ISO/TR 25221:2025
(Main)Electronic fee collection — Image-based tolling systems — Measurable characteristics
Electronic fee collection — Image-based tolling systems — Measurable characteristics
This document analyses the processes of image-based systems to be used for tolling purposes, with the aim to identify their specific characteristics, and where these characteristics can be observed. The study intends to answer the following questions: a) Which are the relevant characteristics of an image-based system used for electronic fee collection (EFC)? b) How can these characteristics be specified?
Perception de télépéage — Systèmes de péage basés sur l'analyse d'images — Caractéristiques mesurables
General Information
Standards Content (Sample)
Technical
Report
ISO/TR 25221
First edition
Electronic fee collection — Image-
2025-01
based tolling systems — Measurable
characteristics
Perception de télépéage — Systèmes de péage basés sur l'analyse
d'images — Caractéristiques mesurables
Reference number
© ISO 2025
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
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 Framework and classifications of discrete tolling systems . 3
5.1 General — Dimensions of the problem .3
5.2 General to processes and functional variables .3
6 Variables in image-based tolling systems . 10
6.1 General .10
6.2 Detection rate .11
6.2.1 Definition .11
6.2.2 Detection of false positives rate .11
6.2.3 Detection of false negatives rate . 12
6.3 Identification rate . 12
6.3.1 Definition . 12
6.3.2 ANPR result verification .14
6.4 Classification rate .14
6.5 Association of identified licence plates with registered vehicles .16
6.6 V erification system added value and verification error .16
7 Testing and performance evaluation . 17
7.1 Generalities .17
7.2 Suitability of characteristic variables .17
Annex A (informative) Image-based systems in contexts different from EFC . 19
Bibliography .20
iii
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out through
ISO technical committees. Each member body interested in a subject for which a technical committee
has been established has the right to be represented on that committee. International organizations,
governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely
with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
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 ISO documents 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).
ISO draws attention to the possibility that the implementation of this document may involve the use of (a)
patent(s). ISO takes 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 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. ISO 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.
This document was prepared by Technical Committee 204, Intelligent transport systems.
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
Introduction
[15]
The European Commission Implementing Regulation (EU) 2020/204 on detailed obligations of European
electronic toll service providers includes among the allowed tolling techniques: “electro-optical imaging
systems at the toll charger’s fixed or mobile equipment at the roadside, providing means for automatic
number plate recognition (ANPR), in EFC systems where the installation and use of an OBE is not required.”
ISO/TR 6026, produced by ISO/TC 204 in collaboration with CEN/TC278, identifies necessary areas of
standardization for image-based tolling. Activities to revise existing EFC standards to support ANPR
technologies have already been started.
It is well known that certified equipment is required, when ANPR is used for purposes other than tolling
(for example, limited traffic zones and speed limit enforcement), and that certification activity requires test
suites. This area has so far not been addressed in the field of EFC.
Also, while some phases in the process of electronic fee collection can be devised as technology independent,
at least the phases of recognition and the identification of vehicles are strictly dependent on the technology
used for tolling, so, in the specific case of ANPR, they depend on the ANPR technology.
Some regional standards (for example, UNI 10772) specify procedures for testing the optical and optical
character recognition (OCR) capabilities of ANPR systems, but the process chain of EFC is much wider than that.
A study is needed to identify characteristics of image-based systems for tolling to be tested for conformance
to specifications and to measure key performance indicators (KPIs).
It is recognized that image-based systems that are suitable for tolling can be used for other purposes.
Although such systems are out of the scope of the present document, informative Annex A is provided with
some examples and case studies.
v
Technical Report ISO/TR 25221:2025(en)
Electronic fee collection — Image-based tolling systems —
Measurable characteristics
1 Scope
This document analyses the processes of image-based systems to be used for tolling purposes, with the aim
to identify their specific characteristics, and where these characteristics can be observed. The study intends
to answer the following questions:
a) Which are the relevant characteristics of an image-based system used for electronic fee collection (EFC)?
b) How can these characteristics be specified?
