Displays
INTRODUCTION
Digital images are viewed, judged and manipulated using display systems. The display technology that has dominated up until now, the cathode ray tube display (CRT), has gradually been replaced by the liquid crystal display (LCD). This has enabled the use of displays in many devices such as portable computers, mobile phones and digital cameras. The two display types have significant technological differences which affect the perceived quality of the displayed images. Other technologies have also emerged, such as organic light-emitting diodes (OLEDs) and flexible displays.
IMAGE DISPLAY
Display of images on a computer monitor is performed via a graphics card. The image data in the computer is in the form of a stream of binary digits (bits). This data is converted by the graphics card into a form suitable for input to the display device. The graphics card communicates with the central processing unit (CPU) of the computer via the motherboard. The necessary power for the graphics card is provided either from the motherboard or from a connection to the power supply of the computer. The graphics card has its own processing unit, which performs the necessary operations for image display. It also has random access memory (RAM), which is used to store the image data. Part of the RAM is a frame buffer in which the whole image is stored before it is displayed on the computer monitor. The size of the frame buffer depends on the number of pixels that comprise an image and the number of bits that are associated with each pixel. The output signal values from the memory are converted via colour look-up tables (LUTs), to analogue signals. This is achieved by the digital-to-analogue converter (DAC). The LUT data range from 0 to (2N − 1) levels, where N is the number of DAC bits. Three analogue RGB video signals are formed and sent to the CRT display. The connection between the graphics card and the monitor is a digital video interface (DVI) for LCDs and video graphics array (VGA) for CRT displays.
CATHODE RAY TUBE (CRT) DISPLAYS
Cathode rays have been studied since the late nineteenth century. They were first investigated by Sir William Crookes, who developed the Crookes tube. Based on the Crookes tube, Professor Karl Ferdinand Braun in 1897 invented the first cathode ray tube where the bending (deflection) of the electron beam, the cathode rays, was controlled electromagnetically. He also used phosphors for light emission. The CRT display in some countries is called a Braun tube. The British physicist Joseph John Thompson also developed a CRT at the same time but the difference from the Braun tube was that it employed two deflection plates producing electrostatic deflection. Thompson had also experimented with other types of deflection.
A CRT display consists of an electron gun, a focusing system, a deflection system and a screen coated with phosphors, which emit light when excited by electrons. A beam of electrons with high velocity, a cathode ray, is produced by the electron gun and is focused on the screen, forming a small dot. The position of the dot on the screen is controlled by a deflection system (see Figure 15.1).
Figure 15.1 A CRT display system. A, cathode; B, modulator; C, focusing system; D, deflection plates (for electrostatic deflection); E, deflection yoke (for electromagnetic deflection); F, phosphor coating; G, faceplate.
The electron gun consists of the cathode, a cylinder with an insulated heating element (filament), a control grid and an anode, which accelerates the electrons. Current passing through the filament causes it to heat, resulting in the emission of electrons in an electron beam (cathode ray). The electron emission is increased by a calcium or strontium coating at one end of the cathode cylinder. There is a minimum time necessary for the filament to be heated and emit electrons. This is known as the warm-up time of the CRT display.
In a colour CRT display, colour is reproduced using the RGB additive system (see Chapter 5). Three electron guns, corresponding to the red, green and blue voltage signals, are focused on separate red-, green- and blue-emitting phosphors. There is no distinction between the three electron guns; each simply controls an electron beam for its corresponding colour. However, in the Trinitron CRT display technology developed by Sony (the first Trinitron colour television entered the Japanese market in 1968), only one electron gun is used for all three signals. The output of one electron gun and therefore the existence of one channel at a given pixel, however, may not be independent of the other channels at that pixel and this may affect the colours displayed on the screen. Research has shown that colour inaccuracies may occur due to limitations of the amplifier or the power supply and it depends on the luminance level. It has also been shown that the higher the luminance, the greater the effect.
The intensity of the electron beam is controlled by the control grid or modulator, from which the electron beam passes through to reach the phosphor coating of the screen. The amount of emitted light by a phosphor depends on the number of electrons that excite it. By varying the voltage of the, slightly negative, control grid the number of electrons (intensity of the electron beam) which excite the phosphors can be controlled, and consequently the luminance of the display. The electrons that exit the control grid pass through the anode cylinder. The anode has a positive potential, which accelerates the electrons.
An electron lens focuses the electron beam to the screen by applying different potentials. The electron lens has similar characteristics and limitations to the optical lens, such as spherical aberration (see Chapters 6 and 10). There are two focusing systems with an electron lens: an electromagnetic (EM) and an electrostatic (ES) system. In the electromagnetic system a magnetic focusing coil external to the CRT envelope is used to focus the electron beam. The resulting spot size is very small, smaller than the spot obtained with the electrostatic system. With the electrostatic system the focusing is achieved using an electrostatic lens, which is a built-in metal cylinder. This system is the most commonly used in commercial colour CRTs. Both focusing systems focus the beam in the centre of the screen. The CRT display screen has a curvature so the focusing distance is not the same for all the points of the screen’s phosphor coating. For this reason it is necessary to use additional focusing systems.
