Analysis of Luminance Contrast Values at Illuminated Pedestrian Crossings in Urban Conditions

Abstract

Lighting pedestrian crossings play an important role in ensuring the safety of pedestrians on the road, especially at night or in conditions of reduced visibility. Currently functioning normative and formal requirements, which are the basis for the design of lighting for pedestrian crossings, are described by criteria of lighting intensity. Each of the existing solutions operates in an urban environment, and at the stage of design and operation, the obtained values of the produced luminance contrast of the pedestrian silhouette and its background are not known. The main purpose of the article is to compare, based on luminance parameters, the three lighting solutions used at pedestrian crossings. This article presents and describes the method of determining the luminance contrast based on luminance measurements of the pedestrian silhouette and its background. Detailed results of measurements of luminance parameters at selected pedestrian crossings are presented. An analysis of the results was made with a breakdown of typical lighting solutions used in urban conditions. The differences between standard lighting were discussed, as well as supplementary and dedicated lighting. The obtained ranges of luminance contrast levels occurring in urban conditions for individual solutions were determined. The energy indicator for each solution was determined.

1. Introduction

Evaluation of the impact of lighting on pedestrian safety indicates that pedestrian crossings are a particularly important and sensitive element of road infrastructure [1]. The visibility of a pedestrian at a pedestrian crossing at night is identified as a key factor affecting the safety of road users [2,3]. Studies indicate the importance of the manner and quality of lighting affecting the distance from which a driver can see a pedestrian approaching the roadway, waiting to cross or crossing the roadway [1,2,4,5,6]. As early as 1978, Polus et al. [7] showed a significant reduction in car and pedestrian accidents at night due to the installation of pedestrian crossing lighting. Improving the quality of pedestrian crossing lighting, based on a study in Australia, reduced the risk of pedestrian accidents at night by almost 60% [8]. According to [9], three times more fatal accidents occurred at unlit or improperly lit pedestrian crossings than at crossings equipped with adequate lighting. Numerous studies [2,10,11,12] show improvements in pedestrian safety after the installation of lighting systems dedicated to pedestrian crossings. Numerous studies [13,14,15,16] have also focused on improving pedestrian safety through the selection of appropriate lighting technical parameters.

In operational terms, the unequivocal lighting parameter documenting the level of illumination of the pedestrian crossing area is illuminance. The effect of the value of the vertical component of the illuminance on the achieved value of visibility at a pedestrian crossing was analyzed [17,18], but without taking into account the conditions of the urban environment. The basis for the design of pedestrian crossing lighting is the lighting classes resulting from the normative recommendations used in Europe, recorded in the documents of the five-sheet standard [19]. It should be mentioned that, until the start of the research work, a uniform and comprehensive procedure for controlling the state of lighting of pedestrian crossings had not been developed and implemented in Poland. Taking into account parameterized lighting factors. There are countries with formal requirements [20,21] and few studies [22] have been carried out on the lighting of pedestrian crossings. Road traffic safety audits and lighting inspections [23,24,25] assessing the condition of pedestrian infrastructure have also been conducted worldwide, but the luminance factor has not been included and described in detail. The international literature has not identified a way to conduct control measurements of lighting parameters of pedestrian crossings using luminance and contrast criteria. Such work has been undertaken in Europe [26,27,28], but, to date, the proposed approach has not been implemented on a larger scale.

Many countries have laws and formal regulations indicating lighting requirements for pedestrian crossings. Some of these derive directly from the recommendations of the International Commission on Illumination, CIE [29], which is a non-governmental, self-financing organization for international cooperation and information exchange on all matters related to light and lighting. In its publications (technical reports), it presents recommendations and guidelines on lighting principles and levels, as well as calculation procedures and ways to carry out measurements. Based on CIE guidelines [29], reports are developed by the European Committee for Standardization (CEN) on, among other things, road lighting (e.g., CEN/TR 13201-1 [19]). CEN members are the national standards bodies of the following countries: Austria, Belgium, Cyprus, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, the Czech Republic, Slovakia, Slovenia, Spain, Sweden, Switzerland and the United Kingdom. Following the standardization process, these countries are using uniform norms for road lighting. In Poland, too, the formal requirements of the road lighting standard have been developed based on the analysis presented in the report [19]. As of 2017, the CIE International Lighting Commission has not provided explicitly defined quantitative recommendations relating to lighting levels for the pedestrian crossing area. Existing qualitative provisions are indirectly discussed in publications on road lighting [30,31,32,33,34]. The guidelines contained in the 2010 publication CIE 115 [29], which is an update of the 1995 report, identify the most important recommendations for road lighting. The above publication distinguishes three basic lighting categories, affecting the selection of applied lighting for areas where pedestrian traffic may occur. These provisions are reflected in Polish standards [19].

