Skip to content

dasec/morph-vuln-prediction

Repository files navigation

Vulnerability of Biometric Systems to Attacks based on Morphed Biometric Information

Implementation of the framework to predict the vulnerability of biometric systems to attacks using morphed biometric information [BMT18]:

License

This work is licensed under license agreement provided by Hochschule Darmstadt (h_da-License).

Instructions

Dependencies

  • seaborn
  • numpy
  • pylab
  • matplotlib
  • argparse
  • csv

Usage

  1. Run evaluateMorphingVulnerability.py

    usage: evaluateMorphingVulnerability.py [-h] [--fmr [FMR]]
                                        	[--figureTitle [FIGURETITLE]]
                                        	[--legendLocation [LEGENDLOCATION]]
                                        	matedScoresFile nonMatedScoresFile
                                        	figureFile
    
    Evaluate the vulnerability of biometric systems to attacks using morphed
    biometric information.
    
    positional arguments:
      matedScoresFile       filename for the mated scores
      nonMatedScoresFile    filename for the non-mated scores
      figureFile            filename for the output figure
    
    optional arguments:
      -h, --help            show this help message and exit
      --fmr [FMR]           FMR in percentage of the verification threshold, if
                            none provided, FMR = 0.1 per cent
      --figureTitle [FIGURETITLE]
                            title for the output figure
      --legendLocation [LEGENDLOCATION]
                            legend location
  2. Input: at least 2 score files (mated and non-mated score examples provided), and the file name for the output figure, and optionally other parameters of the computation and the formatting of the figure.

    The score files contain the subject IDs being compared followed by the resulting comparison score, separated by blank spaces. They should be stored as a csv or text file, which will be processed with the csv package. Examples are provided.

  3. Output: figure with score distributions and resulting Pmorph.

References

More details in:

  • [BMT18] M. Gomez-Barrero, C. Rathgeb, U. Scherhag, C. Busch, "Predicting the Vulnerability of Biometric Systems to Attacks based on Morphed Biometric Information", in IET Biometrics, 2018.

Please remember to reference article [BMT18] on any work made public, whatever the form, based directly or indirectly on these metrics.

About

Implementation of the framework to predict the vulnerability of biometric systems to attacks using morphed biometric information.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages