KEY CONCEPTS - DISPLAY TEST & MEASUREMENT

The following is an overview of key terms in display test and measurement, with emphasis on general concepts and concise description. More rigorous definitions, and complete explanations, are accessible via the companion index of resources.

CONTENTS:

• Photometric Measurements
• Luminance
• Contrast Ratio
• Gamma (Gray Scale/EOTF)
• Color Measurements
• Chromaticity
• Color Temperature (CCT)
• Gamut mapping (RGB plotting)
• Dominant Wavelength & Purity
• White Balance, Gray Balance

          Multidimensional Characteristics

• Uniformity
• Spatial Uniformity (Flat-field)
• Angular Distribution
• Reflection
• Veiling Glare
• Specular and Diffuse Reflectance
• More Complex Metrics
• BRDF (Bidirectional Reflectance Distribution Function)
• MTF (Modulation Transfer Function)

Photometric Measurements

Luminance:  A measurement of the brightness of an area on the display surface, as perceived by a standard human observer.  The definition of the standard human observer is the cornerstone of the discipline of photometry. 

Contrast Ratio:  This metric compares the brightest to the darkest output produced by the display under specified conditions.  The precise test conditions required may vary; light and dark readings may be averaged over several regions; ANSI Contrast compares the maximum display brightness to the minimum brightness achievable with the display powered-on.  Other standards allow the dark measurement to be performed with the display power off….

Gamma:  A function which relates the brightness   (luminance) of a display to the corresponding digital or electronic control parameters (grayscale).  For color displays, gamma characterization is performed for each of the three color channels (red, green, and blue) independently.

Color Measurements

Color:  Though color perception is inherently subjective, it is possible to make objective measurements of the color of a source or object as perceived by a standard human observer.  Various standard human observers are defined in the discipline of Colorimetry.  According to the standard model, the perceived color of a given spot can be reduced to a three-dimensional value.  The three dimensions of color can be described in different ways; in perceptual terms, perhaps the most straightforward approach is to describe color in terms of brightness, hue, and purity, or saturation. 

Chromaticity:  A two-dimensional description of color, which corresponds to the combination of hue and purity, omitting the third dimension of brightness.  The luminance (brightness) and chromaticity of a spot on a display, taken together, provide a complete description of its color.

Gamut mapping (RGB plotting):  All colors produced by a display are created by some combination of three primary colors:  Red, Green, and Blue.  In fact, each display color can be described in terms of the amount of R, G, and B primaries present.  (This description is alternative, but equivalent, to a description in terms of brightness, hue, and purity.)  If the chromaticity coordinates of the three primaries are plotted in a chromaticity diagram (see Fig. 4), the triangle enclosed by these points represents the full range of colors reproducible by the display.  This range is the display’s color gamut. 

Correlated Color Temperature (CCT): This metric is used to describe the color of a white light (such as a display backlight) by comparing its chromaticity to that of an idealized incandescent source, known as a black body. The color of an incandescent source (which glows due to heat) depends upon its temperature; lower temperature sources are more red or yellow; higher temperature sources are more blue.  The CCT of a white light is the temperature of the black body which most closely matches its chromaticity.

White Balance, Gray Balance:  After characterizing the chromaticity and gamma functions for each of a display’s three color channels (R, G, B), it is possible to calculate the amount of R, G, and B required to reproduce any color within the display’s color gamut.  A particularly important feature is the location of the white point:  The chromaticity of a specified white light, to which the observer is assumed to be adapted.  Light sources with different CCT have slightly different chromaticities, but once the location of the desired white point is specified, the proportion of R, G, and B primaries required to reproduce it is known as the display’s white balance.  The chromaticity of a neutral gray is the same as that of the white point, but since the gamma functions of a display are typically non-linear, a different gray balance may be required to reproduce the same chromaticity at a lower luminance level. 

Dominant Wavelength & Purity:  These values, taken together, represent an alternative description of chromaticity.  Dominant wavelength corresponds to hue, while purity corresponds to saturation.  The relationship between wavelength and hue can be understood in terms of the colors of the visible spectrum, as observed in the rainbow:  Shorter wavelengths correspond to violet and blue hues; medium wavelengths to greens and yellow; longer wavelengths to orange and red hues.  In the chromaticity diagram (figure 3), hues change in an arc about the central white point, moving clockwise from violet through blue, cyan, green, yellow, orange, and red.  Purity increases as chromaticity moves from the central white point to the outer limit of the spectrum locus: The horseshoe-shaped curve representing the chromaticities of pure spectral light.  Thus, for example, an intensely-saturated red would be plotted near the edge of the diagram, while a pink color of the same hue (or dominant wavelength) would fall near the center.

