Information technology - Conformance testing methodology for biometric data interchange formats defined in ISO/IEC 19794 - Part 5: Face image data

This document specifies elements of conformance testing methodology, test assertions, and test procedures as applicable to two-dimensional face images defined in the ISO/IEC 19794-5:2005 biometric data interchange format standard for face image data. This document establishes - test assertions of the structure of the face image data format as specified in ISO/IEC 19794-5:2005 (Type A Level 1 as defined in ISO/IEC 29109-1:2009), - test assertions of internal consistency by checking the types of values that may be contained within each field (Type A Level 2 as defined in ISO/IEC 29109-1:2009). This document does not establish - tests of conformance of 3D face records defined in ISO/IEC 19794-5:2005, 5.7.1, codes 0x80, 0x81, and 0x82, - tests of conformance of CBEFF structures required by ISO/IEC 19794-5:2005, - tests of consistency with the input biometric data record (Level 3), - tests of conformance of the image data to the quality-related specifications of ISO/IEC 19794-5:2005, - tests of conformance of the image data blocks to the respective JPEG or JPEG 2000 standards, - tests of other characteristics of biometric products or other types of testing of biometric products (e.g., acceptance, performance, robustness, security).

Technologies de l'information — Méthodologie d'essai de conformité pour les formats d'interéchange de données biométriques définis dans l'ISO/IEC 19794 — Partie 5: Données d'image de la face

General Information

Status
Published
Publication Date
20-May-2019
Current Stage
9093 - International Standard confirmed
Start Date
06-Sep-2024
Completion Date
30-Oct-2025
Ref Project

Relations

Overview

ISO/IEC 29109-5:2019 defines a conformance testing methodology specifically for two‑dimensional face image data as defined in ISO/IEC 19794‑5:2005. The standard specifies the framework, test assertions, and test procedures used to verify that binary face image records conform to the structure and internal consistency requirements of the ISO/IEC 19794‑5 face biometric data interchange format. It targets Type A Level 1 (structure) and Type A Level 2 (internal consistency) tests as defined in ISO/IEC 29109‑1:2009.

Key topics and requirements

  • Conformance testing methodology: Applies the general conformance framework in ISO/IEC 29109‑1 for face image records, detailing how tests should be organized and executed.
  • Test assertions (Level 1): Assertions that validate the structural elements of the face image record (header, blocks, multi‑byte ordering, record length and counts).
  • Test assertions (Level 2): Assertions that verify internal consistency and data types of fields (e.g., numeric formats, field value ranges and bit masks).
  • Mandatory versus optional components: Supports reporting which optional components of ISO/IEC 19794‑5 are implemented by the item under test (IUT) via an Implementation Conformance Statement.
  • Scope exclusions: Explicitly does not include tests for 3D face records, CBEFF structure conformance, Level 3 consistency with input biometric data, image quality requirements, conformance of JPEG/JPEG2000 image blocks to their respective image standards, or other product‑level tests (performance, robustness, security).

Applications and who uses it

ISO/IEC 29109‑5:2019 is intended for:

  • Conformance testing laboratories and certification bodies creating test suites for face image data interchange.
  • Biometric developers and vendors who need to validate that their face image record implementations conform to ISO/IEC 19794‑5.
  • Systems integrators and procurement teams requiring interoperability assurances for biometric components used in border control, digital ID, access control, and law enforcement systems.
  • Standards and compliance engineers designing test methods, test harnesses, or automated validators for face image data.

Practical benefits include reliable biometric interoperability, repeatable conformance test results, and a standardized approach for declaring supported features and limitations.

Related standards

  • ISO/IEC 19794‑5:2005 - Biometric data interchange formats - Part 5: Face image data (base format)
  • ISO/IEC 29109‑1:2009 - Generalized conformance testing methodology for ISO/IEC 19794 series
  • CBEFF (Common Biometric Exchange Formats Framework) - contextual framework used with ISO face records

Keywords: ISO/IEC 29109‑5:2019, face image data, conformance testing, biometric data interchange formats, ISO/IEC 19794‑5, test assertions, interoperability.

Standard
ISO/IEC 29109-5:2019 - Information technology — Conformance testing methodology for biometric data interchange formats defined in ISO/IEC 19794 — Part 5: Face image data Released:5/21/2019
English language
26 pages
sale 15% off
Preview
sale 15% off
Preview

Frequently Asked Questions

ISO/IEC 29109-5:2019 is a standard published by the International Organization for Standardization (ISO). Its full title is "Information technology - Conformance testing methodology for biometric data interchange formats defined in ISO/IEC 19794 - Part 5: Face image data". This standard covers: This document specifies elements of conformance testing methodology, test assertions, and test procedures as applicable to two-dimensional face images defined in the ISO/IEC 19794-5:2005 biometric data interchange format standard for face image data. This document establishes - test assertions of the structure of the face image data format as specified in ISO/IEC 19794-5:2005 (Type A Level 1 as defined in ISO/IEC 29109-1:2009), - test assertions of internal consistency by checking the types of values that may be contained within each field (Type A Level 2 as defined in ISO/IEC 29109-1:2009). This document does not establish - tests of conformance of 3D face records defined in ISO/IEC 19794-5:2005, 5.7.1, codes 0x80, 0x81, and 0x82, - tests of conformance of CBEFF structures required by ISO/IEC 19794-5:2005, - tests of consistency with the input biometric data record (Level 3), - tests of conformance of the image data to the quality-related specifications of ISO/IEC 19794-5:2005, - tests of conformance of the image data blocks to the respective JPEG or JPEG 2000 standards, - tests of other characteristics of biometric products or other types of testing of biometric products (e.g., acceptance, performance, robustness, security).