Over the years, Poland has had various standards [35,36] defining the rules and conditions for the use of road lighting and lighting used in the area of pedestrian crossings. These regulations defined the general rules for the use of lighting solutions applied in a conflict area, which is considered a pedestrian crossing. Currently, road lighting is regulated by the five-sheet European standard PN-EN 13201:2016 road lighting [19]. The standard, about the area of pedestrian crossings, does not specify lighting requirements but only gives general guidance on the solutions to be used.

Some countries apply additional lighting requirements for the lighting of pedestrian crossings, which operate in parallel with road lighting requirements and lighting standards. Separate regulations are used by countries such as Belgium [37] including Flanders [38], Czech Republic [21], Norway [39], Germany [20,40], Sweden [41], Switzerland [42], Italy [43], the UK [44], USA [9,45], Australia and New Zealand [46]. The cited regulations take a varied approach in terms of detail and scope of requirements.

According to the available literature, to date, the rules for proper lighting of pedestrian crossings have not been standardized in the EU. Existing regulations generally point to an approach that enforces the achievement of a high level of contrast—doable or negative [30] (Figure 1), with a low or zero contrast value as the worst case. The need to distinguish the silhouette of the pedestrian is emphasized, mainly by the level of illumination, or by the colour of the light.

Countries such as Germany [20], the Czech Republic [21] or Australia and New Zealand [46] have introduced detailed requirements for dedicated solutions described by illuminance values (luminance of the roadway in front of and behind the crossing [20]). However, based on the illuminance criterion, one cannot conclude the created luminance contrast level of a pedestrian crossing the roadway against the background of the crossing environment. The luminance of the pedestrian crossing environment must be taken into account. The luminance parameter is the best criterion for determining the conditions for a vehicle driver to observe a pedestrian.

In practice, this should mean using the luminance method to assess the illumination of the human figure at a pedestrian crossing. This study uses luminance and luminance contrast parameters [25,33,47]. An effective way of evaluating the characteristics of the lighting used at a pedestrian crossing is the field tests of luminance distribution on pedestrian silhouettes [48].

2. Lighting Solutions Used at Pedestrian Crossings

In urban areas, the assessment of the condition of the lighting infrastructure is very important from the point of view of maintaining the existing condition or improving lighting conditions affecting road traffic safety [23,49,50]. This article undertakes a comparison of the obtained results of measurements of lighting parameters obtained with different types of lighting solutions. The technical solutions used to illuminate pedestrian crossings can be divided into three types: A (street), B (supplementary) and C (dedicated)—the nomenclature adopted for the minor article.

City streets are illuminated in different lighting classes (from M1 to M6) [19]. Thus, pedestrian crossings are illuminated at the same level as the roadway, using standard road lighting located in the lighting system geometry adopted for the road. In this solution (Type A), a high level of illumination must be produced behind the pedestrian crossing in such a way that the pedestrian has a significantly lower luminance value than the illuminated roadway behind the pedestrian. Thus, the illuminated section of the roadway must be long enough, even though the standard [19] does not indicate this parameter. The luminaires used are characterized by a typical, light distribution for roadway (Figure 2).

Such a lighting situation is implemented with street lighting luminaires arranged in one of several standard configurations along the street string. The solution is characterized by the randomness of the location of the lighting pole and luminaire about the geometry of the pedestrian crossing. For most installations, the streetlight lantern is located directly adjacent to the pedestrian crossing, but this is not the rule. As indicated in the simulation, pedestrians are usually observed by drivers as dark silhouettes against a lighter roadway background, in negative contrast (Figure 3) or low positive contrast (Figure 4). It is also possible that the luminance of the object and the background are equalized and thus the contrast is zeroed out.

Another solution (Type B) is to leave the lighting pole in place, change the jib and use two luminaires with changed light distribution characteristics instead of a single road luminaire (Figure 5).

