle**,
**nomRightRing** consider just one finger per subject. Therefore, a total
number of 60 subjects will be considered for the experiments in these cases.
In all these protocols, the two finger vein images from the first session are
used for enrolment and the two from the second session as probe samples.
Citation: All documents and papers that report on research that uses the UTFVP
database must acknowledge the use of the database by including a citation of
the paper [Ton+13].
.. [Ton+13] *B. Ton and R.N.J. Veldhuis*. **A High Quality Finger Vascular Pattern Dataset Collected Using a Custom Designed Capturing Device**. In: 6th IAPR International Conference on Biometrics (ICB), pp. 1-5, June 4-7, Madrid, Spain, 2013.
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment