U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

Report
This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-HRT-04-144
Date: December 2005

Enhanced Night Visibility Series, Volume XIII: Phase III—Study 1: Comparison of Near Infrared, Far Infrared, High Intensity Discharge, and Halogen Headlamps on Object Detection in Nighttime Clear Weather

PDF Version (1.01 MB)

PDF files can be viewed with the Acrobat® Reader®

CHAPTER 3—RESULTS

Results included in this report are based on statistically significant effects at an alpha symbol = 0.05 level except where otherwise stated. In main effect graphs, means with the same letter are not significantly different based on the SNK post hoc test. Bars above and below the means indicate standard error.

OBJECTIVE MEASURES

Table 7 shows the results for the 6 (VES) by 3 (Object) by 3 (Age) mixed factorial design ANOVA (i.e., object group ANOVA), which was conducted on the objective measures of detection distance for the three groups of objects (i.e., obstacles, pedestrians, and retroreflective). Significant age, VES, and object main effects were found along with a significant VES by Object Group interaction.

Table 8 shows results for the object group ANOVA on recognition. Table 9 shows significant VES and object group main effects along with a significant VES by Object interaction.

Table 7. Object group ANOVA summary table for the dependent measurement: detection distance.
Source DF SS MS F value P value  
TOTAL 323 650512408.7       
Between
Age 2 5675797.6 2837898.8 4.64 0.027 *
Subject/Age 15 9175174.3 611678.3     
 
Within
VES 5 12651827.2 2530365.4 23.76 <0.0001 *
VES by Age 10 1028099.1 102809.9 0.97 0.4804  
VES by Subject/Age 75 7986752.0 106490.0     
 
Object Group 2 531867870.5 265933935.2 1038.61 <0.0001 *
Object Group by Age 4 2677855.3 669463.8 2.61 0.0549  
Object Group by Subject/Age 30 7681454.0 256048.5     
 
VES by Object Group 10 50053432.0 5005343.2 40.51 <0.0001 *
VES by Object Group by Age 20 3180129.6 159006.5 1.29 0.1960  
VES by Object Group by Subject/Age 150 18534017.1 123560.1     

 

Table 8. Object group ANOVA summary table for the dependent measurement: recognition distance.
Source DF SS MS F value P value  
TOTAL 323 442874305.8       
Between
Age 2 4516545.624 2258272.812 2.08 0.1591  
Subject/Age 15 16264060.3 1084270.7     
 
Within
VES 5 7017614.386 1403522.877 10.8 <0.0001 *
VES by Age 10 603581.26 60358.126 0.46 0.9077  
VES by Subject/Age 75 9746314 129950.9     
 
Object Group 2 331270458.6 165635229.3 277.43 <0.0001 *
Object Group by Age 4 3563344.7 890836.2 1.49 0.2295  
Object Group by Subject/Age 30 17911080.7 597036     
 
VES by Object Group 10 24746359.56 2474635.96 14.75 <0.0001 *
VES by Object Group by Age 20 2075484.13 103774.21 0.62 0.8943  
VES by Object Group by Subject/Age 150 25159462.5 167729.8     

 

Table 9. Summary of significant main effects and interactions for object group analysis.
Source Significant
Detection
Significant
Recognition
Between
Age x  
Subject/Age    
 
Within
VES x x
VES by Age    
VES by Subject/Age    
 
Object Group x x
Object Group by Age    
Object Group by Subject/Age    
 
VES by Object Group x x
VES by Object Group by Age    
VES by Object Group by Subject/Age    

Age

The main effect of age was significant (p < 0.05) for object detection. For all the VESs and all the object groups, the younger and middle-aged groups detected objects at greater distances than did the older group. Figure 30 shows the object detection means for the three different age groups.

Bar graph. Object detection means for the three age groups. Click here for more detail.
Figure 30. Bar graph. Object detection means for the three age groups.

The Age by Object Group interaction p value of 0.0549 for detection distance is not significant at p < 0.05, but it indicates that further consideration of the effect of age is warranted. For this reason, age will continue to be included in subsequent analysis and discussion.

Object Group and VES Main Effects

The main effects of object group and VES were statistically significant (p < 0.05) in this model for object detection and recognition measurements. An object group main effect is expected because of the dramatic detection and recognition differences between a retroreflective object, for example, and a small, low-contrast obstacle such as a tire tread. The main effect of VES is similarly limited for interpretation at this object group level because the effect is influenced by the number of different objects selected for investigation. This summary effect could generate an advantage or disadvantage for a given system based on the type of experimental objects tested. Further description of this can be found in the beginning of the Obstacle Object Group Analysis section later in this report.

VES by Object Group Interaction

The VES by Object Group interaction was found to be significant (p < 0.05) for both the distance at which participants detected objects with a VES and the distance at which participants recognized an object with a VES. This interaction indicates that some VESs performed better with one object group than with another object group (i.e., type of object). For example, detection of pedestrians as a group and obstacles (i.e., tire and dog) using the FIR vehicle occurred earlier (i.e., at a greater distance) than detection of pedestrians or obstacles with any of the other VESs; however, the mean detection distance for retroreflective objects with the FIR vehicle was later (i.e., shorter distance) than for any of the other VESs. Figure 31 indicates the mean detection distance for each of the VESs for each of the three object groups. A standard error bar is included at the top of each mean bar.

Bar graph. Mean detection values for each VES for each of the three object groups. Click here for more detail.
Figure 31. Bar graph. Mean detection values for each VES for each of the three object groups.

The NIR 1 configuration performed well detecting retroreflective objects and people; however, for detecting obstacles (i.e., dog and tire tread), it was surpassed by the better headlamp-based systems (i.e., HLB and HID 1). The HID 1 vehicle provided similar detection distances to the benchmark HLB vehicle for obstacles and retroreflective objects, but it had lower detection distances than the HLB vehicle for the pedestrian group overall. The HID 2 vehicle had the lowest mean detection distance of the VESs tested for pedestrians. For detecting obstacles, the HID 2 vehicle was surpassed by the other VESs, except for the NIR 2 and NIR 1 vehicles. For retroreflective objects, the HID 2 vehicle was surpassed by NIR 1, HLB, and HID 1 vehicles; however, the HID 2 vehicle had better mean detection distances for retroreflective objects than did the FIR vehicle.

Similar results were present when considering the mean recognition distances of the different VESs for the object types. Figure 32 indicates the mean recognition distances for each VES and object group.

Bar graph. Mean recognition distances for each VES for each of the three object groups. Click here for more detail.
Figure 32. Bar graph. Mean recognition distances for each VES for each of the three object groups.

