Federal Highway Administration

by John Arens and Mark Reilly

You're on a road trip, speeding along a lonely stretch of rural highway on a moonless, early summer's night. Windows rolled down, the rush of cool air is filled with the hum of night insects. Relaxed, carefree, and with nothing in particular on your mind, you miss a very important road sign — a sign you could not see with your headlights, the one indicating an intersection and a stop sign ahead. Now you see the headlights of a car approaching rapidly from the right, and you hit the brakes. But it's too late.

Aside from such a dire scenario, most of us have wondered at one point or another while driving at night on some unknown highway or backroad whether we saw the right sign to turn or exit — the sign that would have prevented us from continuing in the wrong direction and would have saved us the time and frustration of having to turn around and backtrack. And why did we “miss” the sign? Maybe we simply were not paying close attention. But maybe our lights did not properly reflect the sign because the reflective qualities of the sign itself were deficient.

The Photometric and Visibility Building at Turner-Fairbank Highway Research Center, in McLean, Va.

Being able to see what’s ahead at night and during other periods of limited visibility is a problem that has dogged motorists since the invention of the automobile. And signs and curves in the road are not the only things for which we need to be on the lookout. Pedestrians are just as important, if not more so.

Thanks to investigations and development of new retroreflective materials — materials that direct, or return, the reflected rays in directions close to the illuminating source — and the exploratory investigation of various lighting sources, including ultraviolet headlamps and fluorescent materials, at the Federal Highway Administration’s (FHWA) Photometric and Visibility Laboratory, “reading” the road and seeing people while driving under inclement conditions and at night are becoming much easier.

The Photometric and Visibility Laboratory (P&V Lab) at the FHWA’s Turner-Fairbank Highway Research Center in McLean, Va., is housed in a long, non-descript wooden building, nicknamed by many as the “chicken coop” or “bowling alley.” Built in 1991 and visited by some 250 people a year, the building contains an office; photometric measurement and calibration room; a small workshop and storage room; and an unobstructed laboratory space, running the length of the building, painted completely black to minimize light reflections during lab tests and human factors studies under conditions simulating a nighttime environment. Here, many of the experiments simulating perception under various conditions of visibility are conducted, and signing and marking materials are evaluated.

The interior of the Photometric and Visibility Laboratory is painted black to simulate nighttime visibility.

However, the P&V Lab is concerned with more than just the application and efficacy of different traffic control devices. It is also concerned with light, color, and retroflection — the visual characteristics that determine what one can see and comprehend on the road ahead. The P&V researchers want to make sure drivers are able to travel as safely at night and in rain, fog, snowfall, and other traffic situations in which oncoming cars produce headlamp glare as they can in the sunshine.

Currently, the P&V Lab has three objectives or research activities. The first area of study is measurements, which is divided into photometric and colorimetric measurements. The second area, material studies, concerns the use of materials that are used to enhance night visibility. The third area is in-house research, involving many different investigations.

One recently completed study was a worldwide laboratory intercomparison of measuring the spectral coefficient of retroreflectors under nighttime viewing conditions. This work was sponsored by the International Commission on Illumination (CIE) to determine the accuracy with which certain physical characteristics can be measured using different instruments and laboratory systems.

Photometric Measurements
Photometry is a branch of physics concerned with the measurement of the properties of light. Light is defined as electromagnetic radiation having a wavelength from about 380 to 780 nanometers, which can be perceived by the unaided human eye.

The laboratory is large enough to accommadate automobiles and small trucks in this research.

In addition to being able to see within the range of the normal human eye, the P&V Lab's photometric instruments can also measure radiation slightly outside the visual spectrum - in the long-wave region of ultraviolet (UV-A) and short-wave infrared areas (near-IR).

The P&V Lab has some rather sophisticated setups that allow the precise measurement and evaluation of retroreflective materials. These retroreflective materials are used for signs and marking roads, in safety clothing, and on bicycles and many other objects to improve the chances of being seen during darkness.

