ISO/IEC 29794-1:2016
(Main)Information technology — Biometric sample quality — Part 1: Framework
Information technology — Biometric sample quality — Part 1: Framework
ISO/IEC 29794-1:2016, for any or all biometric sample types as necessary, establishes the following: - terms and definitions that are useful in the specification and use of quality metrics; - purpose and interpretation of biometric quality scores; - encoding of quality data fields in biometric data interchange formats; - methods for developing biometric sample datasets for the purpose of quality score normalisation; - format for exchange of quality algorithm results; - methods for aggregation of quality scores. The following are outside the scope of ISO/IEC 29794-1:2016: - specification of minimum requirements for sample, module, or system quality scores; - performance assessment of quality algorithms; - standardization of quality algorithms.
Technologies de l'information — Qualité d'échantillon biométrique — Partie 1: Cadre
General Information
Relations
Buy Standard
Standards Content (Sample)
INTERNATIONAL ISO/IEC
STANDARD 29794-1
Second edition
2016-01-15
Corrected version
2016-09-15
Information technology — Biometric
sample quality —
Part 1:
Framework
Technologies de l’information — Qualité d’échantillon biométrique —
Partie 1: Cadre
Reference number
ISO/IEC 29794-1:2016(E)
©
ISO/IEC 2016
---------------------- Page: 1 ----------------------
ISO/IEC 29794-1:2016(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO/IEC 2016, Published in Switzerland
All rights reserved. Unless otherwise specified, 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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO/IEC 2016 – All rights reserved
---------------------- Page: 2 ----------------------
ISO/IEC 29794-1:2016(E)
Contents Page
Foreword .iv
Introduction .vi
1 Scope . 1
2 Conformance . 1
3 Normative references . 1
4 Terms and definitions . 2
5 Abbreviated terms . 3
6 Biometric sample quality criteria . 4
6.1 Reference model . 4
6.2 Quality components: character, fidelity, utility . 5
6.3 Usefulness of quality data . 5
6.3.1 Real-time quality assessment . 5
6.3.2 Use in different applications . 6
6.3.3 Use as a survey statistic . 6
6.3.4 Accumulation of relevant statistics . 6
6.3.5 Reference dataset improvement . 6
6.3.6 Quality-based conditional processing . 6
6.3.7 Interchange of quality data by disparate systems . 7
7 Data interchange format field definition . 7
7.1 Binary encoding . 7
7.2 XML encoding . 9
7.3 Quality score .10
7.3.1 Purpose .10
7.3.2 Data transformation considerations .10
7.3.3 Failure modes .10
7.3.4 Resolution .10
7.3.5 Summarisation .10
7.4 Quality algorithm identification .10
7.4.1 Overview .10
7.4.2 Methodology .11
7.5 Standardized exchange of quality algorithm results .11
8 Normalisation .12
Annex A (informative) Example of encoding a biometric quality record .13
Annex B (informative) Example of standardized exchange of quality algorithm results .14
Annex C (informative) Procedures for aggregation of utility-based quality scores .16
Bibliography .19
© ISO/IEC 2016 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO/IEC 29794-1:2016(E)
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. In the field of information technology, ISO and IEC have established a joint technical committee,
ISO/IEC JTC 1.
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).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the meaning of ISO specific terms and expressions related to conformity
assessment, as well as information about ISO’s adherence to the WTO principles in the Technical
Barriers to Trade (TBT) see the following URL: Foreword - Supplementary information.
The committee responsible for this document is ISO/IEC JTC 1, Information technology, Subcommittee
SC 37, Biometrics.
This second edition cancels and replaces the first edition (ISO/IEC 29794-1:2009), which has been
technically revised to revise Clause 8 and Table 2, which describes the structure of quality record.
ISO/IEC 29794 consists of the following parts, under the general title Information technology —
Biometric sample quality:
— Part 1: Framework
— Part 4: Finger image data
— Part 5: Facial image data [Technical Report]
— Part 6: Iris image data
ISO/IEC 29794 series is prepared to accommodate new, additional parts that address other modalities
specified by ISO/IEC 19794, with part numbers and titles aligning appropriately. However, as Part 1
is intended for use by all modalities, a modality does not necessarily need a modality-specific part in
order to make use of quality scores.
It is anticipated that a future version of each part of the ISO/IEC 19794 series will reference this part of
ISO/IEC 29794 normatively, and their respective data fields will be updated as required.
This corrected version of ISO/IEC 29794:2016 incorporates the following corrections.
1. “as given in Formula (C.1)” has been deleted from C.2 a).