The mean recognition distance for retroreflective objects with the FIR vehicle was shorter than that of all the other VESs except HID2; however, recognition of pedestrians with the FIR vehicle was the best of the VESs. When detecting obstacles, FIR had the longest mean detection distance of the VESs. In recognizing obstacles, that is, identifying the obstacle ahead, the FIR vehicle had results similar to the HLB and HID 1 vehicles. This same effect occurred with the NIR 1 vehicle in detecting and recognizing retroreflective objects. In detecting retroreflective objects, the NIR 1 had the longest mean distance; however, in recognizing the object as an RRPM, sign, or turn arrow, NIR 1 was similar to HLB and HID 1.

Obstacle Object Group Analysis

For the obstacle object group, which included the tire tread and dog, a 6 (VES) by 2 (Object) by 3 (Age) mixed factorial design was conducted. The results of this analysis are shown in table 10; they indicate significant main effects of age, VES, and object. A significant interaction between VES and object was also found.

Table 10. Obstacle group ANOVA summary table for the dependent measurement: detection distance.
Source DF SS MS F value P value  
TOTAL 214 2263772.9       
Between
Age 2 129037.4 64518.7 3.97 0.0414 *
Subject/Age 15 243975.1 16265.0     
 
Within
VES 5 270960.0 54192.0 7.95 <.0001 *
VES by Age 10 69654.8 6965.5 1.02 0.4337  
VES by Subject/Age 75 511294.5 6817.3     
 
Object 1 72816.6 72816.6 10.52 0.0055 *
Object by Age 2 28501.3 14250.7 2.06 0.1621  
Object by Subject/Age 15 103813.1 6920.9     
 
VES by Object 5 188522.9 37704.6 4.83 0.0007 *
VES by Object by Age 10 68043.9 6804.4 0.87 0.5624  
VES by Object by Subject/Age 74 577153.3 7799.4     

Table 11 indicates that for recognition distance of the two obstacles, there were statistically significant main effects of VES and object.

Table 11. Obstacle group ANOVA summary table for the dependent measurement: recognition distance.
Source DF SS MS F value P value  
TOTAL 214 1528904.2       
Between
Age 2 60231.55242 30115.77621 1.97 0.1734  
Subject/Age 15 228872.3114 15258.1541     
 
Within
VES 5 77110.49642 15422.09928 2.66 0.0288 *
VES by Age 10 31662.52511 3166.25251 0.55 0.8519  
VES by Subject/Age 75 434969.4358 5799.5925     
 
Object 1 121761.6172 121761.6172 35.45 <.0001 *
Object by Age 2 11960.9061 5980.4531 1.74 0.2089  
Object by Subject/Age 15 51515.4452 3434.363     
 
VES by Object 5 45491.44406 9098.28881 1.58 0.1762  
VES by Object by Age 10 39240.80808 3924.08081 0.68 0.7382  
VES by Object by Subject/Age 74 426087.6993 5757.9419     

Table 12 provides a summary of the statistically significant main effects and interactions in the analysis of the obstacle object group.

Table 12. Significant main effects and interactions summary for obstacle object group analysis.
Source Significant
Detection
Significant
Recognition
Between
Age x  
Subject/Age    
 
Within
VES x x
VES by Age    
VES by Subject/Age    
 
Object x x
Object by Age    
Object by Subject/Age    
 
VES by Object x  
VES by Object by Age    
VES by Object by Subject/Age    

Age

The main effect of age that was identified in the object group model discussed previously was evident in the analysis of the obstacles object group as well. For the tire and dog, the mean detection distances were 50.3 m and 48.5 m (165 ft and 159 ft) for the young and middle-aged groups. For the older age group, the mean detection distance was 33.5 m (110 ft). SNK analysis indicates the older group had shorter detection distances than the middle and younger groups (figure 33).

Bar graph. Mean detection distances for the age main effect for the obstacle group. Click here for more detail.
Figure 33. Bar graph. Mean detection distances for the age main effect for the obstacle group.

The main effect of object in this analysis describes differences in the detection distances of the dog versus the tire tread. The mean detection distance for the dog was 38.4 m (126 ft) and the mean detection distance for the tire tread was 50.0 m (164 ft). The main effect of VES is best understood by consideration of the VES by Object interaction, which is discussed next.

VES by Object Interaction

The ANOVA for the detection and recognition distances when approaching either the tire tread or the dog had a significant VES by Object interaction for detection distance. SNK tests performed on detection distances were used for the two obstacles to identify differences between the VESs for each object. In figure 34, SNK grouping for detection distance is shown with an uppercase letter. SNK grouping for recognition distance is shown with a lowercase letter. Bar height indicates the mean detection or recognition distance for a specific VES, with a tall bar being better than a short bar. SNK grouping applies only to a specific object, not to both objects shown on the chart together. Standard error bars are also included around the means to illustrate the associated variance.

Bar graph. Tire and dog mean detection and recognition distances by VES. Click here for more detail.
Figure 34. Bar graph. Tire and dog mean detection and recognition distances by VES.

The mean distance at which participants were able to detect the tire with the different VESs ranged from 43.0 m to 56.7 m (141 ft to 186 ft). Recognition distances for the tire ranged from 45.4 m to 50.6 m (149 ft to 166 ft). There were no statistically significant differences at the alpha symbol = 0.05 level between the VESs for detecting or recognizing the tire tread. The left side of figure 34 shows the mean detection and recognition distances for the tire, with standard error bars provided around the means.

When approaching the dog object, the FIR vehicle provided a mean detection distance of 78.9 m (259 ft). This was statistically different at the alpha symbol = 0.05 level from all of the other vehicles, which had mean detection distances ranging from 23.5 m to 41.1 m (77 ft to 135 ft). Recognition distances for the dog indicated differences between the FIR vehicle, with a mean recognition distance of 47.5 m (156 ft), and NIR 2, with a mean recognition distance of 19.2 m (63 ft). The mean recognition distances for the other four VESs fell between the means for the NIR 2 and the FIR and were not significantly different from these two vehicles.

Pedestrian Object Group Analysis

The next set of results is from the analysis of the pedestrian object group. This was a 6 (VES) by 12 (Object) by 3 (Age) mixed factorial design. In this analysis, the object variable is composed of the 12 pedestrian scenarios. As discussed in the Independent Variables section, these scenarios involved pedestrians standing in different locations in relation to the participant vehicle and wearing black or denim clothing.

Table 13 indicates that for detection distance within the pedestrian object group, all main effects and interactions were significant. Table 14 indicates that for recognition distance in the pedestrian object group, statistically significant main effects were found for age, VES, and object. Statistically significant interactions were found for Object by Age and VES by Object. Table 15 provides a summary of the statistically significant main effects and interactions in the analysis of the pedestrian object group.