The lab is set up for a 15-meter retroreflectivity experiment. The "observation angle positioner" shown in the foreground, holds the light source, detector, and retroreflectance of the sample on the goniometer at the far end of the laboratory.

To make a retroreflective evaluation on the luminance of certain objects, the P&V Lab uses three devices together to represent actual spacial driving conditions. A fully automated 15-meter retroreflectometer system uses a collimated light source to simulate, for example, the headlamp of an automobile, motorcycle, or bicycle. A detector simulating the human eye receives the light reflected by the object under investigation; the detector can be moved to change the geometry between the light source, test specimen, and detector. A goniometer acts as a holding device for the test object.

The goniometer can move the test object vertically and horizontally and can also rotate it around its center. Because it can move the object in three planes, it is also called a three-axis goniometer. Its function is to present the test sample to the light source and the detector in various geometries simulating the spacial relationship between car, sign or road markings, and driver. Just as the observation angle between headlamp, sign, and observer changes as the driver changes position relative to the sign, the goniometer changes the presentation of the test specimen relative to the light source and detector.

The entire system is controlled by a computer, which also evaluates the test object for its retroreflective characteristics. The computer contains calibration values, controls the desired geometric coordinates of the detector and the test sample, and provides a printout of the physical characteristics along with the geometric settings.

Aside from signing materials, this three-way setup is also used to evaluate road-marking materials, retroreflective markers, headlamps, and other small light sources. And with the P&V Lab's participation in the Measurement Assurance Program of the National Institute of Standards and Technology, the absolute accuracy within the measurement system’s range is known.

In addition to testing and evaluating existing and new materials for in-house and contract research, the P&V Lab provides technical assistance and support to other state and local governmental laboratories as well as to FHWA's field staff. These supporting functions include, but are not limited to, the calibration of photometric instruments and the measurement of all types of retroreflective materials.

Colorimetric Measurements
Studying and measuring color, particularly for the regulation of traffic control devices — objects drivers universally recognize by shape, legend, and color — is the other half of the P&V Lab's photometric capability.

Without getting too technical, perhaps this is a good time to present a brief review on color. The classic definition of color is the property of an object reflecting the light of a particular wavelength. The distinct colors of the spectrum are red, orange, yellow, green, blue, indigo, and violet. Thus, an apple looks red because it reflects the red wavelength of the color spectrum. These general terms work well when selecting the color of a dress or car, but the transmittal of exact color information of technical materials between manufacturers and specifiers requires a more precise measurement.

Such color regulations have already been established by the Federal Highway Administration as well as other organizations responsible for specifying; manufacturing; installing; and maintaining the color of traffic signals, stop signs, yellow and orange signs, road markings, and other traffic-regulating objects. Therefore, manufacturers must provide products with colors within specified tolerances, regardless of whether the products were made in Phoenix or Philadelphia and installed in Atlanta or Anchorage.

But this is not all there is to measuring or determining an appropriate color for traffic control devices. In the P&V Lab, a spectroradiometer is used for precise color measurements of light sources. It not only behaves much like the human eye in the sense that it can see the various colors of the visible spectrum, but it breaks the visible spectrum into individual wavelength segments, similar to a prism breaking sunlight into the colors of the rainbow. It can evaluate the amount of energy, or light, contained in each of up to 200 discrete bands. It can also measure the amount of radiation reflected by an object as well as the transmittance of materials, such as filters, under diverse lighting conditions. Transmittance is the fraction of radiant energy that travels through an absorbing matter.

Again, with the use of an integral computer, the "raw" color of the object perceived by the spectroradiometer can then be obtained. Multiplying this data by the known mathematical description of any light source, the P&V Lab can essentially determine what the color of virtually any object will look like under any light source. This information, in terms of x and y coordinates, can then be located on the CIE Chromaticity Diagram. The x and y coordinates for the colors of traffic control devices are thus determined and incorporated into all applicable color standards.

