ISO/IEC 20059:2025

Information technology - Methodologies to evaluate the resistance of biometric systems to morphing attacks

This document establishes a methodology to evaluate the resistance of BSs to morphing attacks, including multiple identity attacks. The document is limited to image-based morphing attacks. The term "image-based" includes modalities such as face, iris and finger image data. The document establishes: - a definition of biometric sample modifications and manipulation with a specific focus on manipulations that constitute a multiple identity attack. This can be, for instance, an enrolment attack with face image morphing; - a methodology to measure the morphing attack potential of a morphing method. The document also describes how morphing algorithms can be used for system evaluation.

Technologies de l'information — Méthodologies pour l'évaluation de la résistance des systèmes biométriques aux attaques par morphing

General Information

• Status: Published
• Publication Date: 24-Aug-2025
• ICS: 35.030 - IT Security
• ICS: 35.240.15 - Identification cards. Chip cards. Biometrics
• Technical Committee: ISO/IEC JTC 1/SC 37 - Biometrics
• Drafting Committee: ISO/IEC JTC 1/SC 37 - Biometrics
• Current Stage: 6060 - International Standard published
• Start Date: 25-Aug-2025
• Due Date: 09-Feb-2026
• Completion Date: 25-Aug-2025

Overview

ISO/IEC 20059:2025 - "Information technology - Methodologies to evaluate the resistance of biometric systems to morphing attacks" defines a standardized methodology to assess how image-based biometric systems resist morphing attacks, including multiple identity attacks (MIA). The standard is limited to image-based modalities such as face, iris and finger image data and provides definitions, evaluation metrics and procedures to measure the morphing attack potential (MAP) of morphing methods. It also describes how morphing algorithms can be used to evaluate biometric systems and includes informative annexes for reference implementation and visualization.

Key technical topics and requirements

• Scope and terminology: Clear definitions of image modification, image manipulation, biometric morphing, and related attack/detection concepts (e.g., S-MAD, D-MAD).
• Morphing attack potential (MAP): A methodology and matrix-based metrics to quantify a morphing method’s capability to deceive one or more biometric systems across multiple verification attempts.
• Detection and error rates: Standardized metrics for morphing attack detection performance, including BSCER (bona fide sample classification error rate) and MACER (morphing attack classification error rate), with procedures to report these values.
• Multiple contributing subject generalization: Treatment of morphed samples that combine multiple identities and how to measure their effect on system vulnerability.
• Benchmarking and influencing factors: Guidelines for benchmarking morphing methods, visualizing morphed samples, and understanding factors that impact MAP (e.g., number of systems, thresholds, probe attempts).
• Reporting and reproducibility: Guidance for consistent measurement, visualization (examples in Annex B), and a reference implementation (Annex A) to support reproducible evaluation.
• Limitations: The standard focuses on evaluation methodology and explicitly notes that resistance evaluation is not a full security evaluation.

Practical applications

• Evaluating the robustness of automated border control (ABC) gates, passport issuance workflows and enrolment systems against image-based morphing attacks.
• Benchmarking morphing and morph-detection algorithms in laboratories and testbeds.
• Integrating MAP and MACER metrics into procurement, conformance testing and risk assessments for biometric systems.
• Supporting research into morphing generation and detection methods and informing operational policies for trusted capture and enrolment.

Who should use ISO/IEC 20059:2025

• Biometric system vendors and integrators
• Border control and identity document authorities
• Testing laboratories and certification bodies
• Security architects, risk managers and policy makers
• Researchers developing morphing or morph-detection algorithms

Related standards

Normative references include:

• ISO/IEC 2382-37 (Biometrics vocabulary)
• ISO/IEC 30107-1 and ISO/IEC 30107-3 (Biometric presentation attack detection frameworks and testing)
• ISO/IEC 39794-5 (Face image data interchange formats)

ISO/IEC 20059:2025 provides a consistent, repeatable framework for assessing image-based morphing threats and integrating quantitative MAP metrics into biometric security programs.