This document specifies elements of conformance testing methodology, test assertions, and test procedures as applicable to two-dimensional face images defined in the ISO/IEC 19794-5:2005 biometric data interchange format standard for face image data. This document establishes - test assertions of the structure of the face image data format as specified in ISO/IEC 19794-5:2005 (Type A Level 1 as defined in ISO/IEC 29109-1:2009), - test assertions of internal consistency by checking the types of values that may be contained within each field (Type A Level 2 as defined in ISO/IEC 29109-1:2009). This document does not establish - tests of conformance of 3D face records defined in ISO/IEC 19794-5:2005, 5.7.1, codes 0x80, 0x81, and 0x82, - tests of conformance of CBEFF structures required by ISO/IEC 19794-5:2005, - tests of consistency with the input biometric data record (Level 3), - tests of conformance of the image data to the quality-related specifications of ISO/IEC 19794-5:2005, - tests of conformance of the image data blocks to the respective JPEG or JPEG 2000 standards, - tests of other characteristics of biometric products or other types of testing of biometric products (e.g., acceptance, performance, robustness, security).

ISO/IEC 29109-5:2019 is classified under the following ICS (International Classification for Standards) categories: 35.240.15 - Identification cards. Chip cards. Biometrics. The ICS classification helps identify the subject area and facilitates finding related standards.

ISO/IEC 29109-5:2019 has the following relationships with other standards: It is inter standard links to ISO/IEC 29109-5:2014. Understanding these relationships helps ensure you are using the most current and applicable version of the standard.

You can purchase ISO/IEC 29109-5:2019 directly from iTeh Standards. The document is available in PDF format and is delivered instantly after payment. Add the standard to your cart and complete the secure checkout process. iTeh Standards is an authorized distributor of ISO standards.

Standards Content (Sample)


INTERNATIONAL ISO/IEC
STANDARD 29109-5
Fourth edition
2019-05
Information technology —
Conformance testing methodology for
biometric data interchange formats
defined in ISO/IEC 19794 —
Part 5:
Face image data
Technologies de l'information — Méthodologie d'essai de conformité
pour les formats d'interéchange de données biométriques définis dans
l'ISO/IEC 19794 —
Partie 5: Données d'image de la face
Reference number
©
ISO/IEC 2019
© ISO/IEC 2019
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: +41 22 749 01 11
Fax: +41 22 749 09 47
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO/IEC 2019 – All rights reserved

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms . 2
5 Conformance . 2
6 Conformance testing methodology . 2
6.1 Overview . 2
6.2 Table of requirements in the base standard . 2
6.3 ISO/IEC 19794-5:2005 Level 1 and 2 test assertions .12
Bibliography .26
© ISO/IEC 2019 – 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).
Attention is drawn to the possibility that some of the elements of this document may be the subject
of patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent
rights. Details of any patent rights identified during the development of the document will be in the
Introduction and/or on the ISO list of patent declarations received (see www .iso .org/patents) or the IEC
list of patent declarations received (see http: //patents .iec .ch).
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.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology, SC 37,
Biometrics.
This fourth edition cancels and replaces the third edition (ISO/IEC 29109-5:2014), which has been
technically revised.
The main changes since the last edition include technical revision of the tables.
A list of all the parts in the ISO/IEC 29109 series, can be found on the ISO website.
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.
iv © ISO/IEC 2019 – All rights reserved

Introduction
ISO/IEC 19794-5:2005 specifies a data record interchange format for storing, recording, and
transmitting one or more face images within a Common Biometric Exchange Formats Framework
(CBEFF) data structure. Each image is accompanied by subject-specific and image-specific metadata
contained in a header record. This part of the ISO/IEC 29109 series establishes tests for checking the
correctness of the binary record.
The objective of ISO/IEC 19794-5:2005 cannot be completely achieved until biometric products can
be tested to determine whether they conform to those specifications. Conforming implementations
are a necessary prerequisite for achieving interoperability among implementations; therefore there
is a need for a standardized conformance testing methodology, test assertions, and test procedures
as applicable to specific modalities addressed by each part of ISO/IEC 19794. The test assertions will
cover as much as practical of the ISO/IEC 19794 requirements (covering the most critical features),
so that the conformity results produced by the test suites will reflect the real degree of conformity of
the implementations to ISO/IEC 19794 Data Interchange Format records. This is the motivation for the
development of this conformance testing methodology.
This document supports those applications that require use of face image data according to ISO/IEC
19794-5:2005. It defines a testing methodology to ensure conformance of a vendor’s application or
service to the base ISO/IEC 19794-5:2005 specification. Thus this document is intended to:
— establish elements of the Conformance Testing Methodology framework that are specific to the Face
Image-based Data Record requirements of ISO/IEC 19794-5:2005 conformance testing;
— define requirements and guidelines for specifying conformance test suites and related test
methods for measuring conformity of products and services to the Face Image-based Data Record
requirements of ISO/IEC 19794-5:2005; and
— define test procedures to be followed before, during, and after conformance testing.
This document is applicable to the development and use of conformity test method specifications,
conformity test suites for ISO/IEC 19794-5:2005 records, and conformance testing programs for ISO/
IEC 19794-5:2005 conformant products. It is intended primarily for use by testing organizations, but
may be applied by developers and users of test method specifications and test method implementations.
© ISO/IEC 2019 – All rights reserved v

