Verification of the Applicability of Obstacle Recognition Distance as a Measure of Effectiveness of Road Lighting on Rainy and Foggy Roads

Abstract

Adverse weather conditions at night are very fatal to drivers, causing serious traffic accidents. Road lighting is a facility that can alleviate these dangerous situations. Nevertheless, road lighting has only rarely been studied during adverse weather. The reason is that the current road lighting performance evaluation method is presented based on normal weather. The current road lighting performance evaluation method uses a luminance meter to measure the road surface, which is not suitable due to scattering during adverse weather such as rain and fog. Therefore, this study proposes obstacle recognition distance as a measure of effectiveness to evaluate the performance of road lighting during adverse weather. There is a lack of actual research on whether obstacle recognition distance can be used as a measure of effectiveness for road lighting during adverse weather. Therefore, in this study, 30 subjects were used to measure the subjects’ obstacle recognition distance according to changes in weather conditions, road lighting grade, and road lighting color temperature. As a result, it was analyzed that there was a clear trend of change in obstacle recognition distance depending on the change in each condition. It was found that, under the same road lighting performance conditions, there was a difference of up to 72.86% by weather condition; under the same weather conditions, there was a difference of up to 22.75% by road lighting grade; and by color temperature, there was a difference of up to 21.87%. In addition, a statistical significance test was performed to support the existence of a difference, and the results were synthesized to suggest that obstacle recognition distance can be used as a performance measure of effectiveness of road lighting in adverse weather.

1. Introduction

Adverse weather conditions such as rain or fog pose various visual restrictions to drivers in maintaining a safe driving environment. In night situations, the risk is greater, the driver’s workload increases, and the severity of traffic accidents is greater than on a clear day. In adverse weather at night, road lighting is almost the only means of providing direct visibility of the road surface to drivers, except for car headlights. Therefore, drivers rely on road lighting in adverse weather at night to determine the condition of the road surface and take driving measures according to the situation. However, all road lighting is developed and evaluated based on normal weather conditions. In fact, there is almost no development of road lighting to ensure visibility in adverse weather situations, and there is no related research. One of the reasons may be that there is no way to evaluate road lighting during adverse weather.

The International Commission on Illumination (CIE) measures the road surface luminance, overall/longitudinal uniformity, threshold increment on dry roads, and overall uniformity of wet road surfaces [1] and classifies road lighting classes according to each value to judge the performance of road lighting [2]. However, there is no official method to evaluate road lighting performance in foggy and rainy situations. The reason is that, under normal clear weather conditions, the luminance of the road surface illuminated by road lighting is measured through a luminance meter. However, it is believed that during adverse weather such as fog or rain, atmospheric scattering occurs due to fog or rain particles in the atmosphere, making it impossible to measure the luminance value of the road surface [3]. In reality, the road environment that is fatal to drivers is adverse weather, so although road lighting performance evaluation measures should be actively carried out in adverse weather at night, little related research is being carried out due to the difficulty of implementing adverse weather environments. There are cases of experiments using obstacle recognition distance as a measure of effectiveness to evaluate road lighting during some adverse weather, and CIE 115:2010 also specifies that the purpose of road lighting is to make obstacles on the road visible [2]. Considering this, this study proposed obstacle recognition distance as a measure of effectiveness for evaluating road lighting during rain and fog. In order to verify whether obstacle recognition distance is appropriate as a measure of effectiveness for road lighting performance evaluation in adverse weather, an empirical experiment using subjects in various environments was conducted and comparative results were presented.

2. Literature Review

Because drivers obtain 90% of road information through vision [4], securing driver visibility is important for safe driving. However, adverse weather such as rain and fog have a serious effect on driver visibility [5].

During rain, photons of light that hit water droplets may be absorbed or scattered. The scattered light forms a luminous veil, reducing the luminance contrast visible through the rain. As a result, the direct effect of light in the atmosphere reduces visibility [6], and reduced visibility has a serious effect on driver safety [7]. The reason why visibility is impaired during fog is because the light from road lighting is absorbed and scattered by the fog in the atmosphere. Because of this, some of the light is lost or reflected back towards road users in the form of a veil of fog that obscures details of the road ahead [8]. This decrease in visibility due to fog results in a decrease in the driver’s visual and operating abilities, threatening safe driving [9,10].

