ISO/IEC 29794-4:2024

Information technology - Biometric sample quality - Part 4: Finger image data

ISO/IEC 29794-4:2017 establishes - terms and definitions for quantifying finger image quality, - methods used to quantify the quality of finger images, and - standardized encoding of finger image quality, for finger images at 196,85 px/cm spatial sampling rate scanned or captured using optical sensors with capture dimension (width, height) of at least 1,27 cm × 1,651 cm.

Technologies de l'information — Qualité d'échantillon biométrique — Partie 4: Données d'image de doigt

General Information

• Status: Published
• Publication Date: 29-Sep-2024
• ICS: 35.240.15 - Identification cards. Chip cards. Biometrics
• Technical Committee: ISO/IEC JTC 1/SC 37 - Biometrics
• Drafting Committee: ISO/IEC JTC 1/SC 37 - Biometrics
• Current Stage: 6060 - International Standard published
• Start Date: 30-Sep-2024
• Due Date: 08-Oct-2024
• Completion Date: 30-Sep-2024

Relations

• Revises: ISO/IEC 29794-4:2017 - Information technology - Biometric sample quality - Part 4: Finger image data
• Effective Date: 06-Jun-2022

Overview

ISO/IEC 29794-4:2024 - "Information technology - Biometric sample quality - Part 4: Finger image data" defines how to measure, encode and exchange quality information for plain finger (fingerprint) images. The standard applies to 8‑bit greyscale finger images captured or scanned at a spatial sampling rate of 196.85 px/cm (commonly from optical area sensors or inked cards) and provides standardized terminology, normative quality measures, and a machine‑readable encoding for quality values. The 2024 second edition adds normalization algorithms, new quality algorithm identifiers, and an updated conformance test set.

Key technical topics and requirements

• Scope and data model
  • Targets plain fingerprint images with 196.85 px/cm sampling and 8‑bit depth; capture dimensions and foreground trimming rules specified for consistent processing.
• Local and global quality components
  • Normative measures include Orientation Certainty Level (OCL), Local Clarity (LCL), Frequency Domain Analysis (FDA), Ridge-Valley Uniformity (RVU), Orientation Flow (OFL), and minutiae‑based counts/qualities (MU, MMB).
• Feature composition and unified score
  • Procedures for composing a quality feature vector and combining components into a unified quality score are specified, including mapping methods and training considerations.
• Encoding & interoperability
  • Defines binary and XML encodings for embedding quality values in biometric interchange formats and assigns quality algorithm identifiers for unambiguous interpretation.
• Conformance
  • Conformance levels and test assertions are specified (including the revised Annex A). A reference implementation (NFIQ/NFIQ 2) is cited for practical use.

Practical applications and users

Who benefits:

• Biometric system vendors and sensor manufacturers - to certify and report fingerprint image quality consistently.
• System integrators and developers - to implement quality‑based enrollment, capture feedback, quality thresholds and routing.
• Government agencies, border control, ID programs and law enforcement - to ensure images meet interoperability and forensic quality requirements.
• Test labs and certification bodies - to run conformance tests and benchmark algorithms.

Common use cases:

• Real‑time capture feedback (accept/reject) and operator guidance during enrollment.
• Automated quality gating and score‑based matching thresholds to reduce false accepts/ rejects.
• Interchange of quality metadata in ISO/IEC 19794‑4 / CBEFF formats for cross‑vendor compatibility.
• Comparative evaluation of quality assessment algorithms (e.g., NFIQ 2).

Related standards

• ISO/IEC 29794‑1 (sample quality framework)
• ISO/IEC 19794‑4, ISO/IEC 39794‑4 (finger image interchange formats)
• ISO/IEC 19794‑2, 19794‑3, 19794‑8 (feature/template formats)

Keywords: ISO/IEC 29794-4:2024, biometric sample quality, finger image quality, fingerprint image quality, NFIQ, orientation certainty, local clarity, ridge valley uniformity, frequency domain analysis, conformance.

