Division 8: Image Technology

Status Reports for 24th Session of CIE in Warsaw Poland, June , 1999

TC8-01: Color Appearance Models for Color Management Applications

by Gary Dispoto, Hewlett Packard Co., USA - TC Chair

1.1 List of Members

Two new members have joined TC8-01 since our last report: Paula Alessi of Eastman Kodak Company (USA) and Noaya Katoh of Sony Corporation (Japan).  Our membership now consists of the following:

·        Gary Dispoto, TCC, , USA

·        Dr. Paula J. Alessi, CIE TC1-27 Chair,  USA

·        Nathan Moroney, USA

·        Dr. Ronnier Luo, UK

·        Dr. R.W.G. Hunt, UK

·        Dr. Mark Fairchild, CIE TC1-34 Chair, USA

·        Naoya Katoh, CIE TC8-04 Chair, Japan

·        Hirohisa Yaguchi, Japan

·        Mark Mahy, Belgium

·        Nenad Vujovic, USA

1.2 Progress since last report

Drs. Hunt and Luo have developed a revision of the CIECAM97s model that addresses many of the concerns outlined in the previous quarterly report.  In particular, the revision makes the lightness correlate, J, zero when the Y tristimulus value is zero, for all values of surround.  It also modifies the chromatic induction factor (Nc) from 1.10 to 0.95 for the dim surround, to address CRT gamut size peculiarities first pointed out by Katoh and presented by Moroney at the Sixth Color Imaging Conference.  The revision has been described in a paper "A Revision of the CIECAM97s Model," which has been submitted to the IS&T Seventh Color Imaging Conference.

1.3 Plans for next quarter

The TC will be organizing itself over the next weeks to do two things: test the CIECAM97s revision developed by Hunt and Luo and develop initial parameter setting recommendations. Additionally, we have been asked to explore the question of whether an imaging medium's white and black should be "forced" to neutral, since apparently some of the gamut mapping algorithms being developed in TC8-03 assume this to be the case. The ICC's Reference Implementation Working Group's beta code is being investigated as a possible software test bed for all TC8-01 work. The chair will be developing a revised timetable and plan for these activities.

TC8-02: Colour difference evaluation in images

by Michael Stokes, USA - TC Chair

3.1 List of Members

Several new members have joined TC8-02 since our last report. We are continuing to solicit members for additional national bodies for participation.  Our membership now consists of the following:

·        Michael Stokes, TCC, USA

·        Paula J. Alessi, USA

·        Osvaldo Da Pos, Italy

·        Gary Field, USA

·        Bernhard Hill, Germany

·        R.W.G. Hunt, UK       

·        Naoya Katoh, Japan      

·        Ronnier Luo, UK

·        Mark Mahy, Belgium

·        Michael Pointer, UK

·        Kazuhiko Takemura, Japan

·        JA Stephen Viggianno, USA

·        Hirohisa Yaguchi, Japan

1.2 Progress since last report

Committee members Da Pos, Hunt, Luo, Stokes, Takemura, and Viggianno and TC1-34 TCC Fairchild contributed draft text to a rough outline provided by the TCC. These contributions were combined to provide a very rough first working draft of a technical committee report. This rough draft of an initial technical report has been circulated to the committee members for comment. 

1.3 Plans for next quarter

It is expected that this very rough report will be developed to a reasonable rough draft by late summer 1999.

TC8-03: Gamut Mapping

by Ján Morovic, UK - TC Chair

3.1 List of Members

Bodrogi Péter and Raimondo Schettini have joined the TC.

• Ján Morovic - chairman, UK
• Gus Braun, USA
• Bodrogi Péter, Hungary
• Fritz Ebner, USA
• Mark D. Fairchild, USA
• Patrick Herzog, Germany
• Tony Johnson, UK
• Naoya Katoh, Japan
• M. Ronnier Luo, UK
• Marc Mahy, Belgium
• Gabriel Marcu, USA
• John McCann, USA
• Ethan Montag, , USA
• Hideto Motomura, USA
• Todd Newman, USA
• Raimondo Schettini, Italy
• Geoff J. Woolfe, USA

3.2 Current revision of time table/plan of work

3.2.1 Work Programme

§         Review current gamut mapping algorithms and algorithms for calculating gamut boundaries.

§         Provide guidelines for the evaluation of gamut mapping algorithms.

§         Review results of gamut mapping algorithm evaluation done according to guidelines.

§         Recommend a standard gamut mapping algorithm or a small number of algorithms and rules for deciding when each of them is to be used.