According to Korea’s 2022 traffic accident statistics, the fatality rate from traffic accidents on a clear day is 1.3 people/100 cases, the fatality rate during rain is 2.0 people/100 cases, and the fatality rate during fog is 6.3 people/100 cases [11].

The fatality rate during rain was found to be 53.8% higher than on a clear day, and during fog, the fatality rate was 384.6% higher than on a clear day. In the case of traffic accident fatality rate by time zone, the fatality rate from 06:00 to 18:00, corresponding to daytime, is 1.3 people/100 cases, and the fatality rate from 18:00 to 06:00, corresponding to nighttime, is 2.4 people/100 cases. It was analyzed that the traffic accident fatality rate at night was 85.8% higher than during the daytime [11].

In the United States, the National Highway Traffic Safety Administration (NHTSA) provides normal and rain fatality rate statistics between the lighted section and the non-lighted section. According to 2020 statistics, the fatality rate was 0.52/100 for the section with lights in rain while 1.01/100 for the section without lights, indicating that the fatality rate of the non-lighted section in rainfall is 48.6% higher than that of the lighted section [12].

In other words, the severity of traffic accidents is very high at night and in adverse weather, and road lighting is considered a measure that can reduce the fatality rate of traffic accidents during adverse weather at night. Road lighting has great potential to improve traffic safety and, most importantly, reduce the severity of road accidents [13]. However, most studies related to road lighting are conducted under normal weather conditions.

Wood et al. [14] investigated the effect of adjusting the level of road lighting on night driving performance. The level of road lighting was adjusted to four levels using 14 subjects, and the recognition distance and reaction time for moving targets and walking pedestrians were investigated. As a result, it was verified that the brighter the road lighting, the longer the recognition distance to the target and the shorter the reaction time.

Davidovic et al. [15] investigated the preference between 3000 K and 4000 K road lighting from the driver’s perspective through a survey, and, as a result, 3000 K road lighting was evaluated as preferable.

Cenani et al. [16] investigated the interaction between road lighting and car headlights in terms of target detection distance and presented results showing that target recognition distance changes according to changes in road lighting intensity.

Jin et al. [17] conducted a dark adaptation study using LEDs with different color temperatures in the process of finding the performance of designated road lighting, and the results suggested that the dark adaptation time increases as the color temperature increases.

In this way, several studies have verified that conditions such as the intensity of road lighting or color temperature can change the driver’s visibility, but there is a limitation in that the experiment was conducted only under normal conditions.

Even though road lighting is the only solution in adverse weather conditions at night, in fact, little research has been conducted on road lighting visibility in lighting environments such as rain and fog [18]. Most studies are conducted on the relationship between adverse weather and traffic accidents based on past records. The reason may be economic considerations, and it is presumed that there is a lack of related research because there are few facilities that can implement real-scale rain and fog.

In fact, the fact that there is currently no official method to evaluate road lighting during adverse weather may be seen as a further reason for research and development on road lighting during adverse weather. CIE proposes a method for measuring and calculating the luminance of road surfaces using a luminance meter in CIE 114-2000 [1] and sets standards for classifying road lighting grades based on the luminance of road surfaces measured in CIE 115-2010 [2]. However, this standard is based on normal conditions and is not suitable for adverse weather. The method presented by CIE suggests that the higher the road surface luminance, the higher the class of high-quality road lighting, but this measurement method is not suitable when applied in adverse weather. To analyze the effect of weather on road luminance, there is a previous study that measured road surface luminance under various weather conditions [19]. As a result, it was confirmed that road surface luminance increased significantly on wet, snowy, and foggy roads compared to dry roads. Even though the luminance of the road surface increased compared to the dry road surface, it is difficult to say that the situation is one in which the driver has better visibility than the dry road surface situation. In other words, the existing evaluation method may be judged to be unsuitable in adverse weather. The reason why it is difficult to evaluate road lighting performance in the same way as under normal conditions during adverse weather is because the level of road surface luminance varies depending on the measurement point due to diffuse reflection of the road surface in adverse weather conditions, and it is impossible to extract accurate road surface luminance values due to light scattering in the atmosphere.

CIE 115-2010 states ‘The purpose of road lighting is to provide revealing obstacles’ and ‘there is a need to reveal extraneous objects that suddenly appear on the road’; it is expressed that good quality road lighting provides an appropriate distance to recognize obstacles [2].

