. .

Project Sponsors

Bundesamt für Sicherheit in der Informationstechnik (BSI)
Norwegian Research Council (RCN)

Project Partners

Hochschule Darmstadt (CRISP) NTNU (NBL) secunet Security Networks AG (SEC)
Salzburg University (Wavelab)
National Institute of Standards and Technology (NIST)

Contact

Christoph Busch
Email: christoph.busch@h-da.de

Christian Rathgeb
Ulrich Scherhag
Raghu Ramachandra
Kiran Raja
Marta Gomez-Barrero
Sergey Isadskiy
Daniel Fischer

Morphing Attack Detection (MAD)

Morphing Attacks and Impact

Biometrics refers to the automated recognition of individuals based on their biological and behavioral characteristics. Due to the strong link between subjects and their biometric samples, the wide acceptance, and their user convenience, biometric systems become increasingly popular and important for border control.
Even though the security of biometric systems is increasing, specifically face recognition systems are vulnerable with regard to the unique link between the biometric facial sample and its subject. By enrolling an artificial sample, generated by merging samples of two or multiple subjects in image domain, the contributing subjects might be verified successfully against the manipulated reference. This morphing attack can be done, for instance, in the passport application process, where in most European countries the applicant brings his own printed photograph. This way, the unique link between individuals and their biometric reference data is annulled. The impact of this vulnerability has been illustrated in this video
To prevent the aforementioned face morphing attacks, an automatic detection of morphs is required at border control as well as early in the application process. This must be considered as the most challenging research task to date and long-term research will be need to fully address the challenge. However on this page reports recent progress in this research area.

Application Scenarios

Focusing on the workflow of a generic biometric system, two Morphing Attack Detection (MAD) tasks can be distinguished: (1) detection during enrolment, e.g. the passport application process, where the MAD mechanism processes a single image, referred to as no-reference morphing attack detection and depicted in Figure (a); and (2) detection at the time of authentication, e.g. the usage of Automated Border Control (ABC) gates at borders, where a live capture from an authentication attempt serves as additional source of information for the MAD mechanism, referred to as differential morphing detection and depicted in Figure (b).

picture 1

History of Morphing Attacks

The problem of morphing attacks has been addressed in the biometric research community only recently, despite that the issue was discussed back in 2004 in the presentation
Matthew Lewis, Philip Statham: "CESG Biometric Security Capabilities Programme: Method, Results and Research challenges", in Proceedings Biometrics Consortium Conference (BCC), (2004)
and was also classified as vulnerability of a biometric system as early as 2009 in Clause 8.3.8.1 of ISO/IEC FDIS 19792 stating: "… Examples of abnormal characteristics could include those with unusually large or small numbers of features. Such characteristics may not be representative of any human biometric characteristic but could be synthesised and copied to an artefact. Alternatively a synthesised characteristic could be injected electrically during a replay attack or planted in the reference database. - feature sets comprising amalgamations of biometric features from 2 or more individuals, e.g. morphed facial images”
The feasibility of such morphing attacks was first shown for face recognition systems in 2014 in the FIDELITY project by M. Ferrara, A. Franco and D. Maltoni in her IJCB paper "The magic passport".

Detection of Morphing Attacks

With our research we address the problem of morphing attacks. The work has started only recently. In 2016 the first Morphing Attack Detection (MAD) mechanism was proposed by R. Raghavendra, K. Raja and C. Busch in the BTAS paper "Detecting Morphed Facial Images". Since we have started numerous efforts to address this research challenge.

