ISO/IEC 15444-4:2021

Information technology - JPEG 2000 image coding system - Part 4: Conformance Testing

This Recommendation | International Standard specifies the framework, concepts, methodology for testing, and criteria to be achieved to claim compliance to Rec. ITU-T T.800 | ISO/IEC 15444-1 or Rec. ITU-T T.814 | ISO/IEC 15444-15. It provides a framework for specifying abstract test suites (ATSs) and for defining the procedures to be followed during compliance testing. This Recommendation | International Standard: ‒ specifies compliance testing procedures for encoding and decoding using Rec. ITU-T T.800 | ISO/IEC 15444-1 and Rec. ITU-T T.814 | ISO/IEC 15444-15; ‒ specifies codestreams, decoded images, and error metrics to be used with the testing procedures; ‒ specifies ATSs; ‒ provides guidance for creating an encoder compliance test This Recommendation | International Standard does not include the following tests: Acceptance testing: the process of determining whether an implementation satisfies acceptance criteria and enables the user to determine whether or not to accept the implementation. This includes the planning and execution of several kinds of tests (e.g., functionality, quality, and speed performance testing) that demonstrate that the implementation satisfies the user requirements. Performance testing: measures the performance characteristics of an implementation under test (IUT) such as its throughput and responsiveness, under various conditions. Robustness testing: the process of determining how well an implementation processes data which contains errors.

Technologies de l'information — Système de codage d'images JPEG 2000 — Partie 4: Tests de conformité

General Information

• Status: Withdrawn
• Publication Date: 28-Oct-2021
• ICS: 35.040.30 - Coding of graphical and photographical information
• Technical Committee: ISO/IEC JTC 1/SC 29 - Coding of audio, picture, multimedia and hypermedia information
• Drafting Committee: ISO/IEC JTC 1/SC 29/WG 1 - JPEG Coding of digital representations of images
• Current Stage: 9599 - Withdrawal of International Standard
• Start Date: 13-May-2024
• Completion Date: 30-Oct-2025

Relations

• Revised: ISO/IEC 15444-4:2024 - Information technology - JPEG 2000 image coding system - Part 4: Conformance testing
• Effective Date: 24-Sep-2022
• Revises: ISO/IEC 15444-4:2004 - Information technology - JPEG 2000 image coding system: Conformance testing - Part 4:
• Effective Date: 17-Jul-2021

Standards Content (Sample)

INTERNATIONAL ISO/IEC
STANDARD 15444-4
Third edition
2021-10
Information technology — JPEG 2000
image coding system —
Part 4:
Conformance Testing
Technologies de l'information — Système de codage d'images JPEG
2000 —
Partie 4: Tests de conformité
Reference number
© ISO/IEC 2021
© ISO/IEC 2021
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
ii
© ISO/IEC 2021 – All rights reserved

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.
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent rights. Details
of any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www.iso.org/patents) or the IEC list of patent
declarations received (see https://patents.iec.ch).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to the World
Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT)
see www.iso.org/iso/foreword.html. In the IEC, see www.iec.ch/understanding-standards.
This document was prepared by ITU-T (as ITU-T REC. T.803) and drafted in accordance with its editorial
rules, in collaboration with Joint Technical Committee ISO/IEC JTC 1, Information technology,
Subcommittee SC 29, Coding of audio, picture, multimedia and hypermedia information.
This third edition cancels and replaces the second edition (ISO/IEC 15444-4:2004), which has been
technically revised.
The main changes are as follows:
— addition of the criteria to be achieved to claim compliance with Rec. ITU-T 814 | ISO/IEC 15444-15.
A list of all parts in the ISO/IEC 15444 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 2021 – All rights reserved iii