Table 13. Pedestrian object group ANOVA summary table for the dependent measurement: detection distance.
Source DF SS MS F value P value  
TOTAL 1276 121500521.1       
Between
Age 2 5106529.9 2553265.0 5.30 0.0182 *
Subject/Age 15 7226283.5 481752.2     
 
Within
VES 5 37856755.0 7571351.0 50.46 <.0001 *
VES by Age 10 3178104.1 317810.4 2.12 0.0331 *
VES by Subject/Age 75 11252753.1 150036.7     
 
Object 11 9169261.9 833569.3 28.11 <.0001 *
Object by Age 22 1323152.7 60143.3 2.03 0.0066 *
Object by Subject/Age 165 4893476.1 29657.4     
 
VES by Object 55 12450098.9 226365.4 7.47 <.0001 *
VES by Object by Age 110 4625001.0 42045.5 1.39 0.0080 *
VES by Object by Subject/Age 806 24419105.1 30296.7     

 

Table 14. Pedestrian object group ANOVA summary table for the dependent measurement: recognition distance.
Source DF SS MS F value P value  
TOTAL 1276 89384933.9       
Between
Age 2 4640310.902 2320155.451 4.78 0.0247 *
Subject/Age 15 7278259.15 485217.28     
 
Within
VES 5 25849407.49 5169881.5 44.85 <.0001 *
VES by Age 10 1991478.05 199147.8 1.73 0.09  
VES by Subject/Age 75 8644631.59 115261.75     
 
Object 11 6180821.269 561892.843 26.18 <.0001 *
Object by Age 22 952324.79 43287.49 2.02 0.007 *
Object by Subject/Age 165 3541686.7 21464.77     
 
VES by Object 55 10829295.69 196896.29 9.15 <.0001 *
VES by Object by Age 110 2141637.54 19469.43 0.91 0.7411  
VES by Object by Subject/Age 806 17335080.69 21507.54     

 

Table 15. Summary of significant main effects and interactions for pedestrian group analysis.
Source Significant
Detection
Significant
Recognition
Between
Age x x
Subject/Age    
 
Within
VES x x
VES by Age x  
VES by Subject/Age    
 
Pedestrian x x
Pedestrian by Age x x
Pedestrian by Subject/Age    
 
VES by Pedestrian x x
VES by Pedestrain by Age x  
VES by Pedestrian by Subject/Age    

Differences in performance due to age were evident in the distances at which participants detected pedestrians. The main effect of age indicated the younger participants were detecting pedestrians at longer distances than the older participants. The Age by VES interaction indicated differences in how the age groups performed with one VES versus another. This interaction is presented in figure 35.

Bar graph. Mean detection distances for the Age by VES interaction for pedestrian scenarios. Click here for more detail.
Figure 35. Bar graph. Mean detection distances for the Age by VES interaction for pedestrian scenarios.

For the headlamp-based systems, the younger and middle-aged groups tended to be similar, while the older age group tended to have shorter detection distances. For the FIR and NIR 2 VESs, the younger group had longer detection distances than the older group, while the middle-aged group was somewhere between the other two. For the NIR 1 vehicle, there was no difference between the middle and older age groups. The older age group had longer detection distances for pedestrians with the NIR 1 vehicle than with the FIR or NIR 2 vehicles.

VES by Object Interaction and VES by Object by Age Interaction

The statistically significant VES by Object and VES by Object by Age interactions for the pedestrian scenarios indicate a large number of possible differences. Due to the number of interactions present and the interest in how all the VESs performed on each of the objects, separate Age by VES ANOVAs were conducted for each of the pedestrian scenarios. Only the main effect of VES for each of these objects is discussed, allowing the reader to evaluate the effect of every VES on each pedestrian individually. The Discussion section provides a comparison of the stopping distance achievable for each object for each VES; a separate discussion on the effect of age is also provided there.

Pedestrian, Denim Clothing

Mean detection and recognition distances for pedestrians wearing denim clothing and standing on the left and right side of a straight section of road are shown in figure 36.

Bar graph. Mean detection and recognition distances for pedestrians in denim on straight—left and right side. Click here for more detail.
Figure 36. Bar graph. Mean detection and recognition distances for pedestrians in denim on straight—left and right side.

The FIR vehicle mean detection distance may have been limited by the available sight distance. (Further explanation appears in the Data Analysis section of this report.) When pedestrians were standing on the left side of a straight section of road, the FIR vehicle had the longest mean detection distance of 259.4 m (851 ft), followed by the NIR 1 vehicle, with a mean detection distance of 215.6 m (707 ft). The HLB vehicle and the NIR 2 vehicle were not statistically different from each other, with mean detection distances of 137.8 m and 124.7 m (452 ft and 409 ft), respectively. The HID 1 and the HID 2 vehicles had shorter detection distances than the other four vehicles; however, the HID 1 and the HID 2 vehicles were not statistically different from each other, with mean detection distances of 68.0 m and 40.2 m (223 ft and 132 ft), respectively.

Mean recognition distances were longest for the FIR vehicle and the NIR 1 vehicle. The mean recognition distances were 235.9 m and 199.9 m (774 ft and 656 ft) for the FIR and the NIR 1, respectively. These two vehicles were not statistically different from each other. The next longest mean recognition distances were for the HLB and the NIR 2 vehicles. These distances were 121.0 m and 112.2 m (397 ft and 368 ft), respectively. The shortest mean recognition distances were for the HID 1 and the HID 2 vehicles, with mean recognition distances of 65.5 m and 39.3 m (215 ft and 129 ft), respectively. These distances were not statistically different from each other.

When pedestrians were standing on the right side of the road, the mean detection distances of 272.5 m (894 ft) for FIR and 240.2 m (788 ft) for NIR 1 were the longest of the VESs. Detection distances of 182.6 m (599 ft) for HLB, 157.6 m (517 ft) for HID 1, and 138.7 m (455 ft) for the NIR 2 were grouped together. Because of the limited available sight distance, the FIR mean detection distance may have been less than its full potential. The detection distance of 81.7 m (268 ft) for the HID 2 vehicle in this scenario was the shortest of the VESs.

Recognition distances for the FIR of 227.1 m (745 ft) and for the NIR 1 of 206.0 m (676 ft) were again the longest of the VESs. The HLB recognition distance of 161.5 m (530 ft) was longer than the 114.3 m (375 ft) for the NIR 2 vehicle and 77.1 m (253 ft) for the HID 2 vehicle. The HID 1 recognition distance of 146.9 m (482 ft) was not statistically different from the HLB or the NIR 2 vehicles. The HID 2 vehicle recognition distance was the shortest of all the VESs.