In addition to the spectroradiometer, the P&V Lab has other instruments that can make similar color measurements by different, usually simpler, means. However, some accuracy is traded for faster data collection. Also, the CIE x-y system is not the only one in use; over the past 10 years or so, several new color spaces and descriptions have been developed with the hope of simplifying and visually providing a more understandable color description. While some of these newer color spaces have found their followers, the CIE x-y color system is still the most common and accepted method for describing and specifying traffic control device coloration.

Material Studies
The lab also investigates and determines the physical characteristics of various materials that can be used to enhance visibility at night.

Over the last several years, the lab has studied the life performance of almost 200 retroreflective signing materials representing all manufacturers, colors, and types of materials. These materials were installed on specially built racks positioned to face south with the samples inclined 45 degrees. Their retroreflective values were then measured once a year for six years, and their total color changes over the period were determined. This field test very closely resembles real-world conditions, albeit at an accelerated time rate, to simulate degradation and changes in color over a 10-year period.

The data from this study have been assembled, but a final report has not yet been published. Work currently underway to determine the end-of-life color of signing materials has made use of this test data to develop so-called end-of-life color boxes that specify the amount colors can shift before they must be replaced.

Another material study involved the investigation of the life, color shift, and color rendition of metal-halide lamps used in automobile headlamps. These lamps are gradually entering the high-end or luxury car market. They differ fundamentally from the traditional incandescent or tungsten-halogen headlamps in that they produce light by an electric discharge, or arc, through a gas, rather than by heating a filament to incandescence. While the incandescent lamp produces a high amount of light in the red region, the metal-halide lamp produces a spectrum much closer to daylight — that is, with more blue.

Because FHWA wants to make sure motorists can recognize traffic signs by their color as well as by their shape and legend, the lab investigated the effect of these new metal-halide headlamps on the colors of existing traffic signs and road markings. The investigation showed that some colors, especially in the red and orange areas, do appear slightly different under the metal-halide headlamp as compared to the incandescent headlamp. However, for all colors used for traffic signs, the measured color under a metal-halide lamp is much closer to the color as seen under daylight, and thus, metal-halide headlamps do not create any color confusion.

The P&V Lab also tracks the performance over time of fluorescent materials used in some signs, delineator posts, and white and yellow pavement-marking materials. When fluorescent signs and markings are seen under ultraviolet light, they become considerably more visible. Fluorescent signs in daylight become brighter and more vivid, especially at dawn and dusk and under overcast skies. Fluorescent delineator posts and road markings can be detected at much greater distances when the standard low-beam headlamps of a car are augmented with auxiliary UV headlamps.

Interestingly, ultraviolet light does not only interact with fluorescent materials or pigmentation. Light-colored clothes made of cotton or synthetic fabrics and conventional laundry detergent residue embedded in clothing produce a similar luminosity on a pedestrian. But the visibility of a dark wool suit will not improve under UV lighting.

Unfortunately, the P&V Lab confirmed that most currently available fluorescent pigments have a relatively short life span in an outdoor environment. However, the lab is testing other materials that have a predicted useful life of more than five years. Such a life span is very important for the envisioned ultraviolet/fluorescent application in road markings and sign posts because it is very impractical and much too costly to restripe and/or change signs every six to 12 months.

On a similar front, the lab is also investigating the potential safety benefits of auxiliary headlamps that emit ultraviolet light. Because ultraviolet radiation is not visible to the human eye, ultraviolet auxiliary headlamps on vehicles can be designed to provide a beam pattern similar to a "high beam." Thus, the distance at which traffic control devices and pedestrians with fluorescent additives/clothing can be seen increases substantially.

Additional experiments showed that ultraviolet headlamps also provide better nighttime visibility during rain, fog, and snow because the invisible ultraviolet light does not cause the backscatter, or halo effect in front of the car, normally experienced when driving with conventional headlamps. However, to benefit from this new feature, the conventional headlamps and lamp-spacing on the front of cars must be modified. Results from a small field study of 38 subjects are so encouraging that two larger research contracts were awarded to further investigate and evaluate the potential safety benefits of ultraviolet lighting and fluorescent additives.

In-House Research
The P&V Lab also continues to do in-house research.