TABLE OF CONTENTS
Page
1 Scope . 1
2 References . 1
3 Definitions . 1
4 Abbreviations and symbols . 4
4.1 Abbreviations . 4
4.2 Symbols . 5
5 Conventions . 6
6 General description . 6
6.1 Profiles, derived sets and compliance classes . 7
6.2 Decoders . 8
6.3 Encoders and codestreams . 8
6.4 Implementation compliance statement . 8
6.5 Abstract test suites . 9
6.6 Encoder compliance testing procedure . 9
6.7 Decoder compliance testing procedure . 9
7 Copyright . 9
8 Compliance files availability and updates . 9
Annex A Decoder compliance classes . 10
A.1 Compliance class parameter definitions . 10
A.1.1 Profile: codestream guarantees. 10
A.1.2 H, W, C: Image size guarantees . 10
A.1.3 N : Code-block parsing guarantee . 11
cb
A.1.4 N : Component parsing guarantee . 11
comp
A.1.5 L : Coded data buffering guarantee . 11
body
A.1.6 M: Decoded bit-plane guarantee . 12
A.1.7 P: 9-7I precision guarantee . 12
A.1.8 B: 5-3R precision guarantee . 12
A.1.9 T : Transform level guarantee . 12
L
A.1.10 L: Layer guarantee . 12
A.1.11 Progressions . 12
A.1.12 Tile-parts . 13
A.1.13 Precincts . 13
A.1.14 M : Magnitude bound guarantee . 13
MAGB
A.2 Compliance class definitions . 13
A.3 Lossless encoding and decoding . 14
Annex B Decoder compliance testing procedures . 15
B.1 General . 15
B.2 Decoder test procedure . 15
B.2.1 Files for testing . 16
B.2.2 Decoder settings . 16
B.2.3 Output file format conversion . 16
B.2.4 Compare decoded and formatted components with reference components . 18
B.2.5 Compare error metrics with specification . 18
B.2.6 Reference components file format . 19
Annex C Compliance tests . 20
C.1 Abstract test suite (informative) . 20
C.1.1 Syntax and compressed data order . 20
C.1.2 Arithmetic entropy encoding . 20
C.1.3 Coefficient bit modelling . 21
C.1.4 Quantization . 21
C.1.5 Discrete wavelet transform . 21
C.1.6 DC level shift and multiple component transform . 21
C.1.7 Region of interest . 21
C.1.8 JP2 file format . 22
Rec. ITU-T T.803 (06/2021) v
© ISO/IEC 2021 – All rights reserved

C.1.9 High throughput cleanup pass coding . 22
C.1.10 HT refinement pass coding . 22
C.1.11 Placeholder passes . 22
C.1.12 Mixing of HT and J2K code-blocks within HTJ2K codestreams . 22
C.1.13 JPH File format . 22
C.2 Executable test suite . 23
C.2.1 Class 0 Profile-0 . 23
C.2.2 Class 0 Profile-1 . 27
C.2.3 Class 1 Profile-0 . 28
C.2.4 Class 1 Profile-1 . 29
Annex D Encoder compliance test procedure . 31
D.1 General . 31
D.2 Reference decoder . 31
D.3 Compliance requirement and acceptance . 31
D.4 Encoding compliance test procedure . 31
Annex E Decoder implementation compliance statement . 33
E.1 General . 33
E.2 Decoder implementation compliance statement . 33
E.3 Extended support . 33
Annex F Encoder implementation compliance statement . 36
F.1 General . 36
F.2 Encoder description . 36
Annex G JP2 and JPH file format reader compliance testing procedures . 38
G.1 General . 38
G.2 JP2 file compliance requirement and acceptance . 38
G.3 Reading a JP2 file compliance test procedure . 38
G.4 JP2 file format test codestreams and images . 39
G.4.1 Test files . 39
G.4.2 Reference decoded images . 39
G.4.3 Tolerances . 39
G.4.4 Additional information regarding the JP2 test files . 40
G.5 JPH file format test codestreams and images . 41
G.5.1 Test files . 41
G.5.2 Relationship between the JP2 and JPH test files . 41
vi Rec. ITU-T T.803 (06/2021)
© ISO/IEC 2021 – All rights reserved