Pedestrian, Black Clothing

Mean detection and recognition distances for pedestrians wearing black clothing and standing on the left and right side of a straight section of road are shown in figure 37.

Bar graph. Mean detection and recognition distances for pedestrian in black on straight—left and right side. Click here for more detail.
Figure 37. Bar graph. Mean detection and recognition distances for pedestrian in black on straight—left and right side.

When a pedestrian dressed in black appeared on the left side of a straight road section, the FIR vehicle had the longest mean detection distance measured at 176.48 m (579 ft). The NIR 1 vehicle had the next longest mean detection distance of 132.89 m (436 ft). The mean detection distance with the HLB vehicle of 82.0 m (269 ft) was not statistically different from the NIR 2 vehicle measurement of 67.1 m (220 ft). The mean detection distances of 35.1 m (115 ft) with the HID 1 vehicle and 27.4 m (90 ft) with the HID 2 vehicle were very similar and were not statistically different from NIR 2 VES. The FIR, the NIR 1, and the HLB vehicles all had longer mean detection distances than the HID VESs. Though the detection distance for the FIR and the NIR 1 vehicles were statistically different, the recognition distances of 162.5 m (533 ft) and 130.8 m (429 ft), respectively, did not show significant statistical differences. These were the longest mean recognition distances for the tested VESs. The HLB vehicle had a longer mean recognition distance
(80.5 m (264 ft)) than both the HID 1 vehicle (32.3 m (106 ft)), and HID 2 vehicle (26.2 m (86 ft)). The NIR 2 vehicle (63.7 m (209 ft)) was not statistically different from the HID VESs or the HLB.

When a pedestrian dressed in black appeared on the right side of a straight road section, the FIR and the NIR 1 had the longest mean detection distances of 189.3 m and 161.2 m (621 ft and 529 ft), respectively. Recognition distances for these two VESs were 170.7 m and 148.1 m (560 ft and 486 ft), respectively. Neither the detection distances nor the recognition distances of the vehicles were statistically different from each other. In addition, the mean detection distances for the HLB (114.0 m (374 ft)), HID 1 (106.7 m (350 ft)), and the NIR 2 (100.6 m (330 ft)) were not statistically different from each other; nor were the recognition distances of 112.2 m, 103.0 m, and 89.9 m (368 ft, 338 ft, and 295 ft) for the same VESs, respectively. The HID 2 vehicle had the shortest mean detection (42.7 m (140 ft)) and recognition distances (41.5 m (136 ft)) of the VESs tested.

Pedestrian, Bloom Scenario

In the bloom scenarios, a pedestrian was standing next to a car with its headlamps on while the participant approached in the oncoming lane. Mean detection and recognition distances for these scenarios are shown in figure 38.

Bar graph. Mean detection distances for bloom scenario, left and right side. Click here for more detail.
Figure 38. Bar graph. Mean detection distances for bloom scenario, left and right side.

In the scenario in which the pedestrian was standing to the left of the participant vehicle, the mean detection distance was the longest when driving the FIR vehicle (273.7 m (898 ft)). Because the available sight distance was limited, this FIR mean detection distance may have been reduced in this scenario. The NIR 1 vehicle had the next longest mean detection distance in this scenario (193.6 m (635 ft)). The NIR 2 (88.4 m (290 ft)), the HLB (68.5 m (226 ft)), and the HID 1 (41.5 m (136 ft)) vehicles were not statistically different from each other. At 25.9 m (85 ft), the mean detection distance of the HID 2 vehicle was not statistically different from the HID 1 or the HLB vehicles; however, it was statistically shorter than both the NIR vehicles and the FIR vehicle.

Recognition distances for the different VESs followed the same pattern as the detection distances in this scenario. For example, the mean recognition distance for the FIR vehicle was the longest (268.5 m (881 ft)). The NIR 1 vehicle had the next longest mean recognition distance in this scenario (185.3 m (608 ft)). The NIR 2 (82.9 m (272 ft)), the HLB (47.9 m (157 ft)), and the HID 1 (40.8 m (134 ft)) were not statistically different from each other. The HID 2 mean recognition distance of 25.6 m (84 ft) was not statistically different from either the HID 1 or the HLB; however, it was statistically shorter than both the NIR vehicles and the FIR vehicle.

In general, in the bloom scenario, the performance of the VESs on the left side of the road ordered the VESs similarly as their performance on the right side of the road did. With the pedestrian on the right side of the road, however, the FIR system did not perform as well as it did on the left and so was more similar to the performance of the NIR 1 system. The headlamp-only VESs performed better on the right side than on the left due to more illuminance being provided toward the right than toward the left.

Pedestrian Off Axis

In the pedestrian off axis scenario, a pedestrian stood 9.4 m (31 ft) to the left or right of the participant vehicle centerline. Mean detection and recognition values for these scenarios are shown in figure 39.

Bar graph. Mean detection and recognition distances for pedestrian far off axis, left and right side. Click here for more detail.
Figure 39. Bar graph. Mean detection and recognition distances for pedestrian far off axis, left and right side.

When a pedestrian was located 9.5 m (31 ft) to the left, measured from the center of the participant vehicle’s lane, the FIR vehicle and the NIR 1 vehicle were not statistically different in mean detection distance. The mean detection distances for the two vehicles were 263.7 m and 230.1 m (865 ft and 755 ft), respectively. These two vehicles had the longest mean detection distances of the group. The mean detection distances of the HLB vehicle (141.4 m (464 ft)), the HID 1 vehicle (113.1 m (371 ft)), and the NIR 2 vehicle (97.5 m (320 ft)) could not be statistically distinguished from one another. Moreover, at 51.2 m (168 ft), the mean detection distance for the HID 2 vehicle also was not statistically different from the NIR 2 vehicle or the HID 1 vehicle mean detection distances.

The FIR (208.9 m (685 ft)) and the NIR 1 (187.1 m (614 ft)) had the longest mean recognition distances of the group, and they were not statistically different from each other. The HLB mean recognition distance (134.1 m (440 ft)) was not statistically different from the HID 1 (111.6 m (366 ft)) or from the NIR 2 (80.2 m (263 ft)) but was statistically longer than the HID 2 (50.9 m (167 ft)). The HID 2 mean recognition distance was not statistically different from the NIR 2.

