1、CIE 207 94 m 9006245 0004926 905 )A PJ f ISBN 3 900 734 49 6 COM M ISS ION INTERN AT1 O NALE DE I C LAI RAGE INTERNATIONAL COMMISSION ON ILLUMINATI ON INTERNATIONALE BELEUCHTUNGSKOMMISSION REVIEW OF THE OFFICIAL RECOMMENDATIONS OF THE CIE FOR THE COLOURS OF SIGNAL LIGHTS CIE 107-1994 UDC: 535.625 62
2、8.975 656.054 Descriptor: Coloured signal lights Signal tighting Traffic signals COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE 107 9Li 9006345 0004927 841 W This Technical Report has been prepared by CIE Technical Committee 4-14 of Division 4 Lighting
3、 and Signalling for Transport and has been approved by the Board of Adminstration of the Commission Intemationale de IEclairage for study and application. The document reports on current knowledge and experience within the specific fietd of light and lighting described, and is intended to be used by
4、 the CIE membership and other interested parties. 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 CI
5、E 4-14 de la Division 4 Eclairage et signalisation pour les transporis et a t approuv par le Bureau dAdministration de la Commission Intemationale de IEclairage, pour tude et application. Le document traite des connaissances courantes et de lexprience dans le domaine spcifique indiqu de la lumire et
6、 de lclairage, et il est tabli pour lusage des membres de la CIE et autres groupements intresss. 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 Berich
7、t ist vom CIE-Technischen Komitee 4-1 4 der Division 4 Beleuchtung und Signale fur den Verkehr 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 u
8、nd Beleuchtung; es ist zur Verwendung durch CIE-Mitglieder und durch andere Interessierte bestimmt. Es sollte jedoch beachtet werden, da das Dokument eine Empfehlung und keine Vorschrift ist. Die neuesten CIE-Tagungsberichte oder das CIE NONS sollten im Hinblick auf mgliche sptere nderungen zu Rate
9、gezogen werden. Any mention of organisations or products does not imply endorsement by the CIE. Whilst every care has been taken in the compitation 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
10、CIE. Malgr le soin apport la compilation de tous les documents jusqu la mise sous presse, ce travail ne saurait 1965) as well as experience in the application of the recommendations. The principal revisions were to change the way the restricted colour areas were specified, to introduce a colour area
11、 for violet colours, to extend the blue boundary for white to include the chromaticity of a xenon source, and to make improvements to the text. Figures IA , B and C show the boundaries of the colour areas recommended in Publication 2.2 (1975) for red, yellow, white, green, blue and violet signal col
12、ours. Table 1 summanses the experimental conditions of the three studies that form the basis of the 1975 recommendations. All three studies used colour vision normal male subjects who made observations under dark adapted conditions, simulating observation at night. TABLE 1 Experimental conditions of
13、 the studies underlying the CIE recommendations for the coiours of signai lights Year Number of Adaptation Source size Illuminance Duration subjects (PIX) (SI McNicholas Series f 1936 32 Dark adapted 2,2“ - 22“ 1,5 to 35,75 5 (outside at night) Holmes 1941 varied Dark adapted O,? - 39 1,75 to 989 I
14、(Source size and illuminance differed for each set of experiments) Hill 1947 9 Dark adapted 36“ 30,4 to 0,8 1 ,s “All subjects were rnafes with normal colour vision 2 COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE 107-1994 Figure 1 A Boundaries for col
15、ours of signal lights currently recommended by the CIE (Colors of Signal Lights CIE Publication 2.2-1975) 3 COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE LO7 94 I OObL45 OOOq935 938 I CIE 107-1994 X - Current CIE signal boundaries -=- Current restrict
16、ed CIE signal boundaries O :s 0:s X 0.7 Figure 1 B Boundaries for green, red and yellow signal lights currently recommended by the CIE (Colors of Signal Lights CIE Publication 2.2-1975) 4 COPYRIGHT International Commission on IlluminationLicensed by Information Handling ServicesCIE LO7 94 9006145 00
17、0493b 859 CIE 107-1994 8 o: 1 X 012 - CIE standard obsewer - Current CIE signa) boundaries .- Current restricted CIE signal boundaries A Figure 1 C Boundaries for blue, violet, white and yellow signal lights currently recommended by the CIE (Co/ors of Signal Lights CIE Publication 2.2-1 975) 5 COPYR
18、IGHT International Commission on IlluminationLicensed by Information Handling Services CIE LO7 94 m 9006345 0004937 790 m CIE 107-1 994 4 METHOD A comprehensive bibliography was compiled by computer search of bibliographic data bases, by citation in other references and through the contributions of