The location of the focused spot on the screen is controlled by the deflection system and, again, there are both electromagnetic and electrostatic types of system. The electromagnetic system employs two pairs of electromagnetic coils and an electromagnetic yoke. The two pairs of coils provide horizontal and vertical deflection, which is controlled by varying the current that passes through them. Most CRT displays which are used to view images have electromagnetic deflection systems which provide higher luminance compared to the ones with electrostatic deflection systems and very small spot size (the effective diameter of the spot is defined by its diameter at half maximum of its intensity (FWHM)). CRT displays with electrostatic deflection systems have two pairs of deflection plates for horizontal and vertical deflection inside the cathode ray tube envelope. These systems provide higher speed of deflection but the deflection angle has to be smaller than the angle in the electromagnetic systems, otherwise the electron beam may be defocused. Of course the smaller deflection angle results in longer CRTs.
The screen on which the electron beam is focused is coated with phosphors, materials which luminesce when they are excited by electrons and phosphoresce (continue to glow) once the excitation stops. The afterglow fades slowly and this varies depending on the type of phosphor. There are three types of phosphors according to the duration of afterglow emission: short persistence, medium persistence and long persistence. The persistence of a phosphor is measured as the time it takes for its light emission to decrease to 1% of its maximum intensity. Long-persistence phosphors eliminate the effect of flickering. When the change of the displayed content is rapid, however, such as in that of moving images, an after-image effect is produced. Short-persistence phosphors eliminate this effect. Since in colour CRT displays there are three types of phosphors it is essential that they all have matching persistence. In most CRT displays the phosphor persistence is around 5 ms.
Figure 15.2 Typical spectral power distribution of P22 phosphors.
The maximum light output of the phosphor depends on the acceleration of the electrons of the beam and on the characteristics of the phosphor itself, meaning that different phosphors may emit different amounts of light, even if excited by the same number of electrons with the same acceleration. The age of the phosphor also affects the intensity of the emitted light due to physical degradation caused by the cumulative effect of many electron collisions. In addition, when an electromagnetic deflection system is used, degradation is also caused by ions striking the phosphor surface. The light output of the phosphors can be improved by coating the back of the phosphor surface with a metallic layer, mainly from aluminium or beryllium. The layer of the coating is very thin, around 0.127 mm. Screens which employ this method are called metal-backed phosphor screens.
Phosphors are designated with P values according to the Electronic Industries Association (EIA). A common type of phosphor for commercial colour CRT displays is the P22. This is a set of red (YVO4:Eu – yttrium orthovanadate activated with europium), green (ZnS:Cu, Au, Al – zinc sulphide activated with copper, gold and aluminium) and blue (ZnS:Ag – zinc sulphide activated with silver) phosphors. The red P22 phosphors have their peak emission at 626 nm, the green at 535 nm and the blue at 450 nm (Figure 15.2).
The three phosphors are arranged in triads, with (theoretically) each triad forming a pixel. There are two different types of electron gun arrangement depending on the phosphor triad technology being used, the ‘delta’ and the ‘in-line’ arrangements (Figure 15.3). The magnetically deflected electron beams from the three electron guns with a ‘delta’ arrangement reach the phosphor surface through a steel mask, a shadow mask with dot geometry (Figure 15.3a). With the shadow mask, the beam from the electron gun reaches only the phosphors with the corresponding colour and the other phosphors remain in shadow. It should be noted that a large portion of the electron beam, around 70%, does not pass through the shadow mask holes. The distance between the phosphors is the same and phosphors of the same colour form an equilateral triangle. When the CRT display has an ‘in-line’ electron gun arrangement a slot mask is used with three vertically aligned slots for each triad, as shown in Figure 15.3b. In this technology, used by NEC in the ChromaClear CRT displays and Panasonic in the PureFlat CRT displays, the red, green and blue phosphors are elongated and vertically aligned. In Trinitron technology an aperture grille is used (Figure 15.4), comprising thin vertical wires instead of a metal mask and corresponding phosphor stripes. In this case the screen is less curved compared to the screen of displays using a shadow mask or a slot mask.
Figure 15.3 Electron gun arrangements and shadow mask geometry. (a) ‘Delta’ arrangement and shadow mask with dot geometry. (b) ‘In-line’ arrangement and shadow mask with slot geometry.
Figure 15.4 Trinitron technology with aperture grille.
The distance between two identically coloured phosphors is called the pitch. The smaller the distance, the sharper and brighter the image is. The measurement of pitch depends on the phosphor triad technology. In dot triad technology the pitch is measured diagonally between the centre of two nearest-neighbour, identically coloured phosphor dots (Figure 15.5a). The distance is often called dot pitch and is typically 0.27 mm or less. Due to the fact that CRT display screens have a curvature, the electron beam that passes through a hole in the centre of the shadow mask will form a circle on screen, while if it passes through a hole at the edge of the shadow mask it will form an ellipse. Consequently the dot pitch in the centre of the screen will slightly differ from the dot pitch at its edges. For this reason manufacturers quote an average dot pitch or they may give two values, one for the centre and one for the edges of the screen. The term mask pitch is used to describe the distance between two holes in the shadow mask corresponding to two identically coloured phosphor dots; this has a slightly smaller value than the dot pitch. In displays with slot masks, the pitch is measured as the horizontal distance between two identically coloured phosphor stripes and is called slot pitch (Figure 15.5b). The distance between two phosphor stripes with identical colours in aperture grille technology is referred to as stripe pitch (Figure 15.5c).