Standards Content (Sample)

International
Standard
ISO/IEC 29794-4
Second edition
Information technology —
2024-09
Biometric sample quality —
Part 4:
Finger image data
Technologies de l'information — Qualité d'échantillon
biométrique —
Partie 4: Données d'image de doigt
Reference number
© ISO/IEC 2024
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
© ISO/IEC 2024 – All rights reserved
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Abbreviated terms . 2
5 Conformance . 2
6 Finger image quality measures . 3
6.1 Overview .3
6.1.1 General .3
6.1.2 Methods for mapping to the desired value range .3
6.1.3 Constituent of local quality measures .4
6.1.4 Constituent of global quality measures .4
6.1.5 Image preprocessing .4
6.1.6 Image examples .7
6.2 Normative contributive quality components .7
6.2.1 General .7
6.2.2 Orientation certainty level .7
6.2.3 Local clarity .9
6.2.4 Frequency domain analysis (FDA) .14
6.2.5 Ridge valley uniformity . 15
6.2.6 Orientation flow .17
6.2.7 MU .19
6.2.8 MMB .19
6.2.9 Minutiae count in finger image .19
6.2.10 Minutiae count in centre of mass region . 20
6.2.11 Minutiae quality based on local image mean . 20
6.2.12 Minutiae quality based on local orientation certainty level.21
6.2.13 Region of interest image mean . 22
6.2.14 Region of interest orientation map coherence sum .24
6.2.15 Region of interest relative orientation map coherence sum . 25
6.2.16 Quality feature vector composition . 25
6.3 Non-normative quality measures . 28
6.3.1 General . 28
6.3.2 Radial power spectrum . 29
6.3.3 Gabor filter bank. 30
6.4 Unified quality score .32
6.4.1 Methodology for combining quality components .32
6.4.2 Training method .32
7 Finger image quality block . .33
7.1 Binary encoding . 33
7.2 XML encoding . 33
7.3 Quality algorithm identifiers . 33
Annex A (normative) Conformance test assertions .35
Annex B (informative) Factors influencing fingerprint image quality.57
Annex C (informative) Area consideration .59
Bibliography .60

© ISO/IEC 2024 – All rights reserved
iii
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are
members of ISO or IEC participate in the development of International Standards through technical
committees established by the respective organization to deal with particular fields of technical activity.
ISO and IEC technical committees collaborate in fields of mutual interest. Other international organizations,
governmental and non-governmental, in liaison with ISO and IEC, also take part in the work.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of document should be noted. This document was drafted in accordance with the editorial rules of the ISO/
IEC Directives, Part 2 (see www.iso.org/directives or www.iec.ch/members_experts/refdocs).
ISO and IEC draw attention to the possibility that the implementation of this document may involve the
use of (a) patent(s). ISO and IEC take no position concerning the evidence, validity or applicability of any
claimed patent rights in respect thereof. As of the date of publication of this document, ISO and IEC had not
received notice of (a) patent(s) which may be required to implement this document. However, implementers
are cautioned that this may not represent the latest information, which may be obtained from the patent
database available at www.iso.org/patents and https://patents.iec.ch. ISO and IEC shall not be held
responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO's adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT) see www.iso.org/iso/foreword.html.
In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology,
Subcommittee SC 37, Biometrics.
This second edition cancels and replaces the first edition (ISO/IEC 29794-4:2017), which has been technically
revised.
The main changes are as follows:
— algorithms for normalization of finger image quality components have been added, along with new quality
algorithm identifiers for the unique identification of the quality measures defined in this document;
— Annex A has been technically revised to reflect a new conformance test set.
A list of all parts in the ISO/IEC 29794 series can be found on the ISO and IEC websites.
Any feedback or questions on this document should be directed to the user’s national standards
body. A complete listing of these bodies can be found at www.iso.org/members.html and
www.iec.ch/national-committees.

© ISO/IEC 2024 – All rights reserved
iv
Introduction
This document specifies finger image quality measures. A reference implementation of the normative
measures — NFIQ 2 — is available at Reference [16], which is described in more detail by the developers in
Reference [1].
The quality of finger image data is determined by the degree to which the finger image data fulfils
specified requirements for the targeted application. Information on quality is therefore useful in many
applications. ISO/IEC 19784-1 allocates a quality field and specifies the allowable range for the scores, with
a recommendation that the score be divided into four categories with a qualitative interpretation for each
category. Finger image quality fields are provided in the finger image data interchange formats standardized
in ISO/IEC 19794-4 and ISO/IEC 39794-4. Finger feature data interchange formats standardized in
ISO/IEC 19794-2, ISO/IEC 19794-3, ISO/IEC 19794-8 and ISO/IEC 39794-2 provide finger image quality
fields for the source image. To facilitate the interpretation and interchange of finger image quality scores,
this document specifies how to calculate the finger image quality score of plain finger images with a spatial
sampling rate of 196,85 px/cm and a bit depth of 8 bit for the greyscale pixel intensity values scanned from
inked fingerprint cards or captured using optical area sensors based on frustrated total internal reflection.