Alternatively, the change can consist of supplying a new lighting pole near the pedestrian crossing and using two luminaires on a single V-type boom. The use of this technical solution makes it possible to increase the amount of light on the horizontal surface of the pedestrian crossing and the waiting zones of the crossing (Figure 6). A feature is the different values of the obtained level of illumination on the vertical plane of the crossing for each direction of vehicle traffic (Figure 6 and Figure 7).

This value depends on the location of the luminaire and its position relative to the axis of the pedestrian crossing. Thus, the luminance contrast of the pedestrian silhouette with the background changes depending on the direction of observation. The area of the crosswalk and its surroundings is distinguished by the level of illumination from the street course, but it is difficult to control the distribution of light, which causes the effect of light littering the surroundings of the crosswalk.

The dedicated technical solution (Type C) consists of supplying near the pedestrian crossing, for each direction of vehicle traffic, the use of two luminaires (at a height of about 5.5 m), characterized by the double asymmetry of light distribution (Figure 8).

The use of such a technical solution makes it possible to increase the value of the vertical component of the illumination intensity on the vertical plane of the pedestrian crossing (on the pedestrian silhouettes) and, incidentally, on the horizontal plane of the crossing and in the pedestrian waiting zones of the crossing (Figure 9 and Figure 10).

In the case of this solution, the produced lighting levels are symmetrical for each direction of vehicle traffic. Due to the introduction of the light beam from a lower height, the generated positive luminance contrast of the pedestrian silhouette with the background is high and symmetrical in the cross-section of the passage for each direction of observation. In addition, in this solution, the pedestrian crossing area is distinguished by a higher level of illumination from the entire roadway, and the distribution of the beam is limited to the illuminated area. In addition, the positive contrast obtained is positively enhanced by the lights of an oncoming vehicle.

3. Methods

The authors of the publication propose the use of the luminance criterion to determine the generated luminance contrast of selected solutions used to illuminate a pedestrian crossing. Based on field luminance post measurements, the value of luminance contrast between the silhouette of a pedestrian in the crosswalk and its background (the surroundings of the crosswalk) was determined.

Performing luminance measurements according to the guidelines contained in the report [29] and in the standard [19], especially in continuous operation, is difficult to implement under conditions of heavy urban traffic. Surveys are often practically impossible to carry out without closing a street section. To study and record the luminance distribution on the pedestrian silhouette and in its surroundings, one can use claque luminance meters or high-tech matrix meters. The matrix meter used makes it possible to take pictures and, consequently, images of the luminance of any object without the need to transport and install additional equipment. Thanks to the development of measurement tools that make it possible to take images of pedestrian crossings scaled in luminance levels [51,52], it is possible to calculate the luminance contrast of the silhouette of a person at a pedestrian crossing with the surroundings—the background. This approach makes it possible to take into account the actual lighting conditions present in the surroundings of the pedestrian crossing.

To create uniform comparative conditions related to the dimensions, coefficient and nature of reflection, it was decided to use a measurement object reflecting the geometric and reflective properties of the human silhouette. Based on anthropometric data [53] for the 50th percentile silhouette of an adult male in a lateral position, the dimensions of the test object were developed in the form of a rectangle of 0.25 × 1 m, which reflects the lateral area of a human being at a pedestrian crossing. The view and positions of the object’s location are shown in Figure 11. The centre of the object is 1 m above the road surface. The adoption of such positioning is due to the existence of guidelines [19,29] for this height of measurement, and at the same time takes into account the dimensions of other silhouettes, as in, for example, women, children and people with disabilities who use wheelchairs.

Ongoing studies [45,54,55] show that more than 90% of clothes worn by pedestrians have a reflectance ρ of less than 20%. Figure 12 graphically shows the relative frequency of cumulative variation in the incidence of reflectance of clothing worn by pedestrians.

To perform comparative tests of the luminance of a human silhouette at a pedestrian crossing, we proposed using a test object covered with a material with a reflectance of ρ = 0.2 and reflectance characteristics as close as possible to evenly diffused. Thus, the unfavourable situation of the driver’s observation of the object is assumed—the silhouette of a pedestrian wearing dark clothing. The value of the reflection coefficient ρ is determined as the ratio of the reflected flux to the incident flux. The measurement is made with a reflectance meter. Figure 13 shows the basic geometric parameters and a view of the actual object prepared for future field testing.

The positioning of the test objects at the pedestrian crossing and the directions of observation are shown in Figure 14.