Accurate convergence of the three electron beams is very important for a CRT display. Convergence, however, is not always perfect and this may have an effect on the quality of displayed images. The effect of misconvergence is more pronounced at the edges of the screen, while the best results appear in the central area. Research has shown that the central area is also the most uniform regarding luminance and chromaticity while the largest deviations usually occur at the edges of the screen. One of the reasons is the above-mentioned curvature of the CRT display screen, which results in variability in the distances for the electron beam to travel. Other reasons include non-uniform application of the phosphors on the screen, temperature and magnetic fields. Magnetic fields exist internally in the CRT displays that use electromagnetic deflection. External magnetic fields, however, may cause slight changes in the path of the electron beams but usually the result is not perceptible. Turning off and unplugging nearby electrical appliances can prevent the effect of magnetic fields. Regularly degaussing the monitor also minimizes the effect. In cases where there is a strong magnetic field an external shielding box for the CRT display can be used, or a shielded room. Taking into account the human visual system’s contrast threshold to low spatial frequencies (see Chapter 4), however, the lack of uniformity for most displays is not perceived.
Figure 15.5 Phosphor pitch for the three different phosphor triad technologies: (a) dot pitch; (b) slot pitch; (c) stripe pitch.
Figure 15.6 Variation of luminance and chromaticity during the warm-up time of a CRT display.
When viewing images on a CRT display it is essential to allow the display to warm up. This is usually for at least 30 minutes, so that it reaches its highest level of luminance. Warm-up time is also necessary for stabilized luminance and colour output, essential when viewing images. Figure 15.6 illustrates the variation in luminance and CIE xy-chromaticities of a CRT display which has been allowed to warm up for a period of time. Different types of CRT displays need different times to stabilize after being powered up and it has been estimated to be from 15 minutes to over 3 hours. The stabilization time for a specific device can be determined by repeated measurements of the display luminance and chromaticity over an extended but specified period of time.
LIQUID CRYSTAL DISPLAYS (LCDs)
The discovery of liquid crystals dates back to the late 1800s. It was not until much later, in the 1960s, however, that research by the Radio Corporation of America finally led to the development of the first LCDs. In 1888 the Austrian botanist Friedrich Reinitzer observed the properties of cholesteryl benzoate, a material which changed state upon heating. The change of state occurred at two different temperature points: from solid to cloudy anisotropic liquid (mesophase) at temperature T1 and from cloudy liquid to clear isotropic liquid at a higher temperature point T2. Otto Lehmann, a professor of physics, observed that in meso-phase the material in liquid form showed characteristics of a crystal and for this reason he termed it liquid crystal. These types of liquid crystals are called thermotropic because their state changes with temperature (as opposed to liquid crystals, whose state changes on reaction with water, which are called lyotropic and are studied in fields such as biochemistry). The molecules of thermotropic liquid crystals have either a rod-like shape (calamitic) or a disc-like shape (discotic). In LCDs, liquid crystals with a rod-like shape (their length is approximately 2 nm) are used, but in recent years the discotic type has also been used for improving the viewing angle. Three types of thermotropic liquid crystals exist, known as smectic, nematic and cholesteric (Figure 15.7). Liquid crystals have anisotropic properties, meaning that they have physical properties that are directionally dependent. There are two forms of anisotropy in liquid crystals: optical anisotropy, which refers to the different magnitudes of refractive indices in different directions; and dielectric anisotropy, which refers to the dielectric constant, depending on the axis on which they are measured and specifically on the orientation of the liquid crystal’s molecules.
Most current LCDs are based on the twisted nematic (TN) system, which was described in the early 1970s by Fergason and by Schadt and Helfrich. This system is based on nematic liquid crystals. The molecules of the nematic liquid crystals have parallel order and their favoured orientation is represented by the director, which is a vector n. As mentioned above, they are characterized by their optical and dielectric anisotropy. The optical anisotropy serves to adjust the polarized light in an orientation which is suitable for display. This is as a result of double refraction or birefringence. Dielectric anisotropy is exploited by applying an electric field to change the direction of the liquid crystal molecules; when the electric field is applied crystals with positive dielectric anisotropy will align parallel to the field, while crystals exhibiting negative dielectric anisotropy will align in a perpendicular orientation.
A thin layer (4–10 μm) of liquid crystals with positive dielectric anisotropy is used in the TN system. This layer is between two glass substrates which have conductive electrodes in their internal surfaces. The surfaces have orienting or alignment layers, which are organic or inorganic films used to provide the appropriate orientation of the molecules, resulting in a 90° twist of the director (Figure 15.8). Two cross linear polarizers (see Chapters 2 and 10), usually made of polyvinyl acetate (PVA) with iodine doping, are used to control the light entering and exiting the liquid crystal layer. As described in Chapter 2, light travels in the form of electromagnetic waves. When it passes through two polarizers, the intensity It of the light transmitted through both polarizers depends on the angle θ, the mutual angle between the polarizing axes of the two (i.e. Malus’ law):
Figure 15.7 Types of thermotropic liquid crystals: (a) smectic; (b) nematic; (c) cholesteric.