2. Table 2, row: 5-byte Quality Block, column: Governing Section + Description + Notes:
QAID values of 0 to 32767
is changed to
iv © ISO/IEC 2016 – All rights reserved
---------------------- Page: 4 ----------------------
ISO/IEC 29794-1:2016(E)
QAID values of 1 to 32767
3. A.2, table, row: 5, column: Block 1 Byte 4+5 (QAID)
0
is changed to
10
© ISO/IEC 2016 – All rights reserved v
---------------------- Page: 5 ----------------------
ISO/IEC 29794-1:2016(E)
Introduction
Quality metrics are useful for several applications in the field of biometrics. While ISO/IEC 19784-
1 specifies a structure and gives guidelines for quality score categorization, ISO/IEC 29794 defines
and specifies methodologies for objective, quantitative quality score expression, interpretation, and
interchange. This International Standard is intended to add value to a broad spectrum of applications
in a manner that encourages competition, innovation, interoperability and performance improvements,
and avoids bias towards particular applications, modalities, or techniques.
This International Standard presents several biometric sample quality scoring tools, the use of which
is generally optional but can be determined as mandatory by particular Application Profiles or specific
implementations.
A number of applications can benefit from the use of biometric sample quality data; an example is the use
of real-time quality feedback upon enrolment to improve the operational efficiency and performance of
a biometric system. The association of quality data with biometric samples is an important component
of quality metric standardization. Quality fields as specified in 7.1 and 7.2 will be incorporated into
data interchange formats. If a CBEFF header is present, then CBEFF_BDB_quality may additionally be
used to express quality data. Useful analyses can be performed using quality data along with other data
in order to improve the performance of a biometric system. For example, correlating quality data to
other system metrics can be used to diagnose problems and highlight potential areas of performance
improvement.
This edition introduces encoding of a vector of quality metrics.
vi © ISO/IEC 2016 – All rights reserved
---------------------- Page: 6 ----------------------
INTERNATIONAL STANDARD ISO/IEC 29794-1:2016(E)
Information technology — Biometric sample quality —
Part 1:
Framework
1 Scope
This part of ISO/IEC 29794, for any or all biometric sample types as necessary, establishes the following:
— terms and definitions that are useful in the specification and use of quality metrics;
— purpose and interpretation of biometric quality scores;
— encoding of quality data fields in biometric data interchange formats;
— methods for developing biometric sample datasets for the purpose of quality score normalisation;
— format for exchange of quality algorithm results;
— methods for aggregation of quality scores.
The following are outside the scope of this part of ISO/IEC 29794:
— specification of minimum requirements for sample, module, or system quality scores;
— performance assessment of quality algorithms;
— standardization of quality algorithms.
2 Conformance
A biometric sample quality record shall conform to this part of ISO/IEC 29794 if its structure and data
values conform to the formatting requirements of Clause 7. Conformance to normative requirements
of 7.1 and 7.2 fulfils Level 1 and Level 2 conformance as specified in ISO/IEC 19794-1:2011, Annex A.
Conformance to normative requirements of 7.3 is Level 3 conformance as specified in ISO/IEC 19794-
1:2011, Annex A.
3 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including any amendments) applies.
ISO/IEC 19794–1:2011, Information technology — Biometric data interchange formats — Part 1:
Framework
ISO/IEC 19785–1, Information technology — Common Biometric Exchange Formats Framework — Part 1:
Data element specification
ISO/IEC 2382–37, Information technology — Vocabulary — Part 37: Biometrics
© ISO/IEC 2016 – All rights reserved 1
---------------------- Page: 7 ----------------------
ISO/IEC 29794-1:2016(E)
4 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 2382-37, ISO/IEC 19794-1
and the following apply.
4.1
acquisition fidelity
fidelity (4.6) of a sample attributed to the acquisition process
4.2
character
contributor to quality (4.11) of a sample attributable to inherent properties of the source (4.17)
4.3
environment
physical surroundings and conditions where biometric capture occurs, including operational factors
such as operator (4.9) skill and enrolee cooperation level
4.4
extraction fidelity
component of the fidelity (4.6) of a sample attributed to the biometric feature extraction process
4.5
extrinsic
when used to describe a quality score (4.12), requiring reference to an external source (4.17), such as a
standard, register, or technical specifications for full interpretation (4.8) and normalisation
4.6
fidelity
expression of how accurately a biometric sample represents its source (4.17) biometric characteristic
Note 1 to entry: The fidelity of a sample comprises components attributable to one or more of the processing
steps: acquisition, extraction, signal processing.
4.7
intrinsic
when used to describe a quality score (4.12), conveying fully interpreted (4.8), normalised data without
the requirement for additional extrinsic (4.5)information for quality score normalisation (4.13)
4.8
interpretation
process of analysing a quality score (4.12) along with other data in order to give that score contextual,
relative meaning
4.9
operator
individual who processes a capture subject in a biometric system, performing or supervising capture
and recapture
4.10
performance
assessment of false match rate, false non-match rate, failure to enrol rate and failure to acquire rate of
a biometric system
4.11
quality
degree to which a biometric sample fulfils specified requirements for a targeted application
Note 1 to entry: Specified quality requirements may address aspects of quality such as focus, resolution, etc.