INTERNATIONAL STANDARD ISO/IEC 29109-5:2019(E)
Information technology — Conformance testing
methodology for biometric data interchange formats
defined in ISO/IEC 19794 —
Part 5:
Face image data
1 Scope
This document specifies elements of conformance testing methodology, test assertions, and test
procedures as applicable to two-dimensional face images defined in the ISO/IEC 19794-5:2005
biometric data interchange format standard for face image data.
This document establishes
— test assertions of the structure of the face image data format as specified in ISO/IEC 19794-5:2005
(Type A Level 1 as defined in ISO/IEC 29109-1:2009),
— test assertions of internal consistency by checking the types of values that may be contained within
each field (Type A Level 2 as defined in ISO/IEC 29109-1:2009).
This document does not establish
— tests of conformance of 3D face records defined in ISO/IEC 19794-5:2005, 5.7.1, codes 0x80, 0x81,
and 0x82,
— tests of conformance of CBEFF structures required by ISO/IEC 19794-5:2005,
— tests of consistency with the input biometric data record (Level 3),
— tests of conformance of the image data to the quality-related specifications of ISO/IEC 19794-5:2005,
— tests of conformance of the image data blocks to the respective JPEG or JPEG 2000 standards,
— tests of other characteristics of biometric products or other types of testing of biometric products
(e.g., acceptance, performance, robustness, security).
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 19794-5:2005, Information technology — Biometric data interchange formats — Part 5: Face
image data
ISO/IEC 29109-1:2009, Information technology — Conformance testing methodology for biometric data
interchange formats defined in ISO/IEC 19794 — Part 1: Generalized conformance testing methodology<
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 29109-1 apply.
© ISO/IEC 2019 – All rights reserved 1

ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— IEC Electropedia: available at http: //www .electropedia .org/
— ISO Online browsing platform: available at http: //www .iso .org/obp
4 Symbols and abbreviated terms
For the purposes of this document, the symbols and abbreviated terms given in ISO/IEC 29109-1 apply.
5 Conformance
Biometric data interchange format conformance tests conform to this document if they satisfy all of the
normative requirements related to Clause 6. Specifically, they shall use the test methodology specified
in ISO/IEC 29109-1:2009, Clauses 6, 7 and 8, and all Level 1 and Level 2 tests shall use the assertions
defined in Table 2 of this document.
Implementations of ISO/IEC 19794-5:2005 tested according to the methodology specified shall be able
to claim conformance only to those biometric data record requirements specified in ISO/IEC 19794-
5:2005 that are tested by the test methods established by this methodology.
Implementations of ISO/IEC 19794-5:2005 do not necessarily need to conform to all possible aspects
of ISO/IEC 19794-5:2005, but only to those ISO/IEC 19794-5:2005 requirements that are claimed to
be supported by the implementation in an Implementation Conformance Statement, filled out in
accordance with ISO/IEC 29109-1:2009, Clause 8 and Table 1 of this document.
6 Conformance testing methodology
6.1 Overview
The testing methodology specified in ISO/IEC 29109-1:2009, Clauses 6, 7 and 8 shall apply. The content
of the tables below is based on the conformance testing methodology outlined in ISO/IEC 29109-1 and
shall only be used in the context of that testing methodology.
6.2 Table of requirements in the base standard
The normative requirements of ISO/IEC 19794-5:2005 are listed in Table 1. The supplier of the IUT can
explain which optional components of the standard are supported and the testing laboratory can note
the results of the test.
Under subformat applicability the columns labelled B, F and T indicate the Basic, Full Frontal and Token
Frontal image types.
2 © ISO/IEC 2019 – All rights reserved

Table 1 — Requirements of the base international standard (ISO/IEC 19794-5:2005)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-1 5.2.1 Within the record format and all well-defined data blocks 1 M Y Y Y
therein, all multi-byte quantities are [implied shall] stored in
Big-Endian format. That is, the more significant bytes of any
multi-byte quantity are stored at lower addresses in memory
than less significant bytes. For example, the value 1 025 (2
to the 10th power plus one) would be stored as first byte =
00000100 and second byte =00000001.
R-2 5.2.2 Numeric Values 3C O-3 Y Y Y
All numeric values are fixed-length unsigned integer quanti-
ties, unless otherwise specified.
R-3 5.4.1 Format Identifier 1 M Y Y Y
The (4 byte) Format Identifier shall consist of three ASCII
characters “FAC” followed by a zero byte as a NULL string ter-
minator to identify the record format as the face record format.
R-4 5.4.2 Version Number 1 M Y Y Y
The (4 byte) Version Number block shall consist of three ASCII
numerals followed by a zero byte as a NULL string terminator.
The first and second character will represent the major ver-
sion number and the third character will represent the minor
revision number.
The version number of this specification shall be 0x30313000;
“010” – Version 1 revision 0.
The version number in some implementations may also be
0x30323000, that is, “020”, however this setting is deprecated.
R-5 Table 2 57 ≤ Length of Record ≤ 2 − 1 1 M Y Y Y
R-6 5.4.3 Length of Record 2 M Y Y Y
The (4 byte) Record Length Block shall be the combined length
in bytes for the record. This is the entire length of the record
including the Facial Record Header and Facial Record Data.
R-7 Table 2 1 ≤ Number of Facial Images ≤ 65 535 1 M Y Y Y
R-8 5.4.4 Number of Facial Images 2 M Y Y Y
The (2 byte) Number of Facial Images block shall be the num-
ber of facial images included in the record.
R-9 5.5 The Facial Information block 2 M-2 Y Y Y
The (20 byte) Facial Information block is intended to describe
discrete properties of the individual discernable from the
image, one is included for each facial image included in the
record. The structure of this block is shown in
[ISO/IEC 19794-5:2005] Figure 2.
Zero or more Facial Landmark blocks, one Image Information
block, and one Image Data block follow this block.
R-10 5.5.1 Facial Record Data Length 2 M Y Y Y
The (4 byte) Facial Record Data Length field denotes the sum
of the lengths of the Facial Information block, the Landmark
Point block(s), the Image Information block, and the Image
Data block.
The minimum value of the Facial Record Data Length is
32 bytes plus the size of the Image Data block (in bytes).
R-11 5.5.2 Number of Landmark Points 2 M Y Y Y
The (2 byte) Number of Landmark Points block shall be the
number of Landmark Point blocks that follow the Facial Infor-
mation block. The Landmark Point block is defined in
[ISO/IEC 19794-5:2005] Clause 5.6.
R-12 5.5.3 Gender 1 M Y Y Y
The (1 byte) Gender block shall be specified in accordance
with [ISO/IEC 19794-5:2005] Table 3.
© ISO/IEC 2019 – All rights reserved 3