3.2.2 Timetable

Survey                         02/1999

Guidelines                     08/1999

Review results              03/2000

Write final report          06/2000

3.3 Progress since last report

•  Survey of gamut mapping algorithms (including a brief classification scheme) has been completed.
• Meeting of TC held on 26^th March 1999 in Derby (UK). Of the 18 participants 6 were TC members (Ebner, Johnson, Katoh, Luo, Morovic, Newman) and the most important points of discussion were the reproduction intent for the GMA to be recommended (including a discussion of the meaning of “accuracy”) and what should be included in the guidelines this TC is to prepare. See the minutes of this meeting for more details.

3.4 Plans for next quarter

•  Publish survey of gamut mapping algorithms.
• Prepare guidelines for the evaluation of gamut mapping algorithms.

TC8-04: Adaptation under Mixed Illumination Conditions

by Naoya Katoh, Japan - TC Chair

4.1 List of Members

Two new members* joined our TC; Elibabeth Pirotta from USA, and Elie Khoury from France. We now have eight members; four from Europe, three from USA, and one from Japan.

• Dr. Janos Schanda, Hungary
• Dr. Paula J. Alessi, (CIE TC1-27 Chair), USA
• Dr. Klara Wenzel, Hungary
• Ms. Monica Billger, Sweden
• Dr. Peyma Oskoui, USA
• *Elizabeth Pirrotta, USA
• *Elie Khoury, Alwan France
• Naoya Katoh, Japan - TC8-04 Chair

4.2 Progress since last report

TC chair had sent Katoh's recent paper "Effect of ambient light on the color appearance of softcopy images: Mixed chromatic adaptation for self-luminous displays" on J. Electronic Imaging, Vol. 7, No. 4, pp794-806 (Oct. 1998), to every TC members. He also provided very Preliminary Draft for Experimental guideline as below, based on Katoh's paper.

http://www.colour.org/tc8-04/Experiment_guideline.html

Also, Preliminary Draft for past research survey was provided at URL below;

http://www.colour.org/tc8-04/Past_survey.html

However, no comments regarding how TC members are going to perform cross-lab experiments has been received so far. TC chair plans to provide a "questionnaire" to survey what information is necessary to perform experiments at their lab. According to the answer to the "questionnaire", TC chair will try to update and fill out more information if needed. TC chair hopes every TC members to start the experiments by the end of this year (1999). Dr. Oskouli and Ms. Pirotta will perform Achromatic Color Matching experimets using uniform color patches (not pictorial images), at fixed (to CRT) state of chromatic adaptation.

4.3 Plans for next quarter

Basic idea for Pictorial Image Matching (softcopy vs. hardcopy) is written in Katoh's paper as S-LMS. Possible improvements could be achieved by;

• Applying different chromatic adaptation transformation methods; (Hunt-Pointer-Esteves, Bradford, or others; as in Section 3.2)
• Applying different "incomplete adaptation" methods; (as in Section 3.3.1),
• Applying different for "mixed adaptation" methods; (as in Section 3.3.2)
• Verifying mixed adaptation by using various kinds of images (natural images, CG images, uniform patches, etc).

TC8-05: Gamut Mapping

by Lindsay W. MacDonald, UK - TC Chair

5.1 List of Members

• Rob Buckley, USA
• Geert Carrein, Belgium
• Adrian Clarke, UK
• Phil Green, UK
• Tony Johnson, UK
• Jim King, USA
• Tim Kohler, USA
• Lindsay MacDonald (Chair), UK
• Yoichi Miyake, Japan
• Todd Newman, USA
• Alastair Reed, Canada
• Kevin Spaulding, USA
• Gary Starkweather, USA
• Michael Stokes, USA
• Maureen Stone, USA
• Sabine Susstrunk, Switzerland
• Steve Swen, USA
• Joann Taylor, USA
• Klara Wenzel, Hungary
• Mike Wilsher, UK

5.2 Progress since last report

Only limited progress has been made since the meeting of the TC in Derby on 26th March:

1. Minutes of the meeting were prepared and circulated (thanks to Sabine Susstrunk). 

2. The TC web site was set up (thanks to Tim Kohler). 

3. Attempts have been made to use the List Server (thanks to Todd Newman) but the facility has proven to be frustrating and not really satisfactory, at least from the Chair's viewpoint. 

4. A simple survey was devised and posted on the Web site, and advertised in IS&T Reporter (thanks to Dave McDowell), but so far only a handful of responses has been received.

5.3 Plans for next quarter

The next tasks planned are: - Ask all TC members to respond to survey - Delegate TC members to seek responses from people working in various segments of the imaging industry - Start to prepare a catalogue of color spaces in use together with a taxonomy of color representations and techniques for describing color spaces.