Implicit quality requirements address the likelihood of achieving a correct comparison result.
2 © ISO/IEC 2016 – All rights reserved
---------------------- Page: 8 ----------------------
ISO/IEC 29794-1:2016(E)
4.12
quality score
quantitative expression of quality (4.11)
4.13
quality score normalisation
rescaling of quality scores (4.12) to improve consistency in scale and interpretation (4.8)
4.14
quality score normalisation dataset
QSND
dataset of biometric samples annotated with quality scores (4.12) for use in quality score
normalisation (4.13)
Note 1 to entry: Target quality scores may be assigned on the basis of performance (4.10) outcomes using the
sample in question, or may be based on quality factors recorded in acquisition of the dataset.
4.15
quality score percentile rank
QSPR
percentile rank of the quality score (4.12) of a biometric sample, derived from its own utility score and
those of other samples in an identified control dataset
Note 1 to entry: See QSND (4.14).
4.16
raw quality score
quality score (4.12) that has not been interpreted (4.8), either by the creator or recipient of the score, and
alone may not intrinsically provide contextual information
4.17
source
physical body part or function represented by a biometric sample
4.18
utility
observed performance (4.10) of a biometric sample or set of samples in one or more biometric systems
Note 1 to entry: The character (4.2) of the sample source (4.17) and the fidelity (4.6) of the processed samples
contribute to, or similarly detract from, the utility of the sample.
Note 2 to entry: Utility may combine performance measures such as false match rate, false non-match rate,
failure to enrol rate, and failure to acquire rate.
5 Abbreviated terms
BDB biometric data block
BDIR biometric data interchange record
BIR biometric information record
CBEFF common biometric exchange formats framework (ISO/IEC 19785)
FERET facial recognition technology database
FNMR false non-match rate
QAID quality algorithm identifier
QSND quality score normalisation dataset
QSPR quality score percentile rank
QVID quality algorithm vendor identifier
XML eXtensible Markup Language
© ISO/IEC 2016 – All rights reserved 3
---------------------- Page: 9 ----------------------
ISO/IEC 29794-1:2016(E)
6 Biometric sample quality criteria
6.1 Reference model
In biometrics, the term “quality” is used to describe several different aspects of a biometric sample
that contribute to the overall performance of a biometric system. For the purposes of standardization,
this part of ISO/IEC 29794 defines terms, definitions, and a reference model for distinguishing between
these different aspects of quality, illustrated in Figure 1. Figure 2 illustrates the relationship between
character, fidelity, quality, utility, and system performance.
Image-based
Source Feature-based
Processed
image feature
acquisition Sample Sample
Sample
processing extraction
idelity
idelity idelity
resolution
downsampling
feature quality
lighting
cropping
extraction algorithm
behavior
rotation
compression
character idelity
Figure 1 — Quality reference model illustration
character, idelity
The correlation
quality scoring algorithm between predicted
utility and observed
The observed utility
utility of each sample of a sample relects
is indicative of the its impact on the
quality score (sample)
effectiveness of the
performance of the
quality algorithm system
correlation
predicted utility (sample)
observed utility (sample)
correlation
A quality algorithm
should convey the
predicted utility of
the sample
comparison
observed performance (system)
algorithm
The performance of a biometric
system is a function of the
comparison algorithm
performance and the utility of
all samples in the system
Figure 2 — Relationship between quality and system performance
4 © ISO/IEC 2016 – All rights reserved
---------------------- Page: 10 ----------------------
ISO/IEC 29794-1:2016(E)
6.2 Quality components: character, fidelity, utility
The term “quality” as it is currently used in the field of biometrics has several connotations, depending
on context. Three prevalent uses are to subjectively reflect the following.
a) Character of a sample. An expression of quality based on the inherent properties of the source
from which the biometric sample is derived. For example, a scarred fingerprint has poor character
and blepharoptosis (droopy eyelid) causes poor iris character.
b) Fidelity of a sample to the source from which it is derived. An expression of quality based on fidelity
reflects the degree of its similarity to its source. Sample fidelity is comprised of fidelity components
contributed by different processes.
c) Utility of a sample within a biometric system. An expression of quality based on utility reflects the
predicted positive or negative contribution of an individual sample to the overall performance of a
biometric system. Utility-based quality is dependent on both the character and fidelity of a sample.
Utility–based quality is intended to be more predictive of system performance, e.g. in terms of false
match rate, false non-match rate, failure to enrol rate, and failure to acquire rate, than measures of
quality based on character or fidelity alone. See Table 1.
