 |
Description  |
|
|
RELATED APPLICATION
The invention described herein is related to U.S. patent application Ser.
No. 07/178,949 filed on Apr. 7, 1988, entitled APPARATUS AND METHOD FOR AN
ELECTRONICALLY CONTROLLED COLOR FILTER FOR USE IN INFORMATION DISPLAY
APPLICATION, invented by L. D. Silverstein and A. J. Bernot and assigned
to the assignee of the present application.
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to flat panel color displays and, more
particularly, to displays in which the image is the result of a mosaic of
pixel regions.
2. Description of the Related Art
Liquid crystal mosaic display technology is being developed as a possible
successor to color cathode ray tubes (CRTs) in many display applications,
including those applications in the avionics field. This technology offers
important advantages such as higher reliability along with reduced power,
size and weight. But in the current state of development of the liquid
crystal technology, capability of this technology for the rendering of an
image falls short of the image capability achievable using CRT technology.
This invention addresses three specific problem areas still remaining in
liquid crystal mosaic displays: color definition; image resolution; and
display brightness. In terms of color definition, the liquid crystal
mosaic display color rendition suffers from effects similar to those
observed on a misaligned CRT display tube. The primary hues, the red,
green and blue colors, do not blend properly. A white line, for example,
appears to have multicolored fringes, symptomatic of deficient color
synthesis. Part of the problem can be attributed to the symbol generator
which controls the formation of graphics on the flat panel. However, part
of the problem can also be attributed to the display itself, a
contribution addressed by this invention.
In terms of image resolution, graphic symbols and lines appear excessively
jagged or discontinuous on color mosaic displays, especially when compared
with lines drawn on calligraphic color CRT systems. Again part of the
image resolution problem can be attributed to the symbol generator while
the display panel itself also provides a contribution. A major part of the
contribution from the display panel is the result of the presence of blue
pixels as part of the display pixel mosaic. Referring now to FIG. 1A, the
low degree of spatial sensitivity that the human visual system has for
blue light as compared to the other primary colors is illustrated. The
eye's peak response to blue light occurs at about one half the frequency
of peak response for the red radiation and half again the frequency for
green radiation. This result indicates that blue radiation contributes
only a minor amount to image shape and spatial detail. As a result, blue
pixels on the display surface of the panel tend to degrade the overall
resolution capability of color mosaic displays, a feature addressed by the
present invention.
With respect to display brightness, the origin of the problem can be
attributed to both the pixel arrangement of the panel and the current
backlight technology used in liquid crystal displays. The backlight
technology includes the lamp and the electronics controlling the backlight
lamp. The chief figure of merit for achieving a given level of brightness
is how much power is needed to achieve that brightness level. Research is
being aggressively pursued to make backlight technology more efficient.
The present invention, however, addresses the brightness problem from a
different perspective. Once again, the pixel arrangement on the surface of
the flat panel display can account for a considerable portion of the
problem. Blue pixels contribute little to the total perceived luminance of
the panel display. The photopic response of the eye accounts for this
phenomenon. FIG. 1B illustrates that red and green radiation provide a
larger contribution to perceived brightness than blue radiation. Blue
radiation can typically provide only about a ten percent contribution to
the overall brightness of the panel.
Referring next to FIG. 2, the effect of having blue pixels occupying space
in the pixel arrangement is shown. Wherever a blue pixel is present, the
effect on the pattern of pixels is to occlude the perceivable luminance
passing through the display surface. No appreciable contribution to
luminance capability is available at the sites of the blue pixels. As a
result, these blue pixel regions of FIG. 2 can be considered as black
regions. These regions occupy thirty percent of useful area in a typical
Red/Green/Blue (RGB) pixel mosaic arrangement.
In order to compete successfully with the cathode ray tube technology in a
multiplicity of applications, the liquid crystal mosaic displays must
evolve to the point where they efficiently achieve enough brightness to
prevent bright sunshine from washing out displayed information.
Additionally, they must also exhibit higher resolution and improved color
mixture attributes for higher quality imagery to be displayed. Achieving
these goals has proven difficult in the past.
A wide range of techniques have been implemented in flat panel display
technology to alleviate the problems described above. Listed below is a
description of the principal approaches for solving color definition,
image resolution and display brightness problems in the liquid crystal
mosaic displays.
