1、CIE 75 88 I 9006345 0003342 775 COMMISSION INTERNATIONALE DE LkLAIRAGE INTERNATIONAL COMMISSION ON ILLUMINATION INTERNATIONALE BELEUCHTUNGSKOMMISSION - SPECTRAL LUMINOUS EFFICIENCY FUNCTIONS BASED UPON BRIGHTNESS MATCHING FOR MONOCHROMATIC POINT SOURCES 2O and loo FIELDS PUBLICATION CIE NO 75 (1986)
2、 CENTRAL BUREAU OF THE CIE A-1030 VIENNA, Kegetgasse 27 - AUSTRIA m CIE 75 88 = OObLLt5 00033Lt3 bOL II This Technical Report has been prepared by CIE Technical Committee 1-02 of Division 1 Vision es ist zur Verwendung durch CIE-Mitglieder und durch andere Interessierte bestimmt. Es sollte jedoch be
3、achtet werden, da das Dokument eine Empfehlung und keine Vorschrift ist. Die neuesten CIE- Tagungsberichte oder das neueste CIE-Journal sollten im Hinblick auf mgliche sptere nderungen zu Rate gezogen werden. Any mention of organisations or products does not imply endorsement by the CIE. Whilst ever
4、y 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 la compilation de tous les documents jucqu la mise sous presse, ce travail ne sau
5、rait tre exhaustif. Die Erwhnung von Organisationen oder Erzeugnissen bedeutet keine Billigung diirch 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. CIE 1988 CIE 75 88 I 900bL45
6、0003344 548 III FOREWORD The following members of TCI -02 Luminous Efficiency Functions took part in the preparation of this Technical Report. The Committee comes under CIE Division 1 Vision and Colour. Boynton, R. M. Estevez, O. Fuwa, M. Ikeda, M.(Chairman) Kaiser, P.K. Kinney, J. A. Kokoscbkct, S.
7、 Palmer, D. A. Ronchi, L. R. Sagawa, K. Verriest, G. Yaguchi, H. Consultants Roufs, J. * Hasegawa, T. * Uchikawa, K. Observer Chen, Xiaju USA Netherlands Japan Japan Canada USA Germany, F.R. Great Britain Italy Japan Belgium Canada Netherlands Japan Japan China * Associated with the work of Committe
8、e prior to 1983. CIE 75 8B 900bl145 0003345 484 IV CONTENTS 1 INTRODUCTION AND HISTORY. Page 1 2 PRINCIPLE OF THE HETEROCHROMATIC BRIGHTNESS MATCHING Page 2 3 EXPERIMENTAL CONDITIONS Page 3 4 REMARKS ON TERMINOLOGY Page 3 5 RESULTS Page 3 5.1 Point sources Page 3 5.2 20 field data Page 5 Page 6 5.3
9、loo field data 6 DERIVATION OF FINAL 2O AND loo FUNCTIONS Page 7 7 CONCLUSIONS 7.1 Point sources 7.2 2O field 7.3 loo field 7.4 Discussion Page 9 Page 9 Page 9 Page 9 Page 9 8 REFERENCES Page 11 CIE 75 B 9006145 0003346 310 V SUMMARY Spectral luminous efficiency functions based on heterochromatic br
10、ightness matching for point surfaces, 2 fields and 10 fields are presented. Their averaged functions are derived with the intention that they be used to evaluate the luminous efficiency for the brightness of monochromatic lights. These functions could be used as the basis for further work to evaluat
11、e light sources with compound spectra. RESUME On prsente des fonctions defficacit lumineuse relative spectrale obtenues par galisation htrochrome de luminosit avec des sources ponctuelles, des champs de diametre 2 et des champs de diamtre 10. Les moyennes des fonctions ont t dtermines afin quelles s
12、oient utilises lvaluation de lefficacit lumineuse en termes de luminosit des rayonnements monochroma- tiques. Elles pourraient servir de base des travaux ultrieurs concernant lvaluation de la luminosit des rayonnements complexes. ZUSAMMENFASSUNG Es werden spektrale Hellempfindlichkeitsfunktionen auf
13、 der Grundlage eines heterochromen Heligkeitsabgleichs fr punktartige Lichtquellen sowie fur 2 und 10 Gesichtsfelder dargestellt. Diese aus mehreren Untersuchungen gemittelten Funktionen wurden aufgestellt mit dem Ziel, die Helligkeit mono- chromatischer Strahlung bewerten zu knnen. Sie sind auch di
14、e Grundlage fur weitere Arbeiten mit dem Ziel, die Helligkeit von zusammengesetzten Strahlungen zu bewerten. CIE 75 88 B 7006145 0003347 257 1 1 INTRODUCTION AND HISTORY 1978: CIE Publication No 41 entitled Light as a true visual quantity: principles of measurement was published 1 .I 3. A difference
15、 was clearly shown between the luminous efficiency function of the CIE, V(X), and the function obtained by the direct heterochromatic brightness matching method, the efficiencyofthelatterbeing muchhigherthanthatofthe former at both long and short wavelengths when both functions werenormalizedat570 n
16、m,The need toderive a standard luminous efficiency function for brightness was well recognized, but the data from direct matching experiments were rather limited, and the function was not derived at that time. Thus it was in1978 that an appeal was made that more data be collected using the direct he
17、terochromatic brightness matching method 1.2, 1.31. 1979: Experimental conditions for the data collection were set up as follows: Stimulus size of point sources, 2O and loo arc of visual angle; Photopic illuminance level in t5e range of 50 to 500 td; Observers age between 18 and 50 years old. Some t