© ISO/IEC 2024 – All rights reserved
v
International Standard ISO/IEC 29794-4:2024(en)
Information technology — Biometric sample quality —
Part 4:
Finger image data
1 Scope
This document establishes:
— terms and definitions for quantifying finger image quality;
— methods used to quantify the quality of finger images; and
— standardized encoding of finger image quality;
for finger images at 196,85 px/cm spatial sampling rate and a bit depth of 8 bit for the greyscale pixel
intensity values scanned or captured using optical area sensors in direct contact with friction ridges.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements of this document. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
ISO/IEC 2382-37, Information technology — Vocabulary — Part 37: Biometrics
ISO/IEC 39794-1, Information technology — Extensible biometric data interchange formats — Part 1:
Framework
ISO/IEC 29794-1, Information technology — Biometric sample quality — Part 1: Framework
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/IEC 2382-37, ISO/IEC 29794-1 and
the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1
foreground region
set of all pixels of a finger image that form valid finger image patterns
Note 1 to entry: The most evident structural characteristic of a valid finger image is a pattern of interleaved ridges and
valleys.
3.2
local region
block of m × n pixels of the foreground region, where m and n are smaller than or equal to the width and the
height of the foreground region respectively

© ISO/IEC 2024 – All rights reserved
3.3
finger image quality assessment algorithm
algorithm to calculate a quality measure
Note 1 to entry: “Quality assessment algorithm” and “quality algorithm” are synonyms.
3.4
trim
removal of pixels from the top, left, bottom and right sides of a finger image that do not comprise the
foreground region
Note 1 to entry: The steps for trimming an image to form the foreground region are defined in 6.1.5.2.
4 Abbreviated terms
CBEFF Common Biometric Exchange File Format
DFT discrete Fourier transform
DT determine threshold
FDA frequency domain analysis
FJFX FingerJet Fingerprint Feature Extractor, Open Source Edition
LCL local clarity
NFIQ NIST Fingerprint Image Quality
OCL orientation certainty level
OFL orientation flow
QSND quality score normalisation dataset
RVU ridge valley uniformity
TIR total internal reflection
5 Conformance
A finger image quality assessment algorithm conforms to this document if it conforms to the normative
requirements of Clause 6.
A finger image quality block shall conform to this document if its structure and data values conform to the
formatting requirements of Clause 7 (finger image quality block) and if its quality values are computed using
the methods specified in 6.2 and 6.4.
A finger image quality assessment implementation conformant to this document may use the biometric
organization identifier of ISO/IEC JTC 1/SC 37, which is 257 (101 ), if it has been tested following the
Hex
conformance testing methodology in Clause A.2.
Conformance to normative requirements of Clause 7 is achieved by Level 1 and Level 2 conformance as
specified in ISO/IEC 39794-1:2019, Annex C. Conformance to normative requirements of 6.2 and 6.4 is
achieved by Level 3 conformance as specified in ISO/IEC 39794-1:2019, Annex C.
The conformance test assertion in Annex A shall apply.

© ISO/IEC 2024 – All rights reserved
6 Finger image quality measures
6.1 Overview
6.1.1 General
This clause establishes measures for predicting the utility of a finger image. Image quality measures from a
single image are useful to ensure the acquired image is suitable for recognition.
A complete finger image quality analysis shall examine both the local and global structures of the finger
image. Fingerprint local structure constitutes the main texture-like pattern of ridges and valleys within
a local region while valid global structure puts the ridges and valleys into a smooth flow for the entire
fingerprint. The quality of a finger image is determined by both its local and global structures. This clause
describes the features and characteristics of finger images at both local and global structures that are to be
used as quality components for quantifying finger image quality.
For applying the algorithms as described in 6.2 and 6.3, the finger image shall have a spatial sampling rate
of 196,85 pixels per centimetre (500 pixels per inch), a bit depth of 8 bit for the greyscale pixel intensity
values, with friction ridges represented by greyscale pixel intensity values lower than those for valleys.
The algorithms were developed using images of finger friction ridges in contact with an electronic capture
device and inked fingerprints digitized with an electronic scanner. The imaging devices and scanners are
considered free from geometric distortion and exhibit greyscale linearity and uniformity.
ISO/IEC 29794-1 requires that quality components be mapped to an integer value between 0 and 100,
inclusive.
6.1.2 Methods for mapping to the desired value range
6.1.2.1 Sigmoid function
The mapping of values between 0 and 1 inclusive is accomplished for several quality components with the
sigmoid function as shown in Formula (1):
−1
xx−
  