The term “quality” should not be solely attributable to the acquisition settings of the sample, such as
image resolution, dimensions in pixels, grey scale/colour bit depth, or number of features. Though such
factors can affect sample utility and could contribute to the overall quality score.
Note that the character and utility of an acquired sample depend on the features to be considered by
the comparator. For instance, the same finger image may be of low character and utility with respect
to minutiae recognition (because of too few minutiae), but of high character and utility with respect to
spectral pattern recognition.
Table 1 — Illustration of relationship between fidelity, utility, and character
Fidelity
Low High
Low Low fidelity and low character re- High fidelity and low character results
sults in low utility. Recapture might in low utility. Recapture will not
improve utility. However, if possible, improve utility. Use of other biometric
use of other biometric characteristics characteristics is recommended.
Character
is recommended.
High Samples with high character and low Samples with high character and high
fidelity typically will not demon- fidelity indicate capture of useful sam-
strate high utility. Utility can be ple. High utility is expected.
improved upon recapture or image
enhancement techniques.
6.3 Usefulness of quality data
6.3.1 Real-time quality assessment
Real-time quality data can be used by an operator, automated system, or a user to help improve the
average quality of biometric samples submitted upon enrolment. This feedback might indicate the
character, fidelity, utility, and improvability of a sample. In this way, operational efficiency and overall
system performance can be improved by assisting an operator, or augmenting an automated quality
control system, in decisions to accept the sample, reject the sample, reattempt a capture, or declare
a failure to acquire or failure to enrol. Quality data can be retained for later use in, for example,
determining whether an enrolment sample should be replaced when the next sample is captured.
© ISO/IEC 2016 – All rights reserved 5
---------------------- Page: 11 ----------------------
ISO/IEC 29794-1:2016(E)
6.3.2 Use in different applications
A particular biometric sample might be used for several different applications and therefore, its
associated quality data should be applicable to all of these. This would include both one-to-one and
one-to-many comparisons involving the use of comparison algorithms from different vendors that
would interpret sample features differently and yield different comparison scores. The challenge in
establishing a universal quality standard is in defining a metric that is sufficiently adaptable for use by
all applications for which a particular sample might be used given that metrics of utility vary greatly
between applications. Therefore, it should be recognized that it is a technical challenge to define a
singular metric that accurately conveys the utility of a biometric sample for all applications for which it
may be used, and this should be taken into consideration in defining quality standards. Thus, a quality
metric (ideally predicting performance for a comparator or class of comparators) will likely be designed
to capture only some of the failure modes and sensitivities of only a limited number of biometric
systems. It may be useful to apply more than one quality metric in order to improve predictability of
various failure modes.
It is useful for recipients of quality score data to be able to differentiate between scores generated
by different quality algorithms and capture equipment. This data may be used to enable recipient
software to be configured so that different thresholds or quality classifications can be applied to
scores generated by different algorithms. In addition, by differentiating between scores from different
algorithms, a recipient may isolate results from different algorithms and use the data to optimize
thresholds accordingly.
6.3.3 Use as a survey statistic
Quality scores may be used to monitor operational quality. For example, aggregated quality scores
could be compared with pre-set limits or monitored against an operational requirement. If, for
example, quality scores are generated from biometric samples collected at many sites, or over different
time periods, then they may be used to identify anomalous operation. For example, if face image quality
is computed at the license issuance desks at a Department of Motor Vehicles, then a ranked list of
aggregated quality scores could be used to identify desks that exhibit a lower than average quality, or to
monitor trends over weeks or months.
6.3.4 Accumulation of relevant statistics
Reliable quality scores may be used to survey users and transactions to accumulate statistics giving
conditional probabilities of the kind “given a quality X sample on finger A, what is the likelihood of a
quality Y sample from finger A (or finger B)”. This will inform the system and/or operators over whether
a higher quality sample is likely if another capture is attempted.
6.3.5 Reference dataset improvement
The association of quality data with a sample that is to be entered into a reference dataset is important
for the maintenance and improvement of the reference dataset quality. The tracking of sample quality
can lead to detection of potential deterioration of operator training or it may indicate deterioration
in the performance of the sample capture equipment. Tracking of the sample quality data should be
an important part of biometric systems operating procedures. The quality data may also be used
to improve the quality of the reference file, and hence the performance of the biometric system.
Improvement can be made by the replacement or possible augmentation of the stored information so
as to make use of the best available quality data. Typically, the replacement decisions are linked to the
comparator performance of the system processing the data.
6.3.6 Quality-based conditional processing
Biometric samples can be processed differently based on quality metrics. In particular, poor quality
data can be processed using algorithms or thresholds different from those used for high-quality data.