19、the members of the Committee who drew attention particularly to technical reports not published in the open literature. The references in the bibliography were read and those that provided quantitative information on the recognition of signal colours and which adequately specified the chromaticities
20、 of the colours used in the experiments, were identified as key references that could be of assistance to the task of the Committee. The experiments reported in the key references used a wide range of conditions. Most required subjects to make observations in the dark but a few used light adapted co
21、nditions. Stimulus size, stimulus illuminance in the the plane of the subjects eye and duration of observation time all vaned. A summary of the principal experimental conditions for each of the investigations described in the key references was made and this summary is given in the Appendix. Some si
22、gnals have to be seen at long distances usually at night and have low illuminances at the plane of the observers eye; other signals are intended to be recognised at shorter ranges and often have very much higher illuminances. Table 2 gives typical visual ranges and illuminances for signals used at s
23、ea, for aviation and railway signals and signals used for the control of road traffic. TABLE 2 Signal lights used in transport systems Transport Type of signal Typical intensity Typical viewing Typical system (cd1 distance (km) illuminance (FIX) Maritime Masthead 5 - 94 4-12 0,2 - 0,5 Navigation lig
24、hts 5 -12 4-6 0,2 - 0,3 Lighthouse 200 20 - 30 0,2 - 0,3 Buoys 5 2-5 0,2 - 0,3 Rail Signal lights (i) High intensity 300 - 600 3-5 ?O (i) Suburban 150 - 250 1-2 40 Aviation Signal gun Obstruction lights (i) Steady (i) Flashing Runway lights Approach lights Taxiway lights Navigation lights Presence l
25、ight Road Traffic lights 1 O0 Tail lights 2 30,000 3-6 3 O00 15 - 20 O00 2-5 0,6 - 5 O00 2 O00 - 200 O00 3 - 20 15 - IO O00 01 - 5 60 - 20 O00 0,l - 5 20 - 200 0,l - 0,5 80 - 20 O00 5 - 40 2-5 0,2 - 10 25 5- 10 0,4 - 1 5 - 20 O00 o, 1 o. 1 10 O00 200 Stop light 20 o; 1 2 O00 Source: Adapted from Vin
26、grys (1984) and Smith (1990) Using this information as a guide, the data from the experiments reported in the key references have been grouped according to signal illuminance and adaptation condition: (a) Dark adapted (Night observation) (i) illuminance I 1 plx typical of recognition of signal light
27、s at distances of 2 - 30 km at night, as encountered commonly in maritime signalling and to a lesser degree in aviation signalling. (i) Itluminance 1 to 1 O plx typical of some aviation signals. (iii) Illuminance O to 1 O0 plx typical of railway signals, and some aviation signals. 6 COPYRIGHT Intern
28、ational Commission on IlluminationLicensed by Information Handling ServicesCIE LO7 94 E OObL45 0004938 627 E CIE 107-1994 (iv) Illuminance 0,l to 1 mlx typical of motor vehicle signals and some aviation signals. (v) Illuminance larger than 1 mlx typical of road traffic signals and some aviation sign
29、als. (b) Light adapted (Day observation) As there were only five experiments in this condition, all illuminance levels were grouped together. For each investigation reported in the key references, the signal colours used were plotted on large scale 1931 CIE diagrams together with a notation for each
30、 point indicating the probability of correct recognition of the colour reported by the experiment. For each experiment separately, probability contours were drawn by eye for red, yellow, green, blue and white colours provided there was data for a sufficient number of colours to enable the constructi
31、on of probability contours. In order to compare the results of the different studies, the 90% probability of recognition contours were estimated for red, yellow, green, blue and white signal colours and transferred to a separate 1931 CIE chromaticity diagram for each of the experimental conditions l
32、isted above. These plots are reproduced as Figures 2 to 7 in which the legends list the references from which the information in each figure was obtained. The 90% probability of correct recognition contours were plotted because it is only those colours that are recognised with reasonably high certai
33、nty that are of importance for practical visual signalling and for the purposes of this review. Colours that are called red, yellow, green, blue and white on less than 90% of occasions are of little value as signal colours. For some of the experiments reviewed, there were insufficient data points to
34、 enable a reliable estimate of a 90% contour. In these cases the chromaticity co-ordinates of any individual colour that was recognised with a probability of 90% or more were plotted. 5 DISCUSSION AND RECOMMENDATIONS 5.1 Red signal colours The yellow boundary of red Figures 2 to 7 show that while th