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-13 5.5.4 Eye Colour 1 M Y Y Y
The (1 byte) Eye Colour field shall represent the colour of iris-
es of the eyes according to [ISO/IEC 19794-5:2005] Table 4. If
the eyes are different colours, then right eye colour is to be
encoded.
R-14 5.5.5 Hair Colour 1 M Y Y Y
The (1 byte) Hair Colour field shall represent the colour of the
hair according to the [ISO/IEC 19794-5:2005] Table 5.
R-15 5.5.6 Property Mask 2 M Y Y Y
The (3 byte) Property Mask is a bit mask of 3 bytes and each
bit of the mask position listed in [ISO/IEC 19794-5:2005]
Table 6 shall be set to 1 if the corresponding property is pres-
ent, and set to 0 if absent. The mask position starts from 0 at
the lowest bit. The lowest bit set to 0 shall indicate that prop-
erties are not specified (and all bits shall be zero); the lowest
bit set to 1 shall indicate that all listed properties have been
considered and that a zero value of any property bit indicates
an absence of that property.
Note that a Blink flag set to “1” will indicate non-compliance
with the Frontal, Full Frontal, and Token image types.
R-16 5.5.7 Expression 1 M Y Y Y
The (2 byte) Expression field shall represent the expression of
the face according to [ISO/IEC 19794-5:2005] Table 7.
R-17 5.5.8 Pose Angle 3C O-1 Y Y Y
The (3 multi byte) Pose Angle field (B , B , B ) shall represent
Y P R
the estimate or measure pose of the subject in the image.
Each byte in the field respectively represents pose angles of
yaw, pitch and roll in that order. The pose angle is given by
Tait-Bryan angles.
—  Yaw angle: Rotation about the vertical (y) axis.
—  Pitch angle: Rotation about the horizontal side-to-side
(x) horizontal axis.
—  Roll angle: Rotation about the horizontal back to front
(z) axis.
The angles are defined relative to the frontal view of the sub-
ject, which has angles (0,0,0) as shown in
[ISO/IEC 19794-5:2005] Figure 5. The examples are shown in
[ISO/IEC 19794-5:2005] Figure 6.
As order of the successive rotation around the different axes
does matter, the encoded rotation angle shall correspond to
an order of execution starting from the frontal view. This
order shall be given by Roll (about the front axis), then Pitch
(about the horizontal axis) and finally Yaw (about the vertical
axis). The (first executed) Roll transformation will therefore
always be in the image (x,y) plane.
From the point of view of executing a transformation from the
observed view to a frontal view, the transformation order will
therefore be Yaw, Pitch, and then Roll. Note however that the
encoded angle is from the frontal view to the observed view.
4 © ISO/IEC 2019 – All rights reserved

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-18 5.5.8.1 Pose Angle – Yaw 1 M Y Y Y
The yaw angle, Y, is the rotation in degrees about the y-axis
(vertical axis) shown in [ISO/IEC 19794-5:2005] Figure 5.
Frontal faces have a yaw angle of 0 degrees. Positive angles
represent faces looking to their left (a counter-clockwise
rotation around the y-axis).
“The encoded value, BY, shall be stored in 1 byte with values 0
to 180 computed from a real-valued yaw angle estimate,
−180 ≤ Y < 180, as follows:”
If 180 ≥ Y ≥ 0 and Y is even, then B = Y/2 + 1.
Y
If 180 ≥ Y > 0 and Y is odd, then B = (Y+1)/2.
Y
If −180 ≤ Y < 0 and Y is even, then B = 181 + Y/2.
Y
If −180 ≤ Y < 0 and Y is odd, then By = 181+ (Y-1)/2.
The maximum value of B is 180. If the yaw angle is not speci-
Y
fied, the value of B shall be 0.
Y
R-19 5.5.8.2 Pose Angle – Pitch 1 M Y Y Y
The pitch angle, P, is the rotation in degrees about the x-axis
(horizontal axis) shown in [ISO/IEC 19794-5:2005] Figure 5.
Frontal faces have a pitch angle of 0 degrees. Positive angles
represent faces looking down (a counterclockwise rotation
around the x-axis).
The encoded value, B , shall be stored in 1 byte with values 0
P
to 180 computed from a real-valued pitch angle estimate,
−180 ≤ P < 180, as follows:
If 180 ≥ P ≥ 0 and P is even, then B = P/2 + 1.
P
If 180 ≥ P > 0 and P is odd, then B = (P + 1)/2.
P
If -180 ≤ P < 0 and P is even, then B = 181 + P/2.
P
If -180 ≤ P < 0 and P is odd, then B = 181 + (P − 1)/2.
P
The maximum value of B is 180. If the pitch angle is not
P
specified, the value of B shall be 0.
P
R-20 5.5.8.3 Pose Angle – Roll 1 M Y Y Y
The roll angle, R, is the rotation in degrees about the z-axis
(the horizontal axis from front to back) shown in [ISO/IEC
19794-5:2005] Figure 5. Frontal faces have a roll angle of 0 de-
grees. Positive angles represent faces tilted toward their right
shoulder (counter-clockwise rotation around the z-axis).
The encoded value, B , shall be stored in 1 byte with values 0
R
to 180 computed from a real-valued roll angle estimate, −180
≤ R < 180, as follows:
If 180 ≥ R ≥ 0 and R is even, then B = R/2 + 1.
R
If 180 ≥ R > 0 and R is odd, then B = (R + 1)/2.
R
If -180 ≤ R < 0 and R is even, then B = 181 + R/2.
R
If -180 ≤ R < 0 and R is odd, then B = 181 + (R-1)/2.
R
The maximum value of B is 180. If the roll angle is not speci-
R
fied, the value of B shall be 0.
R
© ISO/IEC 2019 – All rights reserved 5