Figure 15.8 (a) The director is twisted by 90°, resulting in rotation of the polarization direction and light passes through the second polarizer (normally white case). (b) The orientation of the director changes when an electric field is applied. When it becomes perpendicular to the second polarizer’s polarization direction no light can pass through the second polarizer. A, light; B, rear polarizer; C, orientation surfaces; D, liquid crystal molecules; E, front polarizer.
where Ii is the incident light to the first polarizer. It should be noted that, in ideal polarizers, the light that passes through the first polarizer is 50% of the incident light. If the angle θ is equal to 0° then the light transmitted from the first polarizer will pass also through the second polarizer, so 50% of the incident light will be transmitted. If the angle θ is equal to 90° (i.e. orthogonal polarizing axes) no light will be transmitted through the second polarizer. In practice, however, there is some light absorption on the optical axis (dichroism) when the light passes through the polarizers and this has an effect on the contrast ratio of the LCD.
LCDs are backlit; white light first passes through a diffuser, which produces uniform illumination due to scattering, and it then enters the liquid crystal layer via the rear polarizer. As it enters, its polarization rotates with the liquid crystal molecules. If there is no electric field applied, when the light reaches the second (front) polarizer with polarization axis orthogonal to the first one, its polarization direction has rotated by 90° as a result of the twist of the director. It can therefore exit passing through the second polarizer, which is outside the second glass plate. This is known as the normally white mode. When an electric field is applied to the liquid crystals the orientation of the director changes and tends to have an orientation angle greater than 0° with respect to the polarization direction of the second polarizer. In this case the amount of transmitted light is reduced. The light transmission can therefore be moderated by adjusting the applied voltage. When the voltage is sufficiently high the rotation of the liquid crystals can be reduced to 0°. In this case the light cannot pass through the second polarizer and is therefore blocked (Figure 15.8). If the polarizing axes of the two polarizers are parallel, the light exiting from the first polarizer is transmitted through the second polarizer when electric field is applied and is blocked when there is no voltage applied. This is known as the normally black mode.
The first TN LCDs were reflective-type displays, where the ambient light was necessary to read text or numbers. It was not possible to display large images because each pixel had to be individually connected to the addressing circuitry. This meant that for an image array of m rows and n columns, the interconnections would be m × n. Other drawbacks were the parallax errors and difficulty in displaying colours. The display of images became possible with the development of matrix addressing of each pixel, where the number of interconnections was reduced to m + n. These were passive matrix displays, which improved over the years. They had limitations, however, regarding the response time, viewing angle and contrast ratio. A solution to this involved independent control of the voltage of each pixel. This was accomplished by adding a switch at each pixel in a matrix display. This was the active matrix LCD (AMLCD), which is the most common LCD technology today. The on/off switches in the first AMLCDs were CdSe thin-film transistors (TFTs). The materials for TFTs changed through the years to polycrystalline silicon (poly-Si) materials and in the 1990s to amorphous silicone (a-Si) materials, which are used in displays for portable and desktop computers. Poly-Si TFT LCDs started being used again in the late 1990s for small displays. It should also be noted that, for better control of the cells’ twist, nematic crystals are ‘doped’ with cholesteric crystals (see Figure 15.7.).
Figure 15.9 LCD colour filter structure. A, clear substrate; B, black matrix; C, colour filter layer; D, overcoat layer; E, ITO.
In current LCD technology the light sources are cold cathode fluorescent lamps (CCFLs), which have a phosphor coating. The colours of the phosphors are red, green and blue (RGB), resulting in white light emission with distinctive peaks for the three colours and luminance around 3000 cd m−2. Colours are displayed by using a pixel array consisting of three subpixels with RGB filters (additive colour system – see Chapter 5). The filters have suitable surfaces to minimize dispersion of the light (Figure 15.9). In TFT LCDs the filter array is approximately 4–5 μm away from the TFT array (this is the thickness of the liquid crystal layer). The filters consist of a plastic or glass substrate, a black matrix between the colour filters, an RGB layer where the colours are pigments or dyes, a protective overcoat layer which also minimizes any variations in the filter’s thickness, and a film of indium tin oxide (ITO). The black matrix, usually of chromium combined with chromium oxide, prevents leakage of light between the pixels and protects the TFTs from light exposure to ambient light. The spectral transmittance of the RGB filters closely matches the above-mentioned three peaks of the emitted white light from the CCFLs and is one of the parameters that affects the white point of the display. Another parameter is the colour purity of the filters. It is important that the filters have stability, especially during heating and exposure to light. Also, they should have chemical stability because they are exposed to chemical substances during the manufacture of the LCDs.
OTHER DISPLAY TECHNOLOGIES
Plasma display panels (PDPs)
Plasma displays are based on cells filled with gas at low pressure. The gas is neon (Ne) or helium (He) with added 5–15% xenon (Xe). The cells are between two glass plates. Each cell forms a pixel, with three subpixels with red, green and blue phosphors. The phosphors used are of the same type as the phosphors of the CRT displays and cover all the inner surfaces of the subpixels apart from the front. The difference is that in plasma displays the phosphors are excited by ionization of the gas (plasma). Xenon produces ultraviolet light, in a similar way to the neon tubes, with a wavelength of 147 or 173 nm. The ionization is achieved by applying a voltage via the grid of electrodes which address each pixel. The subpixels are isolated with barrier ribs to avoid the effect of crosstalk between them. The phosphors are also insulated by the data electrodes which are on the back glass plate, perpendicular to the scan, and sustained electrodes located on the front glass plate.