The geometry of the location of the observer and the object is shown in Figure 15. Observer A is placed at a distance of 57.28 m from the transverse axis of the pedestrian crossing (Figure 14, section E–F), according to the guidelines [19,29,56], and maintains an observation angle of 1° on the roadway surface. The adopted measurement distance of about 60 m guarantees the maintenance of the correct angle of observation of the roadway and, at the same time, is a distance that guarantees the stopping of the vehicle under pedestrian detection conditions in a real traffic situation. The observer’s eyes are located at a height of 1.5 m above the road level. The axis of observation passes through the centre of the measurement object P.

The measurement should be carried out for each direction of vehicle traffic by measuring the luminance of objects located across the width of the pedestrian crossing, in the section E–F (Figure 14) with an interval between consecutive readings Δx ≤ 1.5 m. Performing luminance measurements in the opposite lane of traffic relative to the assumed direction of travel is necessary. Information about the obtained contrast of the pedestrian’s silhouette with the background is relevant for the entire width of the roadway and is a reflection of the actual traffic situation. The pedestrian may be anywhere on the roadway in the pedestrian crossing area or the waiting zone for crossing. A pedestrian driver approaching a crosswalk must have proper observation conditions in the entire crossing area including the waiting zone or asylum.

Luminance calculations for individual object locations were made based on a series of measurements, and then the luminance of the object (L_T) and the background to the left and right of the pedestrian silhouette (L_B1 and L_B2) were calculated. The luminance measurement plots are shown in Figure 16.

The contrast for each pedestrian crossing object should be calculated from the following Formula (1):

$$C_n=\frac{L_T-L_B}{L_B}$$

(1)

where:

C_n—object contrast (n = 1 do 8),

L_T—object luminance [cd/m²],

L_B—background luminance [cd/m²], remembering that L_B = (L_B1 + L_B2)/2.

For two-way roads, the average value of the contrast obtained is determined from the 8 test objects positioned across the crossing (Figure 11) and the luminance contrast values obtained for them for a given direction of vehicle traffic 1 and 2, from Formulas (2) and (3):

$$C_{AV1}=\frac{{{\displaystyle \sum }}_{n=1}^8{C}_{n1}}{8}$$

(2)

$$C_{AV2}=\frac{{{\displaystyle \sum }}_{n=1}^8{C}_{n2}}{8}$$

(3)

where:

$C_{n1}$—object contrast (n = od 1 do 8) direction 1,

$C_{n2}$—object contrast (n = od 1 do 8) direction 2,

$C_{AV1}$—average contrast value in the direction of 1,

$C_{AV1}$—average contrast value in the direction of 2.

4. Research Subjects

Pedestrian crossings for this study were selected based on a detailed analysis of the results of more than 4000 pedestrian crossings [57,58,59,60]. The focus was on selecting crossings with similar geometric features but that differ in lighting. The pedestrian crossings selected for the study have a typical geometric layout used in the city, with an isle of an aisle and similar dimensions (length and width). Field luminance tests were carried out on 60 pedestrian crossings located in Poland within the city of Warsaw (Figure 17). Traffic lights do not control traffic at the pedestrian crossings selected for the study. The crossings were located on dual carriageways—two-way roads in built-up areas, where street lighting functions. One of the most important criteria for selecting pedestrian crossings for the study was the similar lighting system used. Twenty pedestrian crossings of each type were selected for testing: those illuminated with street luminaires (Type A), additional street luminaires mounted on an additional boom (Type B) and luminaires dedicated to lighting pedestrian crossings (Type C). All pedestrian crossings were illuminated by various road lighting luminaires with discharge sources (sodium and metal halide) and LEDs in various states of condition, with known electrical and lighting parameters.

Tests were conducted at pedestrian crossings with an island 1.5 m wide. The pedestrian crossings had typical geometric dimensions. The dimensions of the test area including waiting zones were as follows: width 4–5 m, length 10–12 m. Pedestrian crossings were located on streets with a speed limit of 50 km/h. This study was conducted in summer, in good weather conditions with good air transparency. The effect of the radiation spectrum on the luminaire was not taken into account. The effect of luminance parameters on drivers’ reactions and behaviour was not analyzed. All tests were performed according to uniform procedures, by a three-person research team.

5. Results

Appendices A, B and C summarize examples of detailed measurement results realized on each type of lighting solution used at pedestrian crossings.

Analyzing the case of illumination of a pedestrian crossing with street lighting fixtures, Type A, presented in Appendix A (Figure A1 and Figure A2,

Table A1. Example of the results for measurement Type A pedestrian crossings in directions 1 and 2.