INTERNATIONAL STANDARD
ITU-T RECOMMENDATION
Information technology –
JPEG 2000 image coding system: Conformance testing
1 Scope
This Recommendation | International Standard specifies the framework, concepts, methodology for testing, and criteria
to be achieved to claim compliance to Rec. ITU-T T.800 | ISO/IEC 15444-1 or Rec. ITU-T T.814 | ISO/IEC 15444-15.
It provides a framework for specifying abstract test suites (ATSs) and for defining the procedures to be followed during
compliance testing.
This Recommendation | International Standard:
‒ specifies compliance testing procedures for encoding and decoding using Rec. ITU-T T.800 |
ISO/IEC 15444-1 and Rec. ITU-T T.814 | ISO/IEC 15444-15;
‒ specifies codestreams, decoded images, and error metrics to be used with the testing procedures;
‒ specifies ATSs;
‒ provides guidance for creating an encoder compliance test
This Recommendation | International Standard does not include the following tests:
Acceptance testing: the process of determining whether an implementation satisfies acceptance criteria and enables the
user to determine whether or not to accept the implementation. This includes the planning and execution of several
kinds of tests (e.g., functionality, quality, and speed performance testing) that demonstrate that the implementation
satisfies the user requirements.
Performance testing: measures the performance characteristics of an implementation under test (IUT) such as its
throughput and responsiveness, under various conditions.
Robustness testing: the process of determining how well an implementation processes data which contains errors.
2 References
The following Recommendations and International Standards contain provisions which, through reference in this text,
constitute provisions of this Recommendation | International Standard. At the time of publication, the editions indicated
were valid. All Recommendations and Standards are subject to revision, and parties to agreements based on this
Recommendation | International Standard are encouraged to investigate the possibility of applying the most recent
edition of the Recommendations and Standards listed below. Members of IEC and ISO maintain registers of currently
valid International Standards. The Telecommunication Standardization Bureau of the ITU maintains a list of currently
valid ITU-T Recommendations.
– Recommendation ITU-T T.800 (2019) | ISO/IEC 15444-1:2019, Information technology – JPEG 2000
image coding system: Core coding system.
– Recommendation ITU-T T.814 (2019) | ISO/IEC 15444-15:2019, Information technology – JPEG 2000
image coding system: High-throughput JPEG 2000.
3 Definitions
For the purposes of this Recommendation | International Standard, the terms and definitions given in
Rec. ITU-T T.800 | ISO/IEC 15444-1, Rec. ITU-T T.814 | ISO/IEC 15444-15 and the following apply.
3.1 abstract test suite (ATS): Generic compliance testing concepts and procedures for a given requirement.
3.2 arithmetic coder: An entropy coder that converts variable length strings to variable length codes (encoding)
and vice versa (decoding).
3.3 big endian: An order of bytes with the most significant byte first.
3.4 bit: A contraction of the term "binary digit"; a unit of information represented by a 0 or a 1.
3.5 bit-depth: The number of bits required to represent an original component of an image.
Rec. ITU-T T.803 (06/2021) 1
© ISO/IEC 2021 – All rights reserved