6 © ISO/IEC 2016 – All rights reserved
---------------------- Page: 12 ----------------------
ISO/IEC 29794-1:2016(E)
6.3.7 Interchange of quality data by disparate systems
Standardized exchange of quality data between disparate systems is useful in retaining the modular
interchangeability of local or remote system hardware and software components, and the integrity of
quality data in the event of such an interchange.
For example, by using standardized exchange of quality data, consumers of quality data from a
component require minimal modification if that component is replaced.
7 Data interchange format field definition
7.1 Binary encoding
A quality record shall consist of a length field followed by zero, one, or multiple 5-byte Quality Blocks,
as shown in Figure 3.
Figure 3 — Location of number of Quality Block fields
Table 2 defines the structu
...
FINAL
INTERNATIONAL ISO/IEC
DRAFT
STANDARD FDIS
29794-1
ISO/IEC JTC 1/SC 37
Information technology — Biometric
Secretariat: ANSI
sample quality —
Voting begins
on: 2015-09-16
Part 1:
Voting terminates
Framework
on: 2015-11-16
Technologies de l’information — Qualité d’échantillon biométrique —
Partie 1: Cadre
RECIPIENTS OF THIS DRAFT ARE INVITED TO
SUBMIT, WITH THEIR COMMENTS, NOTIFICATION
OF ANY RELEVANT PATENT RIGHTS OF WHICH
THEY ARE AWARE AND TO PROVIDE SUPPOR TING
DOCUMENTATION.
IN ADDITION TO THEIR EVALUATION AS
Reference number
BEING ACCEPTABLE FOR INDUSTRIAL, TECHNO-
ISO/IEC FDIS 29794-1:2015(E)
LOGICAL, COMMERCIAL AND USER PURPOSES,
DRAFT INTERNATIONAL STANDARDS MAY ON
OCCASION HAVE TO BE CONSIDERED IN THE
LIGHT OF THEIR POTENTIAL TO BECOME STAN-
DARDS TO WHICH REFERENCE MAY BE MADE IN
©
NATIONAL REGULATIONS. ISO/IEC 2015
---------------------- Page: 1 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO/IEC 2015, Published in Switzerland
All rights reserved. Unless otherwise specified, 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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO/IEC 2015 – All rights reserved
---------------------- Page: 2 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Conformance . 1
3 Normative references . 1
4 Terms and definitions . 1
5 Abbreviated terms . 3
6 Biometric sample quality criteria . 4
6.1 Reference model . 4
6.2 Quality components: character, fidelity, utility . 5
6.3 Usefulness of quality data . 5
6.3.1 Real-time quality assessment . 5
6.3.2 Use in different applications . 6
6.3.3 Use as a survey statistic . 6
6.3.4 Accumulation of relevant statistics . 6
6.3.5 Reference dataset improvement . 6
6.3.6 Quality-based conditional processing . 6
6.3.7 Interchange of quality data by disparate systems . 7
7 Data interchange format field definition . 7
7.1 Binary encoding . 7
7.2 XML encoding . 9
7.3 Quality score .10
7.3.1 Purpose .10
7.3.2 Data transformation considerations .10
7.3.3 Failure modes .10
7.3.4 Resolution .10
7.3.5 Summarisation .10
7.4 Quality algorithm identification .10
7.4.1 Overview .10
7.4.2 Methodology .11
7.5 Standardized exchange of quality algorithm results .11
8 Normalisation .12
Annex A (informative) Example of encoding a biometric quality record .13
Annex B (informative) Example of standardized exchange of quality algorithm results .14
Annex C (informative) Procedures for aggregation of utility-based quality scores .16
Bibliography .19
© ISO/IEC 2015 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
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. In the field of information technology, ISO and IEC have established a joint technical committee,
ISO/IEC JTC 1.
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).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the meaning of ISO specific terms and expressions related to conformity
assessment, as well as information about ISO’s adherence to the WTO principles in the Technical
Barriers to Trade (TBT) see the following URL: Foreword - Supplementary information.
The committee responsible for this document is ISO/IEC JTC 1, Information technology, Subcommittee
SC 37, Biometrics.
This second edition cancels and replaces the first edition (ISO/IEC 29794-1:2009), which has been
technically revised to delete Clause 6 and revise Clause 8 and Table 2, which describes the structure of
quality record.
ISO/IEC 29794 consists of the following parts, under the general title Information technology —
Biometric sample quality:
— Part 1: Framework
— Part 4: Finger image data
— Part 5: Facial image data [Technical Report]
— Part 6: Iris image data
ISO/IEC 29794 series is prepared to accommodate new, additional parts that address other modalities
specified by ISO/IEC 19794, with part numbers and titles aligning appropriately. However, as Part 1
is intended for use by all modalities, a modality does not necessarily need a modality-specific part in
order to make use of quality scores.