Generally, color image synthesis in liquid crystal mosaic displays use
either additive or subtractive techniques (cf. the above identified
related application). Additive techniques use spatial proximity, temporal
superposition or spatial superposition techniques to mix primary hues into
different colors. Additive spatial proximity methods are the most common
approach used in liquid crystal flat panel technology. FIG. 3 illustrates
the basic technique of spatial proximity. Small dots (pixels) of primary
colors, typically red, green and blue, are evenly dispersed across the
surface of the flat panel display. If the dots (pixels) are small enough
and close enough, then the eye fuses or integrates the contribution of
each color dot together with its neighbors. The additive method can
achieve enhanced resolution by making the pixels smaller and more densely
packed. Additionally, the differently colored pixels can be arranged into
different patterns, in hopes of striking a better fit with the
characteristics of the human visual system. Full color imagery is
therefore perceived. Excellent resolution can result because each pixel is
capable of full color control and full luminance control. Additive spatial
proximity, the method generally preferred throughout the industry, suffers
three serious drawbacks, outlined above in the problem discussion. Color
definition is faulty in the case of computer generated imagery (unless
signal processing methods are used) resulting in color fringing and
rainbows effects. As the pixels are made smaller, color integration is
improved but light output is worsened because a greater percentage of the
primary display area gets consumed by address lines and interconnecting
conductors. In addition, blue contributes very little to perceived
brightness yet consumes typically one quarter to one third the active
display area as indicated previously. Blue also detracts from resolution
capability, limiting edge definition and image sharpness. The three
principal problems with this approach then are: (1 ) poor color
integration and (2) wasted luminance and (3) wasted resolution.
In additive temporal superposition methods, the primary hues are rapidly
sequenced before the eye. FIG. 4 shows one possible sequence. First, the
red portion of the image is flashed on the flat panel display, then the
green portion of the image is flashed on the flat panel display and, a
short time later, the blue portion of the image is flashed on the flat
panel display. Successful color synthesis using this temporal additive
technique depends on the limited temporal frequency response of human
vision. If the sequencing occurs rapidly enough, the eye cannot discern
the separate primary hues, but, instead, perceives their overall
integrated image. Temporal superposition suffers from smearing effects,
jitter and image instability as the observer shifts his viewing position
rapidly or vibration induces similar motion. In addition, todays liquid
crystal materials exhibit such slow optical response times, rapid temporal
sequencing using them is virtually impossible.
In additive spatial superposition methods, separate images, each comprised
of only one primary hue, are optically fused into one full color image.
Typically three images, corresponding to red, green and blue hues, are
used. These separate images are formed from three separate image sources.
The output images of these three sources are then fused by optics into one
full color image to be viewed by the observer (cf. FIG. 5). Excellent
resolution is typical of this approach because each pixel is capable of
full color and full luminance control. Brightness can also be high since
three image forming sources are operated in parallel. Additive spatial
superposition techniques suffer from complexity problems and performance
difficulties. These systems also tend to be prohibitively large for many
applications, especially those of the aerospace market. Cost generally
rises due to the fact that three separate imaging devices are needed. Then
additional hardware must be used to combine the three images. Frequently,
this hardware must be extremely precise and rigid to maintain color
purity.
In subtractive display apparatus (illustrated in FIG. 6), white (broad
band) radiation is passed through successive layers of complimentary color
filters, each layer being electrically controlled for absorbing a
well-defined region of the spectrum. By modulating the voltage applied to
each layer, different portions of the white light spectrum can be occluded
or, in the alternative, be allowed to pass through unimpeded. This
spectral control, the ability to withdraw selectively different portions
of the spectrum, can be used to synthesize full color imagery. Resolution
can be excellent with this approach because full color control is
available at the site of each pixel. Subtractive methods suffer from an
expected higher cost, parallax effects and complex methods for color
control. At least three and possibly four separate liquid crystal panels
are needed to make a subtractive superpositional liquid crystal display.
Currently, this additional complexity is viewed as too costly. In
addition, parallax can be troublesome using this technique. As the viewing
angle is changed with respect to the display, each layer of pixels in the
panel is viewed from a slightly different position. Pixels on different
layers of the display will be observed to move with respect to each other.
Lines can vary substantially in perceived thickness, due to head motion
alone. Effects such as these, which are functions of viewing angle, are
unacceptable for many (e.g., aerospace) applications. Finally, color
control has proven to be particularly troublesome to date. Experiments
indicate that, unless better dyes, backlighting or even a fourth layer can
be developed, colors cannot be tracked over a broad range of ambient
lighting conditions. The difficulty lies in the complex interrelationships
between hue and luminance. One interferes with the other in a non-linear
manner which currently has proven very difficult to predict.
A need has therefore been felt for a liquid crystal display unit that
provides increased display brightness, increased image resolution and
better color reproduction.
