U.S. patent application number 09/966527 was filed with the patent office on 2002-01-24 for method and apparatus for displaying computer generated holograms.
This patent application is currently assigned to Nippon Telegraph. Invention is credited to Akimoto, Takaaki, Higuchi, Kazuhito, Horikoshi, Tsutomu, Suzuki, Satoshi.
Application Number | 20020008887 09/966527 |
Document ID | / |
Family ID | 27454887 |
Filed Date | 2002-01-24 |
United States Patent
Application |
20020008887 |
Kind Code |
A1 |
Horikoshi, Tsutomu ; et
al. |
January 24, 2002 |
Method and apparatus for displaying computer generated
holograms
Abstract
A method for displaying computer generated holograms of a
display object is performed by computing fringe patterns produced
by light interference from the display object. The steps are
summarized as follows: three-dimensional data of the display object
are converted into computational data for fringe pattern
generation; a sampling rule for sampling computational data is
selected; computational data are sampled according to a selected
sampling rule; wavefronts generated by light illumination are
computed by assuming that each sampled position has a light source;
fringe patterns generated by computed wavefronts and a reference
beam are computed; fringe patterns are stored as hologram images;
sampling and a wavefront generation are repeated for all data; and
a series of hologram images thus generated are displayed
successively.
Inventors: |
Horikoshi, Tsutomu; (Tokyo,
JP) ; Higuchi, Kazuhito; (Tokyo, JP) ;
Akimoto, Takaaki; (Tokyo, JP) ; Suzuki, Satoshi;
(Tokyo, JP) |
Correspondence
Address: |
PENNIE AND EDMONDS
1155 AVENUE OF THE AMERICAS
NEW YORK
NY
100362711
|
Assignee: |
Nippon Telegraph
|
Family ID: |
27454887 |
Appl. No.: |
09/966527 |
Filed: |
September 28, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
09966527 |
Sep 28, 2001 |
|
|
|
09083687 |
May 21, 1998 |
|
|
|
Current U.S.
Class: |
359/9 ;
359/22 |
Current CPC
Class: |
G03H 2210/441 20130101;
G03H 1/0808 20130101; G03H 2210/30 20130101; G03H 2240/62 20130101;
G03H 2210/452 20130101; G03H 2210/62 20130101; G03H 1/2294
20130101; G03H 2210/56 20130101; G03H 2001/2297 20130101 |
Class at
Publication: |
359/9 ;
359/22 |
International
Class: |
G03H 001/08; G03H
001/26 |
Foreign Application Data
Date |
Code |
Application Number |
May 22, 1997 |
JP |
9-131531 |
May 22, 1997 |
JP |
9-131532 |
Jan 20, 1998 |
JP |
10-008161 |
Mar 2, 1998 |
JP |
10-049093 |
Claims
What is claimed is:
1. A method for computing fringe patterns of a display object
comprised by items and displaying computer generated holograms,
comprising the steps of: converting three-dimensional data of said
display object into computational data for fringe pattern
generation; selecting a sampling rule for sampling computational
data; sampling computational data according to a selected sampling
rule; computing wavefronts by assuming that each position of
sampled three-dimensional data has a light source and generates a
wavefront; computing fringe patterns generated by interference of
computed wavefronts and a reference beam; storing fringe patterns
as hologram images; repeating a step of sampling and a step of
generating a wavefront for all computational data; and displaying
successively a plurality of hologram images thus generated.
2. A method according to claim 1, wherein a process of converting
three-dimensional data into computational data includes generation
of vertex coordinates for surfaces of said display object.
3. A method according to claim 1, wherein a process of converting
three-dimensional data into computational data includes conversion
of said three-dimensional data into voxel data.
4. A method according to claim 1, wherein a process of converting
three-dimensional data into computational data is performed for
each of said items, and said sampling rule is altered in accordance
with an attribute of an item to be displayed.
5. A method for computing interference fringe patterns of a display
object comprised by items and displaying computer generated
holograms, comprising the steps of: inputting three-dimensional
data of said display object into computer means; classifying or
grouping input data and computing a plurality of fringe patterns
for each classified or grouped display object; converting said
plurality of fringe patterns respectively into a plurality of
digital images; decomposing said plurality of digital images into
individual bits to form bit images; synthesizing bit images
obtained for each classified or grouped display object to produce
moving pictures for display; and displaying said moving
pictures.
6. A method according to claim 5, wherein computation of fringe
patterns is performed by classifying said display object according
to an attribute, and computing fringe patterns for each classified
or grouped object.
7. A method according to claim 6, wherein conversion to digital
images include a step of selecting an information content according
to attributes of said items so as to prepare bit images having a
number of attribute bits related to each information content; while
generation of moving pictures for display includes a step of
assigning bit images for each classified or grouped object by
distributing bit images to a plurality of field layers of moving
pictures.
8. A method according to claim 6, wherein conversion to digital
images include a step of selecting an level of information
complexity according to attributes of display items so as to
prepare bit images having a number of attribute bits related to
each information content; while generation of moving pictures for
display includes a step of assigning bit images for each classified
or grouped object by distributing bit images to a plurality of
field layers of moving pictures so as to preserve intensity of
displayed images.
9. A method according to claim 6, wherein generation of moving
pictures for display is carried out by extracting bit images having
high-bit levels to characterize attributes of said items, and
distributing and assigning those bit images to a plurality of field
layers of moving pictures.
10. A method according to claim 5, wherein conversion to digital
images is performed for same levels of attribute bits, and said
moving pictures are produced by adding bit images produced from
digital images for each level of attribute bits so as to prepare a
number of fields of moving pictures corresponding to each level of
said attribute bits, and altering coding mode of field layers in
different fields.
11. A method according to claim 5, wherein generation of moving
pictures is performed by extracting predetermined high-level bit
images from said digital images, and assigning individual pixels of
said high-level bit images to a plurality of field layers to
produce moving pictures for display.
12. An apparatus for displaying computer generated holograms of a
display object comprised by items by computing fringe patterns
produced by light interference comprising: a display object data
input section for inputting three-dimensional data of said display
object; an item managing section for converting three-dimensional
data into computational data for fringe pattern generation, and
determining a selection rule for sampling said computational data,
and sampling said computational data according to a selected
sampling rule; an image generation section for computing a
wavefront for each sampled computational data by assuming that each
sampled computational data has a light source, and producing fringe
patterns generated by light interference of computed wavefronts
with a reference beam as hologram images; and an image display
section for repeating sampling and wavefront generation for all
display items of said display object and successively displaying
hologram images thus produced.
13. A method according to claim 12, wherein said item managing
section converts three-dimensional data into computational data
including a generation of vertex coordinates for surfaces of said
display object.
14. A method according to claim 12, wherein said item managing
section converts three-dimensional data into computational data
including a conversion of said three-dimensional data into voxel
data.
15. A method according to claim 12, wherein said item managing
section converts three-dimensional data into computational data for
each of said items, and said sampling rule is altered in accordance
with attributes of said items.
16. An apparatus according to claim 12, wherein said apparatus is
further provided with an image memory section for storing hologram
images produced by said image generation section, and a
transmission section for successively transmitting stored hologram
images stored in said image memory section.
17. An apparatus for displaying computer generated holograms of a
display object comprised by items, comprising: a display object
data input section for inputting three-dimensional data of said
display object; an image generation section for classifying or
grouping said display object, computing fringe patterns formed by
light interference of a computed wavefront with a reference beam
for each classified or grouped display object, converting a
plurality of computed fringe patterns into respective digital
images, decomposing said respective digital images into individual
bits, and forming moving pictures for display by synthesizing
decomposed bits; and an image display section for successively
displaying said moving pictures for display.
18. An apparatus according to claim 17, wherein said apparatus is
further provided with an item managing section for classifying said
display object according to attributes of said items; wherein said
image generation section is comprised by: an information content
decision section for determining an individual information content
for each display object according to attributes of said display
item; a wavefront computation section for computing fringe patterns
for each display object; a bit decomposing section for converting
said fringe patterns into digital images of different levels of
attribute bits and decomposing bit arrays of individual pixels into
pixel arrays for each bit-level; a display interval decision
section for determining display cycles and sequence of presentation
of objects according to said information content; and wavefront
synthesizing section for synthesizing fringe patterns formed by
pixels from said pixel arrays for each bit-level according to a
pre-determined sequence of display cycle and display sequence; and
said image display section comprises: a display screen
synchronizing section for controlling display timing of synthesized
fringe patterns; and an image display section for successively
displaying synthesized fringe patterns according to a controlled
display timing.
19. A method according to claim 18, wherein said apparatus is
provided with: an image transmission section for transmitting
images of synthesized fringe patterns for static items first and
transmitting images of synthesized fringe patterns of dynamic items
successively later, and an image receiving section for storing
images of said static items and for synthesizing stored images of
static items and successively transmitting images of dynamic items;
wherein said display image synchronizing section controls display
timing for synthesized fringe patterns in said image receiving
section and said image display section displays fringe patterns
synthesized in said image receiving section.
20. An apparatus according to claim 17, wherein said image
generation section comprises: a fine pattern computation section
for computing fringe patterns for display items; a digital image
generation section for converting a computed fringe section into a
digital image and generating a pixel array for each bit-level from
individual pixel bits; and a image sequece generation section for
generating image arrays for displaying moving pictures by selecting
high-level bit pixel arrays of items to be displayed.
21. An apparatus according to claim 17, wherein said image
generation section is comprised by: a gray level image generation
section for generating holograms of gray scale images; a bit image
generation section for decomposing said gray scale images into
individual bits to generate bit images; an image storage section
for storing bit images; an image processing section for processing
images in such a way that a bit image to be presented repeatedly
are processed differently to another bit image to be presented
repeatedly; and said image display section is comprised by: an
image control section for controlling display timing of those bit
images to be repeatedly displayed according to bit-levels of
attribute bits; and an image display section for displaying
processed bit images.
22. A recorded medium for use with computer means to execute
computations to obtain holographic fringe patterns and display
computed hologram images, comprised by programs for: converting
three-dimensional data of a display object into computational data
for fringe pattern generation; determining a sampling rule for
sampling computational data; sampling computational data according
to a selected sampling rule; computing wavefronts form each of
sampled computational data by assuming that each datum position has
a light source; storing computed fringe patterns obtained by
computing interference patterns of wavefronts with a reference beam
as holograms; repeating the steps for sampling and wavefront
computation; and displaying successively a plurality of hologram
images thus produced.