Because the pixels of plasma displays are neither ‘on’ nor ‘off’, the grey levels are controlled by adjusting the fraction of time for which a pixel will be ‘on’ and using eight subfields for addressing the frame.
The colour gamut (range of reproduced colours) of plasma displays is similar to that of the CRT displays. Due to their technology they are also thin and lightweight and are used for large TV sets. They have a very long lifetime, usually over 100,000 hours. However, as in CRT displays, the phosphors age; this affects the displayed luminance, which is reduced over time, and the display colour balance. Power consumption of plasma displays depends on the luminance of each pixel of the displayed images. Although if all pixels are driven with high luminance the power consumption will be very high compared to TFT LCDs, in practice the average pixel luminance is much lower, approximately 20% or less, so the plasma display consumes less energy.
Organic light-emitting displays (OLEDs)
OLEDs are electroluminescent (EL) displays and have two glass substrates, a thin organic layer, and electrodes created by metal and ITO films. Colour is produced by light emission by the OLED materials at different wavelengths. The light emission of an OLED is almost Lambertian (i.e. the radiance is directly proportional to the cosine of the angle, with respect to the direction of maximum radiance, from which the display is viewed) and for this reason the viewing angle is very wide, approximately 170%. With this technology the power consumption is low because it depends on the luminance level of the displayed images (as in plasma displays). Two types of OLEDs exist. These are small-molecule OLEDs and polymer OLEDs (or PLEDs). The PLED technology has received the most interest and development. This is due to limitations of the small-molecule OLED technology for depositing the organic material on large surfaces.
Depending on the way pixels are addressed there are passive matrix OLEDs and active matrix OLEDs (AMOLEDs). Passive matrix OLEDs are not suitable for large-size displays because they are slow and an increase in resolution decreases the luminance to an insufficient level. In an AMOLED there are two TFTs per pixel and the pixels are controlled by adjusting the current (and not the voltage). The AMOLED pixel is illustrated in Figure 15.10. One of the limitations of the OLEDs is that the luminance of the display decreases over time and this also affects the display colours. So far OLEDs have been used as small screens on consumer products such as consumer cameras.
Figure 15.10 Structure of an AMOLED pixel. A, anti-reflection coating; B, cathode; C, organic layers; D, anode; E, glass substrate.
Flexible displays
Flexible display technology (or electronic paper or e-paper) has been developed to approach the appearance of paper. Flexible displays have low power consumption. Based on the thickness of the display, there are flexible displays that curve or very thin ones that can be rolled. This feature makes rollable displays suitable for several applications where easy storage and light weight are important factors. The technology of flexible displays was initially based on liquid crystals. Passive matrix and active matrix flexible displays have been developed, with the glass substrates replaced by flexible materials. As mentioned previously, however, the polarizers significantly reduce the transmitted light so these types of flexible display cannot closely approach the appearance of paper. A different technology, the electrophoretic display, uses microcapsules with white positively charged particles and black negatively charged particles in liquid. The capsules are deposited, with the method of printing, on a plastic film which has a grid of circuits, forming an array of pixels. The black and white dots on the display appear by controlling the voltage. With positive voltage applied, the black particles move to the top of the microcapsule and the viewer sees a dark spot. With negative voltage applied, the viewer sees a white spot because the white particles move to the top of the capsule (Figure 15.11). The resulting white in this technology has reflectance of around 30–50%. The interferometric modulator technology (IMOD) by Qualcomm is based on the principle of constructive and destructive interference (see Chapter 2). In this technology, a thin-film stack and a reflective layer are on a transparent substrate. Incident light is reflected from both the film and the membrane. Depending on the distance between them, there is constructive or destructive interference of the light waves. Red, green and blue colours are produced by varying the distance so that constructive interference will occur only for the wavelengths that correspond to these colours (see Figure 15.12). Black is obtained by applying voltage to the film. This produces electrostatic force that causes the membrane to touch it. In this case the constructive interference is at the, invisible, ultraviolet wavelengths. Greyscale is obtained by spatial or temporal dithering or a combination of both. Flexible OLED is another technology that has been proposed.
CHARACTERISTICS OF DISPLAYS
Refresh rate and response time
Refresh rate refers to CRTs and is the rate (in Hz, i.e. times per second) at which a CRT monitor screen displays the data. Refreshing at a low rate may cause the effect of flickering, which is uncomfortable for the user and can lead to eye fatigue. The reason for flickering in CRTs is the phosphor decay. The electrons emitted by the electron gun excite the phosphors, which emit light. This light begins to decay until the phosphors are bombarded again by electrons. Flicker can be avoided by setting on the computer a refresh rate higher than the critical fusion frequency (CFF). CFF is the frequency higher than which flickering is not perceived. The CFF depends on ambient lighting and on the viewer. The setting at which most people do not perceive flickering is 70 Hz.