18、olerances were allowed so that as much data as possible could be utilized in the analysis. As a result of careful inspection of the data adopted in the 1978 report, that of 12 out of 31 observers were abandoned for various reasons, such as duplication of the same observers, insufficient data points
19、with regard to the wavelength, etc. The data of 19 observers were finally adopted and the revised version of the 1978 report was prepared. 1981: Asa result ofthis worka brightness luminous efficiency function for a 2O field based on 19 observers was achieved 1983: to the previous report. The number
20、of observers had increased The data for point sources, 2O and loo fields were added . from 19 to 37 for the 2O field. Correction factors to derive Vb,lo(x) from Vb,2(X) were presented. However, it was CIE 75 88 I 900b145 0003348 193 I 2 important to determine experimentally the difference between b,
21、2( 1) and Vb,lo(X) rather than assume the macular pigment was the sole cause of this difference. To meet this point it was decided to collect more data, particularly for the loo field, The Vb oint(A) was reported to be the same as CIE V( A) function with Judds modification, tP 1985: Further newly ob
22、tained data for 2O and loo fields was included in a final report based on 6 observers for point sources, 63 observers for the Z0 field, and 76 observers or the loo field. 2 PRINCIPLE OF THE HETEROCHROMATIC BRIGHTNESS MATCHING Whether the field size is small or large, there must be two visual fields,
23、 one for a reference stimulus and the other for a test stimulus of defined wavelength. When the field size is large a bipartite field maybeused, but when it is very small, such as a point source, two separated fields are usually employed, Normally the radiance of the test stimulus is varied and its
24、brightness is made equal tothat of the reference stimulus. The luminous efficiency of the stimulus is proportional to the reciprocal of the radiance at the match condition. The use of a fixationpoint at the time of measurement is optional. An observer can easily view the stimulus foveally even when
25、it appears without the fixation point. The luminous efficiency is not affected by the observer scanning over the field, or fixating at the central portion of the field, unless an intense fixation is made. Such an intense fixation induces a decaying of the field and the luminous efficiency may change
26、, particularly in the case of point sources. With using a 10 field and short wavelength light the observer normally sees a darker spot at the center ofthe visual field because of the macular pigment. The observer should be instructed to neglect the spot and evaluate the brightness in a parafoveal ar
27、ea. CIE 75 88 9006345 0003349 02T = 3 3 EXPERIMENTAL CONDITIONS Some of the experimental conditions are listed in Tables 3.1, 3.2, and 3.3 for point sources, 2O and loo fields, respectively, Most of the researchers presented the stimuli to their observers continuously, but some used a flash, typical
28、ly of one second duration. 4 TERMINOLOGY The notation vb(x) is used to represent the spectral luminous efficiency function based upon brightness matching for monochromatic light. Thus the function for point sources is denoted by vJb,point (A); for the 2O field by Vb,Z(h); and for the IOo field by Vb
29、,lo(h)- It must be clearly recognized that the functions cannot be applied to light sources with compound spectra in order to obtain their luminous efficiencies for brightness, as is done by the CIE V(X) function, because of the failure of additivity for different spectra. 5 RESULTS 5.1 Point source
30、s It is difficult to define how small a stimulus should be to be called a point source: this needs to be defined quantitatively in the future. In this report it is defined as a source presenting almost no perceptual area to the observer. Only a small amount of data are available that satisfy this co
31、ndition as is shown in Table 3.1 C5.1, 5-21. The results are summarized in Table 5.1, values being given in logarithmic units. The data in the columns 2 and 3 are simply averaged, disregarding the number of observers, to give the average data in column 4. These average data are plotted in Fig. 5.1 a
32、s open circles, together with the Judd modified V(A) function as a solid line. This figure shows that the spectral luminous efficiency CIE 75 BB U 900b145 0003350 841 4 -31“ 400 500 600 700 Wavelength in nm Fig. 5.1 Experimentally determined spectral luminous efficiency function for point sources (o
33、pen circles) and the CIE V(X) of Judd modification (solid line). Fig. 5.2 Age distributions of subjects for 2O (circles) and for 10(crosses). function for a point source may be adequately represented by the Judd modified V(h) function. It should be remembered that the amount of datais limited, and t
34、here is the possibilityof revising the Vb,point (A) should more data become available. CIE 75 88 OOblt45 0003351 788 D 5 5.2 2* field data As shown in Table 3.2, the number of observers giving data fora 2Ofieldwas 63andtheirages weredistributed from 20 to 55 years; see Fig.5.2 5.3-5.161. (One subjec