23. A medium according to claim 22, wherein said conversion to
computational data includes generation of vertex coordinates on
surfaces of display object from said three-dimensional data.
24. A medium according to claim 22, wherein said conversion to
computational data includes conversion to voxel data.
25. A medium according to claim 22, wherein said conversion to
computational data is performed for each object comprising said
display object, and said sampling rule is altered in accordance
with an attribute of each object.
26. A recorded medium for use with computer means to execute
computations to obtain holographic fringe patterns and display
computed hologram images, comprised by programs for: inputting
three-dimensional data of a display object into computer means;
classifying and grouping said display object, and computing
interference patterns for each classified or grouped display object
to produce computed fringe patterns; converting a plurality of
computed fringe patterns into respective digital images;
decomposing a plurality of digital images into bit images comprised
by individual bits; synthesizing bit images obtained for each
classified or grouped display object so as to form moving pictures
for display; and displaying successively moving pictures thus
produced.
27. A medium according to claim 26, wherein computation of fringe
patterns is performed by classifying a plurality of display objects
according to an attribute of each display object and computing
fringe patterns for every display object in all
classifications.
28. A medium according to claim 27, wherein conversion to digital
images includes steps of: selecting an information content
necessary to display said display object according to an attribute
of each display object and converting to digital images having
attribute-bits representing a selected information content; while
generation of moving pictures includes a step of distributing and
assigning bit images obtained for each classified or grouped
display object to a plurality of contiguous screens of moving
pictures.
29. A medium according to claim 27, wherein conversion to digital
images includes selecting an information content necessary to
display said display object according to an attribute of each
display object and converting to digital images having
attribute-bits representing a selected information content; while
generation of moving pictures includes distributing and assigning
bit images obtained for each classified or grouped display object
to a plurality of contiguous screens of moving pictures so as to
preserve intensity of images.
30. A medium according to claim 27, wherein generation of digital
moving picture is performed by extracting those bit images having
high bit-levels representing attributes of said display object from
bit images obtained for each classified or grouped display object,
and distributing and assigning said bit images having high
bit-levels moving pictures to a plurality of contiguous image
sequence.
31. A medium according to claim 26, wherein conversion to digital
images is performed so as to form moving pictures having a given
level of attribute-bits, and said moving pictures are produced by
adding bit images produced from digital images for each level of
attribute-bits so that a number of field layers in a frame
corresponds to each bit-level of said attribute-bits, and altering
a coding mode for field layers in different fields.
32. A recording medium according to claim 26, wherein generation of
moving pictures is performed by selecting predetermined high-level
bit images from said digital images, and assigning individual
pixels of said high-level bit images to a plurality of field layers
of moving pictures to produce moving pictures for display.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to a technology for displaying
computer generated holograms on such display medium as electronic
display panel.
[0003] This application is based on Patent Application Nos. Hei
9-131531, Hei 9-131532, Hei 10-008161 and Hei 10-049093 filed in
Japan, the contents of which are incorporated herein by
reference.
[0004] 2. Description of the Related Art
[0005] Conventionally, one method of producing computer generated
holograms is to express an object as a collection of point-light
sources, and obtain holographic fringe patterns of the wavefronts
by computation and display the resulting holograms using
acousto-optical modulator or liquid crystal display.
Acousto-optical modulator suffers from a disadvantage that only
one-dimensional fringes (horizontal parallax only) can be
expressed, but liquid crystal panels has advantages such as its
capability to display two-dimensional images and the ease of
altering the image electrically. However, because it is normally
necessary to electrically control the gray levels of each pixel in
the liquid crystal panel, manufacturing of finer pixel spacing is
limited by the difficulties in control circuit fabrication and
other factors. To display a hologram, it is basically necessary to
provide fine resolution of higher then 1,000 lines/mm, but such
fine resolutions are difficult to achieve in practice so that
holograms can presently display only fairly coarse images.
[0006] That the image resolution and its dynamic range are limited
when using electronic display devices, such as liquid crystal
display, means in effect that there is an upper limit of spatial
frequencies that can be displayed on such devices and it can
display only 256 gray level. In other words, to display one object,
it is necessary to be able to display a certain amount of high
frequency components, but because of the low resolution of the
display devices, it is difficult to clearly display several items
on the same screen. This is because the limiting high frequency
components and dynamic range for display one object overlaps those
of another items, thereby resulting in destruction of the fringe
patterns of the high frequency components. Technically, it leads to
a lack of sufficient data to reproduce the item, resulting in high
value of signal to noise (S/N) ratio, and thereby restricting the
number of items displayable in one view, i.e., limiting the
expressive capability of displaying the details of object
information.
SUMMARY OF THE INVENTION
[0007] It is an aim of the present invention to provide a
technology for displaying computer generated hologram images to
enable to display more detailed shapes of a display object or a
larger number of display objects than is possible by the
conventional hologram display technology, even when using an
electronic display apparatus, such as liquid crystal display, which
has a limited image resolution and dynamic range capability.
[0008] The aim has been achieved in a method for computing fringe
patterns of a display object comprised by items and displaying
computer generated holograms. In this methodology,
three-dimensional data of the display object are converted into
computational data for fringe pattern generation, and a sampling
rule for sampling computational data is determined and
computational data are sampled according to a selected sampling
rule. Wavefronts are generated by assuming that each position of
sampled three-dimensional data has a light source and generates a
wavefront, and fringe patterns generated by interference of
computed wavefronts and a reference beam are obtained and stored as
hologram images. The steps of sampling and generating a wavefront
are repeated for all computational data. The plurality of hologram
images thus generated are displayed successively using the display
apparatus provided in the present invention, which is used in
conjunction with the procedures that are provided through suitable
recording media exemplified.
[0009] The method thus achieves the object of displaying more
detailed shapes of an item or a larger number of items by
distributing the holograms, produced by the steps presented above,
over a plurality of moving picture frames by a sampling technique
appropriate to the nature of the display object.
[0010] The aim has been achieved in another method that can produce
hologram video of a display object. In the moving picture
production methodology, three-dimensional data of the display
object are input into a computer device and input data are
classified or grouped according to attributes of the display
object, and a plurality of fringe patterns are computed for each
classified or grouped display object. The plurality of fringe
patterns are respectively converted into a plurality of digital
images, and the plurality of digital images are decomposed into
individual bits to form bit images. Bit images obtained for each
classified or grouped display object are synthesizing to produce
moving pictures for display. The plurality of hologram images thus
generated are displayed successively using the display apparatus
provided in the present invention, which is used in conjunction
with the procedures that are provided through suitable recording
media exemplified.
[0011] The method thus achieves the aim of displaying a larger
number of items while decreasing the number of items displayed in
one frame by distributing the digital images of the fringe patterns
of a plurality of items, produced by the steps presented above,
over a plurality of scenes in moving pictures, defined by frames
and fields, by a digital processing technique appropriate to the
nature of the display object.
BRIEF DESCRIPTION OF THE DRAWINGS
[0012] FIG. 1 is a flowchart for a hologram display method
incorporating the special sampling methodology.
[0013] FIG. 2 is a block diagram of an apparatus for carrying out
the method shown in FIG. 1.
[0014] FIG. 3 is a flowchart of a first embodiment of the method of
displaying holograms.
[0015] FIG. 4 is an example of three-dimensional input data.
[0016] FIG. 5 is an illustration of a conversion of
three-dimensional data.
[0017] FIG. 6 is an illustration of sampling data.
[0018] FIG. 7 is a block diagram of an apparatus in the first
embodiment.
[0019] FIG. 8 is a flowchart of a second embodiment of the method
of displaying holograms.
[0020] FIGS. 9A, 9B are examples of voxel data to be used in the
above embodiment.
[0021] FIG. 10 is a drawing to illustrate the concept of the
present invention.
[0022] FIG. 11 is a block diagram of an apparatus in the second
embodiment.
[0023] FIG. 12 is a flowchart for a method of displaying holograms
with the special technique for generating moving images.
[0024] FIG. 13 is a block diagram for an apparatus for the method
shown in FIG. 12.
[0025] FIG. 14 is a flowchart of a third embodiment of the method
of displaying holograms.
[0026] FIG. 15 is an illustration of a coordinate system.
[0027] FIGS. 16A.about.16C are examples of field image groups for
each object.
[0028] FIG. 17 is an illustration of an image (i, j) for each
object extracted from a field image array.
[0029] FIG. 18 is an illustration of time-display of an image (i,
j) for each object in a field image array.
[0030] FIG. 19 is an illustration of an example of an image (i, j)
in a synthesized field image array.
[0031] FIG. 20 is an illustration of a method of coupling specified
bits of a synthesized image of different items.
[0032] FIG. 21 is a block diagram of an apparatus for a third
embodiment.
[0033] FIG. 22 is an illustration of image display according to
pulse-width modulation method.
[0034] FIG. 23 is a flowchart for a display method in a fourth
embodiment.
[0035] FIG. 24 is an illustration of different items to be
displayed in the present embodiment.
[0036] FIG. 25 is an example of display sequence for field images
in the present embodiment.
[0037] FIG. 26 is a block diagram of an apparatus for a fourth
embodiment.
[0038] FIG. 27 is a flowchart for a method for displaying holograms
in a fifth embodiment.
[0039] FIG. 28 is an example of weighting used in error diffusion
processing.
[0040] FIG. 29 is a block diagram of an apparatus for a fifth
embodiment.
[0041] FIGS. 30A.about.30F are examples of processed images
according to the present method.
[0042] FIGS. 31A.about.31E are illustrations of the digital display
properties of the computer generated holograms.
[0043] FIG. 32 is a flowchart for a method for displaying holograms
in a sixth embodiment.
[0044] FIGS. 33A.about.33H are examples of the images produced in a
method of the present invention.
[0045] FIG. 34 is a block diagram of an apparatus for a sixth
embodiment.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0046] The present invention of methods and apparatuses for
computed hologram display are embodied in various embodiments which
are presented in the following in such a way that, although each
embodiment is discussed separately, a brief outline of the basic
concept involved is presented before discussing the details of the
respective embodiment.
[0047] The following is an overall outline for Embodiments
1.about.6.
[0048] Embodiments 1 and 2 are related to displaying an image by
distributing the hologram images in a plurality of frames by a
suitable sampling technique of the image data to be displayed so
that the holographic images can be observed as a contiguous display
of moving pictures. In other words, a frame division displaying
technique is utilized to display more detailed shapes or a larger
number of items.