MAD Publications by our Group

  1. R.Raghavendra, S. Venkatesh, K. Raja, C. Busch: "Towards making Morphing Attack Detection robust using hybrid Scale-Space Colour Texture Features“, in Proceedings of 5th International Conference on Identity, Security and Behaviour Analysis (ISBA 2019), Hyderabad, IN, January 22-24, (2019)PDF-Version
  2. L. Debiasi, C. Rathgeb, U. Scherhag, A. Uhl, C. Busch:"PRNU Variance Analysis for Morphed Face Image Detection", in Proceedings of 9th International Conference on Biometrics: Theory, Applications and Systems (BTAS 2018), Los Angeles, US, October 22-25, (2018)PDF-Version
  3. R.Raghavendra, S. Venkatesh, K. Raja, C. Busch: "Detecting Face Morphing Attacks with Collaborative Representation of Steerable Scale-Space Features“, in Proceedings of 3rd International Conference on Computer Vision and Image Processing (CVIP 2018), Japalpur, IN, September 29 - October 1, (2018)PDF-Version
  4. U. Scherhag, D. Budhrani, M. Gomez-Barrero, C. Busch: "Detecting Morphed Face Images Using Facial Landmarks“, in Proceedings of International Conference on Image and Signal Processing (ICISP 2018), Cherbourg, FR, July 2-4, (2018)PDF-Version
  5. U. Scherhag, C. Rathgeb, C. Busch: "Performance Variation of Morphed Face Image Detection Algorithms across different Datasets“, in Proceedings of 6th International Workshop on Biometrics and Forensics (IWBF 2018), Sassari, IT, June 7-8, (2018)PDF-Version
  6. L. Debiasi, U. Scherhag, C. Rathgeb, A. Uhl, C. Busch: "PRNU-based Detection of Morphed Face Images“, in Proceedings of 6th International Workshop on Biometrics and Forensics (IWBF 2018), Sassari, IT, June 7-8, (2018)PDF-Version
  7. U. Scherhag, C. Rathgeb, C. Busch: „Detection of Morphed Faces from Single Images: a Multi-Algorithm Fusion Approach“, in Proceedings if of the 2nd International Conference on Biometric Engineering and Applications (ICBEA 2018), Amsterdam, The Netherlands, May 16-18, (2018)PDF-Version
  8. U. Scherhag, C. Rathgeb and C. Busch: „Towards Detection of Morphed Face Images in electronic Travel Documents“, in Proceedings of the 13th IAPR International Workshop on Document Analysis Systems (DAS 2018), Vienna, Austria, April 24-27, (2018)PDF-Version
  9. M. Gomez-Barrero, C. Rathgeb, U. Scherhag, C. Busch: „Predicting the Vulnerability of Biometric Systems to Attacks based on Morphed Biometric Samples“, in IET Biometrics, (2018)PDF-Version
  10. C. Rathgeb, C. Busch: "On the Feasibility of Creating Morphed Iris-Codes", in Proceedings of International Joint Conference on Biometrics (IJCB 2017), Denver, Colorado, October 1-4, (2017)PDF-Version
  11. R. Raghavendra, K. Raja, S. Venkatesh, C. Busch: "Face Morphing Versus Face Averaging: Vulnerability and Detection", in Proceedings of International Joint Conference on Biometrics (IJCB 2017), Denver, Colorado, October 1-4, (2017)PDF-Version
  12. U. Scherhag, A. Nautsch, C. Rathgeb, M. Gomez-Barrero, R. Veldhuis, L. Spreeuwers, M. Schils, D. Maltoni, P. Grother, S. Marcel, R. Breithaupt, R. Raghavendra, C. Busch: "Biometric Systems under Morphing Attacks: Assessment of Morphing Techniques and Vulnerability Reporting", in Proceedings of the IEEE 16th International Conference of the Biometrics Special Interest Group (BIOSIG), Darmstadt, September 20-22, (2017)PDF-Version
  13. R. Raghavendra, K. Raja, S. Venkatesh, C. Busch: "Transferable Deep-CNN features for detecting digital and print-scanned morphed face images", in Proceedings of 30th International Conference on Computer Vision and Pattern Recognition Workshop (CVPRW 2017), Honolulu, Hawaii, July 21-26, (2017)PDF-Version
  14. M. Gomez-Barrero, C. Rathgeb, U. Scherhag, C. Busch: "Is Your Biometric System Robust to Morphing Attacks?", in Proceedings of 5th International Workshop on Biometrics and Forensics (IWBF 2017), Coventry, UK, April 4-5, (2017)PDF-Version
  15. U. Scherhag, R. Raghavendra, K. Raja, M. Gomez-Barrero, C. Rathgeb, C. Busch: "On The Vulnerability Of Face Recognition Systems Towards Morphed Face Attacks", in Proceedings of 5th International Workshop on Biometrics and Forensics (IWBF 2017), Coventry, UK, April 4-5, (2017)PDF-Version
  16. R. Raghavendra, K. Raja, C. Busch: "Detecting Morphed Facial Images", in Proceedings of 8th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS-2016), September 6-9, Niagra Falls, USA, (2016)PDF-Version

Further Information

top of page