Standards Content (Sample)

International
Standard
ISO/IEC 20059
First edition
Information technology —
2025-08
Methodologies to evaluate the
resistance of biometric systems to
morphing attacks
Technologies de l'information — Méthodologies pour l'évaluation
de la résistance des systèmes biométriques aux attaques par
morphing
Reference number
© ISO/IEC 2025
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
© ISO/IEC 2025 – All rights reserved
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms. 4
5 Morphing attacks . 6
6 Measuring and reporting morphing attack potential . 7
6.1 Morphing attack potential .7
6.2 Multiple contributing subject generalization .9
6.3 Visualisation .9
6.4 Benchmarking of morphing methods and impacting factors . 12
7 Morph detection error rates .12
Annex A (informative) Reference implementation .13
Annex B (informative) Example of morphed sample visualization. 14
Bibliography .16

© ISO/IEC 2025 – All rights reserved
iii
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical activity.
ISO and IEC technical committees collaborate in fields of mutual interest. Other international organizations,
governmental and non-governmental, in liaison with ISO and IEC, also take part in the work.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of document should be noted. This document was drafted in accordance with the editorial rules of the ISO/
IEC Directives, Part 2 (see www.iso.org/directives or www.iec.ch/members_experts/refdocs).
ISO and IEC draw attention to the possibility that the implementation of this document may involve the
use of (a) patent(s). ISO and IEC take no position concerning the evidence, validity or applicability of any
claimed patent rights in respect thereof. As of the date of publication of this document, ISO and IEC had not
received notice of (a) patent(s) which may be required to implement this document. However, implementers
are cautioned that this may not represent the latest information, which may be obtained from the patent
database available at www.iso.org/patents and https://patents.iec.ch. ISO and IEC shall not be held
responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO's adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT) see www.iso.org/iso/foreword.html.
In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology,
Subcommittee SC 37, Biometrics.
Any feedback or questions on this document should be directed to the user’s national standards
body. A complete listing of these bodies can be found at www.iso.org/members.html and
www.iec.ch/national-committees.

© ISO/IEC 2025 – All rights reserved
iv
Introduction
Many application processes for ID documents do not implement trusted capture. For example, as long as
printed biometric samples (e.g. face images) are accepted, morphing attacks, where biometric references
are manipulated to match two or more biometric data subjects submitted during enrolment, pose a threat
to image-based biometric systems (BSs). Morphing attack detection is possible, though the ability to detect
morphing attacks can differ based on the morphing attack techniques.
Not all morphing techniques pose the same risk for an operational BS. This document establishes morphing
attack potential (MAP) as a measure of the capability of a class of morphing attacks to deceive one or more BSs.
The user of this document can simulate a real use case such as issuance of documents or border control.
The use case can consider a variable number of attempts and BSs to determine the MAP against automated
border control (ABC) gates from different vendors.
NOTE The evaluation of the resistance of a BS is not a security evaluation.

© ISO/IEC 2025 – All rights reserved
v
International Standard ISO/IEC 20059:2025(en)
Information technology — Methodologies to evaluate the
resistance of biometric systems to morphing attacks
1 Scope
This document establishes a methodology to evaluate the resistance of BSs to morphing attacks, including
multiple identity attacks. The document is limited to image-based morphing attacks. The term "image-
based" includes modalities such as face, iris and finger image data.
The document establishes:
— a definition of biometric sample modifications and manipulation with a specific focus on manipulations
that constitute a multiple identity attack. This can be, for instance, an enrolment attack with face image
morphing;
— a methodology to measure the morphing attack potential of a morphing method.
The document also describes how morphing algorithms can be used for system evaluation.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements of this document. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
ISO/IEC 2382-37, Information technology — Vocabulary — Part 37: Biometrics
ISO/IEC 30107-1, Information technology — Biometric presentation attack detection — Part 1: Framework
ISO/IEC 30107-3, Information technology — Biometric presentation attack detection — Part 3: Testing and
reporting
ISO/IEC 39794-5, Information technology — Extensible biometric data interchange formats — Part 5: Face
image data
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 2382-37, ISO/IEC 30107-1,
ISO/IEC 30107-3, ISO/IEC 39794-5 and the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1
image content
visual information within an image, such as the face of a subject
Note 1 to entry: Artefacts like an iris shadow, caused by a poor quality morphing method, are potentially contained in
the image content.
Note 2 to entry: Artefacts by compression also belong to the image content.