[0049] Embodiments 3-6 are related to distributing the digital
images of fringe patterns for a plurality of items over a plurality
of frames/fields in a dynamic view so that the number of items per
image layer is decreased but more items can displayed in a given
number of views overall.
[0050] In the presentation which follows, a "frame" refers to one
view for expressing moving pictures, and a "field" refers to image
layers for comprising the frame. Also, a display object may be
comprised by one item or a plurality of items so that these terms
are interchangeable in some cases.
[0051] The following is a summary outline for Embodiments 1 and
2.
[0052] FIG. 1 is a flowchart of a method for displaying computer
generated holograms common to both Embodiments 1 and 2.
[0053] First, three-dimensional data (3D-data hereinbelow) for an
item to be displayed are entered into a computing device, and the
3D-data are converted to computational data for generating the
fringe patterns (steps 11, 12).
[0054] Next, a sampling rule is specified for the converted 3D-data
(step 13).
[0055] Next, converted 3D-data (computational data) are sampled
according to the sampling rule selected, and wavefront data are
computed by assuming that each of the 3D-data sampled has a light
source for emitting waves, and the computed interference patterns
between the wavefronts and the reference beam are stored as
hologram images (steps 14, 15).
[0056] Sampling and wavefront generation steps are repeated, and
the movie-like hologram images thus produced are successively
displayed (steps 16, 17).
[0057] The items are thus distributed over a plurality of moving
pictures by a suitable sampling technique so that the hologram
images can be displayed as contiguous frames, thereby enabling to
observe more detailed shapes or a larger number of items than is
possible by the conventional computed hologram display
techniques.
[0058] FIG. 2 is a block diagram of an apparatus for executing the
method shown in FIG. 1. FIG. 2 shows that the display apparatus is
comprised by: display object data input section 1; item management
section 2; image generation section 3; and an image display section
5. Here, the display object data input section 1 executes step 11
in FIG. 1, and the item management device 2 executes steps 11 to 14
inclusively and step 16. The image generation section 3 executes
step 15 in FIG. 1, and image display section 5 executes step 17 in
FIG. 1.
[0059] An image memory section 4 provided in FIG. 2 is for storing
holographic image data computed by the image generation section 3,
and it is used when displaying the stored images on the image
display section 5 or transmitting the images for display. The image
memory section 4 may also be used as a temporary storage during an
image generation step.
[0060] The sections shown in FIG. 2 may include hardware-driven
devices or software-driven devices to be executed by memory devices
working in conjunction with a central processor unit (CPU) which
are not shown in these drawings.
[0061] The method and apparatus of Embodiment 1 illustrated in
FIGS. 1 and 2 will be explained in more item in the following.
[0062] [Embodiment 1]
[0063] In the first embodiment, 3D-data of the display object are
separated into several sections or items, and each item is sampled
and the computed fringe patterns are displayed on a plurality of
frames, thereby enabling to display more detailed shapes or a
larger number of the items.
[0064] Specific details will be presented in the following.
[0065] Electronic display devices are limited in their image
resolution and dynamic range capability, in other words, even if
attempts are made to present fringe patterns generated by a
plurality of light sources simultaneously, only a limited number of
these fringe patterns can be displayed. Basically, it means that
only "n" pieces of light sources can be displayed at the same time.
As an example, a value of one hundred will be assumed for n,
meaning that the display apparatus is capable of displaying one
hundred point light sources at the same time, and the method will
be illustrated using the flowchart shown in FIG. 3.
[0066] First, 3D-data of the display item are entered into a
computing device (step 121). In this examples, a display object is
comprised of items 101, 102 illustrated in FIG. 4. It is assumed
that each of the display items 101, 102 consist of a collection of
3D-coordinate data and are to be displayed on a display screen
103.
[0067] Here, when the 3D-data of the display object are polygonal
data described by vertices and edges, the surface information can
be expressed by subdividing the surfaces comprising the polygon
into a mesh of finer descriptions. For example, the individual
surfaces in the polygonal data comprising item 102 in FIG. 4 are
subdivided into 16 sectors as shown in FIG. 5, and the vertex
coordinates in each subdivided surface are used as a new 3D-datum.
In FIG. 5, the points 104a are original 3D vertex data, and the
points 104b are additional new points created by the subdivision.
In this step, if the density of the vertex coordinates is already
at a required value, there would be no need to subdivide into a
finer mesh.
[0068] Next, a list for 3D-coordinates (vertex coordinates)
comprising the item is prepared (step 122). In this case, a list is
prepared for each item 101 and 102 to contain the vertex
coordinates of the surfaces created by the sub-dividing process is
prepared for.
[0069] Next, the sampling rule is selected for selecting input data
from the list prepared in step 122 (step 123). For example, based
on the limit of resolution of the display device, and assuming the
number of item-data that can be displayed simultaneously is one
hundred, random sampling will be selected. Sampling rule will be
explained in more detail in later embodiments.
[0070] Next, the specified number (100 pieces) of vertex coordinate
data are selected from the list prepared in step 122 (step 124). In
this case, a total of 100 pieces of vertex coordinate data are
selected from the lists prepared for items 101, 102. As an example,
from the vertex coordinate data for item 101, vertex coordinate
data 105a shown in FIG. 6 will be selected for frame (n) while the
vertex coordinate data 105b will be selected for frame (n+1).
[0071] Next, assuming that each of the vertex coordinate data
selected in step 124 is a point light source, interference fringes
formed by the reference beam on the display screen 103 are computed
(step 125). The obtained fringe patterns are stored temporarily in
memory as hologram images.
[0072] Next, if there are vertex coordinates data still remaining,
steps 124 to 125 are repeated (step 126).
[0073] In step 124, vertex coordinate data 105a shown in FIG. 6 are
selected from the vertex coordinate data for item 101 to be
displayed in frame n. In step 125, interference fringes formed by
the wavefronts and the reference beam on the display screen 103 are
computed, and the results are stored temporarily in memory. Next,
because there are still remaining vertex coordinate data, the
vertex coordinate data 105b remaining in step 124 are selected, and
similar processing is carried out (steps 124.about.126).
[0074] Lastly, hologram images stored temporarily in memory are
successively displayed (step 127).
[0075] The above steps complete the display process of computer
generated holograms.
[0076] Accordingly, holograms of the display item are distributed
in different frames by sampling and these frames are displayed as
image sequences. In other words, frame division display technique
evokes after image effect in human vision, so that more detailed
shapes or a larger number of items can be displayed even on a low
resolution display device. More specifically, using a
two-dimensional square-shaped object illustrated in FIG. 10 as an
example, the ultimate image of the square object, comprised by
fringe patterns from the individual frames n, n+1, n+2, n+3, is
described by a series of coarse images contained in layers such as
231, 232, which are comprised by points generated by wide sampling
intervals. If the frames are displayed continually at a high speed,
human vision perceives them as a single item, (i.e., an item
sampled at a finer sampling interval) as illustrated by a layer
233, of the square shape comprised by densely packed points.
Therefore, even on a low resolution display device, more detailed
shapes and higher number of items can be displayed.
[0077] The flowchart shown in FIG. 1 correspond with that shown in
FIG. 3 as follows: steps 121.about.127 in FIG. 3 correspond to
steps 11.about.17 in FIG. 1.
[0078] In the above example, steps 121 to 126 are carried out first
and step 127 is repeated to produce a moving picture, but it is
also possible to provide real-time viewing based on steps 121 to
127, without resorting to the intermediate storage step. In this
case, before carrying out step 127, the holograms are successively
transmitted to a destination, and the images are displayed on the
destination display device. By adopting such an approach, it
becomes possible to provide progressive transmission by which the
item becomes clear gradually as the number of images transmitted
increase. Because the number of display items in each layer is
relatively low, fringe spacings are relatively coarse. In other
words, compared to the conventional fringe patterns, the spatial
frequency is lower in the present technique so that it is possible
to increase the efficiency for data compression.
[0079] Also, in the above example, vertex coordinate data of the
item are assumed to originate from point light sources to produce a
hologram based on a collection of point light sources, but the type
of light source is not specified in the present embodiment. For
example, it is possible to assume that individual surfaces
(patches) comprising the 3D-polygon data are separate planar light
sources.
[0080] It is also possible to replace individual 3D-data of the
display item with individual voxels 106, illustrated in FIG. 4, to
represent the display space so that the input data now become
volume data to be sampled in each voxel. The concept of voxel will
be explained in detail in Embodiment 2.
[0081] The presentation so far has been based on some given
sampling rule for 3D-data, but in step 122, input data may be
converted item by item so that sampling rule can be altered to suit
the attributes of the item. Specifically, the following steps may
be taken.
[0082] 1) Sampling Based on the Distance Between the Item and the
Screen
[0083] When a scene is comprised by items distributed over a
distance, those items which are further away from the screen can be
sampled at lower intervals while the items closer to the screen are
sampled at finer intervals. The spatial frequencies of fringe
patterns for far items are lower than those of close items such
that, even if the sampling density is raised, the probability of
mutual destruction of fringes is small.
[0084] 2) Sampling Based on Properties of Item
[0085] Sampling density for moving items is made lower than that
for static items. Moving items do not present problems of image
quality compared to static items even at lower sampling densities.
Sampling rule may be chosen so that the faster the speed of the
moving items the lower the sampling density.
[0086] As explained above, by synthesizing wavefronts of differing
spatial frequencies by a sampling rule according to the properties
of the item, mutual destruction of interference fringes is reduced.
Therefore, even a low resolution display device can have many items
displayed simultaneously. It is possible to combine the sampling
rules 1) and 2) discussed above.
[0087] Although the examples above were based on an approach of
altering the sampling rules to suit the properties of the item, the
same effects can be produced by devising a suitable approach in
creating the list for vertex coordinate data in step 122.
Specifically, the following approaches are possible.
[0088] 1) List Creating Rule Based on Distance Between Item and
Display Screen
[0089] Vertex coordinates list is created so that the items which
are far from the display screen will be sampled at a low density of
vertex coordinate data while the items which are near to the
display screen will be sampled at a high density of vertex
coordinate data.
[0090] 2) List Creating Rule Based on Properties of Item
[0091] For moving objects, the density of vertex coordinate data is
made low, and for static objects, the density of vertex coordinate
data is made high. The higher the speed of the moving object, the
lower may be the data density of vertex coordinates.
[0092] Once a sampling rule is chosen in step 123, the same rule is
applied to all the items. Also, the rules 1) or 2) above may be
combined in a suitable way.