sigmoide()xx,,w =+1 xp (1)
0   
w
  
where
x
is a native quality measure value;
x
is the inflection point at which the function has the value 0,5;
w
is a scaling parameter determining the width of the region in which the function transitions from
ε to 1−ε ;
ε
is an infinitesimally small positive quantity.
The values computed from the sigmoid function will be mapped to the target value ranges (0 to 100) in
subsequent clauses.
6.1.2.2 Known ranges
When the range of values for a given quality measure is known (e.g. from 1 to 250, inclusive), the known
range function is used, as shown in Formula (2):
xx−min
 () 
101 (2)
 
maxm()xx− in()+ε
 
© ISO/IEC 2024 – All rights reserved
where
x
is a native quality measure value;
is the floor function giving the greatest integer ≤ x
x
 
 
ε
is an infinitesimally small positive quantity.
6.1.3 Constituent of local quality measures
A finger image is partitioned into local regions such that each local region contains sufficient ridge-valley
information, preferably having at least two clear ridges, while not overly constraining high curvature ridges.
For images with a spatial sampling rate of 196,85 pixels per centimetre (500 pixels per inch), the ridge
[2]
separation usually varies between 8 pixels to 12 pixels. A ridge separation comprises a ridge and a valley.
In order to cover two clear ridges, the local region size has to be greater than 24 pixels in both width and
height. The size for each local region shall be (32 × 32) pixels, which is sufficient to cover two clear ridges.
Instead of Cartesian coordinate, curvilinear coordinate along the ridge can also be used.
σ μ sum
NOTE The size of the local region used during computation of q (6.2.16.3), q (6.2.16.2), q (6.2.14),
OFL COH
OFL
rel
and q (6.2.15) in the reference implementation NFIQ 2 prior to version 2.3.0 deviates from size specified in this
COH
subclause.
6.1.4 Constituent of global quality measures
A global quality measure shall be computed over the whole finger image after trim to assess the utility of the
sample for fingerprint recognition.
6.1.5 Image preprocessing
6.1.5.1 Description
A segmentation process follows where each local region is labelled as background or foreground. There are
several segmentation approaches, such as using the average magnitude of the pixel-intensity gradient in
[2]
each local region.
This document does not prescribe segmentation methods, but notes that performing segmentation
influences several quality components. Constant or near constant areas of the input image shall be removed
according to 6.1.5.2 prior to computing quality using the measures specified in 6.2 and 6.3.
See Annex C for the area consideration.
6.1.5.2 Removal of near constant rows and columns in image
Prior to computing quality components, fingerprint images shall be trimmed to remove near constant rows
and columns on the margins. Pixel intensities take values [0, 255] for an 8-bit greyscale image. As a first
approximation of the region of interest, image columns and rows which are near constant white background
are removed. Using the algorithm specified below, a fixed threshold, T , is set to the greyscale pixel intensity
μ
of 250 to obtain the image without near constant areas.
The algorithm is visualized in Figure 1 and specified as follows:
a) For each row R in Î, starting from the top:
i
1) compute the row arithmetic mean μ ;
row
2) on the first occurrence where μ ≤T set top=R ;
row μ i
© ISO/IEC 2024 – All rights reserved
3) on the last occurrence where μ ≤ T set bottom=R .
row μ i
b) For each column C in Î, starting from the left:
i
1) compute the column arithmetic mean μ ;
col
2) on the first occurrence where μ ≤T set left=C ;
col μ i
3) on the last occurrence where μ ≤T set right=C .
col μ i
c) extract the trimmed region of interest, I, as the pixels of Î encompassed between and including the rows
top (a2) and bottom (a3) and the columns left (b2) and right (b3).
where
Î is the matrix of grey levels corresponding to the pixels of an image;
I is the matrix of grey levels corresponding to the pixels of an image after trim.