Figure 15.11 Electrophoretic display technology. A, top electrode; B, bottom electrode; C, clear fluid.
Figure 15.12 Interferometric modulator technology. A, ambient light; B, thin-film stack; C, air gap; D, deformable membrane; E, glass.
Due to the different technology employed, refresh rate does not apply to LCDs. The equivalent characteristic is response time. It expresses the time needed for the liquid crystals to pass from the aligned to the twisted position and back to the aligned position. Response time is defined as the time needed to change state from white (10% transmittance) to black (90% transmittance) and back to white (10% transmittance), and is usually around 10–50 ms. A long response time results in blurring when fast-moving images are displayed. There are, however, limitations when very short response times are used because they can cause flickering.
Resolution
In a display device resolution expresses the number of pixels per inch (ppi) or the number of pixels in the horizontal and vertical dimensions (e.g. 1024 × 768 pixels) it can display. In the latter case it is also important to know the physical dimensions of the display screen. Addressability expresses the number of points that can be addressed by the graphics card adaptor and in CRT displays it is independent of resolution. There are several factors that affect resolution in CRT displays. One of these factors is the spot size, which depends on the phosphor layer but also on the electron beam current and the optics of the display system. The ratio of resolution to addressability (RAR) can be calculated by taking into account the pixel pitch, p (display height divided by the number of addressed lines), and the spot size, s:
The focusing and deflection systems used also affect effective resolution (see Chapters 19 and 24). The shadow mask is yet another factor. High luminance of the display is related to longer spot diameter, reducing the effective resolution. Due to the Gaussian profile of the light emitted by the phosphors, a CRT monitor can display multiple resolutions by adjustment of the electron beam. It should be noted, however, that changes of resolution have an effect on the refresh rate of the display. Some typical display resolutions are shown in Table 15.1. Reducing or increasing the resolution setting of a CRT display may have an effect on image quality. The effect is scene dependent and it may affect image areas with high frequency where detail may be lost at a lower resolution setting. Changing display resolution also alters the dimensions of the displayed images (at a higher resolution the image appears smaller and vice versa).
LCDs have a matrix of pixels on the screen and for this reason their resolution, which is referred to as native resolution, is the same as the addressability. Optimal image quality is obtained only when the display resolution of the LCD is set to its native resolution. Although the driver may allow alteration of the display resolution setting, in an LCD this may cause blurring of the image due to applied interpolation (see Chapters 23 and 27).
Table 15.1 Typical display resolutions
RESOLUTION |
NUMBER OF PIXELS |
QCIF |
144 × 173 |
QCIF+ |
220 × 176 |
CGA |
200 × 320 |
QVGA |
240 × 320 |
CIF |
288 × 352 |
VGA |
480 × 640 |
NTSC |
480 × 720 |
WVGA |
480 × 800 |
PAL |
576 × 768 |
SVGA |
600 × 800 |
WSVGA |
600 × 1024 |
XGA |
768 × 1024 |
WXGA |
768 × 1280 |
SXGA |
1024 × 1280 |
SXGA+ |
1050 × 1400 |
W-HDTV |
1080 × 1920 |
UXGA |
1200 × 1600 |
WUXGA |
1200 × 1920 |
QXGA |
1536 × 2048 |
WQXGA |
1600 × 2560 |
QSXGA |
2048 × 2560 |
QUXGA |
2400 × 3200 |
WQUXGA |
2400 × 3840 |
Perceived resolution decreases with viewing distance, i.e. displayed spatial frequencies in the visual field become higher and higher as the viewing distance increases. The required resolution for viewing images on a computer monitor where typical distance is approximately 50 cm is up to 180 ppi, and therefore different to that for television viewing at typical distance of 3 m. In that case the required resolution would be up to 30 ppi. Higher resolutions would not improve image quality, provided that distances are kept constant.
Luminance
The level of luminance set for a display depends on the ambient lighting conditions, so it may vary between displays which are viewed in different locations. Typical maximum luminance for a CRT display is around 100 cd m−2. In TFT LCDs the luminance depends on the light intensity of the CCFLs, with typical luminance ranging from 150 to 300 cd m−2. There is a significant reduction of luminance, around 90–95%, compared to the light output from the CCFLs. The reduction occurs due to losses when the light passes through the rear and front polarizers, the black matrix and the colour filters. Although the output luminance of an LCD can be increased by increasing the output of the backlight source, in practice this could result in higher temperature of the display and higher power consumption. Other methods which can be applied to increase the output luminance have been suggested. These include improvements on the light transmission of the colour filters (they transmit around 25% of the incident light), redesign and rearrangement of the colour filters so that they include a white filter, designing the black matrix with higher aperture ratio, and use of brightness enhancement films. In plasma displays the luminance is around 1000 cd m−2.