[0093] Next, an example of the configuration of the display
apparatus to be used with the method according to the flowchart in
FIG. 3 is shown in FIG. 7. The display apparatus comprises: a data
input section 130; a data conversion section 131; a data sampling
section 133; a sampling rule decision section 132; a fringe pattern
computation section 134; and a fringe pattern display section 135.
The operation of the apparatus will be presented in the
following.
[0094] The 3D-data of the display object entered by the data input
section 130 are converted into a data structure to suit the
computational needs, such as dense or sparse 3D-data. In the
sampling rule decision section 132, sampling rule is decided based
on the type of input data, and the data sampling section 133
samples the input data. The fringe pattern computation section 134
computes the fringe patterns, using the sampled 3D-data, to be
displayed as holograms. The computed hologram images are displayed
successively on the fringe pattern display section 135. That is,
the data input section 130 executes step 121 shown in FIG. 3, and
the data conversion section 131 executes step 122. The sampling
rule decision section 132 executes step 123, and data sampling
section 133 executes steps 124, 126. The fringe pattern computation
section 134 executes step 125, and the fringe pattern display
section 135 executes step 127.
[0095] Various sections shown in FIG. 7 correspond to those in FIG.
2 as follows: data input section 130 in FIG. 7 to display object
input section 1 in FIG. 2; data conversion section 131, sampling
rule decision section 132 and data sampling section 133 to item
managing section 2; fringe pattern computation section 134 to image
generation section 3, and fringe pattern display section 135 to
image display section 5.
[0096] Accordingly, an item to be displayed is observed in a
hologram through image sequences that are contiguous frames which
are produced by distributing the item according to a selected image
sampling rule to a plurality of different frames. This is, a frame
division technique is used, in effect, to display more detailed
shapes of an item or a larger number of items in one holographic
image. Because the number of items contained in one layer is
lowered, interference fringes for each item are decreased, thereby
decreasing the S/N ratio to avoid burying the image in the
background noise, and increasing the number of items which can be
clearly displayed in one hologram.
[0097] When a sampling rule based on item properties is chosen,
wavefronts of differing spatial frequencies would be synthesized so
that mutual destruction of fringe patterns can be lessened.
Therefore, the number of items which can be simultaneously
displayed can be increased even on a low resolution display
device.
[0098] In changing the conversion step 122 in FIG. 3 to conversion
to voxel data, if a sampling rule for scanning is according to the
Hilbert curve, for example, the display resolution for the display
object can be described in hierarchically, and it becomes possible
to perform a progressive display of images.
[0099] Further, if the holographic images are to be transmitted,
the number of objects to be displayed in one screen can be
decreased so that the holograms themselves can be represented with
a fewer number of spatial frequencies, resulting that the
efficiency of data compression can be increased. This would be
useful when transmission capacity is limited. Because successive
transmissions of image data are presented in layers of differing
data densities, the hologram image quality becomes dependent on the
capacity of the transmission means. In other words, images are
never made totally invisible but the image resolution would be poor
in a low capacity transmission environment but would be high in a
high capacity transmission environment because there would be less
loss of detailed data.
[0100] [Embodiment 2]
[0101] In Embodiment 1, the approach was to prepare a list of data
vertex coordinates for each object, and vertex sampling was
executed according to the list. In Embodiment 2, the display space
containing the display object is separated into a series of cubes
or "voxels" so that each voxel is sampled by the apparatus.
[0102] The embodiment will be presented with reference to the
drawings.
[0103] First, the method of Embodiment 2 will be explained. FIG. 8
is a flowchart for an example of the method.
[0104] First, a display object, for example polygonal data or
volume data (scanned data such as CT images) is converted into
voxel data 221 such as those shown in FIG. 9A (step 201). Each
voxel is identified by a voxel number (No.) and a voxel containing
an display object has registered coordinates (x, y, z) and
intensity (A). In the case of a table 222 given in FIG. 9B, it can
be seen that voxel Nos. 3, 4, 5 and 6 contain an object.
[0105] The voxel data 221 are sampled under a given sampling rule
(for example, equal distances, such as every 3 boxcells) in steps
202, 203. Those voxels which have been sampled are identified by a
sampled flag (for example, by entering 1 at the end of the count
column) as indicated in table 222. If a voxel contains a display
object (step 204), the intensity of the object is determined and a
corresponding point light source is assigned. Wavefront from this
point light source on the holographic plane is computed (step 205)
and the results are stored in memory (step 207). In table 222 in
FIG. 9B, voxel No. 4 is the object of processing.
[0106] Sampling is continued so that all the wavefronts from each
point light source are computed (step 206) and all the wavefronts
are synthesized, and computed results processed with the wavefronts
of the reference beam are written into the frame memory (step 207).
The data written into the frame memories are displayed as holograms
(fringe patterns) in step 208. Next, remaining voxels are
repeatedly subjected to the same process (steps 203 to 209) in step
209. With reference to table 222 in FIG. 9B, voxels Nos. 2 and 5
become the targets for sampling in the second round, and since an
item exists in voxel No. 5, only voxel No. 5 becomes the target for
wavefront computation. Here, the sequence of steps 209, 208 may be
reversed. This is made possible by providing a plurality of frame
memories so that a plurality of wavefront results for display items
can be computed so that, when they are to be displayed, it is
necessary to execute only data recall step from the frame memories.
This approach enables faster displaying of many frames.
[0107] The steps in FIG. 1 correspond to those in FIG. 8 as
follows: step 201 in FIG. 8 corresponds to step 12 in FIG. 1; step
202 to step 13; step 203 to step 14; steps 204, 206, 209 to step
16; step 205 to step 15; and steps 207, 208 to step 17. In FIG. 8,
a step to correspond to step 11 in FIG. 1 is omitted.
[0108] By repeating the scanning process presented above, using a
two-dimensional square-shaped item illustrated in FIG. 10 as an
example, the ultimate image of the square item, comprised by fringe
patterns from the individual frames n, n+1, n+2, n+3, is
represented by a series of coarse images contained in layers such
as 231, 232, which are comprised by points generated by widely
separated sampling. If the frames are displayed continually at a
high speed, human eyes perceive them as a single item, (like an
item sampled at a finer sampling rate) as illustrated by a file
233, of a square shape comprised by densely packed points.
[0109] Next, the configuration and operation of the apparatus
having the component sections presented above will be explained.
FIG. 11 is a block diagram of the apparatus for displaying
holograms.
[0110] The apparatus is comprised by: a data conversion section
241; a display item managing section 242; a sampling position
decision section 243; a wavefront computation section 244; a fringe
pattern synthesizing section 245; a fringe pattern memory section
246; and a wavefront display section 247.
[0111] The display object input into the data conversion section
241 is converted into voxel data and is stored in the display
object managing section 242. In the sampling position decision
section 243, sampling rule has been pre-selected, and instructs the
wavefront computation section 244 on voxel positions to be sampled.
In the wavefront computation section 244, if the sampled voxel has
an object registered, a point light source is assigned to the voxel
to compute the wavefront on the hologram plane. The computed
wavefront data are registered in the memory in the wavefront
synthesizing section 245. All the relevant voxels are subjected to
the above processing, and wavefront data are successively added to
the memory. When the processing is completed, wavefront
synthesizing section 245 registers memory data into the frame
memory of the fringe pattern memory section 246. The above
processing steps are successively performed and the results are
registered in the fringe pattern memory section 246 accordingly.
The registered fringe patterns are successively called out to the
wavefront display section 247 to be displayed.
[0112] The structures in FIG. 11 correspond to those in FIG. 2 as
follows. The data conversion section 241, the display item managing
section 242 and the sampling position decision section 243 in FIG.
11 correspond to the item managing section 2 in FIG. 2; the
wavefront computation section 244, the fringe pattern synthesizing
section 245 to image generation section 3; the fringe pattern
memory section 246 to image memory section 4; and the wavefront
display section 247 to image display section 5. A section to
correspond to the display object input section in FIG. 2 is omitted
in FIG. 11.
[0113] The control methodology outlined above enables to display
more clear images than conventional images, even on a low
resolution display device, because of the low number of items shown
in each layer constituting a frame of the item to be displayed.
[0114] It should be noted that, although the item is assumed to be
a point light source in the above embodiment, it is possible to
assume that a planar light source having a surface inclination
angle as a parameter exists inside a voxel, therefore, it is not
necessary to limit the type of light source to this embodiment.
[0115] Also, in the present embodiment, equidistant sampling was
used as the sampling method, but other sampling methods may be
used. For example, it is possible to display progressive images, if
a sampling rule based on the Hilbert curve scanning of the item is
chosen, so that the display resolution for the display object can
be sampled in gradually changing layers from coarse image
resolution (low data density) to progressively finer resolution
(high data density). By sampling the space in such layers, it
becomes possible to display images progressively. Also, when there
are many objects in a voxel, sampling by layers or sampling of low
density images may be perfumed in many ways. Selection may be based
on objects of maximum intensity or on an average computed intensity
for all items so that there is no need to specify any one
particular approach. As in previous embodiments, the sampled fringe
patterns may be transmitted successively to a destination, and the
destination image display can be used to refresh the images to
enable progressive image transmission/display.
[0116] Sampling method may be based on a combination of voxels
which are at a far distance from the display screen with those
which are close to the screen. Spatial frequencies obtained from
far items are lower than those from close items so that mutual
destruction of fringe patterns displayed on one screen can be
reduced.
[0117] Furthermore, other sampling methods may include a method
based on lowering the sampling density for those voxels close to
the display screen and raising the sampling density for those far
from the screen. Spatial frequencies for far objects are lower than
those for close objects so that the probability of fringe pattern
destruction is less for the far objects even if the sampling
density is raised.
[0118] Also, regarding the space to be specified initially, it may
not be the whole input data but it may be a partial space to
include only the display object, or it may be a volume data assumed
for each display object to be individually processed. In other
words, for each display object, a volume datum may be defined
locally, thereby enabling to define an optimum degree of resolution
for static as well as moving objects.
[0119] As described above, the present method of holography is able
to display more detailed shapes or more objects compared to the
conventional technologies of hologram display by enabling to view
several frames while decreasing the number of display object
contained in each frame.
[0120] Further, the space division approach enables fringe patterns
to be computed according to a uniform amount of computational
effort, regardless of the complexity of the display object.
[0121] Further, selection of sampling rule enables to achieve the
optimum degree of resolution necessary to display an item for each
frame, thereby enabling to decrease the volume of data necessary
for holographic information transmission or to execute progressive
transmission to suit the changes in the transmission capacity.