It is anticipated that a future version of each part of the ISO/IEC 19794 series will reference this part of
ISO/IEC 29794 normatively, and their respective data fields will be updated as required.
iv © ISO/IEC 2015 – All rights reserved
---------------------- Page: 4 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
Introduction
Quality metrics are useful for several applications in the field of biometrics. While ISO/IEC 19784-
1 specifies a structure and gives guidelines for quality score categorization, ISO/IEC 29794 defines
and specifies methodologies for objective, quantitative quality score expression, interpretation, and
interchange. This International Standard is intended to add value to a broad spectrum of applications
in a manner that encourages competition, innovation, interoperability and performance improvements,
and avoids bias towards particular applications, modalities, or techniques.
This International Standard presents several biometric sample quality scoring tools, the use of which
is generally optional but can be determined as mandatory by particular Application Profiles or specific
implementations.
A number of applications can benefit from the use of biometric sample quality data; an example is the use
of real-time quality feedback upon enrolment to improve the operational efficiency and performance of
a biometric system. The association of quality data with biometric samples is an important component
of quality metric standardization. Quality fields as specified in 7.1 and 7.2 will be incorporated into data
interchange formats. If a CBEFF header is present, then CBEFF_BDB_quality may additionally be used to
express quality data. Useful analyses can be performed using quality data along with other data in order
to improve the performance of a biometric system. For example, correlating quality data to other system
metrics can be used to diagnose problems and highlight potential areas of performance improvement.
This edition introduces encoding of a vector of quality metrics.
© ISO/IEC 2015 – All rights reserved v
---------------------- Page: 5 ----------------------
FINAL DRAFT INTERNATIONAL STANDARD ISO/IEC FDIS 29794-1:2015(E)
Information technology — Biometric sample quality —
Part 1:
Framework
1 Scope
This part of ISO/IEC 29794, for any or all biometric sample types as necessary, establishes the following:
— terms and definitions that are useful in the specification and use of quality metrics;
— purpose and interpretation of biometric quality scores;
— encoding of quality data fields in biometric data interchange formats;
— methods for developing biometric sample datasets for the purpose of quality score normalisation;
— format for exchange of quality algorithm results;
— methods for aggregation of quality scores.
The following are outside the scope of this part of ISO/IEC 29794:
— specification of minimum requirements for sample, module, or system quality scores;
— performance assessment of quality algorithms;
— standardization of quality algorithms.
2 Conformance
A biometric sample quality record shall conform to this part of ISO/IEC 29794 if its structure and data
values conform to the formatting requirements of Clause 7. Conformance to normative requirements
of 7.1 and 7.2 fulfils Level 1 and Level 2 conformance as specified in ISO/IEC 19794-1:2011, Annex A.
Conformance to normative requirements of 7.3 is Level 3 conformance as specified in ISO/IEC 19794-
1:2011, Annex A.
3 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including any amendments) applies.
ISO/IEC 19794–1:2011, Information technology — Biometric data interchange formats — Part 1: Framework
ISO/IEC 19785–1, Information technology — Common Biometric Exchange Formats Framework — Part 1:
Data element specification
ISO/IEC 2382–37, Information technology — Vocabulary — Part 37: Biometrics
4 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 2382-37, ISO/IEC 19794-1
and the following apply.
© ISO/IEC 2015 – All rights reserved 1
---------------------- Page: 6 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
4.1
acquisition fidelity
fidelity (4.6) of a sample attributed to the acquisition process
4.2
character
contributor to quality (4.11) of a sample attributable to inherent properties of the source (4.17)
4.3
environment
physical surroundings and conditions where biometric capture occurs, including operational factors
such as operator (4.9) skill and enrolee cooperation level
4.4
extraction fidelity
component of the fidelity (4.6) of a sample attributed to the biometric feature extraction process
4.5
extrinsic
when used to describe a quality score (4.12), requiring reference to an external source (4.17), such as a
standard, register, or technical specifications for full interpretation (4.8) and normalisation
4.6
fidelity
expression of how accurately a biometric sample represents its source (4.17) biometric characteristic
Note 1 to entry: The fidelity of a sample comprises components attributable to one or more of the processing
steps: acquisition, extraction, signal processing.
4.7
intrinsic
when used to describe a quality score (4.12), conveying fully interpreted, normalised data without the
requirement for additional extrinsic (4.5)information for quality score normalisation (4.13)
4.8
interpretation
process of analysing a quality score (4.12) along with other data in order to give that score contextual,
relative meaning
4.9
operator
individual who processes a capture subject in a biometric system, performing or supervising capture
and recapture
4.10
performance
assessment of false match rate, false non-match rate, failure to enrol rate and failure to acquire rate of
a biometric system
4.11
quality
degree to which a biometric sample fulfils specified requirements for a targeted application
Note 1 to entry: Specified quality requirements may address aspects of quality such as focus, resolution, etc.