© ISO/IEC 2025 – All rights reserved
3.2
image generation
creation of still or animated content with imaging software
Note 1 to entry: For instance, the generation of a synthetic image (e.g. with GANs) and subsequent morphing of the
accomplice image with the synthetic.
Note 2 to entry: Morphing can be a part of that process.
3.3 Terms related to image substitution
3.3.1
image substitution attack
replacement of the printed image in the physical passport booklet with the intention to fool the human
examiner
Note 1 to entry: Could also be used to fool the face recognition system, if the image from the passport data page is
scanned.
Note 2 to entry: The attack takes the target photo and puts parts of it on top of the original printed image on the data page.
Note 3 to entry: The image in the chip is not affected by the attack.
3.3.2
image substitution attack detection
revealing deviations to expected properties of the original portrait image area
Note 1 to entry: An image substitution attack can cause image artefacts in the scanned facial image or its surroundings.
Note 2 to entry: The expected UV pattern cannot be observed.
Note 3 to entry: This is complementary action to other measures, like validating the document numbers.
3.4 Terms related to image manipulation
3.4.1
image modification
act of effect of changing the image content or metadata of an image
Note 1 to entry: Typical signal modifications are beautifications, compression, sharpening, contrast enhancement,
cropping, geometry change.
Note 2 to entry: Typical metadata modification could be change of location, date or time of capturing.
3.4.2
image manipulation
act of effect of intentionally altering the visual appearance or specific properties of an image resulting in
misrepresentation or misinterpretation
Note 1 to entry: The difference to an image modification is the intention of the malicious actor.
Note 2 to entry: A manipulation is a subset of a modification.
3.4.3
digital image manipulation
act or effect of intentionally altering digitally the visual appearance or specific properties of an image
resulting in misrepresentation or misinterpretation
Note 1 to entry: Alteration can be morphing of two parent images or replacing of certain parts/regions of the image.
Note 2 to entry: Manipulation can target elements of metadata (e.g. the capture date field).

© ISO/IEC 2025 – All rights reserved
3.4.4
biometric image manipulation
image modification intended to influence either the output of a biometric system or the decision of a human
examiner, or both
Note 1 to entry: Possible intentions are for criminal attacks (impersonation) or for protecting privacy (avoiding
recognition by means of de-identification).
Note 2 to entry: Using a filter on the face represented in the image, can influence the error rates of the biometric
[5]
system (e.g. increased false reject rate or false accept rate, or both) .
Note 3 to entry: Alteration before the capture process (e.g. manipulating the facial appearance with makeup) is not a
digital manipulation. This is a presentation attack as defined in ISO/IEC 30107-1.
3.4.5
biometric image manipulation attack
submission of an image containing a manipulated representation of a biometric trait to the identity
document application process with the goal of interfering with the operation of either the biometric system
or the human examiner, or both
3.4.6
image manipulation attack detection
detecting traces of image manipulation conducted by either an algorithm or a human examiner, or both
Note 1 to entry: Detection algorithms typically operate on the suspected images, potentially supported with a trusted
live capture image.
Note 2 to entry: The detection algorithm has typically no information about the enrolment process (i.e. the details of
the attack vector).
3.5 Terms related to image morphing
3.5.1
biometric morphing
combining two or more biometric samples into one signal
Note 1 to entry: A biometric sample is defined in ISO/IEC 2382-37.
3.5.2
face image morphing
morphing process executed with facial portrait images
3.5.3
face image morphing attack
biometric image manipulation attack based on morphing two or more facial images
Note 1 to entry: The morphing can be executed on the holistic facial image or on selected areas of interest (e.g. the
periocular region).
3.5.4
morphing attack detection
MAD
observing a biometric morphing attack through either an algorithmic or a human method, or both
Note 1 to entry: The attack detection can be conducted based on a single image (single image morphing attack
detection, i.e. S-MAD) or based on a pair of images (differential image morphing attack detection, i.e. D-MAD).
3.5.5
single image morphing attack detection
S-MAD
morphing attack detection that is based on a single image