Number of Target	Direction 1			Direction 2
	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast
1	0.257	0.310	−0.170	0.162	0.210	−0.231
2	0.297	0.752	−0.605	0.185	0.241	−0.232
3	0.334	0.788	−0.576	0.205	0.370	−0.445
4	0.392	1.126	−0.651	0.243	0.643	−0.621
5	0.374	1.332	−0.719	0.283	2.459	−0.885
6	0.309	0.798	−0.613	0.271	1.170	−0.768
7	0.299	0.937	−0.681	0.281	0.615	−0.543
8	0.226	0.258	−0.125	0.265	0.323	−0.179
average	0.311	0.787	−0.517	0.237	0.754	−0.686

), it can be concluded that the applied solution realizes low luminance levels on pedestrian silhouettes (Figure A3), which, for typical luminance levels of the roadway and crossing environment (Figure A4), in effect generates negative contrast in the entire cross-section of the pedestrian crossing for both directions of observation (Figure A5). Its lowest value (−0.17) occurs in the waiting zones (objects 1 and 8).

For the Type B solution, the results presented in Appendix B (Figure A6 and Figure A7,

Table A2. Example of the results for measurement Type B pedestrian crossings in directions 1 and 2.

Number of Target	Direction 1			Direction 2
	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast
1	0.350	0.284	0.232	1.125	0.252	3.470
2	0.543	0.888	−0.389	1.396	0.478	1.920
3	0.468	1.095	−0.573	1.360	0.374	2.638
4	0.486	0.835	−0.417	1.398	0.443	2.157
5	0.488	1.264	−0.614	1.321	0.887	0.489
6	0.476	0.760	−0.374	1.147	0.490	1.340
7	0.403	1.010	−0.601	0.912	0.359	1.542
8	0.334	0.656	−0.491	1.114	0.569	0.958
average	0.443	0.849	−0.403	1.222	0.481	1.538

), the luminance values of objects at the crossing differ significantly depending on the direction of observation (Figure A8). The solutions were used to illuminate the perimeter background of the pedestrian crossing (Figure A9), so opposite contrast values were obtained depending on the direction of observation (Figure A10). For direction one, the dominant contrast takes on negative values (except for the waiting zone where there is a change in contrast values from positive to negative), while for direction two, high positive contrast values were obtained. For object number 5, located in the central zone of the transition, the obtained contrast value is lower than for the other objects. This is due to the presence of a local maximum in the background luminance.

For the dedicated solution, Type C, the results are presented in Appendix C (Figure A11 and Figure A12,

Table A3. Example of the results for measurement Type C pedestrian crossings in directions 1 and 2.

Number of Target	Direction 1			Direction 2
	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast
1	0.875	2.603	−0.664	0.637	0.426	0.494
2	1.243	0.437	1.841	0.967	0.479	1.018
3	1.335	0.490	1.726	1.729	0.546	2.168
4	2.420	0.515	3.696	3.526	0.917	2.843
5	3.795	1.106	2.430	3.968	2.169	0.829
6	2.909	0.510	4.707	2.863	0.896	2.195
7	2.459	0.494	3.976	2.685	0.549	3.887
8	2.514	0.584	3.306	2.479	0.980	1.529
average	2.194	0.842	2.627	2.357	0.870	1.708

). The luminance values of objects in the passage do not differ significantly depending on the adopted direction of observation (Figure A13). The solutions used illuminate the background of the pedestrian crossing in a controlled manner (Figure A14), so similar contrast values were obtained depending on the direction of observation (Figure A15). For the first and second directions, the dominant contrast takes on do-date values, except for waiting zone 1, where there is a change in contrast values from positive to negative due to the very high luminance of the background. Also, for object number 5, located in the central zone of the transition, the obtained contrast value is lower than for the other objects. The occurrence of a local maxima of background luminance causes this. The presented lighting solution provides drivers with similar conditions for observing a pedestrian located both on the roadway and in the waiting zone, regardless of the direction of observation.

Table 1. A combined summary of the results of luminance measurements and contrast calculations for the three types of pedestrian crossings studied for the two directions of observation.

