on top

Indigo 1x1 base

Public dataset of copy detection patterns (CDP) 

 Download

 

Indigo 1x1 base data set is not for any commercial usage, distribution or reproduction. It is solely for academic usage. The authors do no make any warranty or assume any legal liability or responsibility for the accuracy, completeness, or usefulness of any information (either isolated or in the aggregate) contained, or represent that its use would not infringe privately owned rights.

 

Digital template

Original HP 55

Original HP 76

indigo_based_t.png

indigo_based_o55.png

indigo_based_o76.png


Fake 55 / 55

Fake 55 / 76

 

indigo_based_f55_55.png

indigo_based_f55_76.png


Fake 76 / 55

Fake 76 / 76

 

indigo_based_f76_55.png

indigo_based_f76_76.png

 

30%

  

35%

  

40%

  

45%

  

50%
indigo_based_dens30.png   indigo_based_dens35.png   indigo_based_dens35.png   indigo_based_dens45.png   indigo_based_dens50.png

 

General info

  • Originals:
    • HP Indigo 5500 DS (HPI55), 812.8 dpi
    • HP Indigo 7600 DS (HPI76), 812.8 dpi
    • Density: 30%, 35%, 40%, 45%, 50%
    • Acquisition: Epson Perfection V850 Pro, 2400 ppi
  • Machine learning fakes:
    • Fakes 55 / 55: original printed at HPI55 / fakes printed at HPI55
    • Fakes 55 / 76: original printed at HPI55 / fakes printed at HPI76
    • Fakes 76 / 55: original printed at HPI76 / fakes printed at HPI55
    • Fakes 76 / 76: original printed at HPI76 / fakes printed at HPI76
    • Density: 50%
    • Acquisition: Epson Perfection V850 Pro, 6400 ppi

Citing | PDF

@inproceedings { Chaban2021wifs,
author = { Chaban, Roman and Taran, Olga and Tutt, Joakim and Holotyak, Taras and Bonev, Slavi and Voloshynovskiy, Slava },
booktitle = { IEEE International Workshop on Information Forensics and Security (WIFS)},
title = { Machine learning attack on copy detection patterns: are 1x1 patterns cloneable? },
address = { Montpellier, France },
month = { December },
year = { 2021 }
}