Visibility Performance UnderVisibility Performance Under New Lighting Technologies

Dr. Ronald B. GibbonsGroup Leader, Lighting and Infrastructure

Technologygy

Anchorage Alternative Li htiLighting

This project was the investigation of the impact of broad spectrum light sources on driver visibility.

Visibility was rated based on the detection distance of ll t t th id f th da small target on the side of the road

Luminance and Illuminance was measured in-vehicle during the experimentg p

Experimental Designp g

Variable DescriptionLighting five alternative light sources (Dimming HPS, Beta LED, Kim

Induction, Lumecon LED, Kim LED), one group of the existing condition (HPS 400W Non-Dimming)

Lighting Level High and Low (Dimmed) *Except existing conditionLighting Level High and Low (Dimmed) Except existing condition

6 different lighting systems were tested along an urban street in Anchorageurban street in Anchorage• HPS, LED, Induction

2 Dimming Levels2 Dimming Levels• Attempting to investigate adaptive lighting

Lighting Layoutsg g y

Participantsp27 Participants from the public were tested for object detection and public opinion• The participants were invited by the Anchorage

M ffi t t k t iMayors office to take part in a survey• After a briefing at the Public Library, participants were

taken to the test area in buses where they were allowed yto rate the installations

• During the subjective rating sessions, 3 participants d t d f thwere drawn at random from the group.

MethodsAfter Pickup, the Participants were then driven to the beginning of the test area and instructed on the taskEach participant was given a button to press when they were sure that they could see the target on the side of the roadthe side of the road.• Two targets were located in each test area

The button presses were recorded in the data fileThe button presses were recorded in the data file

Equipmentq pThe VTTI RLMMS was installed on a rental vehicle and used during the data collection• The measurement system utilizing GPS, a Luminance

d ill i t d h t t icamera and illuminance meters records photometric and visual information at 10 Hertz.

• The system, also measures:y ,• Spectroradiometer• Button Presses• Color Video Camera• Vertical Illuminance• Linking to Automobile J1850 interface for SpeedLinking to Automobile J1850 interface for Speed

and Throttle information

System Layouty y

Data Collection

Luminance Camera12 bit Point Grey Di it l Fi iDigital Firewire camera.• Calibrated against a

Prometric Still L i CLuminance Camera

Varying shutter and gain values determine gthe range of luminance measured• 2 cameras can be• 2 cameras can be

coupled to increase dynamic response

Individual images areIndividual images are stored for later analysis

Luminance Camera

ResultsThe results were analyze using ANOVA techniques

Source F value P value SignificantLighting Type 5.00 0.0011 *Lighting Level 10.24 0.0037 *Lighting Type* Lighting Level 3.00 0.0225 *Lighting Class 4.06 0.0004 *

All factors and their interaction were significant

Lighting Level Comparisong g p

70Lighting Section Comparison

B

A

60

70

(m)

C

B, C B, C B, C

40

50

ion

Dis

tanc

e

20

30

Mea

n D

etec

ti

0

10

250W Dimming Beta LED Kim Induction Lumecon LED Kim LED 400W Non

M

250W Dimming HPS

Beta LED Kim Induction Lumecon LED Kim LED 400W Non-Dimming HPS

Lighting Section

Lighting Levelg g

70Lighting Intensity Comparison

60

70

m)

40

50

on D

ista

nce

(m

20

30

Mea

n D

etec

tio

0

10

M

0High Low

Lighting Intensity

Lighting Type and Levelg g yp

70Lighting Class Comparison

A

A, B

A

A, B50

60

70

ce (m

)

C

B, CB, C B, C B, C

30

40

ectio

n D

ista

nc

C

10

20

Mea

n D

ete

0250W

Dimming HPS

Beta LED Kim Induction

Lumecon LED

Kim LED 250W Dimming

HPS

Beta LED Kim Induction

Lumecon LED

Kim LED

Low High

Lighting Class

Practical Differences400W HPS and Alternative Lighting Comparisons

-5

0250W Dimming

HPSBeta LED Kim Induction Lumecon LED Kim LED

)

Lighting Section

-20

-15

-10

Dis

tanc

e (m

)

-30

-25

20

in D

etec

tion Lower 95%

MeanUpper 95%

-40

-35

Diff

eren

ce

-50

-45

Illuminance Results25

Mean Illuminance Levels For Each Lighting Section

20

15

nce

(lux)

Left

5

10

ean

Illu

min

an

Front

Right

0250W

Dimming HPS

Beta LED

Kim Induction

Lumecon Kim LED

250W Dimming

HPS

Beta LED

Kim Induction

Lumecon Kim LED

400W Non-Dimming

HPS

Me

HPS HPS HPS

Low High Control

Lighting Class

Relationship to Illuminancep80

Mean Detection Distance and Illuminance Levels

60

70

30

40

50

Mean Illuminance Level (Lux)

0

10

20Mean Detection Distance (Meters)

0

W D

imm

ing

HPS

Bet

a LED

Kim

Indu

ctio

n

Lum

econ

LED

Kim

LED

400

W H

PS

W D

imm

ing

HPS

Bet

a LED

Kim

Indu

ctio

n

Lum

econ

LED

Kim

LED

400

W H

PS

250W

250W

Low High

Anchorage Observationsg400 Watt HPS provided the longest detection distance250 Watt HPS was the shortest detection distanceThe Broader Spectrum sources provided an improved visibility at a average lower horizontal ill i l lilluminance level.The lighting level significantly impacted D t tiDetection

Lighting Considerationsg gMesopic Effect• The eye is more sensitive to blue spectrum light

sources at low luminance levels

I h i b d h b dIs the improvement based on the broader spectrum sources a result of Light Sensitivity or more available information in the visualmore available information in the visual environment?• Further investigation is requiredg q

Further AnalysisyWe will be looking at the distances of detection for both the pedestrians and the objectsThe luminance, illuminance and contrast will all be assessed based on the measured results of the systemGl ill b d b d i hGlare will be attempted to be measured using the interior illuminance meter

Luminance Metrics

Applying multiple contrastApplying multiple contrast metrics to imagesSemi-automated processp1. Accesses database of

images for analysis2. User selects target3. Automatically calculates

contrast metricscontrast metrics

Luminance MetricsResults in luminance and contrast information

Mean Luminance of Target 0.677cd/m2 0.961cd/m2Mean Luminance of Target 0.677cd/m 0.961cd/mMean Luminance of Background 1.579cd/m2 1.492cd/m2

Weber Contrast -0.571 -0.356Simple Contrast 2.331031 1.552046 Mi h l C 0 399585 0 216315Michelson Contrast 0.399585 0.216315