Contrast ratio
Contrast ratio (CR) is the ratio between the maximum luminance (peak white), Lmax, to the minimum luminance (black), Lmin, that a display system can produce (see also Chapter 21):
There are two methods of assessing contrast ratio: small-area (pixel) and large-area (group of pixels) measurements. Contrast ratio, however, is affected by the ambient lighting conditions and the environmental conditions. It also depends on the screen’s reflectance (screen glass and phosphors for CRTs) or diffusion of the incident ambient light. Other parameters, for LCDs, are the viewing angle and the wavelength (depending on whether the LCD is in normally white or normally black mode). Measurements may be carried out for a range of viewing angles and the results plotted in an isocontrast ratio diagram. When measured in a totally darkened room the contrast ratio is called intrinsic contrast ratio. The contrast ratio may also be measured with the presence of ambient lighting (extrinsic contrast ratio). In that case it is essential to quote the measured ambient light illuminance and the angle of incident light in addition to the luminance values of the display. For CRTs the extrinsic contrast ratio can be expressed by the following equation:
where Lr is the reflected light from the glass and r is the phosphor reflectivity, typically 70%. A typical value for the intrinsic contrast ratio of a CRT display is around 100:1, while for an LCD it can be around 500:1. Plasma displays have a higher intrinsic contrast ratio, 5000:1. With ambient light present the contrast ratio of a CRT display is reduced and with typical lighting conditions is around 20:1. There are several different methods used for the measurement of contrast ratio by manufacturers, however, so the contrast ratio values quoted may not be comparable.
Viewing angle
The reproduced range of grey levels, the contrast ratio and the colour in LCDs depend on the viewing angle. The effect is more pronounced on the vertical angle, while the horizontal viewing angle is more symmetric. This is a result of the different orientation of the liquid crystals’ director for positive and negative vertical angles. The liquid crystals are tilted when voltage is applied and their direction is different when the screen is viewed from an upper or lower angle. In the normally white mode an inverted image may be observed when the vertical viewing angle changes. This is more obvious in the lower viewing angles. The effect of the viewing angle can be significant when viewing and manipulating images using an LCD, and also when the display is viewed simultaneously by several viewers, or when a large display is viewed by a single user. The viewing angle should also be taken into account when conducting colorimetric or luminance measurements. A colorimeter specially designed for LCDs should be used, measuring at a very narrow angle (see Chapters 5 and 23).
In recent years several methods have been developed to improve viewing of LCDs from different angles. One of these methods is the in-plane switching (IPS) mode, introduced by Hitachi. The IPS mode is based on the normally black mode. Two orthogonal polarizers are used but the liquid crystals are aligned when there is no electric field applied and thus the light cannot pass through the second polarizer (Figure 15.13a). When an electric field is applied the orientation of the director changes and light is transmitted. Maximum light transmission occurs when the angle between the director and the polarizer is 45°(Figure 15.13b). The liquid crystals are parallel to the glass substrates so there is no variation in the orientation of the director. The response times of the IPS LCDs used to be slow in the early models, but this has been improved. The aperture ratio of the cell is also lower, resulting in the use of more powerful backlight sources. Recent improvements of this technology include the True White IPS by LG Philips and Super IPS (S-IPS) by Hitachi. An alternative is vertical alignment (VA), where the liquid crystals are at right angles to the glass substrate when there is no electric field (Figure 15.14). Another method for viewing angle improvement is the use of compensation films with discotic liquid crystals, developed by Fuji Corporation. With the compensation films the retardation of the light that passes through the liquid crystals does not depend on the viewing angle, thus improving the contrast ratio. CRT and plasma displays have a very wide viewing angle duo to the fact that phosphors have Lambertian light emission. Flexible displays also have a very wide viewing angle.
Figure 15.13 In-plane switching technology. (a) When there is no electric field applied the liquid crystals are aligned and no light passes through the second polarizer (normally black case). (b) When electric field is applied light is transmitted because the orientation of the director changes. A, light; B, rear polarizer; C, orientation surfaces; D, liquid crystal molecules; E, front polarizer.
Colour
The colour gamut of a CRT display depends on the spectral characteristics and luminance of the RGB phosphors. The colour gamut produced by the P22 set of phosphors is illustrated in Figure 15.15.
In LCDs the colour gamut depends on the spectral transmittance of the RGB filters and on the spectral characteristics and luminance of the backlight source. Perceived colour also depends on the colour filter arrangement (Figure 15.16). Each arrangement has advantages and limitations. The vertical line arrangement is the most popular for computer displays and is less complicated than the other two. Colour integration, however, is not as good as with the RGB diagonal or ‘delta’ arrangements.
Figure 15.14 Vertical alignment method for LCDs. A, front polarizer; B, colour filter plate; C, ITO; D, liquid crystal molecules; E, ITO; F, TFT plate; G, rear polarizer.
Figure 15.15 Colour gamut of the P22 phosphors.
The number of colours addressed by the graphics card is related to the number of bits (and therefore grey levels) that correspond to each pixel. For an 8-bit system, typical for a computer monitor, the grey levels are equal to 28 per colour while high definition LCDs employ a 10-bit system resulting in 210 grey levels per colour. In general the number of colours in an LCD is 23n, where n is the number of bits. In an optimum situation, the number of colours the display is capable of reproducing should correspond to the number of addressable colours. However, the number of displayed colours depends on the contrast ratio of the display and the chromaticity of the display primaries. If the contrast ratio is high enough and the primaries are saturated enough, each addressable colour value can be displayed as a different chromaticity and/or luminance level. This is rarely the case. Generally, for a specific set of display primaries, smaller gamuts are reproduced when the contrast ratio is reduced.