3.6 bit-plane: A two-dimensional array of bits. In this Recommendation | International Standard, a bit-plane
refers to all the bits of the same magnitude in all coefficients or samples. This could refer to a bit-plane in a component,
tile- component, code-block, region of interest, or other.
3.7 bitstream: The actual sequence of bits resulting from the coding of a sequence of symbols. It does not include
the markers or marker segments in the main and tile-part headers or the end of codestream marker. It does include any
packet headers and in stream markers and marker segments not found within the main or tile-part headers.
3.8 box: A portion of the file format defined by a length and unique box type. Boxes of some types may contain
other boxes.
3.9 byte: Eight bits.
3.10 Cclass: Defines a level of performance for a decoder. Also provides guidance for encoders to produce
codestreams that are easily decodable by compliant decoders.
3.11 code-block: A rectangular grouping of coefficients from the same sub-band of a tile-component.
3.12 coder: An embodiment of either an encoding or decoding process.
3.13 codestream: A collection of one or more bitstreams and the main header, tile-part headers, and the end of
codestream required for their decoding and expansion into image data. This is the image data in a compressed form with
all of the signalling needed to decode. This does not include the file format.
3.14 coding pass: A procedure accessing coefficients in a code-block where the context and bit are determined.
Typically, there are three different coding passes for each bit-plane, each coefficient will be represented in exactly one
of the three passes. For an encoder a coding pass examines coefficients and augments a bitstream. For a decoder a
coding pass reads a bitstream and updates coefficients.
3.15 coefficient: The values that are the result of a transformation.
3.16 component: A two-dimensional array of samples. An image typically consists of several components (e.g.,
red, green, and blue).
3.17 compressed image data: Part or all of a codestream. Can also refer to a collection of bitstreams in part or all
of a codestream.
3.18 compliance: Fulfilment of the specified requirements, as defined in this Recommendation | International
Standard, for a given Profile and Cclass.
3.19 compliance test procedure: The process of assessing compliance.
3.20 context: Function of coefficients previously decoded and used to condition the decoding of the present
coefficient.
3.21 decoder: An embodiment of a decoding process, and optionally a colour transformation process.
3.22 decoding process: A process that takes as its input all or part of a codestream and outputs all or part of a
reconstructed image.
3.23 decomposition level: A collection of wavelet sub-bands where each coefficient has the same spatial impact or
span with respect to the source component samples. These include all sub-bands of the same two-dimensional sub-band
decomposition. For the last decomposition level, the LL sub-band is also included.
3.24 discrete wavelet transformation (DWT): A transformation that iteratively transforms one signal into two or
more filtered and decimated signals corresponding to different frequency bands. This transformation operates on
spatially discrete samples.
3.25 encoder: An embodiment of an encoding process, and optionally a colour transformation process.
3.26 encoding process: A process that takes as its input all or part of a source image data and outputs a
codestream.
3.27 executable test suite (ETS): Set of executable test cases that support the abstract test cases.
3.28 file format: A codestream and additional support data and information not explicitly required for the
decoding of the codestream. Examples of such support data include text fields providing titling, security and historical
information, data to support placement of multiple codestreams within a given data file, and data to support exchange
between platforms or conversion to other file formats.
3.29 fully decode: Applying Rec. ITU-T T.800 | ISO/IEC 15444-1 to produce an image from a codestream where
all coded data in the codestream has been used to produce the image.
2 Rec. ITU-T T.803 (06/2021)
© ISO/IEC 2021 – All rights reserved