35、e CIE colour area for red generally lies within the 90% probability contours, many of the 90% contours cross the red colour area inside the yellow boundary of red suggesting that orange-reds within the recommended CIE boundary are not reliably recognised as red. It must also be borne in mind that th
36、e contours are for 90% correct recognition and for most practical applications a higher degree of reliability is expected. Most users of the CIE recommendations will assume that any red colour lying within the recommended colour area will be reliably recognised as red but in fact this is not the cas
37、e: orangish-reds may be incorrectly identified on more than 10% of occasions. The present yellow boundary of red is misleading by suggesting that it defines the coiour area of reds that are reliably recognised. There is therefore good reason to consider moving the yellow boundary of the red colour a
38、rea to a longer dominant wavelength. On the other hand it can be argued that orange-reds should continue to be included in the recommended cotour area for red signals because: 7 COPYRIGHT International Commission on IlluminationLicensed by Information Handling Services CIE LO7 94 9006145 0004939 5b3
39、 CIE 107-1994 Legend CIE standard observer Current CIE signal boundaries - - - - HiIl(1947), ROYGBW, E=0,386 pix, O, t=1,5 s -.-.- Hi11(1947), ROYGBW, E=0,77 pix, 0,6, t=1,5 s Kinney( 1979), RGW or “Not visible“, E=O, 12-15,5 plx 3-10“, t=1,5 s + Hofmann(1975) border of test colours I . -.- Hofmann(
40、l975) RYGBW, E=0,67 pixl i, 2 s Hofmann(l975) RYGBW, E=0,67 plx, IO, 2 s r.-#.-.-. 455 nm 470 nm o Figure 2 Recognition of coloured signal lights of illuminance s 1 microlux at night 1931 CIE chromaticity diagram showing 90 % correct recognition contourc In this and subsequent figures the sign repre
41、sents “minute of arc“. COPYRIGHT International Commission on IlluminationLicensed by Information Handling Services CIE 107-1994 Legend b CIE standard observer Current CIE signal boundaries Holmes(l941) border of test COloUrS .,., , :.:.:.:. Holrnes(lO41) RYGBWP, E=1,74 pix, 0,7-40, I s x=-x=-=: Holm
42、es(l941) RYGW, E=2,91 fix, 0,740, 1 s -.- Holrnes(l941) RYGN, E=2,91 PIX, 0,7-40, Is McNicholas(l936), ROYGBWP, E varied, 2,2-22“, 5s CrCCCCC*t=* Halsey(l959-11), RYGBW, E vaned, 46“-710“, 2 s Das(l966), RYGB ,.,. Hofmann(l975), RYGBW, Ea7 I, 0,04 s flashing 4 Halsey et al(1951) Colours numbered, E=
43、377 pix, 45, tvanabie * Verniest(l979), ROG, E*-7 mix, size 8 time varied X Figure 7 Recognition of coloured signal lights of illuminance 0,2 to 70 O00 microlux during the day 1931 CIE chromaticity diagram showing 90 % correct recognition contours 13 COPYRIGHT International Commission on Illuminatio
44、nLicensed by Information Handling ServicesCIE 107-1994 Author Signal system Colour Defective simulated vision dour normal vision observers observers Farnsworth (1946)* Maritime o; o; o o; 3; 1 Heath and Schmidt(l959) Aviation O II Nathan et al( 1964) Road 1 OP5 Freedman et al (1 985) Road O 20-; 30
45、Farnsworth (1 946) Maritime O 2 Sloan and Habel (I 955) instrument panel lights O 3 Heath and Schmidt(l959) Aviation O 19 Nathan et al( 1964) Road O 2 Heath and Schmidt(l959) Aviation O 22 Nathan et al( 1964) Road 8 10 Freedman et al (1985) Road O 27-; 58 Farnsworth (1946) Maritime 8; 9 11; 15 Freed
46、man et al (1985) . Road -0 ,33*; 6Q I (a) red filter glasses with long dominant wavelength have low luminous transmittance and it is difficult to achieve signal intensities that will give the desired visual range with these colours. However, modern light sources (such as halogen tungsten filament la
47、mps, xenon lamps, etc.) are more efficient than the sources used in the past and high intensities can be more readily achieved even with filters of low luminous transmittance. High intensity sources consume more power and this is a consideration for signals that cannot be connected to a mains power
48、supply. However, there are now power generators (eg solar) that make this a less critical concern. In any event the loss in luminous transmittance from the use of deeper reds compared to orange-reds is not unduly great (Cole and Brown, 1966). if for example, the yellow limit of red is moved from y =
49、 0,335 to y = 0,315, the maximum practical luminous transmittance is decreased from 30% to 25%, a 17% reduction whtch will only reduce the visual range by about 10%. some 2% of men have a form of defective colour vision which markedly reduces their ability to see red signals. Since their handicap is greater for the long wavelength reds and least for (b) orangish-reds, they will be more likelyto be able to see orange-red signals (Cole and Ving
