1、CIE 101 93 I 9006345 0004714 321 ISBN 3 900 734 38 O COM M ISSION INTERNATIONALE DE LCLAIRAGE INTERNATIONAL COMMISSION ON ILLUMINATION INTER NATI O NALE BELEU CHTU NGSKO M M ISS1 ON PARAMETRIC EFFECTS IN CO LO U R-D I FFERE N CE EVALUATION Pub. No. CIE 101 a Ist Edition 1993 UDC: 535.65 Descriptor:
2、Colour measurement, colorimetric methods and apparatus COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE LOL 93 I 900bL45 0004735 268 II This Technical Report has been prepared by CIE Technical Committee 1-28 of Division 1 Vision and Colour and has been a
3、pproved by the Board of Adminstration of the Commission Internationale de IEclairage for study and application. The document reports on current knowledge and experience within the specific field of light and lighting described, and is intended to be used by the CIE membership and other interested pa
4、rties. It should be noted, however, that the status of this document is advisory and not mandatory. The latest CIE proceedings or CIE NEWS should be consulted regarding possible subsequent amendments. Ce rapport technique a t prpar par le Comit Technique CIE 1-28 de la Division 1 Vision et Couleur e
5、t a t approuv par le Bureau dAdministration de la Commission Internationale de IEclairage, pour etude et application. Le document traite des connaissances courantes et de lexprience dans le domaine spcifique indique de la lumire et de lclairage, et il est tabli pour lusage des membres de la CIE et a
6、utres groupements interesses. II faut cependant noter que ce document est indicatif et non obligatoire. Pour connaitre dventuels amendements, consulter les plus rcents comptes rendus de la CIE ou le CIE NEWS. Dieser Technische Bericht ist vom CIE-Technischen Komitee 1-28 der Division 1 Sehen und Far
7、be ausgearbeitet und vom Vorstand der Commission Internationale de IEclairage gebilligt worden. Das Dokument berichtet ber den derzeitigen Stand des Wissens und Erfahrung in dem behandelten Gebiet von Licht und Beleuchtung; es ist zur Verwendung durch CIE-Mitglieder und durch andere Interessierte be
8、stimmt. Es sollte jedoch beachtet werden, da das Dokument eine Empfehlung und keine Vorschrift ist. Die neuesten CIE-Tagungsberichte oder das CIE NEWS sollten im Hinblick auf mgliche sptere nderungen zu Rate gezogen werden. Any mention of organisations or products does not imply endorsement by the C
9、IE. Whilst every care has been taken in the compilation of any lists, up to the time of going to press, these may not be comprehensive. Toute mention dorganisme ou de produit nimplique pas une prfrence de la CIE. Malgr le soin apport a la compilation de tous les documents jusqua la mise sous presse,
10、 ce travail ne saurait tre exhaustif. Die Erwhnung von Organisationen oder Erzeugnissen bedeutet keine Billigung durch die CIE. Obgleich groe Sorgfalt bei der Erstellung von Verzeichnissen bis zum Zeitpunkt der Drucklegung angewendet wurde, ist es mglich, da diese nicht vollstndig sind. O CIE 1993 C
11、OPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE 101 73 W 7006345 00047Lb IT4 W III FOREWORD The following members of TC 1-28, Parameters affecting colour-difference evaluation, took part in the preparation of this technical report. The committee comes un
12、der CIE Division 1, Vision and Colour. Members R. S. Berns F. W. Billmeyer, Jr. K. Ikeda A. R. Robertson K. Witt Consultants M. R. Lu0 T. O. Maier USA USA Japan Canada Germany (chairman) Great Britain USA COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE
13、101 93 9006345 O004737 030 IV Table of contents Foreword Table of contents Summary Zusammenfassung Rsum 1 .INTRODUCTION 1.1 Scope 2. PARAMETRIC FACTORS 2.1 Observer uncertainty 2.1 .i Variability of colour matching 2.1.2 Variability of judgments on constant stimuli 2.2 Physical parameters 2.2.1 Samp
14、le size 2.2.2 Sample separation 2.2.3 Texture 2.2.4 Colour of background 2.2.5 Luminance of sample 2.2.6 Lightness of sample 2.2.7 Unidentified factors 3. CONCLUSIONS IV V V V 1 2 3 4 4 4 4 5 5 6 7 8 8 9 10 4. FUTURE WORK 11 5. REFERENCES 12 COPYRIGHT International Commission on IlluminationLicensed
15、 by Information Handling ServicesCIE LO1 93 I 90ObLY5 00047L Ti7 m V PARAMETRIC EFFECTS IN COLOUR-DIFFERENCE EVALUATION SUMMARY The variability in correlating visual colour discrimination with colorimetric measures may be understood better if parameters that affect the correlation are quantified. A
16、review is given of known parametric effects of a human and physical nature, and advice is given for a tentative modification of colour-difference evaluation techniques based on characterized parametric effects. Human factors such as variability in and between observers are very important random fact
17、ors, and may be a major source of uncertainty. Some physical changes in the presentation of colour-discrimination tasks are shown to produce significant shifts of colour-difference sensitivity. In some cases these changes have been quantified, however, these results were carried out using quite diff