3.30 guard bits: Additional most significant bits that have been added to sample data.
3.31 header: Either a part of the codestream that contains only markers and marker segments (main header and tile
part header) or the signalling part of a packet (packet header).
3.32 image: The set of all components.
3.33 image data: The component samples making up an image. Image data can refer to either the source image
data or the reconstructed image data.
3.34 implementation: A realization of a specification.
3.35 implementation compliance statement (ICS): Statement of specification options and the extent to which
they have been implemented by an implementation under test.
3.36 implementation under test (IUT): An implementation that is being evaluated for compliance.
3.37 irreversible: A transformation, progression, system, quantization, or other process that, due to systemic or
quantization error, prevents lossless recovery.
3.38 JP2 file: The name of a file in the file format described in this Recommendation | International Standard.
Structurally, a JP2 file is a contiguous sequence of boxes.
3.39 JPEG: Joint Photographic Experts Group – The joint ISO/ITU committee responsible for developing
standards for continuous-tone still picture coding. It also refers to the standards produced by this committee:
Rec. ITU-T T.81 | ISO/IEC 10918-1, Rec. ITU-T T.83 | ISO/IEC 10918-2, Rec. ITU-T T.84 | ISO/IEC 10918-3 and
Rec. ITU-T T.87 | ISO/IEC 14495-1.
3.40 LL sub-band: The sub-band obtained by forward horizontal low-pass filtering and vertical low-pass filtering.
This sub-band contributes to reconstruction with inverse vertical low-pass filtering and horizontal low-pass filtering.
3.41 layer: A collection of compressed image data from coding passes of one, or more, code-blocks of a tile-
component. Layers have an order for encoding and decoding that has to be preserved.
3.42 lossless: A descriptive term for the effect of the overall encoding and decoding processes in which the output
of the decoding process is identical to the input to the encoding process. Distortion-free restoration can be assured. All
of the coding processes or steps used for encoding and decoding are reversible.
3.43 lossy: A descriptive term for the effect of the overall encoding and decoding processes in which the output of
the decoding process is not identical to the input to the encoding process. There is distortion (measured
mathematically). At least one of the coding processes or steps used for encoding and decoding is irreversible.
3.44 main header: A group of markers and marker segments at the beginning of the codestream that describe the
image parameters and coding parameters that can apply to every tile and tile-component.
3.45 marker: A two-byte code in which the first byte is hexadecimal FF (0xFF) and the second byte is a value
between 1 (0x01) and hexadecimal FE (0xFE).
3.46 marker segment: A marker and associated (not empty) set of parameters.
3.47 packet: A part of the codestream comprising a packet header and the compressed image data from one layer
of one precinct of one resolution level of one tile-component.
3.48 packet header: Portion of the packet that contains signalling necessary for decoding that packet.
3.49 parser: Reads and identifies components of the codestream down to the code-block level.
3.50 partial decoding: Producing an image from a subset of an entire codestream.
3.51 precinct: A rectangular region of a transformed tile-component, within each resolution level, used for limiting
the size of packets.
3.52 precision: Number of bits allocated to a particular sample, coefficient, or other binary numerical
representation.
3.53 progression: The order of a codestream where the decoding of each successive bit contributes to a "better"
reconstruction of the image. What metrics make the reconstruction "better" is a function of the application. Some
examples of progression are increasing resolution or improved sample fidelity.
3.54 profile: A subset of technology, from Rec. ITU-T T.800 | ISO/IEC 15444-1, that meets the needs of a given
application with limits on parameters within a selected technology. This is a codestream limitation.
Rec. ITU-T T.803 (06/2021) 3
© ISO/IEC 2021 – All rights reserved

3.55 quantization: A method of reducing the precision of the individual coefficients to reduce the number of bits
used to represent them. This is equivalent to division while compressing and multiplying while decompressing.
Quantization can be achieved by an explicit operation with a given quantization value (scalar quantization) or by
dropping (truncating) coding passes from the codestream.
3.56 reconstructed image: An image that is the output of a decoder.
3.57 reference grid: A regular rectangular array of points used to define other rectangular arrays of data. The
reference grid is used to determine the number of samples in tile-components for example.
3.58 region of interest (ROI): A collection of coefficients that are considered of particular relevance by some
user-defined measure.
3.59 reversible: A transformation, progression, system, or other process that does not suffer systemic or
quantization error and therefore allows for lossless signal recovery.
3.60 reversible filter: A particular filter pair used in the wavelet transformation which allows lossless
compression.
3.61 sample: One element in the two-dimensional array that comprises a component.
3.62 selective arithmetic coding bypass: A coding style where some of the code-block passes are not coded by
the arithmetic coder. Instead, the bits to be coded are appended directly to the bitstream without coding.
3.63 shift: Multiplication or division of a number by powers of two. Division of an integer via shift implies
truncation toward minus infinity of the non-integer portion.
3.64 sign bit: A bit that indicates whether a number is positive (value 0) or negative (value 1).
3.65 sign-magnitude notation: A binary representation of an integer where the distance from the origin is
expressed with a positive number and the direction from the origin (positive or negative) is expressed with a separate
single sign bit.
3.66 source image: An image used as input to an encoder.
3.67 sub-band: A group of transform coefficients resulting from the same sequence of low-pass and high-pass
filtering operations, both vertically and horizontally.
3.68 testing: The process of evaluating compliance.
3.69 tile: A rectangular array of points on the reference grid, registered with an offset from the reference grid
origin and defined by a width and height.
3.70 tile-component: All the samples of a given component in a tile.
3.71 tile-part: A portion of the codestream with compressed image data for some, or all, of a tile. The tile-part may
include one or more packets that make up the coded tile.
3.72 tile-part header: A group of markers and marker segments at the beginning of each tile-part in the
codestream that describe the tile-part coding parameters.
3.73 transformation: A mathematical mapping from one signal space to another.
3.74 transform coefficient: A value that is the result of a transformation.
4 Abbreviations and symbols
4.1 Abbreviations
For the purposes of this Recommendation | International Standard, the abbreviations given in Rec. ITU-T T.800 |
ISO/IEC 15444-1 and the following apply.
ATS Abstract Test Suite
BSET subset of the ETS consisting of HTJ2K test codestreams that differ only by B value
MAGB
ETS Executable Test Suite
HT High Throughput
HTJ2K High Throughput JPEG 2000
ICC International Colour Consortium
4 Rec. ITU-T T.803 (06/2021)
© ISO/IEC 2021 – All rights reserved