The next TC meeting will held at CIC'99 in Scottsdale.

Reporters

R8-01 Reporter for Grading Colour Instruments - Yoshi Ohno, USA

We have, in Division 2, TC2-42 Colorimetry of Displays as you know. I am wondering if some grading guidelines can be included in the document from this TC. We now have many documents related to measurement of displays from TC100/PT61966. But, as you know, these IEC documents do not go into the details of the sources of errors and measurement uncertainties, etc. I think the TC2-42 documents will address these technical issues for color measurement devices (colorimeter, spectroradiometers, and luminance meters) in general. IEC has asked the CIE to collaborate in developing IEC 61966-1-0 to also help bridge this issue.

On the D8 side, I would like to get more inputs on what kind of performance characteristics of instruments are to be graded or specified.

R8-02 Reporter for Liaison with IEC TC100 PT61966 (Colour Measurement and Management in Multimedia Systems and Equipment ), Hiroaki Ikeda, Japan

 Introduction

The project team 61966 was established on 17^th of September 1996 at the plenary meeting of IEC/TC 100 in Dresden. It has been asked, and committed itself, to develop a series of Inter­na­tional Standards based on the IEC projects 61966.

The scope of the projects 61966 was originally proposed by the Japanese National Committee and discussed by the Task Force during the NP voting.  The expanded scope was approved and informed by the document 100/32/INF.  It has been slightly modified to integrate digital still/video cameras into single part.  The current structure of the series of International Standards is as follows.

IEC 61966: Multimedia Systems and Equipment – Colour Measurement and Management

The Project Team 61966 (PT 61966) consists of approximately 70 experts nominated by the National Committees, liaison representatives endorsed by respective liaison organizations, and associate members who are committed themselves to develop IEC 61966.

IEC 61966 is focusing to realise colour management in global open network systems such as the Internet, but the systems naturally include regional, national, local and site colour production / reproduction systems from the point of view of realising colour management.  Application areas of IEC 61966 include, but not limited, desk top publishing in a room, education materials distribution, electronic commerce, electronic library/museum, colour information ex­change in the World Wide Web, audio-visual teleconferencing, and video on demand.

 Liaison and collaboration

The full liaisons formally recognized by the parent committee are listed below with nominated and endoesed liaison representatives.

It will be benefitial for both groups, IEC/TC 100/PT 61966 and CIE Division 8, to have formal liaisonship in the field of common interests.  In order to establish a bidirectional channel, CIE Division 8 is kindly requested to nominated liaison representatives to PT 61966.  Project leader, Mr. Hiroaki Ikeda, of IEC projects 61966  will act as liaison representaive from PT 61966.

 Current status of the projects 61966

Most updated status with relevant documentation has been posted to the WWW site run by the project leader from the initialtion of the project team at

http://w3.hike.te.chiba-u.ac.jp/IEC/100/PT61966

and a mirror at

http://www.map.chiba-u.ac.jp/IEC/100/PT61966 .

As of the date of this liasion report, the stage of each subproject of the projects 61966 is tabulated below.

Project leader of PT61966: Mr. Hiroaki Ikeda <ikeda@hike.te.chiba-u.ac.jp>

Co-leader: Mr. Michael Stokes <Michael_Stokes@hp.com>

R8-03 Fluorescence, C. S McCamy

Introduction

CIE Division 8 has the responsibility to study processes and prepare guides and standards for optical, visual, and metrological aspects of communication, processing, and reproduction of images, using all types of analog and digital imaging devices, storage media, and imaging media. Central to this mission is the visual comparison of images in different media. In particular, images displayed on electronic devices, such as cathode ray tubes, must be compared to photographs, printed matter, or original art. The issues of importance here are illustrated by the simple case of visual comparison of a paper print to an image on a monitor.

This initial report briefly describes the effect of fluorescence on such comparisons, provides references to the literature and standards in this field, refers to a method of quantifying the effect, and points out why this effect must be kept in mind even though it is not quantified and fully taken into account.

Projected and Luminous Images vs. Paper Prints

The appearance of the image on a monitor is influenced by the adaptation of the eye, which is determined by the illumination and the reflecting properties of materials within the field of view. When the monitor is viewed in the dark, there is nothing visible other than the monitor. Once the eye has adapted to those conditions, the appearance of the image is determined by the properties of the image alone. These are the classic conditions for motion picture and slide projection. The theory is well known.1 For example, it is well known that the measured contrast of such images must be substantially greater than the contrast of a paper print, to be considered acceptable. The extreme contrast of the projected image against the dark surround adjusts the visual system to an expectation of comparatively high contrast. This phenomenon must be accounted for in any comparison of paper prints and projected or luminous images. In addition, the appearance of both projected and printed images depends on the luminance level. For these reasons, the ambient illumination, the illumination of the surround, and the illumination of paper prints must be controlled and standardized.