When the pedestrian was standing 9.5 m (31 ft) to the right of the center of the participant’s lane, the FIR vehicle mean detection distance of 271.0 m (889 ft) was the longest of the group, followed by the NIR 1 mean detection distance of 175.9 m (577 ft). The mean detection distances of the other four vehicles were not statistically different from each other: HLB (95.7 m (314 ft)), HID 1 (79.9 m (262 ft)), NIR 2 (53.6 m (176 ft)), and HID 2 (27.1 m (89 ft)).

At 185.3 m (680 ft), the mean recognition distance of the FIR vehicle was the longest of the group statistically. The mean recognition distance of the NIR 1 (123.4 m (405 ft)) was statistically longer than the NIR 2 (26.5 m (87 ft)) and the HID 2 (27.8 m (88 ft)); however, it was not statistically longer than the HLB (89.3 m (293 ft)) or the HID 1 (75.9 m (249 ft)) vehicles. The HLB, the HID 1, the NIR 2, and the HID 2 vehicles were not statistically different from each other in terms of mean recognition distances in this scenario.

Pedestrian in Right Turn

Mean detection and recognition distances for pedestrians wearing denim clothing and standing on the left and right side of a 1,250-m radius right turn are shown in figure 40.

Bar graph. Mean detection and recognition distances for pedestrian in denim in right turn, left and right side. Click here for more detail.
Figure 40. Bar graph. Mean detection and recognition distances for pedestrian in denim in right turn, left and right side.

When detecting and recognizing pedestrians that were standing on the left side of a 1,250-m radius right-hand turn, the FIR vehicle had the longest mean detection distances, at 212.8 m (698 ft). The mean detection distance for the NIR 1 (163.4 m (536 ft)) was not statistically different from the HLB (152.4 m (500 ft)) or the HID 1 (143.0 m (469 ft)); however, the NIR 1 was statistically longer than the NIR 2 (113.4 m (372 ft)) and the HID 2 (97.3 m (319 ft)). The HLB (152.4 m (500 ft)) and the HID 1 (143.0 m (469 ft)) had statistically longer mean detection distances than the HID 2 (97.2 m (319 ft)), but they were grouped together with the NIR 2 (113.4 m (372 ft)) vehicle. The HID 2 had similar mean detection distances to the NIR 2.

Recognition distances for the systems showed that the FIR (190.2 m (624 ft)), the NIR 1 (149.4 m (490 ft)), and the HLB (149.7 m (491 ft)) were grouped together. Recognition distances for the NIR 1 (149.4 m (490 ft)), HLB (149.7 m (491 ft)), HID 1 (139.0 m (456 ft)), and NIR 2 (108.5 m (356 ft)) were grouped together. At 92.0 m (302 ft), the mean recognition distance for the HID 2 vehicle was statistically shorter than for all the systems except the NIR 2.

Where pedestrians were standing on the right side of a 1,250-m radius right-hand turn, the six vehicles tested were separated into two groups based on detection performance. The mean detection distances of the NIR 1 (134.1 m (440 ft)), FIR (132.3 m (434 ft)), HID 1 (132.0 m (433 ft)), and HLB (126.8 m (416 ft)) placed them in the longer group. The mean detection distances of the NIR 2 (89.6 m (294 ft)) and the HID 2 (80.5 m (264 ft)) placed them in the group with shorter mean detection distances.

This same grouping occurred for recognition distances. The mean recognition distances for the NIR 1, FIR, HID 1, and HLB were 129.5 m, 122.8 m, 132.0 m, and 126.8 m (425 ft, 403 ft, 433 ft, and 416 ft), respectively. For the NIR 2 and HID 2, the mean recognition distances were 58.5 m (192 ft) and 77.7 m (255 ft), respectively, which placed them in a separate group from the other four VESs.

Pedestrian in Left Turn

Mean detection and recognition distances for pedestrians wearing denim clothing and standing on the left and right side of a 1,250-m radius left turn are shown in figure 41.

Bar graph. Mean detection and recognition distances for pedestrian in denim in left turn, left and right side. Click here for more detail.
Figure 41. Bar graph. Mean detection and recognition distances for pedestrian in denim in left turn, left and right side.

When detecting pedestrians standing on the left side of a 1,250-m radius left-hand turn, the mean detection distances from longest to shortest were 152.4 m (500 ft) (NIR 1), 105.5 m (346 ft) (HLB), 84.4 m (277 ft) (HID 1), 66.1 m (217 ft) (NIR 2), 50.9 m (167 ft) (HID 2), and 29.9 m (98 ft) (FIR). Each of these mean detection distances was statistically different from all the other systems.

The recognition distances for the VESs followed a similar arrangement. The order of recognition distances was the same, except for the HID 2 and NIR 2. These VESs were grouped together with mean recognition distances of 49.7 m and 63.0 m (163 ft and 207 ft), respectively. Around these two, the mean recognition distances for the other four vehicles (from longest to shortest) were 148.4 m (487 ft) (NIR 1), 103.32 m (339 ft) (HLB), 82.0 m (269 ft) (HID 1), and 29.0 m (95 ft) (FIR). Each of these was statistically different from each other and from the two grouped VESs.

In the same left-hand 1,250-m curve, when looking at pedestrians standing on the right side of the road, the mean detection distance for the NIR 1 (207.9 m (682 ft)) was grouped with the FIR, (235.0 m (768 ft)). The next longest mean detection distances were for the HLB (125.6 m (412 ft)) and the NIR 2 (117.7 m (386 ft)). These two vehicles were grouped together statistically. The shortest mean detection distance for this scenario was 62.2 m (204 ft), from the HID 2 VES. The mean detection distance for the HID 1 (87.8 m (288 ft)) was not statistically different from the HLB, the NIR 2, or the HID 2 vehicles.

Recognition of the pedestrian standing to the right side of the left curve was longest for the NIR 1 (180.1 m (591 ft)). Mean recognition distances for the FIR (142.7 m (468 ft)), the HLB (121 m (397 ft)), and the NIR 2 (109.4 m (359 ft)) grouped together statistically. The mean recognition distance of the HID 1 (84.4 m (277 ft)) was statistically shorter than for the FIR and NIR 1, but it was not statistically different from the other three VESs. The HID 2 mean recognition distance (56.7 m (186 ft)) was not statistically different from the HID 1, but it was statistically shorter than the recognition distance for the other four VESs in this scenario.

Retroreflective Object Group Analysis

A VES by Retroreflective Object Group ANOVA was used to investigate the effect of the VESs on the detection of roadway guidance objects at night. This object group included the RRPMs, signs, and the turn arrow. Table 16 and table 17 indicate statistically significant main effects in detection and recognition for VES and object and a statistically significant interaction for VES by Object.