ICS Implementation Compliance Statement
ICT Irreversible Component Transform
IDWT Inverse Discrete Wavelet Transformation
IEC International Electrotechnical Commission
ISO International Organization for Standardization
ITU International Telecommunication Union
ITU-T International Telecommunication Union – Telecommunication Standardization Sector
IUT Implementation Under Test
J2K JPEG 2000
JPEG Joint Photographic Experts Group
MAGB Magnitude Bound
MSE Mean Squared Error
RCT Reversible Component Transform
ROI Region Of Interest
sRGB standard Red–Green–Blue
TCS Test Codestream
4.2 Symbols
For the purposes of this Recommendation | International Standard, the following symbols apply.
0x---- Denotes a hexadecimal number
B Bit-depth precision for reversible 5-3
B Magnitude bound parameter for an HTJ2K codestream
MAGB
C Component guaranteed to be decoded
CAP Capabilities
COC Coding style Component
COD Coding style Default
COM Comment
CPF Corresponding Profile
CRG Component Registration
EPH End of Packet Header
EOC End of Codestream
H image Height guarantee
L Layer guarantee
L code data buffering guarantee
body
M decoded bit-plane guarantee
M Magnitude bound decoding guarantee
MAGB
N code-block parsing guarantee
cb
N component parsing guarantee
comp
P irreversible 9-7 Precision guarantee
PLM Packet Length, Main header marker
PLT Packet Length, Tile-part header marker
POC Progression Order Change marker
PPM Packed Packet headers, Main header marker
____________________
As defined in Rec. ITU-T T.800 | ISO/IEC 15444-1.
Rec. ITU-T T.803 (06/2021) 5
© ISO/IEC 2021 – All rights reserved