Direction 1
	Type A			Type B			Type C
No.	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast
1	0.16	0.23	−0.30	0.76	1.64	−0.54	2.58	0.56	3.59
2	0.32	0.33	−0.01	3.08	1.74	0.77	2.30	0.66	2.49
3	0.33	0.43	−0.22	2.27	1.78	0.28	2.75	0.80	2.43
4	0.50	0.46	0.10	0.67	2.66	−0.75	3.05	0.84	2.61
5	0.37	0.49	−0.25	0.62	2.77	−0.78	5.61	0.90	5.20
6	0.11	0.57	−0.80	0.99	1.33	−0.26	4.68	0.91	4.11
7	0.15	0.60	−0.76	1.89	1.34	0.41	3.38	0.94	2.59
8	0.71	0.69	0.02	1.59	1.34	0.18	4.66	1.00	3.67
9	0.48	0.75	−0.37	0.76	1.35	−0.43	3.94	1.00	2.95
10	0.74	0.88	−0.15	0.43	1.35	−0.68	4.39	1.05	3.19
11	0.51	1.07	−0.52	0.34	1.35	−0.75	4.65	1.19	2.92
12	0.43	1.17	−0.63	2.16	1.41	0.53	4.90	1.25	2.93
13	0.82	1.27	−0.35	2.44	1.49	0.64	5.80	1.43	3.05
14	0.62	1.38	−0.55	0.40	1.02	−0.61	3.45	0.86	3.03
15	1.03	1.52	−0.33	2.02	1.02	0.97	3.05	0.84	2.61
16	0.44	1.54	−0.71	2.50	1.03	1.43	4.97	0.90	4.49
17	1.24	1.81	−0.31	2.04	1.05	0.94	3.82	0.91	3.17
18	1.76	2.30	−0.23	0.45	1.07	−0.58	3.38	0.94	2.59
19	0.40	3.07	−0.87	0.27	1.05	−0.74	5.29	1.21	3.38
20	1.51	1.47	0.03	1.30	1.05	0.23	5.54	1.30	3.28
Direction 2
	Type A			Type B			Type C
No.	Target luminance LT [cd/m2]	Background luminance LB [cd/m2]	Contrast	Target luminance LT [cd/m2]	Background luminance LB [cd/m2]	Contrast	Target luminance LT [cd/m2]	Background luminance LB [cd/m2]	Contrast
1	0.12	0.25	−0.52	2.04	1.10	0.86	2.43	0.70	2.48
2	0.19	0.45	−0.58	0.32	1.83	−0.82	2.34	0.83	1.82
3	0.12	0.44	−0.74	0.57	1.02	−0.44	3.00	1.02	1.94
4	0.45	0.36	0.26	6.01	2.10	1.86	3.00	0.90	2.33
5	0.54	0.42	0.28	4.44	1.90	1.34	4.35	0.91	3.78
6	0.45	0.55	−0.19	1.21	1.10	0.10	3.29	1.01	2.25
7	0.21	0.61	−0.65	0.62	1.71	−0.64	5.43	0.91	4.97
8	0.62	0.65	−0.05	1.04	1.11	−0.06	4.29	1.11	2.87
9	0.22	0.74	−0.70	1.53	2.10	−0.27	4.62	1.10	3.20
10	0.32	0.80	−0.60	2.27	0.60	2.79	4.56	1.06	3.30
11	0.74	0.94	−0.21	1.03	0.50	1.05	5.01	0.95	4.28
12	0.77	1.21	−0.36	0.56	1.72	−0.68	4.85	1.72	1.82
13	0.64	1.12	−0.43	0.47	1.42	−0.67	5.63	1.42	2.97
14	1.02	1.41	−0.28	2.02	0.60	2.37	3.51	0.60	4.85
15	1.00	1.53	−0.35	0.38	1.12	−0.66	3.00	1.12	1.68
16	0.88	1.43	−0.38	0.32	1.32	−0.76	4.48	1.32	2.39
17	0.15	1.80	−0.92	0.52	1.32	−0.61	3.80	1.32	1.88
18	0.19	2.24	−0.91	1.53	0.90	0.70	5.43	0.90	5.04
19	2.02	3.03	−0.33	1.37	0.80	0.71	5.67	0.80	6.09
20	0.55	1.61	−0.62	0.50	1.61	−0.69	5.03	1.61	2.13

presents a combined summary of the results of luminance measurements and luminance contrast calculations for the three types of pedestrian crossings studied. Results by direction of measurement and lighting solution are included.

Comparing the obtained results of luminance measurements and contrast calculations for the three types of pedestrian crossings studied together for the two directions of observation (

Table 2. Summary of the obtained results of luminance measurements and contrast calculations for the three types of pedestrian crossings studied, for a total of two directions of observation.