Display artefacts
Artefacts depend on the display technology used. Flickering has already been mentioned for CRTs. It can also appear in LCDs due to the backlight. Temporal aliasing (a stroboscopic effect) can occur if the frame rate of the screen is slower than the rate the data change (moving images). Strobing may also occur if the refresh rate of the CRT display is set to the same rate that fluorescent lamps flicker. To avoid this effect the refresh rate should be higher. Spatial aliasing is an artefact related to the spatial resolution of the display and therefore its ability to display high frequencies, resulting in a moiré effect (see Chapter 7). The point spread function of the pixel is affected by the electron optics. Increase of the spot’s diameter may cause image blurring. Image retention (burn-in) may occur in CRTs or plasma displays due to phosphor degradation when images or text are displayed unchanged for a long time. This results in a ghosting image effect. Image retention may occur in LCDs for different reasons. The DC components which are used to apply voltage to the liquid crystals may have an effect and cause short-term image retention. Long-term image retention may be caused by changes in the alignment layer or the current of the TFTs.
Figure 15.16 RGB colour filter arrangements for LCDs. (a) RGB stripe. (b) RGB diagonal. (c) RGB ‘delta’ or triad. (d) RGBW stripe. (e) RGBW quad.
In LCDs additional artefacts include mura, cross-talk, pixel defects and motion blur. The term mura includes several artefacts caused by the process of LCD manufacturing. It includes the existence of larger or smaller than normal cells in some areas of the screen (causing brighter or darker areas respectively), inhomogeneity in the circuits that control the pixels, and differences in processing of the liquid crystal’s alignment layer. The artefact of crosstalk, mainly in passive matrix LCDs, occurs when the output of one or more pixels is affected by the data of neighbouring pixels (they may be on the same column or row). Pixel defects are subdivided by the International Standards Organization into three categories: hot pixels (always on, resulting in a white spot), dead pixels (always off, resulting in a dark spot) and stuck pixels (one of the subpixels is always on or always off). Due to the response of the human visual system hot pixels in a dark background are easier to detect than dead pixels in a white background. Motion blur may be caused when a fast-moving object is displayed because the voltage applied on a pixel remains until the next refreshing of the data.
EFFECT OF VIEWING CONDITIONS
The viewing conditions have a significant role in the perceived quality of images viewed on displays. The level of ambient illumination and the colour temperature of the illuminant have an effect. As mentioned previously, the contrast ratio decreases with increase of the ambient light illumination due to flare. This makes the discrimination of figure–ground by the human visual system more difficult because when the ambient light is increased, it needs higher rather than lower contrast. Anti-reflection coatings can reduce the effect of flare. A hood can also be attached to some displays, isolating the screen from the ambient lighting. It also serves to isolate the screen from the surround and background, which influence the adaptation of the visual system. The gamma setting of the display is also related to the displayed image contrast (see Chapter 21).
Characterization and calibration of the display devices used for image viewing is essential for accurate colour reproduction of the displayed images. In many cases, however, and especially when images are viewed across the Internet, displays are used in different locations, with different brightness, contrast and white point settings, and under different ambient lighting. The sRGB standard has been developed to ensure accurate colour and tone reproduction of images when viewed on CRT displays under reference display and viewing conditions (see Chapter 23). These include parameters typical to most CRTs such as the white point (phosphor RGB chromaticities), the display luminance level and the ambient light illuminance. Visual adaptation (see Chapter 5) is an important factor in viewing of displayed images. Research has shown that when images were viewed on a CRT display the human visual system was approximately 40% adapted to the ambient light and approximately 60% to the monitor’s white point.
BIBLIOGRAPHY
Collings, P.J., Hird, M., 1997. Introduction to Liquid Crystals – Chemistry and Physics. Taylor & Francis, London, UK.
Damjanovski, V., 2005. CCTV: Networking and Digital Technology, second ed. Elsevier Butterworth-Heinemann, Oxford, UK.
Den Boer, W., 2005. Active Matrix Liquid Crystal Displays – Fundamentals and Applications. Elsevier Newnes, Oxford, UK.
Dorf, R.C., 1997. The Electrical Engineering Handbook. CRC Press, Boca Raton, FL, USA.
Driggers, R. (Ed.), 2003, Encyclopedia of Optical Engineering, Vol. 2. Marcel Dekker, New York, USA.
Holst, G.C. 1996. CCD Arrays, Cameras and Displays. JCD Publishing, Winter Park, and SPIE Optical Engineering Press, Washington, DC, USA.
Hunt, R.W.G., 2004. The Reproduction of Colour, sixth ed. John Wiley, Chichester, UK.
Karim, M. (Ed.), 1992. Electro-Optical Displays. Marcel Dekker, New York, USA.
Kawamoto, H., 2002. The history of liquid-crystal displays. Proceedings of the IEEE Vol. 90 (No. 4) April.
Kelly, S.M., 2000. Flat Panel Displays – Advanced Organic Materials. The Royal Society of Chemistry, Cambridge, UK.
Li, Z., Meng, H., 2007. Organic Light-Emitting Materials and Devices. Taylor & Francis Group CRC Press, Boca Raton, FL, USA.