Table 16. Retroreflective group ANOVA summary table for the dependent measurement: detection distance.
Source DF SS MS F value P value  
TOTAL 322 530670027.8       
Between
Age 2 4942689.2 2471344.6 3.30 0.0647  
Subject/Age 15 11221308.7 748087.2     
 
Within
VES 5 30607560.5 6121512.1 30.55 <.0001 *
VES by Age 10 1143637.5 114363.8 0.57 0.8327  
VES by Subject/Age 75 15027402.4 200365.4     
 
Object 2 417139124.0 208569562.0 918.72 <.0001 *
Object by Age 4 1716496.4 429124.1 1.89 0.138  
Object by Subject/Age 30 6810669.5 227022.3     
 
VES by Object 10 13914411.4 1391441.1 8.45 <.0001 *
VES by Object by Age 20 3607852.1 180392.6 1.10 0.3601  
VES by Object by Subject/Age 149 24538876.1 164690.4     

 

Table 17. Retroreflective group ANOVA summary table for the dependent measurement: recognition distance.
Source DF SS MS F value P value  
TOTAL 322 392627269.9       
Between
Age 2 5636728.812 2818364.406 2.31 0.1338  
Subject/Age 15 18321641.6 1221442.8     
 
Within
VES 5 14875454.76 2975090.95 14.73 <.0001 *
VES by Age 10 1285201.86 128520.19 0.64 0.7783  
VES by Subject/Age 75 15149792.5 201997.2     
 
Object 2 272468249 136234124.5 182.41 <.0001 *
Object by Age 4 2091062.3 522765.6 0.70 0.5981  
Object by Subject/Age 30 22406260.6 746875.4     
 
VES by Object 10 8156580.822 815658.082 4.18 <.0001 *
VES by Object by Age 20 3127735.986 156386.799 0.8 0.71  
VES by Object by Subject/Age 149 29108561.7 195359.5     

Table 18 provides a summary of the statistically significant main effects and interactions found for the retroreflective object group.

Table 18. Summary of significant main effects and interactions for retroreflective group analysis.
Source Significant
Detection
Significant
Recognition
Between
Age    
Subject/Age    
 
Within
VES x x
VES by Age    
VES by Subject/Age    
 
Object x x
Object by Age    
Object by Subject/Age    
 
VES by Object x x
VES by Object by Age    
VES by Object by Subject/Age    

As would be expected, the main effect of object indicates large differences in the detection and recognition distances for the different objects. The mean detection distance for the signs across all the VESs was 776.3 m (2,547 ft). For the RRPM, the mean detection distance was 349.6 m (1,147 ft), and for the turn arrow the mean detection distance was 69.8 m (229 ft). When considering recognition distances of the sign, the RRPMs, and the turn arrow, mean distances were 632.2 m, 273.1 m, and 65.2 m (2,074 ft, 896 ft, and 214 ft), respectively.

In the ANOVA for the retroreflective objects detection distances, the VES by Object interactions were statistically significant, as were the main effects of VES and object.

Object

The significant main effect of object indicates that the signs were detected and recognized at the greatest distances, followed by the RRPMs. The turn arrow was the retroreflective object with the shortest mean detection and recognition distances.

VES

The significant main effect of VES indicates differences between the VESs in both detection and recognition distances of retroreflective objects as a group (figure 42). The NIR 1, HLB, and HID 1 vehicles had the longest detection and recognition distances for retroreflective objects. The mean detection distances for these three vehicles were 598.0 m, 561.4 m, and 551.4 m (1,962 ft, 1,842 ft, and 1,809 ft), respectively. The mean recognition distances were 438.3 m, 466.7 m, and 464.8 m (1,438 ft, 1,531 ft, and 1,525 ft), respectively. The NIR 2 vehicle was the next longest with a mean detection distance of 497.1 m (1,631 ft), followed by the HID 2 (433.7 m (1,423 ft)) and the FIR (322.7 m (1,057 ft)) vehicle.

Bar graph. Mean detection and recognition distances for retroreflective object group by VES. Click here for more detail.
Figure 42. Bar graph. Mean detection and recognition distances for retroreflective object group by VES.

Mean recognition distances for retroreflective objects indicate that the NIR 2 (387.7 m (1,272 ft)) and the HID 2 (350.2 m (1,149 ft)) were similar. The mean recognition distance for retroreflective objects for the FIR was 299.9 m (984 ft), which was the shortest of the VESs. An analysis of the performance of the VESs on each object is provided in the following section.

VES by Object Interaction

The significant VES by Object interaction for the detection distances of retroreflective objects indicates that certain VESs do well on some objects but not on others. The NIR 2 vehicle was comparable to the NIR 1, HLB, and HID 1 in detecting signs; however, for the turn arrow and RRPMs, the NIR 2 had lower performance than did these three other VESs. The NIR 1 vehicle outperformed the HLB benchmark on detecting RRPMs, but it was below the benchmark for the turn arrow.

For retroreflective object recognition, the NIR 1 had recognition distances that were similar to the HLB and HID 1 for the signs, but recognition distances for the RRPMs were shorter than for these two VESs.
Figure 43 and figure 44 illustrate these interactions.

Bar graph. Mean detection distances for the VES by Object interaction for retroreflective group. Click here for more detail.
Figure 43. Bar graph. Mean detection distances for the VES by Object interaction for retroreflective group.

 

Bar graph. Mean recognition distances for VES by Object interaction for retroreflective group. Click here for more detail.
Figure 44. Bar graph. Mean recognition distances for VES by Object interaction for retroreflective group.

For more detailed comparison of the VESs on each of the retroreflective objects, the next sections report on their post hoc comparisons.

Turn Arrow

Mean detection and recognition distances for the six VESs when approaching the retroreflective pavement marking turn arrow are shown in figure 45.

Bar graph. Arrow detection and recognition distances by VES. Click here for more detail.
Figure 45. Bar graph. Arrow detection and recognition distances by VES.

When detecting the turn arrow, the HLB benchmark provided the longest mean detection distance at 89.2 m (295 ft). This was followed by the NIR 1 (77.7 m (255 ft)), the HID 1 (74.4 m (244 ft)), and the HID 2 (65.5 m (215 ft)). These means were not statistically different from each other. The mean detection distances for the FIR and the NIR 2 were 53.0 m (174 ft) and 57.6 m (189 ft), respectively, and were not statistically different from the HID 2 VES.

In comparing recognition distance, the HID 1, the NIR 1, and the HLB vehicles were grouped together, with mean recognition distances ranging from 72.2 m to 79.6 m (237 ft to 261 ft). These were the longest recognition distances for the six VESs. The HID 2 had the next longest mean recognition distance, at 61.3 m (201 ft). The FIR vehicle had the shortest mean recognition distance (48.2 m (158 ft)) for the turn arrow. The NIR 2 had a mean recognition distance of 55.5 m (182 ft), which was not statistically different from the HID 2 or the FIR vehicle.