PPT Packed Packet headers, Tile-part header marker
PRF Profile marker
QCC Quantization Component marker
QCD Quantization Default marker
RGN Region of interest marker
SIZ image and tile Size marker
SOC Start Of Codestream marker
SOP Start Of Packet marker
SOD Start Of Data marker
SOT Start Of Tile-part marker
T Transform level guarantee
L
TLM Tile-part Lengths marker
W image Width guarantee
5 Conventions
The compliance files including test codestreams, JP2 files, JPH files, reference decoded images, and descriptive files
are supplied in the form of a compressed file. File locations given in this Recommendation | International Standard are
expressed relative to the top level of the directory tree. A Unix style file structure and delimiters are assumed.
This Recommendation | International Standard contains instructions for the use of these files. No support can be
provided by ISO | ITU-T beyond that offered in this Recommendation | International Standard.
6 General description
Perhaps the most distinctive feature of JPEG 2000 is its emphasis on and support for scalability. An existing codestream
may be accessed at a reduced resolution, at a reduced quality (higher compression), at a reduced number of components,
and even over a reduced spatial region. Moreover, this Recommendation | International Standard supports a rich family
of information progression sequences whereby the information may be reordered without introducing additional
distortion. This enables a single compressed codestream to serve the needs of a diverse range of applications.
This Recommendation | International Standard also covers compliance for implementations of Rec. ITU-T T.814 |
ISO/IEC 15444-15. To avoid confusion, the terms JPEG 2000 (J2K) and high throughput JPEG 2000 (HTJ2K) are used
in this Recommendation | International Standard, where necessary, to differentiate between JPEG 2000 codestreams that
conform to Rec. ITU-T T.800 | ISO/IEC 15444-1 and those that conform to Rec. ITU-T T.814 | ISO/IEC 15444-15,
respectively. J2K codestreams can be reversibly transcoded to HTJ2K and vice-versa, without any loss in information.
This property allows compliance for HTJ2K and J2K implementations to be treated in a very similar manner. In fact, all
of the HTJ2K test codestreams and JPH files are provided zipped with this Specification at https://www.itu.int/net/itu-
t/sigdb/speimage/ImageForm-s.aspx?val=10100803 or at https://standards.iso.org/iso-iec/15444/-4/ed-3/en have been
obtained by reversibly transcoding corresponding J2K test codestreams and JP2 files available at the same location. The
decoded output from an HTJ2K decoder is expected to conform to the same guidelines as the decoded output from a
J2K decoder, processing the corresponding J2K codestream or JP2 file.
From the perspective of compliance, the main distinction between HTJ2K and J2K is that an HTJ2K codestream does
not generally possess the same quality scalability attributes as the corresponding J2K codestream. An HTJ2K decoder
cannot choose to stop the decoding of a code-block bit-stream at an arbitrary bit-plane, providing a fine grain trade-off
between implementation complexity and reconstructed image quality. Considering this difference between HTJ2K and
J2K, this Recommendation | International Standard provides multiple transcodings of J2K test codestreams, allowing
implementations to be tested at multiple quality operating points.
JPEG 2000 encoders may employ only a fraction of the features supported by Rec. ITU-T T.800 | ISO/IEC 15444-1.
Likewise, some decoders will not support all the features supported by this Recommendation | International Standard. It
is impossible to provide test cases for all possible combinations of tools that an encoder or decoder may choose to
implement. This Recommendation | International Standard provides abstract test procedures for JPEG 2000 encoders
and decoders. A developer may designate the features that have been implemented and determine a set of test cases that
applies to those features. For the greatest level of interoperability, there are explicit decoder test procedures. These tests
are run for a particular Profile (defined in Rec. ITU-T T.800 | ISO/IEC 15444-1) and a particular compliance class
defined herein. Passing the explicit tests allows a decoder to be labelled "Profile-x Cclass-y Compliant".
6 Rec. ITU-T T.803 (06/2021)
© ISO/IEC 2021 – All rights reserved