Uniformity Measurements

An ideal display would render the same luminance and chromaticity for a given output, regardless of position on the screen surface and angle of view.  In practice, all displays are non-uniform to some extent, with output varying as a function of position and angle. Such variation is characterized by mounting a photometric or colorimetric sensor in a positioning stage, which controls the position of the measurement spot and/or the angle of view.

Spatial Uniformity (Homogeneity):  The spatial uniformity of a display is characterized by measuring spots at different positions on the display surface when the image displayed is nominally uniform (same R, G, B values for each pixel).  The variation in actual luminance and/or chromaticity observed is reported and evaluated as a function of position (x,y) on the display screen. 

Angular Uniformity:  The angular uniformity of a display is characterized by measuring a single spot on the display surface from different viewing directions.  The variation in actual luminance and/or chromaticity observed is reported and evaluated as a function of viewing direction.  Viewing direction is specified in terms of two angles:  The zenith angle (q) represents angular displacement from the normal or perpendicular direction; the azimuth angle (f) represents rotation about the normal axis.  (If the display were a clock face, and f were defined as zero degrees at 12:00, then f would equal 30˚ at 1:00, 90˚ at 3:00, 180˚ at 6:00, etc.)

Reflection Measurements

Veiling Glare: A secondary image due to reflection from the display surface that is superimposed upon the primary image rendered by the device.  This secondary image will partially obscure or degrade the primary image.  Such reflections can be specular (mirror-like), diffuse (matte), or something in-between:  Hazy or blurred.  Such glare can reduce the effective contrast produced by a display, or cause an undesired color shift.  For this reason, it is important to characterize the reflectance properties of a display.

Specular Reflection:  Reflection, as from a mirror or polished surface, which preserves the image of the source of the reflected light.  Specular reflectance is a measure of the degree to which a surface will reflect light in this way.  The specular reflectance of a display is measured by illuminating the display surface with a relatively small, well-defined source, and collecting light from the reflected image, which is confined to a specific direction.

Diffuse Reflectance: Reflection, as from a matte or rough surface, in which the reflected light is perfectly scattered, so that no image of the source of reflected light is produced.  Instead, the reflected light renders the entire surface of the display itself visible as a “veil” through which the primary image is seen.  Diffuse reflectance is a measure of the degree to which a surface will reflect light in this way.  Diffuse reflectance of a display may be measured by illuminating the display surface as in the measurement of specular reflectance, and collecting the light reflected in all directions.  The result may include both a specular and a diffuse component; in this case, it is necessary to subtract the contribution of specular reflectance to obtain the diffuse reflectance.

Haze:  Reflection which produces a blurred image of the source of the reflected light.  Since all real images are blurred to some extent, and few surfaces are perfectly diffuse, the practical definition of haze is somewhat arbitrary.  Haze is defined as an intermediate case between specular and diffuse reflection, in which reflected light is scattered by more than some arbitrary small angle (q_min), and less than some arbitrary large angle (q_max) from the ideal specular path.  Like diffuse reflectance, haze may be characterized by making measurements with sources and sensors of different sizes, then separating specular, diffuse, and haze components by analysis.  A more rigorous approach is to use a goniometer to measure the BRDF of the display surface (see below).

More Complex Metrics

Bidirectional Reflectance Distribution Function (BRDF):  A measurement of the reflection properties of a surface for given direction of illumination and direction of view.  As in the measurement of angular uniformity, the viewing direction is specified in terms of two angles:  The zenith angle (q_v) and azimuth angle (f_v).  But here, the direction of illumination (q_i, f_i) must also be specified.  BRDF is measured by mounting a light source and sensor on two positioning stages, and moving each independently with respect to the sample surface. A complete BRDF characterization of any surface is quite data-intensive; it can be both time-consuming to perform, and difficult to interpret.  For this reason, such testing is often abridged, e.g. by limiting the number of directions of illumination.

Modulation Transfer Function (MTF):  For various reasons, the image reproduced by a display may be somewhat blurred; sharp edges between dark and light areas may be softened, with a gray region of transition between them.  If a pattern of fine lines is displayed, the contrast, or modulation, between dark and light areas may be reduced.  Typically, the narrower and more closely-spaced the lines, the more pronounced this effect will be.  The MTF of a display describes such a loss of contrast as a function of spatial frequency – that is, the number of lines in a given distance.  Typically, MTF is measured by displaying an image with a sharp edge, and measuring the luminance of the display at several small, closely-spaced points on either side of that edge.  The result is known as an “edge-spread function.” The MTF of the display can be calculated by analysis of this result, and used to predict the effect of the display upon any other type of image.