In fact, there are research cases that have identified through subject experiments that obstacle recognition distance increases as lighting performance improves [20], and there are studies that have identified through empirical experiments that obstacle recognition distance can change by applying various road lighting during adverse weather [9]. Additionally, there are empirical research results showing that the higher the color temperature of road lighting during fog, the higher the obstacle visibility [21].

Therefore, this study proposed obstacle recognition distance as a measure of effectiveness of road lighting performance evaluation in adverse weather that can replace the existing road surface luminance measurement method, and empirical experiments were conducted to determine whether obstacle recognition distance shows effective differences according to changes in road lighting conditions during adverse weather.

3. Experimental Environment and Method

3.1. Experimental Location

To create an environment similar to an actual road, an experimental environment was created at the Yeoncheon SOC Research Center of the Korea Institute of Civil Engineering and Building Technology. The 200 m long tunnel located within the center allows experiments to be conducted under controlled conditions that are not affected by external illumination, including moonlight and building lighting (Figure 1).

The test site is equipped with rain and fog reproduction equipment in a 200 m section as shown in Figure 2 and Figure 3, so the desired rainfall and fog intensity can be realized. In this study, the rainfall intensity was implemented at 30 mm/h, and the fog was implemented at a visibility distance of 80 m.

In addition, movable pole lights are installed on both sides of the road as shown in Figure 4, and the luminaries can be freely replaced. This lighting system can adjust the height of the luminaire, the spacing between lights, and the overhang and inclination angle of the luminaire through remote control.

3.2. Experimental Road Lighting

In general, road lighting installed on roads is operated with a single brightness or color temperature, so it is difficult to test obstacle recognition distances by road lighting classes and color temperature under the same installation conditions. Therefore, in this study, to confirm the difference in obstacle recognition distance by road lighting classes and color temperature, road lighting with brightness adjustable in 255 levels and color temperature of 4940 K and 3006 K was manufactured. The road lighting can be controlled wirelessly (BLE communication) through a dedicated program. The manufactured road lighting was installed at a height of 9 m, installation angle of 0°, and overhang 3.5 m, and 5 lights were installed at 30 m intervals. Road surface luminance measurement was performed using Techno Team’s LMK6 luminance meter and CIE measurement method. LMK6 is an image-type luminance meter that is capable of analyzing the luminance of the location and area desired by the user in the measured image. By adjusting the brightness of the lighting, the dimming values corresponding to M1, M3, and M5 were set to a level where there was almost no difference in luminance between 4940 K and 3006 K, as shown in

Table 1. Lighting classes and dimming value of experimental road lighting.

Color Temperature	Lighting Classes	Dimming Value	Road Surface
			Dry			Wet
			Average Road Surface Luminance	Overall Uniformity	Longitudinal Uniformity	Overall Uniformity
4940 K	M1	228	2.020	0.764	0.703	0.284
	M3	160	1.000	0.776	0.713	0.289
	M5	5	0.568	0.784	0.721	0.302
3006 K	M1	230	2.010	0.840	0.791	0.275
	M3	161	1.000	0.838	0.791	0.358
	M5	5	0.588	0.842	0.781	0.378

.

3.3. Subject

The experiment consisted of 30 subjects who held a driver’s license. The average age was 41.9 years, and the gender and age groups are as shown in

Table 2. Composition of subjects.

Category	Classification	Number of People	Rate
Gender	Male	25	83.3%
	Female	5	16.7%
Age range	Over 60 years old	1	3.3%
	50~59 years old	12	40.0%
	40~49 years old	4	13.3%
	30~39 years old	6	20.0%
	20~29 years old	7	23.3%

.

3.4. Obstacle

The obstacle is a target that the subjects must check in this experiment, and a black cube measuring 15 cm× 15 cm × 15 cm was used, which was the same size as the target object used in the development of the stopping sight distance. CALTRANS (California Department of Transportation), ODOT (Ohio Department of Transportation), and WSDOT (Washington State Department of Transportation) retained a 15 cm object height for stopping sight distance [22]. As shown in Figure 5, the obstacle was located between the third and fourth lights and at the center of the road.

3.5. Test Procedure

The experiment was conducted in the following order. Step (1) The subject is located on the center line in the same lane at a distance of 100 m from the obstacle. Step (2) Set weather and lighting conditions. Step (3) The subject walks in the direction of the obstacle. Step (4) The subject stops when the obstacle is recognized, checks the distance between the obstacle and the subject by looking at the distance table installed at the edge of the lane, and records the obstacle recognition distance on the sheet. Step 1 to Step 4 were repeated in all conditions, and 30 subjects performed obstacle recognition distance measurements for all experimental conditions.