[0122] Summarizing the above embodiment, the embodied method of
displaying computed holograms is comprised of the steps of:
preparing display data for a display object as voxel data;
specifying a sampling rule for a voxel; sampling an object space
according to a selected sampling rule; determining whether an
object exist in a voxel being sampled, and assuming that a voxel
containing an object is a light source; computing a wavefront
emitted by an object-containing voxel to obtain a fringe pattern as
a hologram image; repeating the steps of sampling a voxel and
computing a wavefront; and successively displaying a series of
hologram images thus produced on a display screen.
[0123] The holographic display apparatus embodied above for
displaying computed holograms is comprised by: a data conversion
section for converting an object into voxel data; a display object
managing section for managing voxel data so converted; a sampling
rule decision section for specifying a sampling rule for sampling
the voxel data; a wavefront computation section for computing a
wavefront generated by an object-containing voxel by assuming the
object-containing voxel to be a light source; a fringe pattern
synthesizing section for combining a plurality of computed
wavefronts for each sampled data to generate a fringe pattern; a
fringe pattern memory section for storing synthesized fringe
patterns; and a wavefront display section for displaying a
holographic image comprised by fringe patterns.
[0124] The apparatus described above may be further provided with a
data transmission section for successively transmitting stored
fringe patterns and replace the display section with a serial
display section for serially displaying successively transmitted
fringe patterns.
[0125] Accordingly, an object to be displayed is observed in a
hologram through movie-like contiguous frames which are produced by
distributing the object according to a selected image sampling rule
to a plurality of different frames. This is, a frame division
technique is used, in effect, to display more detailed shapes or a
larger number of items in one holographic image. Because the number
of data contained in one layer is lessened, interference fringes
for each item are decreased, thereby lowering the S/N ratio to
avoid burying the image in the background noise, and increasing the
number of objects which can be clearly displayed in one
hologram.
[0126] In the step of sampling rule selection for voxel data, by
choosing a sampling rule based on the Hilbert curve for scanning,
it is possible to display progressive images of the object so that
the display resolution for the display object can be described in
gradually changing layers.
[0127] Furthermore; if the holographic images are to be
transmitted, because the number of objects to be displayed in one
layer is decreased, the present method and apparatus are able to
accommodate some limits in transmission capacities. By successively
sending holographic image data presented on layers of differing
data densities, even when the transmission capacity changes,
progressive images are never made totally invisible in the present
invention, but the image resolution would become poor in a low
capacity transmission environment but would become high in a high
capacity transmission environment.
[0128] [Embodiments 3 to 6]
[0129] The Embodiments 3 to 6 presented in the following relate to
methods and apparatuses for distributing digital images of fringe
patterns for a plurality of objects over a plurality of layers of
moving pictures in image layers such as frames/fields so that,
although each layer contains a fewer number of objects, a frame as
a whole, consisting of some given number of field layers, is able
to show a larger number of objects.
[0130] Before explaining the details of the Embodiments 3 to 6,
common features of the display method for computed holograms will
be presented with reference to FIG. 12.
[0131] First, 3D-data of the display object are input into the
apparatus (step 21).
[0132] Then, the input data of the display objects are
classified/grouped, as necessary, and interference fringes formed
by the reference beam are computed for each classified or grouped
display object (step 22).
[0133] Next, the computed fringe patterns are converted into
digital images, by separating into individual bits (step 23).
[0134] Next, image sequences for display are generated by combining
bit images for each classified/grouped display object (step 24),
and the generated moving pictures are displayed under a controlled
timing (step 25).
[0135] Accordingly, by distributing the digital images of a
plurality of fringe patterns for a plurality of items, over a
plurality of layers of bit images, a larger number of items can be
displayed over a frame consisting of a number of layers, although
each layer contains a fewer number of items.
[0136] FIG. 13 is a block diagram of an example of the computed
hologram display apparatus for executing the method shown in FIG.
12. The apparatus is comprised by: a display object input section
1; an object image generation section 7; and an image display
section 8. With reference to the steps shown in FIG. 12, the
display object input section executes step 21, the image generation
section 7 executes steps 22.about.24, and the image display section
executes step 25.
[0137] An image memory section 4 shown in FIG. 13 is for storing
the hologram images computed by the image generation section 7, and
is utilized when displaying or transmitting the stored images for
display. The image memory section 4 can also be used to store
images temporarily while generating display images. The item
managing section 6 is necessary to compose bit images when the
input display object consists of a plurality of objects, and to
execute bit-image synthesizing process to suit the properties of
the objects.
[0138] The sections/devices indicated in FIG. 13 may be comprised
by own dedicated micro-processors, or they may also be application
softwares to be executed by hardwares such as memories and CPU and
the like.
[0139] Detailed methodology and apparatus for Embodiments 3.about.6
will be discussed in the following.
[0140] [Embodiment 3]
[0141] First, normal methods of displaying computed holograms will
be explained. In addition to the methods already mentioned
(acousto-optical modulator and liquid crystal panel), the display
methods include a high-precision display apparatus represented by
digital micromirror device (DMD) method (refer to Larry J.
Hornbeck, "Digital light processing for high-brightness,
high-resolution applications", Electronic Imaging, El'97,
Projection Displays III, an invited paper, 1997.) This method
utilizes drive mirrors attached to those locations corresponding to
individual display pixels, and the radiated beam are directed to
various direction by changing the inclination of the mirrors
thereby controlling the intensity (while/black) of each pixels.
According to this method, intensities of each pixel are expressed
digitally, and the bit arrays for individual pixels are serially
displayed at a high speed, in which the bit arrays are represented
by a plurality of fields. This method is, therefore, a digital
display method and is generally referred to as the pulse-width
modulation method.
[0142] The pulse-width modulation method will be explained with
reference to FIG. 22. As shown in file 351, when expressing a pixel
intensity with an information content of 3-bits, the 2.sup.2-level
bit arrays, 2.sup.1-level bit arrays and 2.sup.0-level bit arrays
are presented separately in succession. For example, in binary
coding, the pixel intensity may be expressed by displaying either
white (1) or black (0). Then, if the intensity in binary coding is
101, decomposed bit arrays for the 22-level array will be a
presentation in the sequence of white-white-white-white (i.e.,
1-1-1-1), followed by black-black (i.e., 0-0) for the 2.sup.1-level
array, followed by white (i.e. 1) for the 2.sup.0-level array. By
assigning individual pixels to separate fields (i.sub.0, i.sub.1 .
. . i.sub.6) and presenting the images in each field sequentially,
the gray levels of the individual pixels can be duplicated as shown
in file 352. If the fields containing images shown in file 353 are
presented sequentially, seven fields can reproduce gray level
images in one frame, as indicated in file 354.
[0143] In essence, digital micromirror method based on pulse-width
modulation is not a conventional analogue gradation display, but is
one of the digital display methods which can express digital images
directly. In the present embodiment, the pulse-width modulation
method was adopted to the computed hologram display method and
apparatus for displaying a plurality of items simultaneously.
[0144] The present embodiment will be presented with reference to
the drawings.
[0145] FIG. 14 is a flowchart for a method based on fields and
frames. The relation of frames and fields in moving picture display
will be explained in detail. A frame refers to a scene in moving
pictures and is composed of a plurality of fields, where each field
contains an image layer. Specifically, if each image layer in a
frame is expressed by 4-bits (attribute-bits), the image layer is
deblocked (decomposed) into 20, 21, 22 and 23 arrays, and the
gradation is expressed by assigning a corresponding number of
fields to each array such that 2.sup.0=1 field, 2.sup.1=2 fields,
2.sup.2=4 fields and 2.sup.3=8 fields, so that one frame would
consist of 15 fields (=1+2+4+8).
[0146] In the beginning, data related to the objects to be
displayed are entered, and individual attributes are examined (step
361). For example, attributes are examined with reference to the
following characteristics;
[0147] 1) attributes of the item itself . . . surface coloring,
gradations, textures etc.;
[0148] 2) dynamic properties . . . shape changes, translation,
rotation etc.; and
[0149] 3) location of items . . . distance from the display screen
etc.
[0150] Next, the amount of information complexity necessary to
display the objects (number of attribute bits and gradation) are
determined on the basis of the attributes of the objects (step
362). Information content necessary for expressing the objects and
the attributes of the objects are pre-defined in a suitable manner,
e.g. a table, and information content is decided according to such
a reference. Qualitative relationships between the information
content and the attributes of the objects are exemplified in the
following list.
[0151] 1) Attributes of the Objects Themselves
[0152] More information is needed for objects having surface
coloring, many gradations and complex shapes.
[0153] 2) Dynamic Properties
[0154] Less information is needed for faster changes in shape,
movement and rotation.
[0155] 3) Location
[0156] Less information is needed for items which are located
further away from the display screen.
[0157] Explanations are provided in the following with reference to
specific examples of processing a display object comprised by three
items. The three items are referred td as items A, B and C with
respective attributes a, b and c. It is assumed that the attribute
"a" is to be expressed by 8-bit data, attribute "b" by 4-bit data
and the attribute "c" by 3-bit data, and these requirements are
already defined in a table to be referenced.
[0158] Next, fringe patterns generated by the wave emitted from
each of the items and the reference beam are computed for each
classified attribute. In this case, conversion to digital image is
performed according to the gradation width in terms of the defined
number of attribute bits (step 363). The digital image thus
generated is expressed by a series of pixels (i, j) as illustrated
in file 300, FIG. 15.
[0159] Next, each digital image is decomposed (deblocked) into
respective field image arrays according to the defined number of
attribute bits (step 364). In this example, the 8-bit digital
images in the attribute "a" group are comprised by 255 layers,
which is derived as (=128+64+32+16+4+2+1) layers of field images;
the 4-bit digital images in the attribute "b" group are comprised
by 31 (=16+8+4+2+1) layers of field images; and the 2-bit digital
images in the attribute "c" group are comprised by 7 (=4+2+1)
layers of field images.
[0160] Here, each pixel in a field image has 1-bit information,
therefore, a field image may be said to represent a special case
(of the bit images) formed by the attribute-bits. This type of
relation between the bit images and the field images apply also to
other embodiments.
[0161] File 301 in FIG. 16A relates to the field image array of
item A, file 302 relates to the array for item B; and file 303
relates to the array for item C. Therefore, a k-th layer of the
field image for item A would be expressed as Akij.
[0162] File 311 in FIG. 17 shows an example of extracting only the
pixels (i, j) in the field image array which show an intensity
value of 129 [(11110001).sub.2] for item A. In the drawing, white
is (0) and black is (1). Files 312, 313 show pixels (i, j) for the
intensity values of 7 and 5, respectively, concerning items B,
C.