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-21 5.5.9 Pose Angle Uncertainty 1 M Y Y Y
The (3 multi-byte) Pose Angle Uncertainty (U , U , U ) repre-
Y P R
sents the expected degree of accuracy of the pose angle yaw,
pitch, and roll. Each byte in the field respectively represents
the uncertainty of yaw, pitch and roll in that order. The
uncertainty is allowed to represent experimental uncertainty
specified by each vendor.
The encoding of Pose Angle Uncertainty is given by bytes (U ,
Y
U , U ) where each byte U in the field (k = Y,P,R) repre-
P R K
sents degree of uncertainty with minimum and maximum
values of 1 and 181 where U =(uncertainty+1). The more
K
uncertain, the value of the uncertainty U shall become larger.
K
If the uncertainty is not specified, then the values of U , U
Y P
and U shall be set to zero (0).
R
R-22 5.6 The Landmark Point Block 1 M Y Y Y
The optional (8 byte) Landmark Point block specifies the type,
code and position of a Landmark Point in the facial image. The
number of Landmark Point blocks shall be specified in the
Number of Landmark Points field of the Facial Information
Block. The structure of this block is shown in
[ISO/IEC 19794-5:2005] Table 8.
R-23 5.6.1 Landmark Point Type 1 M Y Y Y
The (1 byte) Landmark Point Type field represents the type
of the Landmark Point stored in the Landmark Point block.
This field shall be set to 0x01 to denote that the position of
the Landmark Point is represented by the coordinate of the
image. All other field values are reserved for future definition
of Landmark Point types.
R-24 5.6.2 Landmark Point Code 1 M Y Y Y
The (1 byte) Landmark Point Code field shall specify the
Landmark Point that is stored in the Landmark Point block.
The codes of the Landmark Points in [ISO/IEC 19794-5:2005]
Clause 5.6.3, taken from the MPEG4 standard and defined as
MPEG4 Landmark Points, or the additional eye and nostril
Landmark Points in [ISO/IEC 19794-5:2005]
Clause 5.6.4 shall be stored in this block.
Each Landmark Point code is represented by a notation A.B
using a major (A) and a minor (B) value. The encoding of the
Landmark Point code is given by the (1 byte) value of A*16 + B.
R-25 5.6.3 MPEG4 Landmark Points 3C O-1 Y Y Y
[ISO/IEC 19794-5:2005] Figure 7 denotes the Landmark Point
codes associated with Landmark Points as given by Annex C of
ISO/IEC 14496-2.
R-26 5.6.4 Eye and nostril centre Landmark Points 3C O-1 Y Y Y
The eye centre Landmark Points 12.1 (left) and 12.2 (right)
are defined to be the horizontal and vertical midpoints of the
eye corners (3.7, 3.11) and (3.8, 3.12) respectively. The left
nostril centre Landmark Point 12.3 is defined to be the mid-
point of the nose Landmark Points (9.1, 9.15) in the horizontal
direction and (9.3,9.15) in the vertical direction. Similarly, the
right nostril centre Landmark Point 12.4 is defined to be the
midpoint of the nose Landmark Points (9.2, 9.15) in the hori-
zontal direction and (9.3,9.15) in the vertical direction. Both
the eye centre and nostril centre Landmark Points are shown
in [ISO/IEC 19794-5:2005] Figure 8 and values given in
[ISO/IEC 19794-5:2005] Table 10.
R-27 5.7 The Image Information Block 2 M-2 Y Y Y
The (16 byte) Image Information block is intended to describe
digital properties of the facial image, one is included for each
facial image included in the record. The structure of this block
is shown in [ISO/IEC 19794-5:2005] Figure 2. One Image Data
block shall follow this block.
6 © ISO/IEC 2019 – All rights reserved