4. Result

The results of the empirical experiment on obstacle recognition distance according to weather conditions, road lighting grade, and lighting color temperature changes were derived as shown in Figure 6. Based on the derived experimental result data, it was verified whether each experimental condition showed a significant difference in obstacle recognition distance.

4.1. Obstacle Recognition Distance by Road Lighting Classes

One-way ANOVA was conducted to verify whether there was a significant difference in the subjects’ recognition distance depending on the road lighting grade under the same weather conditions and the same color temperature, and the results are shown in

Table 3. Significance of difference in obstacle recognition distance by road lighting classes.

Dependent Variable	Explanatory Variable		n	Mean	SD	F	p	Post Hoc (Tukey HSD)
Obstacle recognition distance (4940 K)	Normal	M1 (a)	60	63.07	16.61	10.966	<0.001	b,c < a
		M3 (b)	60	55.89	15.16
		M5 (c)	60	50.29	12.97
	Fog	M1 (a)	60	19.29	2.95	35.412	<0.001	b,c < a
		M3 (b)	60	15.17	3.61
		M5 (c)	60	15.07	2.79
	Rain	M1 (a)	60	34.53	7.03	9.126	<0.001	b,c < a
		M3 (b)	60	31.40	5.34
		M5 (c)	60	30.07	5.04
Obstacle recognition Distance (3006 K)	Normal	M1 (a)	60	63.07	16.61	10.966	<0.001	b,c < a
		M3 (b)	60	55.89	15.16
		M5 (c)	60	50.29	12.97
	Fog	M1 (a)	60	19.29	2.95	35.412	<0.001	b,c < a
		M3 (b)	60	15.17	3.61
		M5 (c)	60	15.07	2.79
	Rain	M1 (a)	60	34.53	7.03	9.126	<0.001	b,c < a
		M3 (b)	60	31.40	5.34
		M5 (c)	60	30.07	5.04

.

As a result, it was analyzed that the difference in obstacle recognition distance by road lighting classes in 4940 K was statistically significant in normal (F = 10.966, p < 0.001), fog (F = 35.412, p < 0.001), and rain (F = 9.126, p < 0.001) conditions. Similarly, in the lighting of 3006 K, the difference in obstacle recognition distance by road lighting classes showed significant differences in normal (F = 13.303, p < 0.001), fog (F = 10.765, p < 0.001), and rain (F = 13.468, p < 0.001) conditions.

Additionally, Tukey HSD post hoc analysis, which is used when the sample size is equal, was performed. As a result, under the condition of 3006 K lighting in a fog, the obstacle recognition distance of M1 and M3 was found to be longer than that of M5, and in all conditions except this, the obstacle recognition distance of M1 was found to be longer than that of M3 and M5. In order to check the change in obstacle recognition distance according to road lighting grade, when the obstacle recognition distance of class M1 was set to 100% in each weather condition, the ratio in M3 and M5 situations is shown in

Table 4. Obstacle recognition distance ratio compared to M1.

Classification		Obstacle Recognition Distance Ratio Compared to M1
Color Temperature		4940 K			3006 K
Weather		Normal	Fog	Rain	Normal	Fog	Rain
Lighting classes	M1	100.00%	100.00%	100.00%	100.00%	100.00%	100.00%
	M3	88.62%	78.64%	90.94%	85.80%	93.94%	89.66%
	M5	79.74%	78.12%	87.08%	77.25%	85.22%	84.40%

.

Through the analysis results, it was verified that there was a significant difference in obstacle recognition distance by road lighting classes in all weather conditions, and it was confirmed that the obstacle recognition distance became longer as the road lighting class increased. Therefore, it was verified that obstacle recognition distance evaluation is necessary for each road lighting class during adverse weather.

4.2. Obstacle Recognition Distance by Road Lighting Classes

One-way ANOVA was conducted to verify whether there was a significant difference in the obstacle recognition distance according to changes in weather conditions under the same road lighting class and same color temperature, and the results are shown in

Table 5. Significance of difference in obstacle recognition distance by weather.