	Type A			Type B			Type C
	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast	Target Luminance LT [cd/m2]	Background Luminance LB [cd/m2]	Contrast
min	0.11	0.23	−0.92	0.27	0.50	−0.82	2.30	0.56	1.68
max	2.02	3.07	0.28	6.01	2.77	2.79	5.80	1.72	6.09
average	0.60	1.09	−0.39	1.39	1.37	0.14	4.15	1.02	3.16
median	0.49	0.91	−0.36	1.03	1.33	−0.16	4.37	0.95	2.96
quart 1	0.34	0.57	−0.62	0.56	1.03	−0.60	3.31	0.91	2.51
quart 2	0.75	1.48	−0.22	2.02	1.66	0.73	4.98	1.14	3.61

), significant differences were shown between the occurring levels of object luminance (pedestrian silhouettes). The median value of L_T object luminance for each type of solution differs significantly (Figure 18).

Among the studied solutions, the Type C solution stands out, on which luminance levels were obtained more than 5 times higher than for the Type A solution. The levels occurring for the B-type solution doubled in comparison with the A-type solution but did not match the values obtained for the C-type solution.

In the case of the L_B background luminance parameter (on which the pedestrian is observed), the values obtained for the Type C solution (Figure 19) are noteworthy. Their small scatter and relatively low value are due to the ability to precisely direct the light beam in the crossing area without illuminating the surrounding area. The Type B solution illuminates a larger area, which is the background of the pedestrian crossing, increasing background luminance (Table 1). Type A illumination represents cases where the background luminance results from the lighting classes implemented on the roadway. The range of luminance obtained corresponds to the typical range of M luminance classes [19].

Juxtaposing the results of the calculated luminance contrast values for the researched three types of pedestrian crossings together for the two directions of observation (Figure 20) showed significant differences between the occurring contrast levels of objects (pedestrian silhouettes) with the background. The median contrast value for the Type C solution stands out among the other solutions (Figure 20). A high positive contrast value was obtained (Table 2). Among the tested solutions, only solution Type C guarantees the production of high positive contrast values. In the case of solution Type B, unambiguous contrast values were not obtained for both directions of observation. There is a negative contrast for one direction of observation and a positive contrast for the other, and this is due to the geometry of the luminaire suspension concerning the axis of a pedestrian crossing. As expected, the Type B solution realizes mainly negative contrast.

For the surveyed locations, information was obtained on the lighting fixtures used around pedestrian crossings. In Type A, luminaires with an average power of 96 W were used, Type B lighting uses luminaires with an average power of 136 W, while for Type C solutions the average power of the luminaires used is 73 W. To compare the solutions, an energy index was proposed relating what the obtained value of luminance contrast, described by Equation (4). Due to the need to standardize and compare the results for different lighting solutions, it was decided to use the absolute value of contrast.

$$E_{Ci avg}=\frac{P_{avg}}{\left|C_{avg}\right|}$$

(4)

where:

E_{Ci avg}—average Energy Contrast Index [W],

P_avg—average wattage of the luminaire [W],

$\left|C_{avg}\right|$—absolute average value of contrast.

The Type A solution achieved an E_{Ci avg} = 232.1 W, the Type B solution E_{Ci avg} = 177.8 W, and the Type C solution E_{Ci avg} = 23.1 W. These preliminary results indicate that achieving luminance contrast within the pedestrian crossing is the most energy-efficient with Type C lighting. That is, Type C requires the least amount of electricity to achieve unit contrast compared to Type A and B solutions.

6. Discussion

The obtained high value of luminance contrast of the pedestrian silhouette with the background is one of the main factors determining the creation of conditions for the driver to observe the pedestrian silhouette at the pedestrian crossing. As a result of field tests, the values of luminance contrast of the pedestrian silhouette with the background were determined for 60 urban pedestrian crossings where three types of lighting solutions A, B and C are used.

The use of the solution of lighting the pedestrian crossing with street lighting (Type A) is the most common. The study confirmed the achievement of a negative contrast with a low value (C_average = −0.39) for most of the studied locations. Individual contrast values depend on the location of the pedestrian crossing axis relating to the luminaire and the realized lighting class on the roadway. Ambiguous results were obtained for the Type B solution. The obtained contrast value varies depending on the location of the luminaires about the geometry and axis of the pedestrian crossing. This solution generally deepens the contrast (negative for one direction and positive for the other direction), but it is not clearly defined. The technical implementation of this solution requires the use of two luminaires on a single pole, but their total power can be much higher than in the case of a Type C solution [57,58,59,60]. In addition, as shown (in Appendix B), the area outside the pedestrian crossing is illuminated, causing uncontrolled light littering of the road surroundings. The use of this solution is therefore not beneficial for several reasons.