During the past five years, the lab participated in an interlaboratory investigation to develop a measurement system, along with recommendations for the instrumentation, to determine the accuracy of measurements of the spectral coefficient of various retroreflective signing materials viewed under headlamps. This effort was sponsored by CIE with the participation of national laboratories in France, Germany, Italy, Japan, The Netherlands, Norway, Sweden, and the United States. The results of the data from the contributing laboratories were found to be in excellent agreement.

Although the project has been completed for some time, only an internal CIE document, describing the measurement procedures and results, has been published and distributed to the participating laboratories. An abbreviated document without the enormous quantity of data collected is planned for use by laboratories engaged in the measurement of color and retroreflective materials. Worldwide adoption of the findings and recommendations will hopefully lead to better agreement on measurements, regardless of where the retroreflective materials were produced and tested.

The lab started an investigation of the effects of rain, ice, and dew on the visibility of retroreflective signs under automobile headlights. Hopefully, the results will help states to select those materials that are least affected by those conditions. The lab is also concerned with changes in the performance of headlamps and how such changes affect sign visibility.

Another study recently conducted by the lab was the investigation of sign recognition by drivers. Approximately 100 subjects participated — both men and women between the ages of 20 and 85. Each subject viewed a series of 25 traffic signs, scaled to reduced sizes, one at a time, in the darkened lab. Each sign was illuminated for just one second, and the subject was asked to identify it. If the response was incorrect, the sign was shown again during a later trial at a higher lighting level. This study provided data in setting and supporting guidelines for in-service retroreflective values of traffic signs.

Similarly, the conspicuity of traffic signs — the degree to which a sign stands out against its background — has continued to be a topic of safety research because traffic signs located in visually complex backgrounds may easily be overlooked by drivers. A future study will test subjects in the P&V Lab in an attempt to operationally define sign conspicuity and develop a mathematical model to predict the results. To carry out this research, a customized slide projection system is being developed to accurately present the wide range of luminance levels present in the nighttime environment.

After the earlier investigations, which were made from static viewing positions, the P&V Lab added a small electric vehicle, much like a golf cart, to its lab equipment. The vehicle is now used for dynamic testing in which the test subject drives toward a given test stimulus or object and records his or her observations on an on-board laptop computer. In this setup, the lab can record the vehicle’s speed, position, and the exact location — relative to the test object — at which the test subject is able to provide the required information.

The procedure has been used in three studies — two pilots and one final — to determine the minimum sign luminance requirements for legibility of overhead guide signs. The results of this investigation have created a lot more questions than answers because it currently appears that the lab results are not consistent with the predictions of existing visibility models.

Sample instruments and materials were evaluated at the laboratory to determine the range within which all currently available portable retroreflectometers agree when measuring signing materials of all grades, colors, and manufacturers.

Conclusion
The Photometric and Visibility Laboratory at the Turner-Fairbank Highway Research Center is conducting significant research to make driving easier and, more importantly, to make driving safer. The work of the lab will continue to be an important part of FHWA’s Research and Technology Program. The P&V Lab is a first-rate facility for measuring and experimenting with vision and various light sources, the retroreflectivity of traffic control devices and fluorescent materials, and the proper visibility of signs under diverse driving conditions. The lab supports the Federal Highway Administration, including its field staff; other agencies within the Department of Transportation; state and local governments; and other domestic and foreign organizations. The ultimate “customer” of the P&V Lab is each and every driver who, as a result of the lab’s work, is better able during periods of limited visibility to see what’s ahead and to avoid the consequences of “missing” a sign - inconvenience, loss of time, frustration, and potential hazards.

John Arens is the manager of FHWA’s Photometric and Visibility Laboratory. He previously worked in the Lighting Branch of the Office of Traffic Operations, and he spent two years assisting the Saudi Arabian Ministry of Communications in lighting designs and setting up a lighting department. For 22 years, he worked for Westinghouse Electric Corp. in engineering and marketing.

Mark Reilly is a freelance writer from Alexandria, Va.