Download the whole database(58.0MB)
DESCRIPTION OF ATVS-SSig DB
Two datasets of fully on-line synthetic signatures generated according to the original method described and validated in [PR2012-I] and [PR2012-II]:
· DATASET 1: DS1_Modification_TimeFunctions
In this case the duplicated samples are generated modifying directly the time functions of the master signature according to the method described in Sect. 5.1 of [PR2012-I].
It contains 25 signatures of 350 users.
The first 5 signatures follow an intra-session variability and the remaining 20 an inter-session variability.
· DATASET 2: DS2_Modification_LNparameters
In this case the duplicated samples are generated modifying the LogNormal parameters of the master signature according to the method described in Sect. 5.2 of [PR2012-I].
It contains 25 signatures of 350 users.
The first 5 signatures follow an intra-session variability and the remaining 20 an inter-session variability.
· FILES FORMAT
The signatures are stored in txt files with 5 columns, where:
- COLUMN 1: represents the x coordinate.
- COLUMN 2: represents the y coordinate.
- COLUMN 3: this is a synthetic timestamp (it may be neglected).
- COLUMN 4: indicates the penups. It is 0 where p=0, and 1 otherwise.
- COLUMN 5: represents the pressure (p) function.
· FILES NOMENCLATURE
The nomenclature followed in both datasets to name the signature files is as follows: uXXXX_sgYY.txt
XXXX: is the number of the user [1001, 1002, … , 1350]
YY: is the number of the sample [1, 2, … , 25]
REFERENCES
For further information on the database and on different applications where it has been used, including vulnerability assessment of signature verification systems to brute-force attacks, and performance evaluation of different signature recognition systems, we refer the reader to (all these articles are publicly available at http://atvs.ii.uam.es/listpublications.do):
· [PR2012-I] J.Galbally, R. Plamondon, J. Fierrez and J. Ortega-Garcia, "Synthetic on-line signature generation. Part I: Methodology and algorithms", Pattern Recognition, 2012 (DOI: 10.1016/j.patcog.2011.12.011).
· [PR2012-II] J.Galbally, J. Fierrez, J. Ortega-Garcia and R. Plamondon, "Synthetic on-line signature generation. Part II: Experimental validation", Pattern Recognition, 2012 (DOI: 10.1016/j.patcog.2011.12.007).
· [ICDAR2009] J. Galbally, J. Fierrez, M. Martinez-Diaz and J. Ortega-Garcia, "Evaluation of brute-force attacks to dynamic signature verification using synthetic samples", in Proc. IAPR Intl. Conf. on Document Analysis and Recognition, ICDAR, pp. 131-135, Barcelona, Spain, July 2009.
Please remember to reference articles [PR2012-I] and [PR2012-II] on any work made public, whatever the form, based directly or indirectly on any part of the ATVS-SSig DB.