Dependent Variable		Explanatory Variable	n	Mean	SD	F	p	Post Hoc (Tukey HSD)
Obstacle recognition distance (4940 K)	M1	Normal (a)	60	63.07	16.61	266.193	<0.001	b < c < a
		Fog (b)	60	19.29	2.95
		Rain (c)	60	34.53	7.03
	M3	Normal (a)	60	55.89	15.16	278.712	<0.001	b < c < a
		Fog (b)	60	15.17	3.61
		Rain (c)	60	31.4	5.34
	M5	Normal (a)	60	50.29	12.97	279.312	<0.001	b < c < a
		Fog (b)	60	15.07	2.79
		Rain (c)	60	30.07	5.04
Obstacle recognition distance (3006 K)	M1	Normal (a)	60	63.08	16.37	250.252	<0.001	b < c < a
		Fog (b)	60	20.63	3.39
		Rain (c)	60	37.7	6.98
	M3	Normal (a)	60	54.12	15.08	199.725	<0.001	b < c < a
		Fog (b)	60	19.38	3.36
		Rain (c)	60	33.8	5.99
	M5	Normal (a)	60	48.73	14.69	163.672	<0.001	b < c < a
		Fog (b)	60	17.58	4.09
		Rain (c)	60	31.82	5.93

.

As a result, 4940 K lighting showed significant differences in obstacle recognition distances by weather conditions in M1 (F = 266.193, p < 0.001), M3 (F = 278.712, p < 0.001), and M5 (F = 279.312, p < 0.001). Similarly, in 3006 K lighting, it was analyzed that the obstacle recognition distance difference by weather condition showed significant differences in M1 (F = 250.252, p < 0.001), M3 (F = 199.725, p < 0.001), and M5 (F = 163.672, p < 0.001). Additionally, Tukey HSD post-analysis was performed to be used when the number of samples was the same. As a result, under all conditions, the obstacle recognition distance was found to be long in the order of normal > rainfall > fog. In order to check the change in obstacle recognition distance according to changes in weather conditions, when the obstacle recognition distance under normal conditions is set to 100% for each road lighting class, the ratio in rain and fog situations is shown in

Table 6. Obstacle recognition distance ratio compared to normal.

Classification		Obstacle Recognition Distance Ratio Compared to Normal
Color Temperature		4940 K			3006 K
Weather		Normal	Fog	Rain	Normal	Fog	Rain
Lighting classes	M1	100.00%	30.59%	54.75%	100.00%	32.70%	59.77%
	M3	100.00%	27.14%	56.18%	100.00%	35.81%	62.45%
	M5	100.00%	29.97%	59.79%	100.00%	36.08%	65.30%

.

Through the analysis results, it was verified that there was a significant difference in the obstacle recognition distance by weather condition in all road lighting conditions, and it was confirmed that the obstacle recognition distance became longer in the order of normal > rain > fog. Therefore, since there is a difference in obstacle recognition distance depending on weather conditions under the same road lighting conditions, it was verified that obstacle recognition distance can be used as a measure of effectiveness for road lighting according to weather conditions.

4.3. Obstacle Recognition Distance by Color Temperature of Road Lighting

A paired-samples t-test was conducted to verify whether there was a significant difference in the obstacle recognition distance by color temperature of road lighting in the same weather and same road lighting class. The results are shown in

Table 7. Significance of difference in obstacle recognition distance by color temperature.

Weather	Lighting Classes	Color Temperature	n	Mean	SD	t	p
Normal	M1	4940 K	60	63.07	16.61	−0.008	0.993
		3006 K	60	63.08	16.37
	M3	4940 K	60	55.89	15.16	1.240	0.220
		3006 K	60	54.12	15.08
	M5	4940 K	60	50.29	12.97	1.546	0.128
		3006 K	60	48.73	14.69
Fog	M1	4940 K	60	19.29	2.95	−3.826	<0.001
		3006 K	60	20.63	3.39
	M3	4940 K	60	15.17	3.61	−12.360	<0.001
		3006 K	60	19.38	3.36
	M5	4940 K	60	15.07	2.79	−6.751	<0.001
		3006 K	60	17.58	4.09
Rain	M1	4940 K	60	34.53	7.03	−6.885	<0.001
		3006 K	60	37.70	6.98
	M3	4940 K	60	31.40	5.34	−5.867	<0.001
		3006 K	60	33.80	5.99
	M5	4940 K	60	30.07	5.04	−3.943	<0.001
		3006 K	60	31.82	5.93

.