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-28 5.7.1 Face Image Type 1 M Y Y Y
The Face Image Type field shall represent the type of the facial
image stored in the Image Data Block according to
[ISO/IEC 19794-5:2005] Table 12. Note that all Frontal Image
Types are either Full Frontal or Token Frontal. Therefore a
separate Frontal Value is not required.
The Basic Face Image Type is defined in
[ISO/IEC 19794-5:2005] Clause 6. The Frontal, Frontal/Full
and Frontal/Token Face Image Types are defined in
[ISO/IEC 19794-5:2005] Clauses 7, 8, and 9 respectively. Face
Image Types use the notion of inheritance. For example, the
Frontal Face Image Type inherits all of the requirements of
the Basic Face Image Type - the Frontal Face Image type obeys
all normative requirements of the Basic Face Image Type. The
inheritance structure of defined image types is shown in
[ISO/IEC 19794-5:2005] Figure 10.
R-29 5.7.2 Image Data Type 2 M Y Y Y
The (1 byte) Image Data Type field denotes the encoding type
of the Image Data block. Either JPEG (ISO/IEC 10918-1 and
ITU-T Rec. T.81) or JPEG2000 (ISO/IEC 15444-1) shall be
specified. Note that a “Unspecified” Value cannot be encoded.
R-30 5.7.3 Width 2 M Y Y Y
The (2 byte) Width field shall specify the number of pixels in
the horizontal direction.
R-31 5.7.4 Height 2 M Y Y Y
The (2 byte) Height field shall specify the number of pixels in
the vertical direction.
R-32 5.7.5 Image Colour Space 2 M Y Y Y
The (1 byte) Image Colour Space field indicates the colour
space used in the encoded Image Data block according to the
values in [ISO/IEC 19794-5:2005] Table 14. The values of 128-
255 are vendor specific. Application developers may obtain
the values for these codes from the vendor.
R-33 5.7.6 Source Type 1 M Y Y Y
The (1 byte) Source Type field denotes the classification of the
source of the captured image and is given in
[ISO/IEC 19794-5:2005] Table 15.
R-34 5.7.7 Device Type 1 M Y Y Y
The (2 byte) Device Type field denotes the vendor specific cap-
ture device type ID. A value of all zeros will be acceptable and
will indicate that the capture device type ID is unspecified.
Application developers may obtain the values for these codes
from the vendor.
R-35 5.7.8 Quality 1 M Y Y Y
The (2 byte) Quality field shall be reserved for future defini-
tion to represent a quality of the facial image. This field shall
be set to the value 0 indicating “unspecified”.
R-36 5.8.1 Data structure 2 M Y Y Y
The (variable byte) Image Data block shall be the image data
encoded by either the JPEG or JPEG2000 standards.
© ISO/IEC 2019 – All rights reserved 7

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
Basic Face Image Type
R-37 6.2 Image data encoding requirements for the Basic Face Image 2 M Y Y Y
Type
One of two possible encodings is to be used for all image
types (Basic)
1)  The JPEG Sequential baseline (ISO/IEC 10918-1) mode
of operation and encoded in the JFIF file format (the JPEG
file format)
2)  The JPEG-2000 Part-1 Code Stream Format (ISO/IEC
15444-1) and encoded in the JP2 file format (the JPEG2000 file
format).
R-38 6.4.1 Facial Header [for the Basic Face Image Type] 1 M Y Y Y
The Format Identifier, Version Number, Length of Record, and
Number of Faces fields shall be specified.
R-39 6.4.2 Facial Information [for the Basic Face Image Type] 1 M Y Y Y
The Block Length and Number of Landmark Points fields shall
be specified.
R-40 6.4.3 Image Information [for the Basic Face Image Type] 1 M Y Y Y
The Face Image Type field shall be specified with value 0x00.
R-41 6.4.3 The Image Data Type, Width, and Height fields shall be 1 M Y Y Y
specified.
The Frontal Face Image Type
R-42 7.2.2 Pose 3C O-1 N Y Y
Thus, the full-face frontal pose shall be used. Rotation of the
head shall be less than ±5° from frontal in pitch and yaw (ref
5.5.8). Pose variations that lead to an in-plane rotation of
the head can be more easily compensated by automated face
recognition systems. Therefore, the rotation of the head shall
be less than ±8° from frontal in roll (ref 5.5.8).
[ISO/IEC 19794-5:2005] Figure 13 shows an example of ±8°
rotation in roll.
The best practice recommendation as outlined in A.2.2 is that
the rotation of the head should be less than ±5° from frontal
in roll.
This constraint refers to the pose of the subject associated
with the face image format data for all applications that call
for this format to be used.
R-43 7.2.4 Assistance in positioning the face 3C O-1 N Y Y
In no cases will any other face be captured in the Frontal image.
See informative annex A.2 for best practices on this topic.
R-44 7.2.5 Shoulders 3C O-1 N Y Y
Shoulders shall be “square on” to the camera. “Portrait style”
photographs where the subject is looking over one shoulder
are not acceptable.
R-45 7.2.7 Subject and scene lighting 3C O-1 N Y Y
Lighting shall be equally distributed on the face.
R-46 7.2.7 There shall be no significant direction of the light from the 3C O-1 N Y Y
point of view of the photographer, as further described in
[ISO/IEC 19794-5:2005] Clauses 7.2.8 and 7.2.9.
R-47 7.2.8 Shadows over the face 3C O-1 N Y Y
The region of the face, from the crown (as defined in
section 4.6) to the base of the chin, and from ear-to-ear, shall
be clearly visible and free of shadows.
8 © ISO/IEC 2019 – All rights reserved