© ISO/IEC 2025 – All rights reserved
3.5.6
differential image morphing attack detection
D-MAD
morphing attack detection that is based on a pair of images
Note 1 to entry: D-MAD can be implemented by quantifying the similarity between a suspected morph image and a
trusted-capture (i.e. bona fide image).
3.5.7
multiple identity attack
MIA
biometric morphing attack with the intention of obtaining an identity document that can be successfully
used by multiple subjects
3.5.8
morphing attack potential
MAP
measure of the capability of a class of morphing attacks to deceive one or more biometric systems using
multiple recognition attempts
3.5.9
bona fide sample classification error rate
BSCER
proportion of bona fide samples incorrectly classified as morphed samples in a specific scenario
3.5.10
morphing attack classification error rate
MACER
proportion of morphed samples incorrectly classified as bona fide samples in a specific scenario
3.5.11
morphing attack classification error rate at a given morphing attack potential
MACER
MAPr[],c
MACER computed on the subset of morphed images that can successfully reach a match decision with both
contributing subjects in at least r verification attempts by at least c biometric systems
Note 1 to entry: See Clause 7 for further details.
4 Symbols and abbreviated terms
The following abbreviated terms and symbols are used in this document.
ABC automated border control
BS biometric system
BSCER bona fide sample classification error rate
D-MAD differential image morphing attack detection
FAR false accept rate
FRR false reject rate
FMMPMR fully mated morph presentation match rate
MACER morphing attack classification error rate
MAD morphing attack detection
© ISO/IEC 2025 – All rights reserved
MAP morphing attack potential
MIA multiple identity attack
MMPMR mated morph presentation match rate
S-MAD single image morphing attack detection
V
a generic vulnerability indicator
D
a set of morphed samples
M
M
a morphed sample
a condition that the morphed sample M must satisfy in the computation of the vul-
CM()
V
nerability indicator V
D
a set of probe samples
P
P
a probe sample
D
a set of biometric systems
F
F
a biometric system
the similarity (or dissimilarity) score between the morphed sample M and the probe
sM ,P
()
F
sample P using the biometric system F
the threshold used by the biometric system F to determine whether a similarity (or
τ
F
dissimilarity) score indicates that two biometric samples have the same biometric
source
a function that returns the comparison decision made by the biometric system F
mM ,,PF
()
when comparing the morphed sample M and the probe sample P
mc MD,,F a function that counts the number of probe samples in D that are successfully ver-
()
P P
ified against the morphed sample M by the biometric system F
m
in the computation of the MAP metric, it is the number of probe samples for each
contributing subject
n
the number of biometric systems used in the computation of the MAP metric
the value of the element of the MAP matrix at row r and column c
MAPr ,c
[]
fmcM(),,DD ,r a function that returns the number of biometric systems in D for which at least r
PF F
probe samples in D are successfully verified against the morphed image M
P
N
the number of subjects involved in the generation of a morphed sample
S
the weight factor assigned to all values in the column c of a MAP matrix in the com-
wc[]
col
putation of the robustness curve and the weighted average MAP
the weight factor assigned to all values in the row r of a MAP matrix in the compu-
wr
[]
row
tation of the generality curve and c
R
the robustness curve derived from a MAP matrix
MAP[]r
G
the generality curve derived from a MAP matrix
MAP[]c
© ISO/IEC 2025 – All rights reserved
MAP
the weighted average MAP
Avg
N
the number of bona fide samples used to compute the BSCER and MACER metrics
BF
N
the number of morphed samples used to compute the BSCER and MACER metrics
M
th
Res
i
a function used to compute the BSCER and MACER metrics that returns 1 if the i
sample is classified as morphed and 0 if it is classified as a bona fide
D the subset of morphed image in D for which fmcM,,DD ,r is above or equal to
()
MAP,rc M PF
[]
c for both contributing subjects
5 Morphing attacks
Biometric recognition is nowadays widely in use in border control applications, both automat
...