As a result, it was found that there was no difference in obstacle recognition distance between 4940 K lighting (white light) and 3006 K lighting (yellow light) under normal conditions. On the other hand, in the fog, there was a significant difference in M1 (t = −3.826, p < 0.001), M3 (t = −12.360, p < 0.001), and M5 (t = −6.75, p < 0.001). Likewise, in the rain, there was a significant difference in M1 (t= −6.885, p < 0.001), M3 (t = −5.867, p < 0.001), and M5 (t = −3.943, p < 0.001). For comparison of obstacle recognition distances by color temperature in groups with significant differences in obstacle recognition distances, when the 3006 K is set to 100%, the ratio of 4940 K is as shown in

Table 8. Obstacle recognition distance ratio compared to 3006 K.

Classification		Obstacle Recognition Distance Ratio Compared to 3006 K
Weather		Fog		Rain
Color Temperature		3006 K	4940 K	3006 K	4940 K
Lighting classes	M1	100.0%	93.5%	100.0%	91.6%
	M3	100.0%	78.3%	100.0%	92.9%
	M5	100.0%	85.7%	100.0%	94.5%

.

As a result of the analysis, the difference in the color temperature of road lighting is not significant under normal weather conditions, but in rain and fog situations there is a significant difference in obstacle recognition distance between 3006 K (yellow light) and 4940 K (white light), and it was found that the obstacle recognition distance was longer than that of 4940 K lighting. Therefore, since there is a difference in obstacle recognition distance according to color temperature in rain and fog situations, it was verified that evaluation by color temperature in rain and fog is necessary using obstacle recognition distance as a measure of effectiveness.

5. Conclusions

The severity of traffic accidents is high in bad weather situations such as rain and fog, but there is no standard for evaluating road lighting in bad weather, and there is little research related to road lighting that can reduce this. The existing road lighting performance evaluation method is based on normal standards measures of road surface luminance through a luminance meter, which is a method that cannot be used in bad weather. Therefore, in this study, obstacle recognition distance was proposed as a way to evaluate the performance of road lighting in bad weather, and the following results were derived through an empirical experiment using subjects to determine whether obstacle recognition distance is a valid measure of effectiveness in adverse weather.

• Under the same road lighting conditions, there is a difference in obstacle recognition distance according to weather conditions, and the obstacle recognition distance is longer in the order of normal > rain > fog; the difference is found to occur up to 72.86%. Through these results, it was verified that performance evaluation of road lighting according to weather conditions is necessary.
• Under the same weather conditions, there is a difference in obstacle recognition distance by road lighting grade, and the obstacle recognition distance is longer in the order of M1 > M3 > M5, so it was found that road lighting with a higher road lighting grade is more advantageous in securing visibility in rain and fog. The difference was found to occur up to 22.75%.
• Under normal weather conditions, the difference in obstacle recognition distance by color temperature is not significant when the road lighting level is the same, but in rain and fog situations, 3006 K (yellow light) was found to have a longer obstacle recognition distance than 4940 K (white light), and the difference was up to 21.87%.

As this study was an experiment conducted in a relatively limited environment, it may be difficult to conclude that the resulting obstacle recognition distance values for each experimental condition are absolute values. However, since the tendency of the experimental results appears to be very clear, it is judged that it has been sufficiently verified that obstacle recognition distance can be used as measure of effectiveness of road lighting performance evaluation in adverse weather. In the future, it is expected that more specific experimental results will be derived if the test subjects’ conditions are subdivided by age or gender and the rain and fog conditions are further subdivided. In addition, although this study conducted experiments on foot to confirm feasibility, it is expected that results more similar to reality will be possible if experiments are conducted using actual vehicles. When driving in an actual vehicle, it is assumed that the distance at which the driver perceives obstacles will be more similar to the actual phenomenon due to the speed of the vehicle, and the difference in recognition distance may become clearer depending on weather conditions and road lighting conditions. Driving in rain and fog at night causes a lot of workload and can lead to traffic accidents. In the long term, in order to reduce traffic accidents through road lighting during rain and fog at night, measures of effectiveness applicable to rain and fog situation are needed. Based on the results of this study, we hope to improve the method of deriving obstacle recognition distances under various and more specific experimental conditions in the future and to actively conduct performance evaluation of road lighting during adverse weather.