Even with the explicit decoder tests, it is expected that some decoders may not decode all of the information that was
originally incorporated into the codestream by an encoder. This is the only way to truly exploit the scalability of
Rec ITU-T T.800 | ISO/IEC 15444-1. It is desirable to allow decoders to ignore information that is not of interest to
their target application. While this flexibility is one of the strengths of JPEG 2000, it also renders inappropriate some of
the conventional compliance testing methodologies that have been applied to non-scalable or less scalable compression
standards.
Many approaches to compliance could be taken. At one extreme, decoder implementations might be allowed to decode
any portion of the codestream that is of interest to them. At the other extreme, they might be required to correctly
decode the entire codestream. The first approach offers content providers and consumers no guarantee concerning the
quality of the resulting imagery. The other approach is also inappropriate because it offers the implementer no
guarantee concerning the resources that may be required for decoding, and in many cases the codestream may contain
information that is of no interest to the application.
This Recommendation | International Standard describes compliance for JPEG 2000 decoders in terms of a system of
guarantees. These guarantees serve to discourage encoders from producing codestreams that will be exceedingly
difficult or impossible for a decoder to process, to encourage decoders to provide quality images from any reasonable
codestream, and to encourage use of the flexibility and scalability of JPEG 2000 codestreams.
Profiles define a subset of technology, from Rec. ITU-T T.800 | ISO/IEC 15444-1, that meets the needs of a given
application with limits on parameters within a selected technology. Profiles limit bitstreams. Decoders define
capabilities for all bitstreams within a profile. Encoders achieve quality guarantees for particular decoders by encoding
bitstreams which meet a particular profile definition. Compliance classes (Cclass) define guarantees of a given level of
image quality for a decoder and guidance for encoders to produce codestreams that are easily decodable by compliant
decoders.
Essentially, if a JPEG 2000 encoder produces a codestream with certain properties, then a decoder of a certain Cclass
will be capable of producing an image with some defined level of quality. The compliance class of a decoder is based
solely on passing certain tests. The tests in this Recommendation | International Standard are designed to require a
compliant decoder to be capable of decoding all codestreams with a set of defined properties.
6.1 Profiles, derived sets and compliance classes
Two profiles are defined in Rec. ITU-T T.800 | ISO/IEC 15444-1, labelled Profile-0 and Profile-1. The two profiles
describe bitstream constraints for a Rec. ITU-T T.800 | ISO/IEC 15444-1 encoder. Profile-0 is a subset of Profile-1.
Hence, any implementation capable of decoding Profile-1 test streams shall be capable of passing the compliance tests
for Profile-0 of the same Cclass.
For HTJ2K codestreams, no profiles are defined in Rec. ITU-T T.814 | ISO/IEC 15444-15. All of the HTJ2K test
codestreams and JPH files are provided zipped with this specification at https://www.itu.int/net/itu-
t/sigdb/speimage/ImageForm-s.aspx?val=10100803 or at https://standards.iso.org/iso-iec/15444/-4/ed-3/en include a
CPF marker segment, that indicates the profile of the J2K codestream from which they were transcoded.
Rec. ITU-T T.814 | ISO/IEC 15444-15 defines constrained codestream sets that partition the space of all HTJ2K
codestreams as a function of the capabilities, complexity or throughput of an HTJ2K decoder. One important example is
the HTONLY set, which consists of HTJ2K codestreams that use only HT code-blocks. Codestreams that do not belong
to the HTONLY set may use the J2K block coding algorithm for some or all of their code-blocks. In particular, the
MIXED set of HTJ2K codestreams involve tile-components that use a mixture of HT and J2K code-blocks.
This International Recommendation | Standard provides test procedures and codestreams for HTJ2K decoders that are
derived from the same compliance points as those for J2K decoders, by means of a transcoding step. The derived
HTJ2K test codestreams are classified into the four derived sets defined in Table-1. Each of the two DS0 sets is a subset
of the corresponding DS1 set, being derived from J2K profiles 0 and 1, respectively. Also, the HT derived sets are each
subsets of the corresponding HM set.
NOTE – There are many other ways to construct and classify test codestreams for HTJ2K decoders that can be useful for specific
applications. The principles embodied by this Recommendation | International Standard can be readily extended to the testing of
HTJ2K decoders whose capabilities are classified in a different manner to that set out in Table 1.
Annex A defines three compliance classes (Cclass) for J2K codestreams, and three derived compliance classes for
HTJ2K codestreams. These Cclasses define levels of image quality guarantees for decoders and guidance for encoders
to produce codestreams that are easily decodable by compliant decoders. Cclass guarantees increase with the increasing
Cclass numbers.
Rec. ITU-T T.803 (06/2021) 7
© ISO/IEC 2021 – All rights reserved

Table 1 – HTJ2K derived sets employed in this Recommendation | International Standard
Derived set Definition
DS0_HT HTJ2K codestreams with a CPF marker segment that identifies Profile-0 as the compatible profile, that
belong to the HTONLY set of HTJ2K codestreams
DS1_HT HTJ2K codestreams with a CPF marker segment that identifies Profile-1 as the compatible profile, that
belong to the HTONLY set of HTJ2K codestreams
DS0_HM HTJ2K codestreams with a CPF marker segment that identifies Profile-0 as the compatible profile
...