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-48 7.2.8 Special care shall be taken in cases when veils, scarves or 3C O-1 N Y Y
headdresses cannot be removed for religious reasons to
ensure these coverings do not obscure any facial features and
do not generate shadow. In all other cases head coverings shall
be absent .
R-49 7.2.9 Shadows in eye-sockets 3C O-1 N Y Y
There shall be no dark shadows in the eye-sockets due to
the brow.
R-50 7.2.9 The iris and pupil of the eyes shall be clearly visible. 3C O-1 N Y Y
R-51 7.2.10 Hot Spots 3C O-1 N Y Y
Care shall be taken to avoid "hot spots" (bright areas of light
shining on the face). These artefacts are typically caused
when one, high intensity, focused light source is used for
illumination.
R-52 7.2.10 Instead, diffused lighting, multiple balanced sources or other 3C O-1 N Y Y
lighting methods shall be used.
A single bare “point” light source is not acceptable for imaging.
Instead, the illumination should be accomplished using other
methods that meet requirements specified in this
[ISO/IEC 19794-5:2005] Clause.
R-53 7.2.11 Eye Glasses 3C O-1 N Y Y
If the person normally wears glasses then they should wear
glasses when their photograph is taken. Glasses shall be clear
glass and transparent so the eye pupils and irises are clearly
visible.
R-54 7.2.11 Permanently tinted glasses or sunglasses are acceptable only 3C O-1 N Y Y
for medical reasons (and shall otherwise be removed). In cases
where tinted glasses or sunglasses are worn, the specification
of dark glasses in the header structure is recommended.
R-55 7.2.11 Care shall be taken that the glasses frames do not obscure 3C O-1 N Y Y
the eyes.
R-56 7.2.11 There shall be no lighting artefacts or flash reflections on 3C O-1 N Y Y
glasses. This can typically be achieved by increasing the angle
between the lighting, subject and camera to 45º
(degrees) or more.
R-57 7.2.12 Eye Patches 3C O-1 N Y Y
The wearing of eye patches is allowed only for medical
reasons. In these cases, the specification of the patch, in the
header structure is recommended.
R-58 7.3.2 No Over or Under Exposure 3C O-1 N Y Y
For each patch of skin on the person’s face, the gradations in
textures shall be clearly visible. In this sense, there will be no
saturation (over or under exposure) on the face.
R-59 7.3.3 Focus and Depth of Field 3C O-1 N Y Y
The subject’s captured image shall always be in focus from
nose to ears and chin to crown.
R-60 7.3.3 All images will have sufficient depth of focus to maintain 3C O-1 N Y Y
greater than two millimetre resolution on the subject’s facial
features at time of capture.
R-61 7.3.4 Unnatural Color 3C O-1 N Y Y
Unnaturally coloured lighting, yellow, red, etc. is not allowed.
Care shall be taken to correct the "white balance" of image
capture devices.
R-62 7.3.4 The lighting shall produce a face image with natural looking 3C O-1 N Y Y
flesh tones when viewed in typical examination environments.
R-63 7.3.4 “Red-eye” is not acceptable. 3C O-1 N Y Y
© ISO/IEC 2019 – All rights reserved 9

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-64 7.3.5 Color or Grayscale Enhancement 3C O-1 N Y Y
A process that overexposes or under-develops a colour or
greyscale image for purposes of beauty enhancement or artis-
tic pleasure is not allowed. The full spectrum shall be repre-
sented on the face image where appropriate. Teeth and whites
of eyes shall be clearly light or white (when appropriate) and
dark hair or features (when appropriate) shall be clearly dark.
R-65 7.3.6 Radial Distortion of the Camera Lens 3C O-1 N Y Y
The fish eye (ref. 4.11) that is associated with unusually large
noses in the image is not allowed.
R-66 7.4.1.1 Pixel aspect ratio 3C O-1 N Y Y
Digital cameras and scanners used to capture facial images
shall produce images with a pixel aspect ratio of 1:1. That is,
the number of pixels per inch in the vertical dimension shall
equal the number of pixels per inch in the horizontal direction.
R-67 7.4.1.2 Origin at Upper Left 3C O-1 N Y Y
The origin of coordinates shall be at the upper left given by
coordinate (0,0) with positive entries from left to right (first
dimension) and top to bottom (second dimension).
R-68 7.4.2.3 Colour space 2 O-1 N Y Y
Frontal images shall be represented as one of the following
a)  The 24-bit RGB colour space where for every pixel, eight
(8) bits will be used to represent each of the Red, Green, and
Blue components.
b)  An 8-bit monochrome colour space where for every pixel,
(8) bits will be used to represent the luminance component.
c)  The YUV422 colour space where twice as many bits
are dedicated to luminance as to each of the two colour
components. YUV422 images typically contain two 8-bit Y
samples along with one 8-bit sample of each of U and V in
every four bytes.
R-69 7.4.3 Video interlacing 3C O-1 N Y Y
Interlaced video frames are not allowed for the Frontal Image
Type. All interlacing must be absent (not simply removed, but
absent).
R-70 7.5.2 Image Information [for the Frontal Image Type] 1 M N Y Y
Frontal Images are either Full Frontal or Token Frontal images
and the Face Image Type field shall be set accordingly (ref.
8.5.2, 9.3.2).
The Full Frontal Image Type
R-71 8.3.1 Introduction 3C O-1 N Y Y
In addition to the requirements of 8.3.2 to 8.3.6, the face from
chin to crown as defined in 8.3.5 and with the full width as
defined in 8.3.4 shall be visible in the image.
R-72 8.3.2 Horizontally centred face 3C O-1 N Y N
The approximate horizontal midpoints of the mouth and of
the bridge of the nose define the imaginary line AA (usually
the symmetry axis of the face). Furthermore, the imaginary
line BB is defined as the line through the centres of the left
and right eye. The intersection of AA and BB defines the point
M as the centre of the face. The x-coordinate M of M shall be
x
between 45 % and 55 % of the image width.
R-73 8.3.3 Vertical position of the face 3C O-1 N Y N
The y-coordinate M of M shall be between 30 % and 50 % of
y
the image height.
10 © ISO/IEC 2019 – All rights reserved