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/TS 17573-2, Electronic fee collection — System architecture for vehicle related tolling — Part 2: Vocabulary
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO/TS 17573-2 and the following
terms and definitions 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
automated number plate recognition
ANPR
technology to automatically read vehicle registration plates
Note 1 to entry: A vehicle registration plate typically contains the indicator or the code of the country that issued the
vehicle registration plate.
Note 2 to entry: Optical character recognition techniques are typically part of the technology associated with
automated number plate recognition.
[SOURCE: ISO 17573-2:2020, 3.18]
3.2
enforcement
means to identify and pursue violators of laws, regulations or rules
3.3
false negative
incorrect reporting of a failure when in reality it is a pass
[SOURCE: ISO/IEC TR 29119-11:2020, 3.1.34]
3.4
false positive
incorrect reporting of a pass when in reality it is a failure
[SOURCE: ISO/IEC TR 29119-11:2020, 3.1.35]
3.5
formally valid licence plate
licence plate that has been correctly identified as for the nationality of the vehicle, the characters and the
numbers, and the associated format
3.6
free-flow tolling system
collection of tolls on toll roads without the use of physical toll barriers
3.7
constrained tolling system
collection of tolls on toll roads that impose restrictions (in road lanes or speeds, or both) on vehicles where
tolls are collected
Note 1 to entry: This covers, among others, all toll booths and toll plazas based tolling systems.
3.8
true negative
correct reporting of a failure when it is a failure
[SOURCE: ISO/IEC TR 29119-11:2020, 3.1.82]
3.9
true positive
correct reporting of a pass when it is a pass
[SOURCE: ISO/IEC TR 29119-11:2020, 3.1.83]
4 Symbols and abbreviated terms
A association rate
r
A number of correct ANPR results (true positives)
tp
C classification rate
r
C number of correctly classified vehicles (true or semi-true positives)
m
D number of detected false positives
fp
D number of detected false negatives
fn
D , detection of false negatives rate
nr
D detection of false positives rate
pr
D detection efficiency
r
Im ratio between the number of vehicles correctly identified by the secondary
secondary
system and the total number of correctly identified vehicles by the primary
and secondary systems
Ip is the number of vehicles correctly identified only by the secondary system
secondary
ID identification rate
r
P number of formally identified licence plates
f
P number of identified licence plates corresponding to existing real identified
r
vehicles, that combines the results of both the primary and the secondary
systems
V is the number of detected vehicles
d
V number of passed vehicles
t
AI artificial intelligence
ANPR automatic number plate recognition
DSRC dedicated short range communication
EFC electronic fee collection
KPI key performance indicator
LP licence plate
LPN licence plate number
OCR optical character recognition
RSE roadside equipment
5 Framework and classifications of discrete tolling systems
5.1 General — Dimensions of the problem
This document considers the characteristics of tolling systems where tolling is based on a number of
geographically fixed identification points where vehicle passages and characteristics of the vehicles are
observed. This is in contrast to tolling systems where tolling is based on the continuous recognition of how
long (in time or space) a vehicle has travelled in an area or how many times has it crossed borders between
defined areas. The considered tolling systems are known as discrete tolling systems.
An initial classification of discrete tolling systems can be made based on the geometrical characteristics
of the tolling points, by roughly dividing the systems into free-flow tolling systems and constrained
tolling systems. Another dimension, that can have an impact on the system’s performance is the presence
of multiple tolling technologies (e.g. DSRC manual payments, etc.), and their relevance to the processes
incorporated in the tolling system (e.g. process of toll calculation). These and other dimensions add to the
physical characteristics (e.g. communication, optical or OCR capabilities) of the tolling devices to form the
body of variables to be considered, measured, and ultimately tested, to evaluate the tolling system.
The characteristics of discrete tolling systems that are described in Clause 5 are independent of the
technology that is used for tolling.
5.2 General to processes and functional variables
[3]
In the US Department of Transportation's classification of congestion pricing technologies, the generic
tolling process, independent of the used technologies, can be divided into 7 sub-processes, each one
characterized by the set of variables.