Implicit quality requirements address the likelihood of achieving a correct comparison result.
4.12
quality score
quantitative expression of quality (4.11)
2 © ISO/IEC 2015 – All rights reserved
---------------------- Page: 7 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
4.13
quality score normalisation
rescaling of quality scores (4.12) to improve consistency in scale and interpretation (4.8)
4.14
quality score normalisation dataset
QSND
dataset of biometric samples annotated with quality scores (4.12) for use in quality score
normalisation (4.13)
Note 1 to entry: Target quality scores may be assigned on the basis of performance (4.10) outcomes using the
sample in question, or may be based on quality factors recorded in acquisition of the dataset.
4.15
quality score percentile rank
QSPR
percentile rank of the quality score (4.12) of a biometric sample, derived from its own utility score and
those of other samples in an identified control dataset
Note 1 to entry: See QSND (4.14).
4.16
raw quality score
quality score (4.12) that has not been interpreted, either by the creator or recipient of the score, and
alone may not intrinsically provide contextual information
4.17
source
physical body part or function represented by a biometric sample
4.18
utility
observed performance (4.10) of a biometric sample or set of samples in one or more biometric systems
Note 1 to entry: The character (4.2) of the sample source (4.17) and the fidelity (4.6) of the processed samples
contribute to, or similarly detract from, the utility of the sample.
Note 2 to entry: Utility may combine performance measures such as false match rate, false non-match rate,
failure to enrol rate, and failure to acquire rate.
5 Abbreviated terms
BDB biometric data block
BDIR biometric data interchange record
BIR biometric information record
CBEFF common biometric exchange formats framework (ISO/IEC 19785)
FERET facial recognition technology database
FNMR false non-match rate
QAID quality algorithm identification
QSND quality score normalisation dataset
QSPR quality score percentile rank
XML eXtensible Markup Language
© ISO/IEC 2015 – All rights reserved 3
---------------------- Page: 8 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
6 Biometric sample quality criteria
6.1 Reference model
In biometrics, the term “quality” is used to describe several different aspects of a biometric sample
that contribute to the overall performance of a biometric system. For the purposes of standardization,
this part of ISO/IEC 29794 defines terms, definitions, and a reference model for distinguishing between
these different aspects of quality, illustrated in Figure 1. Figure 2 illustrates the relationship between
character, fidelity, quality, utility, and system performance.
Image-based
Source Feature-based
Processed
image feature
acquisition Sample Sample
Sample
processing extraction
idelity
idelity idelity
resolution
downsampling
feature quality
lighting
cropping
extraction algorithm
behavior
rotation
compression
character idelity
Figure 1 — Quality reference model illustration
character, idelity
The correlation
quality scoring algorithm between predicted
utility and observed
The observed utility
utility of each sample of a sample relects
is indicative of the its impact on the
quality score (sample)
effectiveness of the
performance of the
quality algorithm system
correlation
predicted utility (sample)
observed utility (sample)
correlation
A quality algorithm
should convey the
predicted utility of
the sample
comparison
observed performance (system)
algorithm
The performance of a biometric
system is a function of the
comparison algorithm
performance and the utility of
all samples in the system
Figure 2 — Relationship between quality and system performance
4 © ISO/IEC 2015 – All rights reserved
---------------------- Page: 9 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
6.2 Quality components: character, fidelity, utility
The term “quality” as it is currently used in the field of biometrics has several connotations, depending
on context. Three prevalent uses are to subjectively reflect the following.
a) Character of a sample. An expression of quality based on the inherent properties of the source
from which the biometric sample is derived. For example, a scarred fingerprint has poor character
and blepharoptosis (droopy eyelid) causes poor iris character.
b) Fidelity of a sample to the source from which it is derived. An expression of quality based on fidelity
reflects the degree of its similarity to its source. Sample fidelity is comprised of fidelity components
contributed by different processes.
c) Utility of a sample within a biometric system. An expression of quality based on utility reflects the
predicted positive or negative contribution of an individual sample to the overall performance of a
biometric system. Utility-based quality is dependent on both the character and fidelity of a sample.
Utility–based quality is intended to be more predictive of system performance, e.g. in terms of false
match rate, false non-match rate, failure to enrol rate, and failure to acquire rate, than measures of
quality based on character or fidelity alone. See Table 1.
The term “quality” should not be solely attributable to the acquisition settings of the sample, such as
image resolution, dimensions in pixels, grey scale/colour bit depth, or number of features. Though such
factors can affect sample utility and could contribute to the overall quality score.
Note that the character and utility of an acquired sample depend on the features to be considered by
the comparator. For instance, the same finger image may be of low character and utility with respect
to minutiae recognition (because of too few minutiae), but of high character and utility with respect to
spectral pattern recognition.