a)  Image prior to trimming near b)  Results of following the steps in
constant rows and columns on the 6.1.5.2
margins
Key
top
bottom
left
right
NOTE 1  In Figure 1 a), the area overlayed in green is a visualisation of Î.
NOTE 2  In Figure 1 b), the area overlayed in green is a visualisation of I.
NOTE 3  Each subfigure within Figure 1 contains a dashed black border.
Figure 1 — Example of removing near constant white rows and columns from an image

© ISO/IEC 2024 – All rights reserved
6.1.5.3 Foreground segmentation based on local standard deviation
For quality components that require a foreground mask to indicate regions containing the fingerprint, an
algorithm using local standard deviation is adopted.
The algorithm is specified as follows:
a) Normalize I to zero mean and unit standard deviation to produce I’.
b) For each local region V in I’:
1) compute the standard deviation of V as σ ;
v
2) mark the corresponding local region in I as foreground if σ >01, .
mask v
6.1.5.4 Computing the dominant ridge flow orientation for a local region from pixel-intensity
gradients
The dominant ridge flow orientation is determined by computing the pixel-intensity gradient information
and then determining the orientation of the principal variation axis.
The numerical gradient of the local region is determined using finite central difference for all interior pixels
in x-direction and y-direction, as shown in Formulae (3) and (4):
IIxy+11,,−−xy
() ()
f = (3)
x
II()xy,,+11−−()xy
f = (4)
y
With f and f , the dominant ridge flow orientation, angle()V , is determined analytically using the sine
x y
and cosine doubled angle determined from the arithmetic means of the pixel-intensity gradient covariances,
as shown in Formulae (5) to (12):
af= (5)
x
bf= (6)
y
cf= f (7)
xy
ac
 
C = (8)
 
cb
 
dc=+()ab− (9)
c
sin θ = (10)
()
d
ab−
cos()θ = (11)
d
sin θ
1  () 
angle V = arctan (12)
()
 
2 cos()θ
 
NOTE In Formulae (5), (6) and (7), the use of the overbar indicates the mean of the value.

© ISO/IEC 2024 – All rights reserved
6.1.6 Image examples
For algorithms operating in a block-wise manner the trimmed input image is subdivided into local regions
according to an overlay grid. This is demonstrated in Figure 2 b), in which the local region V (6,1) is used as
an example in local processing and is marked up using a bold blue line. Figure 2 c) shows an enlarged view
of V (6,1) and Figure 2 d) shows V (6,1) rotated according to its dominant ridge orientation computed using
Formula (12).
a) Input finger image b) Division into local c) Enlarged view of d) V (6,1) rotated
regions V (6,1) according to its
dominant ridge
orientation as
determined using
Formula (12)
Figure 2 — Example of computing the dominant ridge flow orientation for a local region
6.2 Normative contributive quality components
6.2.1 General
Subclause 6.2 specifies algorithms for computing finger image quality components that contribute to the
ISO/IEC 29794-4 quality feature vector and to the computation of the unified quality score.
6.2.2 Orientation certainty level
6.2.2.1 Description
[3]
The orientation certainty level (OCL) of a local region is a measure of the consistency of the orientations of
the ridges and valleys contained within this local region. The feature computes local quality and operates in
a block-wise manner.
The finger image within a (32 × 32) pixels local region [as shown in Figure 2 c)] generally consists of dark
ridge lines separated by white valley lines along the same orientation. The consistent ridge orientation and the
appropriate ridge and valley structure are distinguishable local characteristics of the fingerprint local region.
The pixel-intensity gradient (dx, dy) at a pixel describes the direction of the maximum pixel-intensity change
and its strength. By performing principal component analysis on the pixel-intensity gradients in a local
region, an orthogonal basis for the local region can be formed by finding its eigenvalues and eigenvectors.
The resultant first principal component contains the largest variance contributed by the maximum
total gradient change in the direction orthogonal to ridge orientation. The direction is given by the first
eigenvector and the value of the variance corresponds to the first eigenvalue, λ . On the other hand, the
max
resultant second principal component has the minimum change of gradient in the direction of ridge flow
which corresponds to the second eigenvalue, λ . The ratio between the two eigenvalues thus gives an
min
© ISO/IEC 2024 – All rights reserved
indication of the strength of the energy concentrated along the dominant direction with two vectors pointing
to the normal and tangential direction of the average ridge flow respectively.
6.2.2.2 Computing the eigenvalues and local orientation certainty
From the covariance matrix C [Formula (8)] the eigenvalues λ and λ are computed as shown in
min max
Formulae (13) and (14):
2 2
ab+− ab− +4c
()
λ = (13)
min
ab++ ()ab− +4c
λ = (14)
max
This yields the local orientation certainty level shown in Formula (15):
λ