The identified sub-processes are as follows (the order is not significant):
— Information and registration — This process is related to all communication aspects of both the tolling
system towards its users (signs, barriers, etc.), and the users towards the system (plate registration,
installation and personalization of OBE, etc.).
— Passage detection — This process recognizes a vehicle’s passage. The process is highly influenced by the
geometry of the identification points (free-flow, constrained, etc.).
— Vehicle identification — This process uniquely identifies a vehicle, e.g. by recognizing its licence plate,
or by reading its OBE identifier. The process is dependent on the used technology. It can use the same
technology as for the passage detection, or a different one.
— Classification — This process classifies an identified vehicle according to the toll regime vehicle classes.
This process can be performed with the same technologies used for passage detection or vehicle
identification.
— Verifications and reliability — Information collected by the above sub-processes can be verified by
further independent processes to enhance its reliability.
EXAMPLE The passage of a vehicle, that is recognized and classified by means of a DSRC transaction, can be
verified by reading its licence plate or by the recognition of its axles and dimensions by laser sensors.
— Payment — The payment process is generally independent of the technologies that are used to identify
and classify vehicles. However, it can be the case that further evidence is necessary for payment of a
toll. For example, it can be necessary that a picture of the licence plate, associated with the time and
geographical coordinates of the passage, is associated with a DSRC transaction.
— Enforcement— Enforcement is often associated with a technology alternative to that used for tolling. A
typical example is ANPR used to enforce a DSRC-based tolling system.
Not all the above listed sub-processes are necessarily always present in a tolling system. Also, the existence
and execution of one sub-process can in some cases influence the behaviour of other sub-processes.
The above sub-processes are listed without any temporal ordering. Figure 1 depicts the sub-processes by
outlining, in a grey rectangle, those that characterize a specific EFC system by the tolling technology used.
Figure 1 — EFC sub-processes
In Figure 1, two kinds of sub-processes are identified.
a) Core sub-processes, i.e. those sub-processes that are always present in any type of tolling system.
b) Auxiliary sub-processes, i.e. those sub-processes that can be present, e.g. to improve reliability (like
“verification and reliability”), or must be present, e.g. due to local regulations (like “registration”).
Some EFC systems organize their processes in a sequential manner, as it is presented in Figure 2. Others
feature some parallelism, as presented in Figure 3.
Figure 2 — Sequential ordering of EFC sub-processes
Figure 3 — Parallelism in EFC sub-processes
In all cases presented in Figure 3, it is assumed that, apart from a possible previous registration, the passage
detection is the first sub-process that happens when an EFC system is triggered by the passage of a vehicle.
However, it can happen that a single indivisible sub-process manages both detection, classification and
identification of vehicles, as shown in Figure 4.
Figure 4 — Detection, classification and identification as a single sub-process
Ordering of sub-processes has implications on the definition of the functional variables that characterize
any single sub-process, as shown in Clause 6.
Due to the different possible architectures of the processes implemented in any EFC system, specifying
characteristics that can be tested for a system as a whole is very difficult. This document intends to identify
the testable characteristics of each one of the above listed sub-processes independently of the other sub-
processes.
All identified sub-processes can be described in terms of one or more characterizing variables. The set of all
characterizing variables defines the tolling system dimensions and allows for identifying its performance
indicators and critical aspects. Some variables are qualitative, others are not orthogonal with each other,
so that the resulting analysis cannot be performed with pure numerical methods. More information on
variables for image-based tolling systems is given in Clause 6.
Table 1 lists the characteristic variables associated to each sub-process. Note that the order of the rows in
the table (i.e. the ordering of processes) is of no significance.
Table 1 — Tolling processes and associated variables
Sub-process or Variables Type Range of values Comments
main function-
ality
Information and Required registra- Qualitative YES/NO Registration improves vehicle
registration tion identification.
Passage detection System geometry Qualitative Example: Free-flow, System geometry affects identifi-
slow and go, stop and cation, classification and detec-
go, etc. tion.
Detection technol- Qualitative Examples: None, laser If detection is separate from iden-
ogy trigger, DSRC, camera, tification, its efficiency, including
etc. false positives, must be consid-
ered.