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-74 8.3.3 A single exception is allowed for children under the age of 3C O-1 N Y N
11 years, in which case the higher limit shall be modified to
60 % (i.e. the centre point of the head is allowed to be lower in
the image for children under the age of 11). The origin O of the
coordinate system is defined to be in the upper left corner of
the image.
R-75 8.3.4 Width of head 3C O-1 N Y N
The width of a head is defined as the distance between the
two imaginary lines parallel to the line AA; each imaginary
line is drawn between the upper and lower lobes of each ear
and shall be positioned where the external ear connects to the
head. The head width is shown as length CC in Figure 14.
R-76 8.3.4 To ensure that the entire face is visible in the image, the head 3C O-1 N Y N
width (CC) shall be between 50 % and 75 % of the image
width (A).
R-77 8.3.5 Length of head 3C O-1 N Y N
The length of a head is defined as the distance between the
base of the chin and the crown measured on the imaginary
line AA. This is shown as length DD in Figure 14. The crown is
defined as the top of the head ignoring any hair.
In order to assure that the entire face is visible in the image,
the minimum image height shall be specified by requiring that
the crown-to-chin portion (DD) of the full frontal image pose
shall be between 60 % and 90 % of the vertical length of the
image (B).
R-78 8.3.5 A single exception is allowed for children under the age of 11 3C O-1 N Y N
years, in which case the lower limit shall be modified to 50 %.
R-79 8.4.1 Resolution [Digital requirements for the Full Frontal Face 3C O-1 N Y N
Image Type]
For an image for optimal human examination and permanent
storage, the resolution of the full images shall be at least
180 pixels of resolution for the width of the head, or roughly
90 pixels from eye centre to eye centre. See informative annex
section A.3.1.1 for best practices on this topic.
R-80 8.5.2 Image Information 1 M N Y N
The Face Image Type field shall be specified with value 0х01.
The Token Face Image Type
R-81 9.2.2 Eye positions 3C O-1 N N Y
To create a Token Face image, the eye socket centres, or simply
eye positions, defined as Landmark Points 12.1 and 12.2, shall
be determined. For the determination of eye positions, it is
possible:
1)  to use computer inspection
2)  to use human visual inspection
3)  to use computer and human visual inspection.
R-82 9.2.3 Token image geometric format 2 M N N Y
A Token image is a colour or greyscale image with image
dimensions and eye position coordinates given by
[ISO/IEC 19794-5:2005] Table 16. Note that
[ISO/IEC 19794-5:2005] Clause 5.2.3 specifies conversion of
values to integer.
Image Height W/0,75 where W is the Image Width.
R-83 9.2.3 Y coordinate of Eyes 0,6 * W 3C O-1 N N Y
R-84 9.2.3 X coordinate of First (right) Eye 0,375 * W 3C O-1 N N Y
R-85 9.2.3 X coordinate of Second (left) Eye = (0,625 * W) − 1 3C O-1 N N Y
R-86 9.2.3 Width from eye to eye (inclusive) 0,25 *W 3C O-1 N N Y
© ISO/IEC 2019 – All rights reserved 11

Table 1 (continued)
Require- Ref. in Requirement summary Level Sta- Subformat IUT Sup- Test
ment ID base tus applicabil- sup- ported re-
std ity port range sult
B F T
R-87 9.2.4 Minimum Width Token Image 2 M N N Y
The minimum required image width is 240 pixels.
R-88 9.2.4 The distance from eye to eye (inclusive) in this case is there- 3C O-1 N N Y
fore [at least] 60 pixels. This example is shown in Figure 16.
R-89 9.2.4 Coordinates are relative to the top left corner of the image 3C O-1 N N Y
(0,0) and all measurements are in units of pixels.
R-90 9.2.5 Padding 3C O-1 N N Y
The normative practice shall be to fill any undefined set of
pixels with any colour. See informative annex A.4.3 for best
practices on this matter.
R-91 9.3.2 Image Information 1 M N N Y
The Face Image Type field in the Image Information structure
shall be specified with value 0x02.
NOTE 1 Level 3C conformance test assertions are not specified. In all cases, the reason for this is that no
method has been defined to test the conformance of the IUT or BDIR for this mandatory requirement of the
base standard. For the purposes of this document, the requirements are optional ("O") until an appropriate test
method is established. The 3C label indicates that conformance testing of this Level 3 conformance requirement
is beyond the scope of the present version of the conformance testing standard containing the table. However,
most of the requirements can be tested by means of suitable image processing operations. For example, there is a
large academic literature on pose estimation (see [5] for a review).
NOTE 2 These Level 2 requirements appear in clauses which define the block structure of a conformant ISO/
IEC 19794-5:2005 record. The requirements do not appear in Tables 2, 3, and 4 below because other conformance
test assertions cover these requirements.
NOTE 3 This is not a Level 1 requirement because testing requires more information than is contained in the
field. It can be addressed on a case-by-case basis by the requirements / assertions pertaining to those fields.
6.3 ISO/IEC 19794-5:2005 Level 1 and 2 test assertions
A subsystem that claims to produce face image records that conform to the Level 1 and Level 2
requirements of the Basic Face Image Type of ISO/IEC 19794-5:2005 Face Image Format for Data
Interchange shall be assessed by execution of all of the specific test assertions listed in Table 2. The test
notes which follow Table 2 are normative.
A subsystem that claims to produce face image records that conform to the Level 1 and Level 2
requirements of the Full Frontal Face Image Type of ISO/IEC 19794-5:2005 Face Image Format for Data
Interchange shall be assessed by execution of all of the specific test assertions listed In Table 3. The
test notes which follow Table 3 are normative.
A subsystem that claims to produce face image records that conform to the Level 1 and Level 2
requirements of the Token Face Image Type of ISO/IEC 19794-5:2005 Face Image Format for
...

Questions, Comments and Discussion

Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.

Loading comments...