min
10−>, if λ
max

local
λ
q = (15)
 max
OCL

0,.otherwise

which is a ratio in the interval [0,1] where 1 is highest certainty level and 0 is lowest.
NOTE The orientation certainty level fails to predict recognition performance when some marks or residue exist
in the samples that have strong orientation strength, such as those exhibited by latent prints left by the previous user
of a capture device.
6.2.2.3 OCL algorithm
For each local region V in I:
a) compute the pixel-intensity gradient of V with the centred differences method [Formulae (3), (4)];
b) compute the covariance matrix C [Formula (8)];
local
c) compute the eigenvalues of C to obtain q [Formulae (13), (14), (15)].
OCL
Figure 3 visualizes the processing steps.
local
a) Current local region with the ratio between
b) Original, untrimmed image and its q
OCL
eigenvalues marked as ellipse
values, mapped to values 0-255
Figure 3 — Processing steps of orientation certainty level quality algorithm

© ISO/IEC 2024 – All rights reserved
6.2.3 Local clarity
6.2.3.1 Description
[4]
Good quality fingerprints exhibit clear ridge-valley structure. Thus, the local clarity (LCL), which is the
measure of the ridge-valley structure clarity, is a useful indicator of the quality of a fingerprint. The feature
computes local quality and operates in a block-wise manner.
To perform ridge-valley structure analysis, the foreground of the finger image is quantized into local
[3]
regions of size (32 × 32) pixels. Inside each local region, an orientation line, which is perpendicular to the
ridge direction, is computed. At the centre of the local region along the ridge direction, a local region of size
(32 × 16) pixels shall be extracted and transformed to a vertically aligned local region.
A linear regression (or least square fitting) is applied to calculate the determine threshold (DT), which is a
line positioned at the centre of the local region V used to segment the local region into the ridge and valley
regions. Regions with grey levels lower than DT are classified as ridges. Otherwise, they are classified as
valleys.
Since good finger images cannot have ridges that are too close or too far apart, the nominal ridge and valley
thickness can be used as a measure of the quality of the finger image captured. Similarly, ridges that are
unreasonably thick or thin indicate that the finger image has not been captured properly (due to pressing
too hard or too softly, for example), or that the image is a residual sample. Thus, the finger image quality can
be determined by comparing the ridge and valley thickness to each of their nominal range of values. Any
value out of the nominal range can imply a bad quality ridge pattern. To normalize the range of the thickness
values, a pre-set maximum thickness is used. The maximum ridge or valley thickness, W , for a good
max
finger image is estimated at 20 pixels for a 196,85 pixels per centimetre (500 pixels per inch) scanner spatial
sampling rate. The pre-set value of 20 pixels for a 196,85 pixels per centimetre (500 pixels per inch) scanner
[2]
spatial sampling rate is obtained from the median of the typical ridge separation of (8 – 12) pixels, and
assuming that any ridge separation will not exceed twice the median value. This will ensure that the pre-
set value is indeed the maximum to limit the value of the normalized ridge and valley thickness between 0
and 1. The ridge thickness, W , and valley thickness, W , are then normalized with respect to the maximum
r v
thickness, W .
max
With the ridge and valley separated as above, a clarity test can be performed in each segmented rectangular
2D region.
For local regions with good clarity, the pixel-intensity distribution of ridges and the pixel-intensity
local
distribution of the valleys have a very small overlapping area and thus q is high. The following factors
LCL
affect the size of the total overlapping area:
a) noise on ridge and valley;
b) water patches on the image due to wet fingers;
c) incorrect orientation angle due to the effect of directional noise;
d) scar across the ridge pattern;
e) highly curved ridges;
f) ridge endings, bifurcations, delta and core points;
g) incipient ridges, sweat pores, and dots.
Factors a) to c) are physical noise found in the image. Factors d) to g) are actual physical characteristics of
the fingerprint. See Annex B for the factors influencing fingerprint image quality.