35、t exceeded the range set originally but the data were included in the final analysis.) but their data were also included. Mean luminous efficiency values for each research group are listed in Table 5.2. They are given in logarithmic units and normalized at 570 nm - ar1 arbitrary wavelength. Some res
36、earchers used a field size of less than 2O, The procedure used to derive the final Vb,z(A) was as follows: (i) All calculations were done using logarithmic units. (ii) All individual data were tabulated in 10 nm steps. Linear interpolation was used to complete the table where data were available onl
37、y in 20 nm steps. unity (or log vb 2 = 0.00) at 570 nm. (iii) These values were simply averaged without giving specific weight to any observer. column 13 of Table 5.2. These values may differ from means of averages given in the foregoing columns in Table 5.2 because the number of observers used diff
38、ered among researchers. The data were normalized to I The results are tabulated in the Fig. 5.3 Scheme to illustrate the cascade method to obtain efficiency at spectral end. CIE 75 48 S 9006345 0003352 614 6 (iv) The number of observers differ depending on the wavelengths of the test stimulus with f
39、ewer observers being used at each end of the spectral range. Because of this the simpleaverage cannotbeusedas a finalfunctionanda cascade method was used as follows: A scheme to derive the efficiency at lzI where the number of observers decreased, compared with those at X1 is shown in Fig. 5.3. obse
40、rvers available at As (the filled circle at 121, and log vbIz(l) A,. Then the difference was calculated and the efficiency at 12, log Vb,2(h2), was obtained using the following equation: Let log Vb,2(X2)* denote the average based on the all * the average based on the same observers but at * * Av1-2
41、= log vb,Z(xl) - log vb,2( 12) log Vb,2(12) = log vb,2(x1) - AV1-2. vb (11) represents the efficiency at A1 from all the data available at that wavelength. I This cascade method was applied step-by-step until the end ofthe spectral rangewas reachedandtheresults are given in column 14 of Table 5.2. T
42、hese data are plotted as open circles in the lower section of Fig. 5.4 and should be considered to be the spectral luminous efficiency function for a 2O field. 5.3 ,Oo field data As shown in Table 3.3 the number of observers giving data for a 100 field was 76. r5.9, 5.10, 5.12-5.15, 5.19-5.21 3 Two
43、subjects greatly exceeded the age range specified (Fig. 5.21, but it was found that there was no systematicchange in the luminous efficiency function with advancing age. Thus their data were also included in the analysis. One researcher used an 8O field and these data were also included. The analysi
44、s was carried out in the same way described for the 2O field and the results are given in column 11 of Table 5.3. They are plotted as crosses in Fig. 5.4. 7 Wavelength in nrn Fig. 5.4 Lower section; experimentally determined spectral luminous efficiency functions of 2O(circIes) and 1 Oo( crosses). U
45、pper section; difference between IOo and 2O functions. 6 DERIVATIONS OF FINAL 2O AND IOo FUNCTIONS When the luminous efficiency functions vb,2(xl and b8 1 o (A) are plotted together, as in Fig. 5-48 it appears that the difference between the two functions is rather small for wavelengths greater than
46、 540 nm, and significant only at short wavelengths. By applying the t test to the mean value at each wavelength it was confirmed that the difference is significant, to a confidence level of 0.05, at wavelengths from 420 nm to 600 nm, except at 540 nm. If the confidence level is increased to 0.01 tha
47、n the difference is significant from 420 nm to 520 nm, and at 560, 580 and 590 nm. This is indicated by asterisks .in column 4 of Table 6.1. The significance found at the three latter wavelengths is probably due to the normalization procedure using a wavelength of 570 nm. The standard deviations of
48、inter-observer data become very small at around 570 nm as shown in Table 5.2 and 5.3; consequently the corresponding t- 8 values became rather large. 20and10 field sizes for wavelengthsgreaterthan 530 nm, and at 400 nm, by averaging the two original functions. Two separate functions are retained for
49、 the other wavelengths. The final recommended values of log Vb,l o( A) and log vb,2(x) are tabulated in Table 6.2, and plotted in Fig. 6.1 by crosses and open circles, respectively. The difference between the two functions, AVpi, is given in column 4 of Table 6.2 and plotted at the top of Fig, 6.1. For simplicity a single function has been derived for the +p0 +O OOOOo +o O +o + oo O O +o O0 +O O b vb modified o T. 63s + 10 765 -3 O O O O Wavelength in nm Fig. 6.1 Lower section; Proposed luminous effici
copyright@ 2008-2019 麦多课文库（www.mydoc123.com）网站版权所有
备案/许可证编号：苏ICP备17064731号-1