FEATURES OF THE INVENTION
It is an object of the present invention to provide an improved color
display device.
It is a feature of the present invention to provide an improved flat-panel
mosaic display device.
It is a further feature of the present invention to provide for green and
red color components by spatial proximity of pixels while the blue color
component is provided by pixel subtraction.
It is another feature of the present invention to provide a mosaic display
device in which a first panel transmitting red and green light through a
mosaic of pixels have a second panel aligned therewith through which blue
light is transmitted in a mosaic of pixels.
SUMMARY OF THE INVENTION
The aforementioned and other features are accomplished, according to the
present invention, by providing a liquid crystal display system in which a
first panel has liquid crystal pixel elements that control the
transmission of red and green image components and a second panel, aligned
with the first panel for which liquid crystal pixel elements control the
transmission blue image components therethrough. The first panel controls
the red and green color components by additive spatial proximity
techniques. The second panel controls the blue image component by
subtractive techniques. Because of the reduced sensitivity of the eye to
blue color components, the pixel array of the second panel can have
diminished resolution and can have a diminished refresh rate compared to
the first panel.
These and other features of the invention will be understood upon reading
of the following description along with the drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1A illustrates the eye's lower spatial frequency for blue as compared
to other primary colors.
FIG. 1B illustrates the reduced sensitivity to blue radiation as compared
to radiation of the other primary colors.
FIG. 2 illustrates the regions (blue pixels) that do not contribute to the
mosaic display luminance.
FIG. 3 illustrates how the eye integrates neighboring pixels to provide a
full color spectrum from primary hues.
FIG. 4A and FIG. 4B illustrate temporal integration of a sequence of
primary color images to provide a complete image.
FIG. 5 is a block diagram illustrating the development of an image using
spatial superposition of image portions.
FIG. 6 illustrates a controllable filter for creating a color image by
removing selected portion of broad band optical transmission passing
therethrough.
FIG. 7 is a schematic diagram of a first panel of an additive/subtractive
display panel of the present invention.
FIG. 8A illustrates the passband characteristics of a magenta filter; FIG.
8B shows the passband characteristics of a cyan filter; and 8C illustrates
the passband characteristics of dichroic filter for transmitting blue
light.
FIG. 9 is a schematic diagram of a second panel of the additive/subtractive
display of the present invention.
FIG. 10 is a cross-sectional view of the additive/subtractive display
system according to the present invention.
FIG. 11 is a table illustrating the colors available with the
additive/subtractive display panel of FIG. 10.
FIG. 12 is a CIE diagram illustrating the colors that can be achieved using
the additive subtractive display system of the present invention.
DESCRIPTION OF THE PREFERRED EMBODIMENT
1. Detailed Description of the Figures
Referring now to FIG. 7, a first panel 1 of the additive/subtractive
display is shown. The liquid crystal display has an MxN matrix of pixels,
each pixel being addressed by one of M column conductors and one of N row
conductors. The M column conductors are selected by x-axis column bus
drive unit 2 in response to groups of address signals, Wx, and the N row
conductors are selected by y-row bus drive unit 3 in response to groups of
address signals, Wy. The intersection of activated column conductors and
an activated row conductor activates the associated pixels along the row.
Either active matrix methods or multiplexing methods can be used to
activate the pixels, techniques known in the related art. In the active
matrix approach, active devices such as thin film transistors or metal
insulator metal diodes are used as switching or non-linear devices to
control the storage of charge across each pixel. In multiplexing methods,
no active device is present. The relationships of voltage signals on row
bus lines with respect to voltages present on column bus lines controls
charge storage across each pixel which in turn controls the state of the
pixel (liquid crystal) optically active material. This method depends on
the sharp electro-optical threshold of the liquid crystal material itself
rather than on the threshold behavior of intervening electronic devices
used in active matrices.
Still referring to FIG. 7, the color pixel arrangement of the top layer of
the additive subtractive display is illustrated. Pixels of one
complimentary primary hue, (magenta filter) pixel 4, for example, are
distributed in a checkerboard pattern with pixels of another complimentary
primary hue (cyan filter) pixel 5. The checkerboard pattern is used by way
of example and is not meant to be limiting. The optical passband
characteristics of the magenta and cyan filters are shown in FIG. 8A and
8B. No blue primary hue pixels are present in the pixel pattern of the top
surface. The use of only magenta and cyan filter pixels leads to the
resolution and brightness advantages described earlier in the disclosure.