RRPMs

Mean detection and recognition distances for the six VESs when approaching an RRPM are shown in figure 46.

Bar graph. RRPM detection and recognition distances by VES. Click here for more detail.
Figure 46. Bar graph. RRPM detection and recognition distances by VES.

The detection distance provided by the NIR 1 of 541.6 m (1,777 ft) was statistically longer for the RRPMs than those of the HLB (422.2 m (1,385 ft)), the NIR 2 (288.7 m (947 ft)), the HID 2 (230.4 m (756 ft)), and the FIR (160.6 m (527 ft)). The HID 1 (455.1 m (1,493 ft)) and the HLB were not statistically different from each other, and the NIR 2 and the HID 2 were not statistically different from each other. In addition, the FIR was not statistically different from the HID 2; however, the NIR 2 had statistically longer detection distances than did the FIR.

When considering recognition distances for the RRPMs, HLB and HID 1 had the longest recognition distances at 416.7 m and 410.9 m (1,367 ft and 1,348 ft), respectively. The mean recognition distance of 268.8 m (882 ft) for the NIR 1 was statistically longer than the FIR of 143.9 m (472 ft). The mean recognition distances of 190.5 m (625 ft) for the NIR 2 and 208.8 m (685 ft) for HID 2 were not statistically different from those of the FIR or the NIR 1.

Signs

When detecting the signs ahead, the mean detection distances for HLB, NIR 1, HID 1, and NIR 2 were 866.9 m, 886.7 m, 837.9 m, and 821.1 m (2,844 ft, 2,909 ft, 2,749 ft, and 2,694 ft), respectively. These distances were not statistically different from each other. The FIR had the shortest mean detection distance (537.7 m (1,764 ft)). For signs, the HID 2 detection distance of (709.0 m) 2,326 ft was statistically shorter than that of the long-distance group, but longer than that of the FIR.

Mean recognition distance for the signs generated similar performance grouping. The mean recognition distances for the long-distance group ranged from 652.6 m to 705.0 m (2,141 ft to 2,313 ft). The FIR mean recognition distance was 504.1 m (1,654 ft). The HID 2 mean distance of 557.5 m (1,829 ft) was not statistically different from the others. Figure 47 shows the mean detection and recognition distances for signs, with standard error bars provided around the means.

Bar graph. Sign detection and recognition distances by VES. Click here fore more detail.
Figure 47. Bar graph. Sign detection and recognition distances by VES.

Recognition of Sign Type

A 6 (VES) by 3 (Age) mixed factorial ANOVA was conducted to investigate differences in when the participants could recognize (i.e., discriminate between) a stop or a yield sign. Table 19 indicates that age was a statistically significant factor in distinguishing the stop sign from yield sign (p = 0.0239).

Table 19. ANOVA summary table for recognition of sign type (stop versus yield).
Source DF SS MS F value P value  
TOTAL 107 25238655.8       
Between
Age 2 5533941.2 2766970.6 4.84 0.0239 *
Subject/Age 15 8575531.6 571702.1     
 
Within
VES 5 1240877.9 248175.6 1.98 0.092  
VES by Age 10 465657.7 46565.8 0.37 0.9556  
VES by Subject/Age 75 9422647.3 125635.3     

At 327.1 m (1,073 ft), the mean detection distance for the younger group was longer than the mean distances for the middle (242.9 m, 797 ft) and older (211.5 m, 694 ft) groups. Analysis of the mean distance at which the participants could recognize a sign as either a stop sign or a yield sign did not indicate statistical differences among the VESs at the alpha symbol < 0.05 level (p = 0.092). The distances at which the participants recognized the sign type ranged from 232.0 m to 299.3 m (761 ft to 982 ft) across the VESs, with a mean of 260.6 m (855 ft). The mean distance for each of the VESs is presented in figure 48.

Bar graph. Sign recognition distances of stop versus yield sign by VES. Click here for more detail.
Figure 48. Bar graph. Sign recognition distances of stop versus yield sign by VES.

A 6 (VES) by 3 (Age) mixed factorial ANOVA of when participants could read the speed limit shown on the speed limit sign also did not indicate differences between the systems. Table 20 shows the results of this ANOVA.

Table 20. ANOVA summary table for when participants could read speed limit.
Source DF SS MS F value P value  
TOTAL 107 4594742.9       
Between
Age 2 1056334.747 528167.373 3.64 0.0513  
Subject/Age 15 2173934.533 144928.969     
 
Within
VES 5 125850.2558 25170.0512 1.6 0.1696  
VES by Age 10 61307.6299 6130.763 0.39 0.947  
VES by Subject/Age 75 1177315.704 15697.543     

Figure 49 provides the mean distances at which participants could read the speed limit signs for each of the VESs.

Bar graph. Mean distances at which speed limit signs were read. Click here for more detail.
Figure 49. Bar graph. Mean distances at which speed limit signs were read.

Distances at which the signs could be read ranged from 108.5 m to 130.5 m (356 ft to 428 ft), with a mean of 122.2 m (401 ft). Using the 25.7-cm (10.125-inch) number height of the speed limit, a range from 4.2 m/cm to 5.1 m/cm with a mean of 4.8 m/cm (35 ft/inches to 42 ft/inches with a mean of 40 ft/inches) is obtained.

Though not statistically significant at the alpha symbol = 0.05, it appears that an age effect may occur for reading a speed limit sign. The mean distances at which the younger, middle, and older groups read the signs were 151.8 m, 112.5 m, and 102.7 m (498 ft, 369 ft, and 337 ft), respectively. Figure 50 provides the mean distances at which the different age groups could read the speed limit.

Mean distance at which speed limit signs were read by each age group. Click here for more detail.
Figure 50. Bar graph. Mean distance at which speed limit signs were read by each age group.

SUBJECTIVE RATINGS

ANOVAs were conducted on the participant responses for each of the eight Likert-type scale statements presented below:

  1. This VES allowed me to DETECT objects sooner than my regular headlights.
  2. This VES allowed me to RECOGNIZE objects sooner than my regular headlights.
  3. This VES helped me to stay on the road (not go over the lines) better than my regular headlights.
  4. This VES allowed me to see which direction the road was heading (i.e., left, right, straight) beyond my regular headlights.
  5. This VES did not cause me any more visual discomfort than my regular headlights.
  6. This VES allowed me to read signs beside the road sooner than my regular headlights.
  7. This VES makes me feel safer when driving on the roadways at night than my regular headlights.
  8. This is a better VES than my regular headlights.