Particularly noteworthy is the lighting of pedestrian crossings realized with dedicated luminaires (Type C). Studies have confirmed that the use of supplemental lighting (Type C) at pedestrian crossings is beneficial from the point of view of drivers, providing them with higher levels of positive contrast (C_average > 3.16) necessary to recognize the silhouette of a pedestrian wearing clothing with a low (ρ = 0.2) reflectance.

The luminance measurement method proposal presented in the article makes it possible to determine the contrast value of the pedestrian silhouette with the background at the pedestrian crossing and in the waiting zone. This is the author’s proposal, which can be implemented universally and find practical application. The presented test results for three types of technical solutions, however, should not be treated unequivocally, generalizing the results for other crossings. The presented test results refer to the tested pedestrian crossings and one should be aware of the limitations that occur. The luminance parameters depend on the season, weather conditions, visibility limitations, the organization of the crossing environment and the objects present in the surroundings, including light-emitting objects and light obstructions. Therefore, each pedestrian crossing should be treated individually. Each of the studied crossings is illuminated was different road lighting fixtures, with different power and luminous flux, luminous body, in different degrees of operation, and has different ambient lighting conditions, especially in urban conditions.

This article assumes a constant value of the reflectance of the material in which the pedestrian is dressed, valid for the above comparative analysis. Note that for materials with different reflectance, the obtained contrast value will be different. The value of the obtained luminance contrast is directly dependent on the reflectance of the clothing in which the pedestrian is dressed and is essentially unpredictable for the actual traffic situation.

The proposed method of measuring luminance takes into account the existence of the driver and the associated geometry of the measurement. The measurements assume the existence of the observer (driver) but ignore the existence of the vehicle’s lights, which can change the contrast value. This issue requires further research. The issue presented above does not exhaust the entire spectrum of problems related to lighting installed at pedestrian crossings. The research presented in this article will continue on a larger number of pedestrian crossings characterized by greater variability of technical characteristics, including geometric, lighting and energetic characteristics. Traffic engineering parameters will be taken into account, including the criterion of speed and traffic conflicts, which will be used to assess the behaviour of traffic participants at pedestrian crossings illuminated by different technical solutions.

7. Conclusions

When designing pedestrian crossing lighting, it is necessary to take into account several legal and formal requirements arising from general and local regulations. Despite the existence of normative documents [19], guidelines relating to the issue of road lighting [31,32] and pedestrian crossings [30] have been implemented in Poland.

In the design process of pedestrian crossing lighting, both illuminance and luminance criteria should be considered. To produce the required luminance contrast value on the pedestrian silhouette (object), it is necessary to take into account the background luminance value. Knowledge of the background luminance value of the pedestrian crossing can influence the correct selection of the value of the actual illuminance level and the applied beam distribution of the used luminaire. The luminance of the background generally results from the lighting class that is implemented on the roadway, behind and in front of the pedestrian crossing and its surroundings. Therefore, it is important to take into account the level of street lighting when designing pedestrian crossing lighting, which in most cases determines the background luminance at which a pedestrian is perceived by a driver. It would be most beneficial to take luminance measurements (luminance photos) of the roadway (over a distance of about 60 m) before and after the designed pedestrian crossing. According to the study, producing a positive contrast value C > 2.5 is possible for Type C solutions used in the city.

Type C solutions use LED luminaires with specific light distribution. The luminaires are available in a wide range of wattages, luminous flux and types of distribution and are characterized by high luminous efficiency and durability. This allows for the selection of luminaires for specific lighting needs. In addition, the possibility of using additional luminous flux control systems allows individual adjustment of the level of emitted luminous flux at the stage of luminaire installation. By controlling the luminance level and luminance contrast, the lighting effects assumed at the design stage can be confirmed. In this way, the luminous flux and power of luminaires can be optimized. Taking into account the criteria of contrast efficiency about the power of installed luminaires, the dedicated system (Type C) has the highest energy efficiency while providing the best level of pedestrian observation and ensuring control of light beam distribution.