18、erent experimental techniques (such as colour matching using a visual colorimeter or judgments of constant stimuli using physical samples) and are based upon very limited numbers of colour centres and observations. Hence, more research is needed for a full understanding of their influence. PARAM ETR
19、ISCHE EINFLUSSE AUF DIE FARBABSTANDSBEWERTUNG ZUSAMMENFASSUNG Die Schwankungen im Grad der bereinstimmung von visuell bewerteten Farbunterschieden mit den farbmetrischen Mewerten knnen besser verstanden werden, wenn die Parameter, die diese bereinstimmung bestimmen, quantifiziert werden. Es wird ein
20、 berblick ber bekannte parametrische Effekte von menschlicher und physikalischer Art gegeben, und es werden Vorschlge zur versuchsweisen Modifikation der Verfahren zur Farbdifferenz-Bewer- tung gemacht. Menschliche Faktoren, wie die Variabilitt der Urteile eines Beobachters oder der zwischen Beobach
21、tern, sind sehr wichtige statistische Faktoren und knnen eine groe Quelle der Unsicherheit sein. Einige physikalische nderungen in der Gestaltung von Versuchen zur Farbunterschei- dung werden gezeigt, die wesentliche nderungen in der Empfindlichkeit fr Farbunterschiede bewirken. In einigen Fllen wur
22、den diese nderungen quantifiziert, jedoch wurden diese Ergeb- nisse bei ganz verschiedenen experimentellen Verfahren erhalten (wie Farbabgleiche mit einem visuellen Farbmegert oder Beurteilung gleicher Farbreize an physikalischen Proben), und sie beruhen auf einer sehr begrenzten Anzahl von Farbrter
23、n und Beobachtungen. Daher ist mehr Forschungsarbeit notwendig, um ihren Einflu voll zu verstehen. PARAMETRES DANS LEVALUATION DES DIFFERENCES DE COULEUR RESUME La corrlation entre les discriminations visuelles de couleur et leurs mesures colorimtriques est affecte dune variabilit qui peut tre mieux
24、 comprise si les paramtres qui interviennent dans la corrlation sont valus numriquement. On passe en revue laction des paramtres connus, de nature humaine et de nature physique. On donne ensuite des indications pour un essai dintroduction de ces parametres dans lvaluation des diffrences de couleur.
25、Les facteurs humains, tels que la variabilit dun observateur ou la variabilit entre observateurs sont des lments alatoires trs importants qui peuvent tre une source majeure dincertitude. On montre que quelques modifications physiques dans la prsentation des tches de discrimination de couleur produis
26、ent des changements significatifs de la sensibilit aux diffrences de couleur. Dans quelques cas ces changements ont pu tre quantifis. Cependant ces rsultats ont t obtenus A partir de techniques experimentales assez diffrentes (telles que des galisations colorimtriques avec un colorimtre visuel, ou s
27、ur des chantillons reels par une mthode de stimulus constants) et sappuient sur un nombre trs limit de centres de couleurs et dobservations. De ce fait des recherches complmentaires sont ncessaires pour une connaissance plus complte de leur influence. COPYRIGHT International Commission on Illuminati
28、onLicensed by Information Handling ServicesCIE 303 93 H 7006345 0004739 903 1 1. INTRODUCTION A major problem in colour-difference evaluation is the heterogeneity of experimental results relating psychological responses to psychophysical measures. Their variability may to a large extent depend on so
29、me of the parameters underlying the experimental organisation of a visual task. Fig. 1 demonstrates the variability of perceptual colour-difference ellipses in the chroma- ticity plane at a defined colour centre. The results shown are for different studies l .l of contour lines of equal colour-diff
30、erence perception weighted for equal size of ellipse area. 0,015 0,Ol 0,005 AY O -0,005 -0,Ol -0,015 I I I I I I I I I 1 II II DI II I I I I I I I I I I -0,015 -0,Ol -0,005 O 0,005 0,Ol 0,015 Ax Fig. I: Perceptual colour-difference chromaticity ellipses at CIE red colour centre. A comprehensive colo
31、ur-difference formula is desired by industry and should include some adjusting factors to allow for the changes of viewing parameters such as different sample sizes, separations, textures, background colours, luminances and lightnesses of samples, if they significantly change perceived colour differ
32、ence. The root of this thinking can be traced back to Judds development of the NBS colour-difference formula, which included parameters accounting for the gap between adjacent samples and for field size 1.2. CIE TC-1.3 (Colorimetry) has already described some physical parameters limiting the cases t
33、o which the C1E 1976 colour-difference formulae should be applied 1.3. They are “.intended to apply to comparison of differences between object colours of the same size and COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE 101 93 U 9006145 0004720 625 2 s