Detection rate Quantitative Percentage Percentage of detected over
passed vehicles.
False positives Quantitative Percentage Percentage of false positives over
passed vehicles. Includes dupli-
cates, like trailers counted as
separate vehicles.
Classification Separation from Qualitative YES/NO If separate, classification rate is a
identification useful measurement.
Classification rate Quantitative Percentage Percentage of correctly classified
vehicles over detected vehicles.
Tolling classes Qualitative Defined tolling classes The more tolling classes, the more
difficult it is to assign a correct
classification.
Identification of Technology Qualitative Examples: DSRC, ANPR, Used to evaluate separate systems
vehicles etc. with different technologies.
Identification rate Quantitative Percentage Percentage of automatically
identified vehicles over correctly
detected vehicles.
Association rate Quantitative Percentage Percentage of actually registered
vehicles over formally identified
vehicles.
Verifications and Verification sys- Quantitative 0-Max Additionally identified vehicles
reliability tem added value when a secondary system is used.
It is a measure of the added value
provided by a secondary system.
Verification error Quantitative Percentage Rate of incorrect vs. total verifica-
tions.
Payment Payment type Qualitative — Immediate direct
debit
— Postponed direct
debit
— Postponed indirect
debit
— Etc.
Enforcement Presence of an en- Qualitative YES/NO
forcement system
Technology of en- Qualitative
forcement system
6 Variables in image-based tolling systems
6.1 General
ISO/TR 6026 intends to clarify some basic concepts for image-based EFC systems and to identify a number
of possible standardization activities for those systems. The document contains a description of the tolling
process but does not identify the variables that characterize image-based systems and that can be used to
determine their critical aspects.
Of the variables used to characterize general EFC systems listed in Table 1, the following ones, both
qualitative and quantitative, are relevant for classifying and measuring the efficiency of image-based
systems:
— required registration (qualitative);
— system geometry (qualitative);
— detection technology (qualitative);
— detection rate (quantitative);
— classification rate (quantitative);
— tolling classes (qualitative);
— identification technology (qualitative);
— identification rate (quantitative);
— association rate (quantitative);
— verification system added value (quantitative);
— verification error (quantitative).
The qualitative variables of the list given in 6.1 can only be used to correctly classify and group together
different image-based tolling systems into uniform categories. However, although some of these qualitative
variables can influence the values of quantitative variables, their effects are rarely measurable, nor are they
precisely predictable.
Subclauses 6.2 to 6.4 analyse only the quantitative variables. Most of these variables are independent from
each other. In the case where they are dependent, that dependency is clearly indicated. The implications of
using any of the following variables as KPIs, or as characteristics of a given system subject to test are also
highlighted. Commonly, a reference system is used to determine the baseline data (also known as ground
truth) when measuring KPIs.
An image-based tolling system is made by different components and its performance depends on specific
conditions, both intrinsic (e.g. geometry and redundancy, presence of an external trigger, etc.) and external
(e.g. weather conditions, general conditions of the circulating vehicles, lighting and latitude, etc.). Such
integrated and composite nature makes it impossible to identify a generalized type of architecture for
image-based tolling systems, thus rendering metrics measurement not feasible on overall system level.
6.2 Detection rate
6.2.1 Definition
The detection efficiency D of the detection sub-process is defined as the ratio between the number of
r
detected vehicles and the number of vehicles that have passed through the detection point:
V
d
D =
r
V
t
where:
V is the number of detected vehicles;
d
V is the number of passed vehicles.
t
Detection of vehicles can be performed by different technologies, which can be integrated in the image
recognition system, or be separated from it. Some of detection technologies that are used by image-based
tolling systems are (the list is not exhaustive):
— Laser based triggers — A laser scanner detects an object moving towards the detection point, and by
that it starts the image recognition system.
— Radar based triggers — Same as for laser-based triggers, the difference being that incoming objects are
recognized by a radar.
— Image recognition — The detection and recognition of incoming vehicles is performed by the same system.
Detection rate depends on several components and, at the same ti
...
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...