Participants were asked to indicate their agreement or disagreement on a seven-point scale ranging from “Strongly Agree” to “Strongly Disagree.” ANOVA tables for the analyses of these responses are shown in appendix G.

Table 21 indicates significant main effects for VES (alpha symbol < 0.05) were found for participant responses to all of the statements except number 8, in which the participants indicated agreement with the statement “This is a better VES than my regular headlights.” No other statistically significant main effects or interactions were found at the alpha symbol < 0.05 level.

Table 21. Summary of significant main effects and interactions for the Likert-type scales.
Source 1 2 3 4 5 6 7 8
Between
Age
Subject/Age
 
Within
VES x x x x x x x  
VES by Age
VES by Subject/Age

For those statements where a main effect of VES was found, SNK post hoc analyses were conducted to identify differences between the different VESs.

When participants were asked which VES allowed them to detect objects sooner than their regular headlights, they indicated that the NIR 1 VES was the best, with an average rating of 1.17 on a scale of 1 to 7 (1 representing “Strongly Agree” and 7 representing “Strongly Disagree”) (figure 51). Based on participant responses, the HID 2 VES faired worst, with an average rating of 3.67. The mean values of the remaining systems were between 2.06 and 2.61 on the same scale.

Bar graph. Mean subjective ratings by VES for Statement 1: This vision enhancement system allowed me to detect objects sooner than my regular headlights. Click here for more detail.

Figure 51. Bar graph. Mean subjective ratings by VES for Statement 1: This vision enhancement
system allowed me to detect objects sooner than my regular headlights.

With respect to recognition, participants responded that the NIR 1 VES allowed them to recognize objects sooner as compared to their regular headlights (figure 52). The average rating for the NIR 1 was 1.67 on a scale of 1 to 7 (1 representing “Strongly Agree” and 7 representing “Strongly Disagree”) (alpha symbol = 0.05). The FIR, HID 1, and the NIR 2 were not statistically different from the NIR 1 in this measure, with mean response ratings between 2.61 and 2.44. The HLB, with a mean rating of 2.83, was not statistically different from the previous three vehicles. The HID 2 VES received the lowest average rating, at 3.78.

Bar graph. Mean subjective ratings by VES for Statement 2: This vision enhancement system allowed me to recognize objects sooner than my regular headlights. Click here for more detail.

Figure 52. Bar graph. Mean subjective ratings by VES for Statement 2: This vision enhancement
system allowed me to recognize objects sooner than my regular headlights.

When asked which VES helped them to stay on the road better than their regular headlights, participant responses showed the HID 1 to be the most effective, with an average rating of 3.39 (alpha symbol = 0.05). This vehicle was not statistically different from the HLB, the NIR 1, or the HID 2. The NIR 2 (mean 4.4) and the FIR (mean 4.7) had mean values on the “Disagree” side of the response scale, indicating that participants felt these systems were not as helpful in staying on the road as their regular headlamps (figure 53).

Bar graph. Mean subjective ratings by VES for Statement 3: This vision enhancement system helped me to stay on the road (not go over the lines) better than my regular headlights. Click here for more detail.

Figure 53. Bar graph. Mean subjective ratings by VES for Statement 3: This vision enhancement
system helped me to stay on the road (not go over the lines) better than my regular headlights.

When asked which VES allowed the driver to see which direction the road was heading beyond regular headlights, the NIR 1 received the most favorable average rating, at 2.33 (figure 54). The HID 2, with a wide beam profile, faired worst, with an average rating of 3.89. The remaining systems averaged between 2.72 and 3.67 on a 1–7 scale (1 representing “Strongly Agree” and 7 representing “Strongly Disagree”).

Bar graph. Mean subjective ratings by VES for Statement 4: This vision enhancement system allowed me to see which direction the road was heading (i.e., left, right, straight) beyond my regular headlights. Click here for more detail.

Figure 54. Bar graph. Mean subjective ratings by VES for Statement 4: This vision enhancement system allowed
me to see which direction the road was heading (i.e., left, right, straight) beyond my regular headlights.

The HID 1 VES was rated the best with respect to causing the least amount of visual discomfort in comparison to normal headlights. The NIR 2 VES received the lowest mean rating of 2.9. The remaining systems averaged between 1.9 and 2.7 on the same scale (figure 55).

Bar graph. Mean subjective ratings by VES for Statement 5: This vision enhancement system did not cause me any more visual discomfort than my regular headlights. Click here for more detail.

Figure 55. Bar graph. Mean subjective ratings by VES for Statement 5: This vision enhancement
system did not cause me any more visual discomfort than my regular headlights.

When asked which VES allowed the participant to read signs beside the road sooner than regular headlights, responses showed that participants felt that, overall, the tested headlamps were better than the IR vehicles (figure 56). Specifically, the HID 1 VES was rated best among the vehicles, with an average rating of 3.11 on a scale of 1 to 7 (1 representing “Strongly Agree” and 7 representing “Strongly Disagree”).

Bar graph. Mean subjective ratings by VES for Statement 6: This vision enhancement system allowed me to read signs beside the road sooner than my regular headlights. Click here for more detail.

Figure 56. Bar graph. Mean subjective ratings by VES for Statement 6: This vision enhancement
system allowed me to read signs beside the road sooner than my regular headlights.

With respect to safety while driving at night, responses indicated that participants felt that the NIR 1 VES made them feel safer in relation to normal headlights, with an average rating of 2.2 on a scale of 1 to 7 (1 representing “Strongly Agree” and 7 representing “Strongly Disagree”). While still on the positive side of the scale, the HID 2 VES had the lowest mean rating of 3.7. The remaining systems received average ratings between 2.6 and 2.8 on the same scale (figure 57).

Bar graph. Mean subjective ratings by VES for Statement 7: This vision enhancement system makes me feel safer when driving on the roadways at night than my regular headlights. Click here for more detail.

Figure 57. Bar graph. Mean subjective ratings by VES for Statement 7: This vision enhancement
system makes me feel safer when driving on the roadways at night than my regular headlights.

When asked to respond to the statement, “This is a better VES than my regular headlights,” no significant main effect was found for VES (p = 0.1061) (figure 58).

Bar graph. Mean subjective ratings by VES for Statement 8: This is a better vision enhancement system than my regular headlights. Click here for more detail.

Figure 58. Bar graph. Mean subjective ratings by VES for Statement 8: This is a better vision
enhancement system than my regular headlights.

 

Previous    Table of Contents    Next
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101