Enter Fluorescence

Given well-controlled illumination, the appearance of the monitor is well controlled, but the appearance of paper prints involves the further complication of fluorescence. Most papers and some inks are fluorescent. They absorb invisible ultraviolet energy and emit that energy in the visible spectrum. The main reason for this practice is that paper has a tendency to be slightly yellowish and the addition of dyes that fluoresce in the blue region of the visible spectrum make the paper look white. Obviously, the extent of the whitening effect depends not only on the nature of the paper and the fluorescent whitening agent (also called a "fluorescent brightener"), but also on the relative amount of ultraviolet power in the spectrum of the illuminant.

Illuminants

We distinguish a source form an illuminant. A light source is an object, such as the sun or a lamp, that emits light. For our purposes, an illuminant is the quantity and spectral quality of light incident on a specimen. It depends not only on the source, but also on any reflectors, lenses, diffusers, filters, or other optical components that influence the light. There is a substantial amount of ultraviolet power in natural daylight. It is somewhat diminished when that light passes through a glass window. There is little ultraviolet power in incandescent lamp light, but a considerable amount in the spectra of fluorescent lamps and some gas discharge lamps. Even though all of these illuminants may be considered white, the ultraviolet component differs markedly. The general problem must be considered with respect to normal operating conditions in typical offices and well-controlled and standardized conditions for critical appraisal of colors. Illuminants that simulate daylight, for appraisal of colored materials, were introduced in 1915 and have been well standardized for many years.2, 3 Viewing conditions for critical appraisal in photography and graphic arts were nationally standardized over thirty years ago and internationally standardized soon thereafter.4, 5 These standards are being revised. Practical illuminators that simulate natural (daylight) and artificial (incandescent lamps, fluorescent lamps, etc.) illuminants are commercially available.

The Quality of Illuminant Simulators

A method of testing the quality of the spectra of daylight simulators was published by the CIE and has been in widespread use for many years.6, 7,8,9 A CIE technical committee is now writing a standard based on that publication and recent research.10, 11 A standard method of assessing the spectral quality of standard artificial illuminants is under development in ASTM.12 An understanding of these publications and the ensuing standards will be essential to a reasoned approach to controlling the variables in inter-media comparisons of images.

A Pitfall for Young Players

At a recent conference, the theory of managing the colors of images was well described. I asked if there were plans to take fluorescence into account. The speaker replied, "No. We must learn to walk before we try to run." That response is quite understandable. We can't solve all the problems of the world in one fell swoop; we take them on one at a time, the simplest first. However, fluorescence effects must be known and kept in mind from the outset, because they can be overriding variables in experiments and can cause untold confusion in the interpretation of data, if ignored. Reference 6 provides a good introduction to the method of computing the amount of fluorescence and combining it with the amount of reflection to find the total effect of the two processes.

References

1. R.W.G. Hunt, "The Reproduction of Colour", 2nd ed., Fountain Press, London, 1967. Chap. 9.

2. C.S. McCamy, Simulation of daylight for viewing and measuring color, Color Res. Appl., 19, 437-445 (1994).

3. ASTM D 1729-96 Standard Practice for Visual Appraisal of Colors and Color Differences of Diffusely-Illuminated Opaque Materials.

4. ANSI Standard NAPM IT2.30-1994, Viewing Conditions - photographic prints, transparencies, and photomechanical reproductions.

5. ISO Standard 3664 - Viewing Conditions for Photographic Prints, Photomechanical Reproductions, and Transparencies.

6. CIE Publication 51, A Method for Assessing the Quality of Daylight Simulators for Colorimetry.

7. Japanese Industrial Standard JIS Z 8720-1983 Standard Illuminants and Sources for Colorimetry.

8. C.S. McCamy, Metamers for assessing the quality of CIE D50 simulators, Color Res. Appl., 21, 236-238 (1996).

9. Supplement to CIE Publication 51. (Based on Reference 7.)

10. C.S. McCamy, New metamers for assessing daylight simulators and the method of evaluating them, Color Res. Appl., 24, (1999), in press.

11. CIE Technical Committee TC1-53 Standard Method of Assessing the Spectral Quality of Daylight Simulators, C.S. McCamy, chairman.

12. ASTM Committee E-12 Color and Appearance, Subcommittee E-12.11Visual Methods, contact: C.S. McCamy.