Referring next to FIG. 9, the second panel 6 of the additive/subtractive
display is shown. The matrix is shown with half the number (M/2.times.N/2)
of rows and columns used in the first panel 1. This degraded resolution is
possible because of the lower spatial resolution capability human vision
has for blue light modulation as compared to red and green light
modulation. Except for this spatial resolution difference, the drive units
and address waveforms used for second panel 6 can be identical to those
used for the first panel 1. Another difference is possible, however. The
refresh rate of images displayed on the second panel 6 can be lowered
relative to the refresh rate of the first panel 1 because human vision is
less sensitive to blue light in terms of temporal resolution as well as
spatial resolution. When red/green images are displayed on a display
device whose images decay with time, the red/green images need to be
refreshed periodically at a typical frequency of 60 Hz. This critical
fusion frequency allows the eye to integrate the flashing images into a
steady scene absent of flicker or image decay. Blue light images can be
seen without flicker at much lower refresh frequencies, 10 Hz for example.
This phenomenon can be used to have the overall effect of lowering the
drive requirements for the blue color portion of the system.
Still referring to FIG. 9, the color pixel arrangement of the bottom layer
6 is shown. Each pixel 7 controls the passage of blue light through the
display. Each pixel can block blue light from passing through or,
alternatively, can be energized so blue light can pass through unimpeded.
In order to accomplish this, a guest host dichroic liquid crystal material
(Merck ZLI2010 for example) is used. The passband characteristics of this
liquid crystal material are shown in FIG. 8c. The passband spectral
characteristic is variable and is a function of applied voltage. If
voltage of one value is applied, then the passband is 9 as shown by FIG.
8c. No blue light is allowed to be transmitted. Other spectral components
(red and green), however, can pass through freely, giving this filter a
yellow hue when placed before a broad band light source (white light).
When, on the other hand, another appropriate voltage is applied, then all
wavelengths, including blue, are allowed to pass freely through the
dichroic material. The passband is shown by dotted line 8 plus solid line
9 in FIG. 8c. Therefore, if the second panel 6 is placed before a broad
band (white) light source and all the pixels are activated with the
appropriate voltage, then the light passed through the second panel 6
appears blue-white instead of yellow. To obtain white, one half the pixels
in the lower layer are activated. This activation provides an improved
balance of the blue contribution to the overall spectral output. The
ability of the bottom layer 6 to modulate blue light on a pixel by pixel
basis, yet pass other spectral components is used together with the first
panel 1 to produce full color images.
Referring next to FIG. 10, the cross section of a total
additive/subtractive display unit 100 incorporating both the first panel 1
(of FIG. 7) and the second panel 6 (of FIG. 9) is shown. The
additive/subtractive display unit 100 includes a glass plate 14 and a
glass plate 13 which enclose a region 16 containing dichroic material
combined with a liquid crystal material. The glass plate 13 and filter
plate 15, which includes a mosaic of magenta and cyan filters, enclose a
region 12 containing a liquid crystal. Pixel control devices 10 in region
16 and pixel control devices 11 in region 12 are also shown. These pixel
control devices 10 and 11 can be active matrix control devices (thin film
transistors or metal/insulator/metal diodes for example) or,
alternatively, can represent the intersection points of the row/column
electrodes of multiplexed display technology described above. The blue
control devices 10 are shown with twice the spacing and, therefore, half
the resolution of the red/green control devices 11. This reduced element
spacing reflects the physical differences of human vision resolution for
resolving colors described previously. Alternatively, the resolution of
the second panel 6 can be identical to that of the first panel 1. This
configuration has the effect of adding a higher degree of fault tolerance
capability in addition to increased blue light resolution capability.
Several blue panel control devices and their associated blue pixels can be
activated simultaneously in this configuration to cover the same area as a
lower resolution blue control pixel. Therefore, some measure of redundancy
or immunity to local failure can be achieved. The blue panel control
devices 10 provide control over each pixel, enabling the
additive/subtractive display to control the passage of blue light over the
area of each cell or pixel 7 of FIG. 9 in the matrix of the second panel
6. Each pixel can either permit white light or yellow light to pass
through the second panel 6 to the first panel 1. The red/green control
devices 11 control the passage of light through the top layer 1. The
red/green control devices determine whether any light is free to pass
through the red-blue (magenta) 4 and green-blue (cyan) 5 filters located
on the color filter surface of the top panel 1. Consequently, this layer
not only controls hue but, significantly, controls brightness. Each pixel
can be controlled to render gray shades as well as hue. This capability
builds on the advantageous methods for rendering gray shades established
by additive technology and avoids the distinctly complex
luminance/chrominance interrelationships manifest in subtractive
technology, alone.