© ISO/IEC 2024 – All rights reserved
6.2.3.2 Computing the ridge valley signature of a local region
Given the local region V, the ridge valley signature S is obtained according to Formula (16):
V
∑ xy,
y=1
S()x = (16)
where
V is the grey level at point ()xy, ;

xy,
x is the index along x -axis.
6.2.3.3 Determining the proportion of misclassified pixels
Formula (17) specifies the calculation of the proportion of pixels misclassified respectively as valley, and
Formula (18) specifies the calculation of the proportion of pixels misclassified as ridge.
v
B
α = (17)
v
T
r
B
β = (18)
r
T
where
α
is the proportion of pixels misclassified as valley;
β
is the proportion of pixels misclassified as ridge;
v
is the number of pixels in the valley region with intensity lower than DT;
B
v
is the total number of pixels in the valley region;
T
r
is the number of pixels in the ridge region with intensity higher than DT;
B
r
is the total number of pixels in the ridge region.
T
© ISO/IEC 2024 – All rights reserved
6.2.3.4 Determining the normalized ridge and valley width
The normalized valley width W and the normalized ridge width W are determined according to
v r
Formulae (19) and (20):
W
v
W = (19)
v
S ×25, 4
max
W
 
 
W
r
W = (20)
r
S ,×254
max
W
 
 
where
S is the spatial sampling rate in pixels per centimetre of the capture device;
max
W is the estimated ridge or valley width for an image with 49,21 pixels per centimetre
(125 pixels per inch) spatial sampling rate;
W and W
are the observed valley and ridge widths, respectively.
v r
max
According to Reference [2], W =5 is reasonable for 49,21 pixels per centimetre (125 pixels per inch)
S ×25, 4
 
max
spatial sampling rate. By extension, W is 20 for a spatial sampling rate of 196,85 pixels per
 
 125 
centimetre (500 pixels per inch).
6.2.3.5 Computing the local clarity
local
The local quality value q is the constrained average value of α and β with a range between 0 and 1, as
LCL
shown in Formula (21):
αβ+

nmin nmax nmin nmax
1− ,,if WW< vr
 () ()
v v r r
local
q = 2 (21)

LCL

0, ootherwise.

nmin nmin
where W and W are the minimum values for the normalized ridge and valley width, calculated
r v
according to Formulae (22) and (23) respectively:
nmin
W = (22)
r
W
r
nmin
W = (23)
v
W
v
nmax nmax
and where W and W are the maximum values for the normalized ridge and valley width, calculated
r v
according to Formulae (24) and (25) respectively:
nmax
W = (24)
r
W
r
nmax
W = (25)
v
W
v
© ISO/IEC 2024 – All rights reserved
local
NOTE Particular regions inherent in a fingerprint will negatively affect q . For example, ridge endings and
LCL
bifurcations or areas with high curvature such as those commonly found in core and delta points.
6.2.3.6 LCL algorithm
For each local region V in I:
a) rotate V such that dominant ridge flow is perpendicular to x-axis using nearest neighbour interpolation;
b) crop rotated V such that no invalid regions are included;
c) with V obtain the ridge-valley signature S (6.2.3.2);
d) determine DT using linear regression on S;
e) for each element S(x), set threshold T(x) of x being ridge or valley based on DT;
1,if ST()xx< ()
f) classify columns in V as ridge (1) or valley (0) with P()x ={ ;
0,otherwise.
g) determine ridge-valley transition vector C from P;
h) compute the vector W containing ridge and valley widths from C;
i) determine normalized ridge width and valley width W and W (6.2.3.4);
r v
j) determine the proportion of misclassified pixels α and β (6.2.3.3);
local
k) compute the local quality value q (6.2.3.5).
LCL
NOTE The reference implementation fills invalid regions with 0-valued pixels.
Figure 4 visualizes the processing steps.

© ISO/IEC 2024 – All rights reserved
a) Crop of current local region b) Average profile of local region c) Average local region profile
with linear regression line
d) Binarisation mask with e) Pixels determined to be ridge f) Pixels misclassified as valley
ridge and valley regions based on mask based on the threshold
based on regression line
g) Pixels determined to be h) Pixels misclassified as ridge i) Local clarities
valley based on mask based on the threshold
Key
X block x-index
Y mean intensity
Figure 4 — Processing steps of local clarity algorithm