34、hape, viewed in identical white to middle-grey surroundings, by an observer photopically adapted to a field of chromaticity not too different from that of average daylight. Another statement of that document recommends the use of the CIE 1964 supplementary standard colorimetric observer “whenever co
35、rrelation with visual colour matching of fields of angular subtense greater than about 4“ at the eye of the observer is desired.“ These statements hint at possible parameters without quantifying them and do not provide an exact set of experimental conditions to which others should be related. Theref
36、ore the proposal of a new research project in 1978 i ,4 included the study of colour-difference evaluation for various parametric variables in order to get a clearer understanding of their possible influence. An essential topic of that project was the restriction to five colour centres, in order to
37、avoid confusion with varying effects due to differing colours. Within this project some parametric effects on colour-difference perception have been studied, e.g. the effect of a separating gap between a pair of colour samples and the effect of the lightness of a grey background. Additionally some i
38、nformation about other parameters can be gained by looking into older literature on visual colorimeters, though there might be no close correspondence with results from physical samples. Consequently some large differences have been found between results with the two different techniques. Viewing mo
39、de may be an important parameter (.e. aperture mode, object mode, and illuminant mode *). As far as these different modes may be described by varying background lightness, they are assigned to that parameter. Whenever it is possible, aperture mode (black background) is indicated. Results with visual
40、 colorimeters are assigned to object mode, if light backgrounds had been used. No information is given for illuminant mode and its essential parameter(s) separating it from aperture mode. For certain experimental conditions the possible influence of aperture mode compared to object mode on the size
41、of colour-difference perception has generally been found to be not very large I 5, 1.6, 1.71. 1.1 Scope This report is intended to describe the effect of some parameters that may change sensitivity to colour differences and affect colour-difference evaluation. Its goal is to suggest the modifi- cati
42、on of an ideal (or nearly ideal) colour-difference formula by parametric correction factors, if a visual task provokes significant changes of colour-difference sensitivity in relation to some reference task. The report includes studies of different origin and hence for different reference par- amete
43、rs. At its end a set of basic parameters of a visual task is formulated for use in further studies. . *The ASTM Standard Terminology of Appearance defines these modes as follows I .8: aperture mode, - color seen through an aperture which prevents its association with a specific object or source; obj
44、ect mode, - color seen as ascribed to an object; illuminant mode, - color seen as ascribed to a source of illumination. - - - COPYRIGHT International Commission on IlluminationLicensed by Information Handling Services CIE 301 93 e 9006345 0004723 5bL D 3 2. PARAMETRIC FACTORS Parametric factors affe
45、cting colour difference evaluation may be split into two types: one type comprises human factors inherent in the inspection of pairs of colours (excluding colour vision deficiencies); a second type comprises physical changes in the presentation of those colours for inspection (excluding change of il
46、luminant). A significant parametric factor is one that is thought to change sensitivity to colour discrimination in relation to some reference situation, which means a gain or loss in the size of sensitivity steps. Its value may not depend critically on the choice of colour-difference formula within
47、 the broad levels of experimental uncertainty. Assume an ideal colour-difference formula with three main components, so that the total difference is composed of three independent parts, which have physiological meanings, e.g. lightness, chroma, hue. Existing colour-difference formulae depart from id
48、eality more or less and may have components that interact with each other to some degree. These deviations are considered to be of minor importance in relation to large statistical fluctuations of experimental data on parametric effects. Given a measured colour difference A of ideal character, which
49、 can be split into three Cartesian components AL (lightness), Aa (red-/greenness), Ab (yellow- /blueness) or AL, AC (metric chroma), AH (metric hue) (these are the ideal equivalents to the definitions for the CIELAB formula 1.3, which may be used as first approximation of ideality, or else the ideal relations may be deduced from x,y,Y ellipsoids or from some other formula). The scale of colour difference, as defined by the special formula used, is relate