Referring next to FIG. 11, the operating conditions of pixels in each panel
needed to produce the display colors listed is shown. For example, if a
black pixel 12 is desired, then the condition of the second panel 6 is
inconsequential because any light passing through this panel will be
blocked by the off condition of the magenta and cyan pixels in the first
panel 1. If a white region 13 is desired, then one half of the pixels of
the second panel 6 must be in the pass "yellow plus blue" (white) state
and the magenta 4 and cyan 5 pixels in first panel 1 must be fully ON.
Although the panel in actuality produces only discrete magenta and cyan
pixels in this state, the eye fuses them into white because of the close
proximity of these hues. For this fusion to be successful, the pixels must
be small enough and close enough to fall within the integration zone of
the eye. In order to produce a red region 14, the second panel 6 must be
placed in the yellow or minus blue state 9 shown in FIG. 8. The magenta
pixels 4 in the first panel 1 must be ON and the cyan pixels 5 in the
desired region must be placed in the OFF state.
Referring next to FIG. 12, a CIE color chart illustrating the range of
colors which can be produced by gray level control of all pixels. If the
blue control pixels 7 of the second panel 6 are varied from the pass blue
state 8 plus 9 (all wavelengths are passed yielding white) to the block
blue state 9, then each magenta pixel 4 moves from the magenta state to
red along line 15 and each cyan pixel 5 moves from cyan to green along
line 16 shown in FIG. 12. If blue is blocked entirely by pixels on layer
6, and if the cyan pixels 5 are ON and the magenta pixels are varied along
a continuum from OFF to ON, then colors along line segment 17 will be
produced, ending in yellow when the resultant green and red pixels are
finally integrated by the eye. Conversely, if blue is again blocked by the
second panel 6, but, this time, the magenta pixels 4 are fully ON and the
cyan pixels 5 are varied from OFF to ON, then colors along line segment 18
will be produced. Again the final color is yellow. When all pixels are
allowed to vary from one extreme state to the other, all the colors within
region 19 can be produced. In addition, by taking advantage of additive
methods, luminance can be varied as well. Images can be shaded, shaped and
contoured in much the same manner as is used with additive juxtaposition
mechanisms. It is the red/green pixels of additive methods which provide
the brightness and shape information. Similarly, it is the magenta/cyan
pixels of the additive/subtractive display system which provide the same
kind of information.
2. Operation of the Preferred Embodiment
The present invention differs from prior art by tailoring the panel in
better accord with the human visual system. Specifically, the invention
addresses the fact that human vision relies almost exclusively on non-blue
light, the red colors and the green colors, for spatial and intensity
information. Further, the invention takes into account the fact that the
eye uses blue light energy almost exclusively for chromatic information
alone. Because blue light contributes very little to spatial detail and
brightness, the invention removes blue light control from the principal
display surface and dedicates this surface to the brighter and more
resolvable red and green pixels. To achieve a wide range of colors, blue
light control is placed behind the principal imaging layer, using
techniques developed recently for the subtractive superpositional
approach. Together, these two image planes add and subtract light to
synthesize higher resolution, higher brightness images with a broad range
of colors. The invention combines the simplicity and superior luminance
control of one approach (additive juxtaposition) with the resolution
enhancement of the second approach (superpositional subtractive).
By taking better advantage of the human visual system, the display system
of the present invention produces up to one third more luminance and
resolution capability than predominant methods without incurring the
volume and complexity cost of the other methods. Blue pixel control,
useful for color synthesis, but wasteful for brightness and image
sharpness, is relegated to a secondary surface. This control leaves the
primary display surface free to display the highly useful red and green
pixels. These two display surfaces are sandwiched together into one
compact flat panel display. Because imagery is not rapidly sequenced as a
function of primary hue, the invention does not suffer from the temporal
anomalies of temporal superposition approaches. Because it does not
require recombination optics, it has a size advantage over spatial
superposition methods. Finally, because it is primarily an additive
display and uses only two imaging layers with blue on the secondary layer,
it delivers more manageable luminance and color control, less complexity
and parallax stability than the subtractive approach.
The foregoing description is included to illustrate the operation of the
preferred embodiment and is not meant to limit the scope of the invention.
The scope of the invention is to be limited only by the following claims.
From the foregoing description, many variations will be apparent to those
skilled in the art that would yet be encompassed by the spirit and scope
of the invention.
* * * * *
|
|
 |
|
 |
Description |
|