[0163] In this example, it is assumed that gray scale gradations
are represented by 256 levels, which means that an image requires
an 8-bit gray scale, and 255 field images would be presented. If
the information content is expressed in 4- or 3-bit data as
mentioned above, and if only the field images generated by these
attribute-bits are displayed, the intensity level of the item
displayed would be extremely low. Therefore, to preserve the
original intensity values of the items, intensity of each item is
pre-adjusted to correspond to the number of attribute-bits, so that
the intensities of items can be preserved by repeated displays of
relevant field images while all the 256 field images are being
presented. FIG. 18 illustrate this approach, and file 321 shows the
case of displaying of item A in the display time of 255 fields, and
file 322 shows the case of sixteen repetitions (=2.sup.7/2.sup.4)
of item B display within the time interval for displaying item A
field images to maintain the intensity value of item B, and file
323 relates to the case of thirty two repetitions
(=2.sup.8/2.sup.3) of item C display within the time interval for
displaying item A field images.
[0164] In other words, because item B has 4-bits and 31 fields, the
number of fields are ({fraction (1/16)}) of the that for item A.
When adjusting the number of field layers to be presented to
reflect the intensity properly, item B intensity is lowered by
{fraction (1/16)} in the field image array shown in file 302. When
displaying the field images, sixteen repetitions of B field images
are displayed for one display of the field image array for item A.
This approach maintains the degree of intensity of item B.
Similarly, for item C, the intensity is reduced during digital
conversion process and display is repeated thirty two times for one
display of the field image array for item A.
[0165] It should be mentioned that, because the number of field
images with different bit numbers is not an integer, and fractional
remainders are generated. Such fractions are discarded. Approaches
such as adding to an adjoining frame and other techniques are
possible, and this aspect has not been specified.
[0166] Next, fringe patterns from each field for simultaneous
display are synthesized (added) and converted to binary coding to
produce a field image array D for moving picture (steps 365, 366).
Here, each binary field image before adding consists of 0 or 1, but
after the addition of n layers of field images, each field image is
no longer binary but is represented by n-valued coding. Therefore,
they are converted back to binary data and are then processed field
by field. For example, taking a pixel (i, j) in a field image array
shown in FIG. 18, there are may possible processing steps such
as:
[0167] 1) As shown by pixel Dij in the field image array shown in
file 331 in FIG. 19, only those pixels of value 2 or higher after
addition are assigned a value of
[0168] 2) As shown by pixel Dij in file 332 in FIG. 19, a
theoretical sum (OR) of individual pixels in the synthesized images
are obtained: or
[0169] 3) As shown by pixel Dij in file 333 in FIG. 19, obtain a
theoretical product (AND) of the composite images.
[0170] It is obvious that binarization techniques are not limited
to those mentioned above.
[0171] The converted dynamic field image array D are successively
displayed at a high speed (step 367). The observer thus perceives
an item having a span of shading because of the after image of
human vision.
[0172] In this case, steps in FIG. 14 correspond to those in FIG.
12 as follows: steps 361.about.363 in FIG. 14 correspond to steps
22, 23 in FIG. 12, steps 364.about.366 to step 24, and step 367 to
step 25. In FIG. 14, a step to correspond to step 21 in FIG. 12 is
omitted.
[0173] An example of displaying the composite fringe patterns of
the display object will be explained with reference to FIG. 12. The
display object is assumed to be comprised by three items A, B and
C. The presentation cycle for the display object is shown in file
341 in FIG. 20.
[0174] First, fringe patterns for each item are computed (step 22)
and digitized (step 23). Here, the number of bits will not be
specified in the present embodiment, but the following explanations
are based on expressing all three items with 8-bit data.
[0175] As shown in file 341 in FIG. 20, items A, B are displayed
during the time interval t.sub.1-t.sub.2, and items A, C are
displayed during the time interval t.sub.2-t.sub.3. First, during
t.sub.1-t.sub.2, digital images of fringe patterns for item A are
fetched, and those for item B are also successively fetched. Then,
by replacing the lowest level pixel (i, j) for image A with the
highest level pixel (i, j) for image B and repeating this process
for all the pixels, new digital images or successive field image
arrays (for example, Eij, Fij, Gij in FIG. 20 ) are produced in
step 24.
[0176] Specifically, the highest level bit (2.sup.7 level) is left
alone but for all the levels below 26 are replaced with the value
of image B in the 2.sup.7 level. The meaning of the replacement
process, in terms of the pulse-width modulation method for digital
imaging, is that at the 2.sup.7 level, the number of field
presentations is 128 times, and the total number of presentations
at levels below 2.sup.6 is 127 (=64+32+16+8+4+2+1) so that about
the same number of field presentations is achieved for both items A
and B. This method achieves an image quality which is about
equivalent to displaying items A and B at the same time. Another
possible composing method is, after completing the total field
image array for items A and B, to replace a half of the field image
array for item A with a half of field image array for B, as
indicated by field image array Eij in file 342a in FIG. 20.
[0177] Similarly, in the t.sub.2.about.t.sub.3 interval, a high
level bit for item C is switched with a low level bit for item A,
as indicated by Fij in file 342b, and in the interval beyond
t.sub.3, an high level bit for item B is combined with a low level
bit for item A as indicated by Gij in file 342c.
[0178] By successively displaying the moving pictures generated as
explained above (step 25), a number of items can be displayed while
preserving their values of intensity.
[0179] It should be mentioned that degradation in the image quality
is not serious even if only the high level bits are used, but this
will be explained more fully in Embodiment 6.
[0180] Demarcation between the upper and low level bits is made in
the present embodiment by the upper-most level bit generated in
step 23 that separates all the bits which follow. Such demarcation
can be served by time demarcation or combining upper half of bits
from different items, therefore, method of combining bits will not
be specified.
[0181] Also, in the present embodiment, the order of fringe pattern
presentations is made for individual items, but the present method
is applicable so long as the presentation interval is the same
(relevant field presentation frequency) for the same individual bit
level during a given interval, so the sequence of field
presentation at different bit levels will not be specified.
[0182] Also, the amount of items was three in the present
embodiment, but this quantity is dependent on the resolution
capability of the display apparatus, and this value cannot be
specified in the present invention.
[0183] Also, in the present embodiment, field layers and bit
numbers for each item are exemplified by numbers, but
minimum/maximum field layer necessary to express an item and the
number of bits necessary to display individual items are not
restricted.
[0184] Also, in the present embodiment, black/white binary
designations are used to display each field, but it is not
necessary to be limited to such a binary coding. If the display
apparatus is able to switch the fields at a faster speed than
switching speeds normally used for pulse-width modulation method,
multi-valued images may be used. If such approach is possible, even
more items or more clear images can be realized.
[0185] Next, a configuration of the computed hologram display
apparatus having the features described above will be presented in
FIG. 21. The apparatus is comprised by: a display object managing
section 371; a digital image processing section 372; a field image
processing section 373; an image storage managing section 374; and
an image display section 375. The apparatus is operated as
follows.
[0186] The display objects are managed by the display object
managing section 371, and are classified according to the
attributes of the items. In the digital image processing section
372, those items classified by the display object managing section
371 are separately fetched to compute the fringe patterns to
generate holograms, and are converted to digital images according
to the bit-data for the relevant classified attributes. Digitized
images are decomposed into field image arrays in the field image
processing section 373 according to the bit-data, and are stored in
the image storage managing section 374. The field image processing
section 373 successively fetches field images from the image
storage managing section 374, and produces a new field image
containing a plurality of field images, and stores them in the
image storage managing section 374. The image arrays stored in the
image storage managing section 374 are successively displayed on
the image display section 375.
[0187] Various sections in FIG. 21 corresponds to those in FIG. 13
as follows. The display object managing section 371 in FIG. 21
corresponds to item managing section 6; digital image processing
section 372 and field image processing section 373 to image
generation section 7; image storage managing section 374 to image
memory section 4; and image display section 375 to image display
section 8.
[0188] Accordingly, the present invention enables to display more
items, within a given time interval, by selecting the information
content to suit the attributes of the display object (intensity,
movement etc.); controlling the presentation interval according to
the information content; and sequencing frames/field images as
moving pictures; so that as a whole, more items are displayed even
though each one screen (layer) contains fewer items. This approach
enables to relax the strict resolution requirement for the display
apparatus.
[0189] In summary, objects having such gray variations in textures
that require a high information content are expressed by 8-bit
data, for example, and those objects without such gray variations
that require less information content are given a lesser-bit data
(4-bit for example). Poor quality of reproduction of colors or
textures are less noticeable in the images of moving objects so
that a fewer number of bits is adequate to express such moving
objects. Further, because the display intervals are adjusted
according to the number of attribute-bits so that the information
loss caused by burying effects of the added images of other items
can be decreased.
[0190] In the conventional approach, if one item is expressed by
8-bit data, it is necessary to present 255 layers
(=2.sup.7+2.sup.6+2.sup.5+2.s-
up.4+2.sup.3+2.sup.2+2.sup.1+2.sup.0) of fields to express one gray
scale for each pixel in a digital image. That is, for all items to
be displayed, it is necessary to present a uniform number of field
layers. If the item is expressed in 8-bit data, 255 layers are
successively displayed. The difficulty with this approach is that,
if the intensity is higher than 128, the fields after the 128-th
layer are always white (or 1). In other words, after the 128-th
layer, field images presented do not change at all for a given time
interval.
[0191] In the present invention, these no-change sections in the
field image array is replaced with other images. In normal 2D-image
presentation, such addition will result in noise on the display
screen, but in holographic presentation of fringe patterns,
information contains redundancy so that even if some portions of
the fringes are lacking, there is little effect on the quality of
reproduction of the images compared to normal 2D image display. By
inserting information for other items into the time interval of
presentation of unchanging images, it is possible to increase the
number of items to be displayed.
[0192] Specifically, in the present hologram display technology,
field images composed by fringe pattern data of several items are
displayed as a sequence of moving pictures, in such a way that not
only several items can be observed simultaneously but intermediate
tones can be displayed according to light-emitting duration ratios
of individual pixels in the corresponding field images.
[0193] Also, the present invention enables to display more objects
than is possible by the conventional technology, because the number
of display items in one frame can be reduced even when the display
apparatus has a limited capability for displaying different
gradations of gray scale.
[0194] Also, because the information content can be reduced, it
becomes possible to reduce the information content per one
field/layer or one frame, enabling a significant reduction in
required memory capacity for storing holographic information.