Table 1 — Illustration of relationship between fidelity, utility, and character
Fidelity
Low High
Low Low fidelity and low character High fidelity and low character
results in low utility. Recapture results in low utility. Recapture
might improve utility. How- will not improve utility. Use of
ever, if possible, use of other other biometric characteristics
biometric characteristics is is recommended.
Character
recommended.
High Samples with high character Samples with high character and
and low fidelity typically will high fidelity indicate capture
not demonstrate high utility. of useful sample. High utility is
Utility can be improved upon expected.
recapture or image enhance-
ment techniques.
6.3 Usefulness of quality data
6.3.1 Real-time quality assessment
Real-time quality data can be used by an operator, automated system, or a user to help improve the
average quality of biometric samples submitted upon enrolment. This feedback might indicate the
character, fidelity, utility, and improvability of a sample. In this way, operational efficiency and overall
system performance can be improved by assisting an operator, or augmenting an automated quality
control system, in decisions to accept the sample, reject the sample, reattempt a capture, or declare
a failure to acquire or failure to enrol. Quality data can be retained for later use in, for example,
determining whether an enrolment sample should be replaced when the next sample is captured.
© ISO/IEC 2015 – All rights reserved 5
---------------------- Page: 10 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
6.3.2 Use in different applications
A particular biometric sample might be used for several different applications and therefore, its
associated quality data should be applicable to all of these. This would include both one-to-one and
one-to-many comparisons involving the use of comparison algorithms from different vendors that
would interpret sample features differently and yield different comparison scores. The challenge in
establishing a universal quality standard is in defining a metric that is sufficiently adaptable for use by
all applications for which a particular sample might be used given that metrics of utility vary greatly
between applications. Therefore, it should be recognized that it is a technical challenge to define a
singular metric that accurately conveys the utility of a biometric sample for all applications for which it
may be used, and this should be taken into consideration in defining quality standards. Thus, a quality
metric (ideally predicting performance for a comparator or class of comparators) will likely be designed
to capture only some of the failure modes and sensitivities of only a limited number of biometric
systems. It may be useful to apply more than one quality metric in order to improve predictability of
various failure modes.
It is useful for recipients of quality score data to be able to differentiate between scores generated
by different quality algorithms and capture equipment. This data may be used to enable recipient
software to be configured so that different thresholds or quality classifications can be applied to
scores generated by different algorithms. In addition, by differentiating between scores from different
algorithms, a recipient may isolate results from different algorithms and use the data to optimize
thresholds accordingly.
6.3.3 Use as a survey statistic
Quality scores may be used to monitor operational quality. For example, aggregated quality scores
could be compared with pre-set limits or monitored against an operational requirement. If, for
example, quality scores are generated from biometric samples collected at many sites, or over different
time periods, then they may be used to identify anomalous operation. For example, if face image quality
is computed at the license issuance desks at a Department of Motor Vehicles, then a ranked list of
aggregated quality scores could be used to identify desks that exhibit a lower than average quality, or to
monitor trends over weeks or months.
6.3.4 Accumulation of relevant statistics
Reliable quality scores may be used to survey users and transactions to accumulate statistics giving
conditional probabilities of the kind “given a quality X sample on finger A, what is the likelihood of a
quality Y sample from finger A (or finger B)”. This will inform the system and/or operators over whether
a higher quality sample is likely if another capture is attempted.
6.3.5 Reference dataset improvement
The association of quality data with a sample that is to be entered into a reference dataset is important
for the maintenance and improvement of the reference dataset quality. The tracking of sample quality
can lead to detection of potential deterioration of operator training or it may indicate deterioration
in the performance of the sample capture equipment. Tracking of the sample quality data should be
an important part of biometric systems operating procedures. The quality data may also be used
to improve the quality of the reference file, and hence the performance of the biometric system.
Improvement can be made by the replacement or possible augmentation of the stored information so
as to make use of the best available quality data. Typically, the replacement decisions are linked to the
comparator performance of the system processing the data.
6.3.6 Quality-based conditional processing
Biometric samples can be processed differently based on quality metrics. In particular, poor quality
data can be processed using algorithms or thresholds different from those used for high-quality data.
6 © ISO/IEC 2015 – All rights reserved
---------------------- Page: 11 ----------------------
ISO/IEC FDIS 29794-1:2015(E)
6.3.7 Interchange of quality data by disparate systems
Standardized exchange of quality data between disparate systems is useful in retaining the modular
interchangeability of local or remote system hardware and software components, and the integrity of
quality data in the event of such an interchange.
For example, by using standardized exchange of quality data, consumers of quality data from a
component require minimal modification if that component is replaced.
7 Data interchange format field definition
7.1 Binary encoding
A quality record shall consist of a length field followed by zero, one, or multiple 5-byte Quality Blocks,
as shown in Figu
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.