© ISO/IEC 2024 – All rights reserved
6.2.4 Frequency domain analysis (FDA)
6.2.4.1 Description
Frequency domain analysis (FDA) computes local quality and operates in a block-wise manner. A one-
dimensional signature of the ridge-valley structure is extracted and the discrete Fourier transform (DFT) is
[6]
computed on the signature to determine the frequency of the sinusoid following the ridge-valley structure.
The ridge-valley signature of a high quality sample is a periodic signal, which can be approximated either by
a square wave or a sinusoidal wave. In the frequency domain, an ideal square wave should exhibit a dominant
frequency with sideband frequency components (sinc function). A sinusoidal wave consists of one dominant
frequency and minimum components at other non-dominant frequencies.
For each local region, a signature perpendicular to the dominant ridge flow orientation is computed.
The FDA described in 6.2.4 computes the one-dimensional signatures by performing averaging along the
ridge flow direction. The averaging process filters out noise along the ridge and valley flow and provides a
modelling of a smooth changing signal in a direction perpendicular to ridge flow.
6.2.4.2 Computing the local FDA quality component
The local quality value is computed by using Formula (26):
1,ifoFA==rFA

maxm1 ax A


local
AC++AA
q = () (26)
 FF F
FDA
maxmax−+11max
,.otherwise

A//2
A

F
∑

F=1
where
0,3 is the attenuation parameter C;
A
is the amplitude at frequency index x.
x
local
The value of q is set to 1 when the maximum frequency, F , amplitude occurs at index FA= or
FDA max max 1
FA= .
max A
6.2.4.3 FDA algorithm
For each local region V in I:
a) pad V with a 2-pixel border;
b) rotate V with nearest neighbour interpolation such that dominant ridge flow is perpendicular to the
x-axis with nearest neighbour interpolation;
c) crop V such that no invalid regions are included;
d) with V, obtain the ridge-valley signature S (6.2.3.2);
e) compute the DFT of S to obtain the magnitude representation A;
f) discard the DC component (i.e. the first component of A);
g) determine F as the index with the largest magnitude in A;
max
local
h) compute q of V using A and F (6.2.4.2).
FDA max
NOTE The reference implementation fills invalid regions with 0-valued pixels.
Figure 5 visualizes the processing steps.

© ISO/IEC 2024 – All rights reserved
a) Central area of input local region b) Ridge-valley profile
local
c) DFT of ridge-valley profile
d) Map of q
FDA
Key
X1 mean intensity
Y1 block x-index
X2 magnitude
Y2 cycle/pixel
Figure 5 — Processing steps of FDA quality algorithm
6.2.5 Ridge valley uniformity
6.2.5.1 Feature description
[3]
Ridge valley uniformity (RVU) is a measure of the consistency of the ridge and valley widths. . The
expectation for finger image with clear ridge and valley separation is that the ratio between ridge and valley
widths remains fairly constant throughout the finger image.

© ISO/IEC 2024 – All rights reserved
The ratio of ridge thickness to valley thickness should be constant and close to 1 throughout the whole image
for a good quality finger image. The feature computes local quality and operates in a block-wise manner.
NOTE The ridge valley uniformity depends on the spatial sampling rate.
6.2.5.2 RVU algorithm
For each local region V in I:
a) determine dominant ridge flow orientation angle()V of V;
b) rotate V such that angle V is perpendicular to x-axis;
()
c) crop V such that no invalid regions are included;
d) with V , obtain the ridge-valley signature S (6.2.3.2);
e) determine DT using linear regression on S;
f) for each S x , compute threshold T xx=×DT 1+DT 0 ;
() () () ()
g) binarize S using T;
11,if ST()xx− < ()
h) classify ridge and valley in S as P()x ={ ;
0,otherwise.
11,if PPxx− ≠
() ()
i) compute ridge-valley transition vector as C x ={ ;
()
0,otherwise.
j) drop first and last transition from S using C to remove incomplete ridges or valleys and obtain S';
local
k) compute q as the ratio of the width of ridges against valleys in S'.
RVU
Figure 6 visualizes the processing steps.

© ISO/IEC 2024 – All rights reserved
a) Crop of current local region b) Average profile of local region
c) Average profile with regression line d) Local native quality component values displayed
as the ratio of the width of ridges against valleys
Key
X block x-index
Y mean intensity
Figure 6 — Processing steps of ridge valley uniformity quality algorithm
6.2.6 Orientation flow
6.2.6.1 Description
[4]
Orientation flow (OFL) is a measure of ridge flow continuity which is based on the absolute orientation
difference between a local region and its 8-neighbourhood of local regions.
Orientation
...