[0195] Further, because the entire image forming process is
digitized, degradation in image quality caused by synthesis of
wavefronts, data compression and expansion can be prevented.
[0196] It should be noted that it is possible to observe images
similar to the conventional moving pictures (based on 30 frames/s)
when the field presentation period is such that an suitable number
of field layers are presented within {fraction (1/30)} second.
[0197] [Embodiment 4]
[0198] In Embodiment 3, moving pictures for display were generated
by assigning frame image arrays (bit images) to a plurality of
screens in moving pictures under a constraint of "preserve
intensity"; in Embodiment 4, the same will be achieved by simply
distributing bit images "to be assigned by distributing to a
plurality of screens".
[0199] Embodiment 4 will be explained in the following with
reference to the drawings.
[0200] FIG. 23 is a flowchart for the present embodiment. It is
assumed that a display object is a collection of point light
sources. A view in the conventional dynamic display is termed a
frame and a plurality of images comprising a frame are termed
fields. If, for example, each pixel in one image frame is expressed
by 4-bit data of attribute bits, these four bits are arranged as
2.sup.0, 2.sup.1, 2.sup.2, 2.sup.3 so that these attribute bits are
distributed over a total of 15 fields such that 2.sup.0=1 field,
2.sup.1=2 fields, 2.sup.2=4 fields, and 2.sup.3=8 fields.
Therefore, it can be seen that one frame consists of 15 fields.
[0201] First, to display eight objects (421 to 428) such as those
shown in FIG. 24, the intended items are classified according to
their attributes (step 401). For example, if the eight items are
assumed to be classified according to:
[0202] (1) static objects having changes in gray scale or shading
(421);
[0203] (2) static objects having no changes in gray scale (422,
423, 424, 425);
[0204] (3) moving objects (426, 427, 428).
[0205] Then, information contents for the classified items are
determined (step 402). For example, items in (1) would be expressed
by 4-bit data; those in (2) by 1-bit data and those in (3) by 2-bit
data. This method of classification is the same as that explained
in Embodiment 3.
[0206] Next, a wavefront data A#421 formed by the light source 421
on the hologram screen is computed for each those items having
different shading (step 403). This item (421) requires 4-bit data
and each pixel in the wavefront A#421 is converted to a 4-bit
digital image, and the field image (b/w image) thus produced is
stored (step 404).
[0207] Similar processing is carried out for the items under
classification (2), 422, 423, 224, 425, so that wavefronts A#422,
A#423, A#424, A#425, may be computed (step 403). These items can be
expressed by 1-bit data so that digitization is carried out for the
wavefronts A#422, A#423, A#424, A#425 and field images are produced
on the basis of 1 bit-data and stored (step 404).
[0208] Similarly, the obejcts under classification (3), 426, 427,
428) are processed to compute wavefronts A#s 426, 427, 428, which
are digitized to produce field images to be stored (steps 403,
404).
[0209] Individual obejcts are displayed according to a display
sequence such as the one shown in FIG. 25, for example. File 431 is
the display sequence for item 421 (static item with shading) and
uses fifteen fields. File 432 is the display sequence for items
422.about.425 (static item with no shading), and individual objects
are shown separately so that one field contains one object. File
433 is the display sequence for items 426, 427, 428 (moving
objects), and each item requires three fields of 2-bit data. File
434 is the base line for the timing sequence for display of all
fields.
[0210] First, select an image to be shown at field timing t.sub.1
(step 405), and a display item 421 to correspond with field timing
t.sub.1 is selected (step 406), and an image to be displayed in
field-1 (an image formed by the first bit layer in the 2.sup.3
level) is fetched and is written into a hologram array Ht.sub.1 (x,
y) in step 407. Similarly, for display items 422, 426, the images
to be displayed in field-1 (first layer in the 2.sup.0 level for
item 422 and first layer in the 2.sup.1 level for item 426) are
fetched and written into the hologram array Ht.sub.1 (x, y). By
repeating the above steps (steps 406, 407) in step 408, all the
images to be displayed at field timing t.sub.1 are produced. In
other words, at field timing t.sub.1, only three items are targeted
for display.
[0211] Next, images of item 421 to be displayed in field-2 (a
second layer in the 2.sup.3 level) are fetched and written into the
hologram array Ht2 (x,y). Similarly, the images for display items
423, 426 to be displayed in field-2 are processed and written into
the hologram array Ht.sub.2 (x,y), and new wavefronts are generated
at field timing t.sub.2, where only three items are targeted for
display.
[0212] Subsequently, similar image processing operations are
carried out for all the field timings t so that a complete set of
new field images synthesized by the wavefronts and the reference
beam are produced (step 410) and individual field images containing
three display items are successively displayed (step 411), thereby
displaying all eight display items 421 to 428 inclusively.
[0213] The steps in FIG. 23 correspond to those in FIG. 12 as
follows. Steps 401.about.403 in FIG. 23 correspond to step 22 in
FIG. 12; step 404 to step 23; step 404 to step 23; steps
405.about.410 to step 24; and step 411 to step 25. In FIG. 23, a
step to correspond to step 21 in FIG. 12 is omitted.
[0214] Next, the hologram display apparatus having the features
described above will be presented with reference to a block diagram
shown FIG. 26.
[0215] The apparatus is comprised by: an object managing section
441; wavefront computation section 442; an information content
decision section 443; a bit deblocking section 444; a display
interval decision section 445; a wavefront synthesizing section
446; a display section 447; and display screen synchronizing
section 448. The operation of the apparatus is presented in the
following.
[0216] The item managing section 441 manages attributes information
of the display item, such as intensity, color, movement vectors of
each item. The wavefront computation 442 computes the wavefronts of
the individual items, each of which represents a point light
source, formed on the hologram screen. The information content
decision section 443 determines necessary amount of information to
characterize an attribute, and digitize the items accordingly.
Digitized wavefront data are managed as image arrays according to
each bitdata in the bit deblocking section 444. The display
interval decision section 445 manages the items contained in the
fields, and selects an item to be displayed as field images. The
wavefront synthesizing section 446 processes (add, for example) all
the wavefronts of a selected item so that the fringe patterns of
the selected item, acting as a point light source, are produced.
The wavefront synthesizing section 446 computes the wavefront
interference formed by the reference beam, and the results are
displayed on the display section 447. The image screen
synchronizing section 448 fetches wavefront so as to provide a
constant interval for presenting the fields which are synchronized
with the display section 447.
[0217] Sections in FIG. 26 correspond to those in FIG. 13 as
follows. The item managing section 441 in FIG. 26 corresponds to
item managing section 6 in FIG. 13; the wavefront computation
section 442, information content decision section 443, bit
deblocking section 444, display interval decision section 445,
wavefront synthesizing section 446 to image generation section 7;
the image display section 447, display screen synchronizing section
448 to the image display section 8. In FIG. 26, the display object
input section 1 and the image memory section 4 shown in FIG. 13 are
omitted.
[0218] In the present embodiment, presentation sequence of an
object with shading is determined according to the sequence of bit
arrays, but the present invention can be carried out so long as the
presentation intervals for individual bit levels are separated at
the same intervals, so the order of presentation of the fields of
different bit levels is not specified.
[0219] Also, in the present embodiment, explanations are based on
the number of display items as eight, but the number of display
objects/items is dependent on the resolution capability of the
display apparatus, and the number of displayable objects/items is
not specified.
[0220] Also, the number of fields and attribute-bit data for
objects are exemplified with a fixed quantity, but the maximum and
minimum number of fields necessary to express an item and the
information content necessary to express an item are not
specified.
[0221] Also, in the present embodiment, attributes are exemplified
by shading and movement, but other characteristics related to the
item such as color and intensity are acceptable, and methods of
classifying are not specified. Also, information content for
dynamic items is fixed in the present embodiment, but the
information content may be varied according to the magnitude of the
motion vector.
[0222] Also, the fringe patterns produced for each item in the
method and apparatus of the present embodiment can be transmitted
separately to be displayed elsewhere. In this case, static items
are sent first to be stored at the destination, and the moving
objects are forwarded next to be combined with the static items to
be displayed as a whole. This approach enables to reduce the
transmission capacity required to send holographic movie
images.
[0223] Also, in the present embodiment, displays in each field are
expressed in binary (black/white) but it is not necessary to limit
to such a binary mode. If the display device is capable of
presenting images at high speeds, multi-valued images can be
displayed well. By using multi-valued images, even more items or
more clear image can be realized.
[0224] Also, in the present embodiment, the period of presentation
of the items without shading is exemplified with one fixed period,
but it is possible to vary the overall shading by controlling the
cycle width. In other words, if the display interval is lengthened,
the object would appear darker overall, and if the display period
is shortened, the item would appear brighter overall.
[0225] Accordingly, the present invention enables to reduce the
number of display object in one frame so that more items can be
displayed even on a low resolution display apparatus.
[0226] Also, information content required to express one field or
one frame can be reduced so that the transmission capacity can also
be reduced.
[0227] Also, the overall processing is digital so that image
quality degradation caused by wavefront synthesis, data compression
or expansion can be prevented.
[0228] Accordingly, the present method for displaying computed
fringe pattern holograms is carried out by: classifying a display
object according to attributes of the items; computing fringe
patterns generated by classified display objects; determining
information content necessary according to attribute for each
display object; digitizing the generated fringe patterns according
to individual information content; deblocking bit arrays of pixels
of the digitized images into pixel arrays for different bit-levels;
assigning the pixels in the pixel array by distributing the pixels
in a plurality of moving pictures, thereby producing digital moving
pictures having a display interval varying in accordance with
information content of each display object; and displaying the
digital moving pictures of display objects.
[0229] The apparatus for executing the method is comprised by: an
object managing section for managing information on attributes of
display objects; an information content decision section for
deciding information content for each display object according to
attributes of the object; a wavefront computation section for
computing fringe patterns for each display object; a bit
de-blocking section for separating a bit array of pixels into a
pixel array for different bit-levels; a display interval decision
section for determining display period and display level according
to the information content of each display object; and a wavefront
synthesizing section for synthesizing fringe patterns generated by
pixels in the pixel arrays for different bit-levels; a display
screen synchronizing section for controlling display timing of
fringe patterns thus synthesized; a display section for
successively displaying fringe patterns composed according to a
controlled display timing for each item.
[0230] Also, the apparatus may be comprised by: an image
transmission section for transmitting synthesized fringe patterns
for static display object first and sending fringe patterns for
dynamic display objects afterwards; an image receiving section for
storing fringe patterns for static display objects to be combined
with successively transmitted fringe patterns for dynamic display
objects; and the display image synchronizing section is used to
control display timing for displaying fringe patterns produced by
the image receiving section, and the display section displays the
synthesized fringe patterns produced in the image receiving
section.
[0231] Accordingly, the present invention enables to display more
items by choosing a quantity for the information content to suit
the attributes (intensity, movement etc.) of the display object;
controlling the presentation interval according to the information
content; and sequencing frames/field images as moving pictures; so
that as a whole, more items are displayed even though each one view
(image) contains less number of items. This approach enables to
relax the strict resolution requirement for the display
apparatus.
[0232] In other words, items having such a gray scale shading as
textures that require a high information content are expressed by a
high bit-level (8-bit for example), and those items without such
gray variations that require less information content are given a
lower-bit level (4-bit for example). Poor quality of reproduction
of colors or textures are less noticeable in the images of moving
objects so that a fewer number of bits is adequate to express such
moving objects. Further, because the display intervals are adjusted
according to the bit-level so that display objects having less
information content can display more objects.
[0233] In the conventional approach, if one item is expressed by
8-bit data, the resulting digital image to express one gray scale
for one pixel required a presentation of 255 layers
(=2.sup.7+2.sup.6+2.sup.5+2.sup.4+2-
.sup.3+2.sup.2+2.sup.1+2.sup.0) of fields. That is, for each item
to be displayed, it is necessary to present a uniform number of
field layers.
[0234] However, the present invention enables to reduce the number
of layers of presentation fields for lower information content
(small number of bits) so that more objects than is possible by the
conventional technology can be displayed within the same number of
fields.
[0235] Also, by expressing each item by a bit array, and computing
fringe patterns for each bit, shading in fringe patterns can be
expressed in binary, black or white, so that there is no need for
providing an intermediate color tone in the display device to
enable simplifying manufacture of display device applicable to the
present invention.
[0236] It is possible to observe images similar to the conventional
moving pictures (based on 80 frames/s) when the field presentation
period is such that an suitable number of layers are presented
within {fraction (1/30)} second.
[0237] [Embodiment 5]
[0238] The conventional pulse-width modulation method described
above is based on presenting the same binary coded images more
often for binary bit images of higher bit-levels. This method of
hologram display is the same as repeated presentations of binary
holograms. One of the problems with the binary hologram display is
that local bright spots or speckle noise are observed throughout
the image. Making the matter worse for the pulse-width modulation
method, when the same image is repeatedly presented, the presence
of speckle noise is emphasized and the viewer perceives noisy
images.
[0239] Therefore, Embodiment 5 presents a method and apparatus for
resolving such a problem by preparing (adding) the digitized field
image arrays (bit images) according to the attribute bits in such a
way that the number of fields corresponds to the bit-level of the
attribute bits, but the display images are processed using a
different binarization process between the fields of the same
bit-levels. By adopting this approach, the locations of speckle
noise are so altered between the field images that inhomogeniety in
the background shading is eliminated to produce clearer images. The
embodiment will be explained in the following with reference to the
drawings.
[0240] FIG. 27 is a flowchart of the method of Embodiment 5.
[0241] First, the data for the display objects are separated into M
pieces (step 501). There would be many different techniques of
separation, for example, if a display object is represented by an
image layer, the image may be separated into 4 pieces, or if there
are many objects in a 3D-space, each item may be separated from the
other.
[0242] For each separated data, computed holograms are prepared;
for example, fringe patterns having N-bit shading (e.g., 8 bit).
Assume that there are M pieces of data and M layers of holograms
are to be produced (step 502).
[0243] Next, because each pixel in the M-set of fringe patterns is
comprised by N-bits, N layers of images are produced for each pixel
in the fringe patterns of the same bit-level (step 503). In other
words, for each of the M layers of fringe patterns, N layers of bit
images will be prepared.
[0244] Next, because there are M layers for each bit-level, those
pixels belonging to the same bit-level are added. This step
produces N layers of bit images each having 0.about.M shading
gradations (step 504).
[0245] Next, bit images are fetched serially (step 505) for
processing. An example of image processing used in the present
embodiment is error diffusion processing. That is, binarization (0
or M for shading value) is carried out using a threshold value of
shading (N/2 for example). In performing this step, errors caused
by binarization are diffused to the neighboring pixels. For
example, as shown in FIG. 28, weighted values of errors for pixel
21 are added to the neighboring pixels (for example, weighting of
{fraction (3/16)}, {fraction (5/16)}, {fraction (1/16)}, {fraction
(7/16)}) in step 506.
[0246] In carrying out step 506, several variations in the
threshold values, weighting of errors or diffusion direction are
prepared, and the error diffusion processing and binarization are
carried out so that the threshold value, weighting and diffusion
direction are different for each image, and after this processing,
the images are displayed (step 507).
[0247] Steps 505.about.507 are repeated by fetching the same image
to repeatedly present the number of layers corresponding to the
bit-level of the image (step 508). For example, if the image
contains 8-bit shading, an image having the highest bit-level would
be fetched 2.sup.7=128 times.
[0248] The steps shown in FIG. 27 correspond to those shown in FIG.
12 as follows. Steps 501, 502 in FIG. 27 correspond to steps 22, 23
in FIG. 12; steps 503.about.506 to step 24; steps 507, 508 to step
25. In FIG. 23, a step to correspond to step 21 in FIG. 12 is
omitted.
[0249] In the present embodiment, M layers of holograms are all
represented by N-bit data, but it is not necessary to limit to the
same number of bits for all the layers. Layers may have a different
number of attribute bits, and in this case, the number of layers
equal to the maximum number of bits may be prepared (if an image
has no corresponding bit, 0 or black is assigned. Or, by using bit
images from other images, differences in the number of bits may be
overcome).
[0250] Also, in the present embodiment, error diffusion processing
and binarization are carried out in real-time at the time of
displaying the images, but it is also possible to store prepared
images of the same bit-level which have been pre-processed for
error diffusion and binarization, so that the order of processing
is not specified.
[0251] Next, the apparatus for executing the above method will be
presented with reference to the block diagram shown in FIG. 29.
[0252] The apparatus is comprised by: a gray level image generation
section 531; a bit image generation section 532; an image storage
section 533, an image processing section 534; an image display
control section 535; and an image display section 536. The
operation of the apparatus is as follows.
[0253] First, the gray level image generation section 531 produces
a computed hologram of a gray level image which is sent to the bit
image generation section 532. The bit image generation section 532
decomposes the gray level image according to a pre-determined rule
into a plurality of data-sets (gray level images). Or, a plurality
of gray level images may be generated in the image forming stage in
the gray level image generation section 531, and the images are
forwarded to the bit image generation section 532. The decomposed
gray level images are converted to bit images in the bit image
generation section 532, and are stored in the image storage section
533. The image processing section 534 performs error diffusion
processing and binarization to the separated bit images, and the
processed bit images are similarly stored in the image storage
section 533. Or, error diffusion processing and binarization may be
performed in real-time during the display process under the control
of the image display control section 535 to repeatedly display the
same image according to bit-levels of the image.
[0254] The structures in FIG. 29 correspond to those in FIG. 13 as
follows. The gray level image generation section 531, bit image
generation section 532, image processing section 534 correspond to
image generation section 7 in FIG. 13; the image storage section
533 to image memory section 4; the image display control section
535, image display section 536 to image display section 8. In FIG.
26, a step to correspond to display object input section 1 shown in
FIG. 13 is omitted.
[0255] According to the control methodology described above, to
display a file 540 shown in FIG. 30A, a shading fringe pattern
image of a computed hologram will appear as shown by the fringe
pattern image in file 541 in FIG. 30B. The fringe pattern 541 is
decomposed into bits and the resulting image after binarization by
different error diffusion processing techniques are shown in files
542 and 543. The features of the fringe patterns are preserved
while differences in the local shading can be observed.
Accordingly, even though the same original binary coded hologram is
repeatedly displayed, because the binary coded image is processed
with different error diffusion techniques, the locations of
speckles are different for each frame during its reproduction, and
the noise signals are distributed and the overall image of a higher
quality is observed. File 544 shown in FIG. 30E has not be
subjected to different processing frame by frame so that the
speckles are emphasized and the contrast for the original image of
the word G is decreased relative to a clear image containing less
speckle noise shown in file 545 in FIG. 30F which is an example of
an image that has been treated by the method of the present
invention.
[0256] The display apparatus of the present invention is comprised
by: a hard disk or other similar storage device which can store and
freely retrieve image data such as holograms and its bit images;
buffer memories or other related devices required when performing
processes such as generation of shading images and bit images; a
display device such as liquid crystal display panel for displaying
images such as processed digital holograms; and an input device
such as keyboard and mouse. Such devices are controlled by a
computer or other similar control device according to
pre-determined algorithms or a sequence of steps such as that
illustrated in flowchart in FIG. 1. The application programs to
carry out such algorithms and steps can be recorded and distributed
in readable memory devices such as floppy disk, pc card (personal
computer memory card), magneto-optic disk, compact disk and digital
video disk.
[0257] As described above, the present invention enables displaying
of a holographic solid object as digital images so that it becomes
possible to display very clear images that contain fewer speckle
noises.
[0258] Accordingly, the method of the present invention is
comprised by the steps of: generating a holographic image having
shading gradations; decomposing the gradation values of each pixel
in the gray level images into bit arrays; generating bit images
according to individual bit-levels of the gray level images;
processing bit images in such a way that, those bit images to be
repeatedly presented for a time interval corresponding to
bit-levels are subjected to different image processing procedures;
and displaying bit images which have been so processed on a display
device.
[0259] The apparatus to execute the method is comprised by: the
shading image generation section for producing computed holograms
having gradations; a bit image generation section for converting
shading images into bit images; an image storage section; an image
display control section for controlling a time interval for
repeatedly presenting bit images of a specific bit-level; and an
image processing section for providing different image processing
steps for each of the repetitively presented bit images; and an
image display section for displaying processed bit images.
[0260] Recording media may record application program suitable for
executing the present invention for executing the steps of:
generating shading images for comprising holograms having
gradations described N-bits; separating gradations of each pixel of
the shading image into bit arrays; generating a bit image for each
bit-level of the pixels; and image processing the bit images in
such a way that, those bit images, to be repeatedly presented for
an interval time of presentation according to bit-levels, are
subjected to different image processing procedures; and displaying
images which have been so processed on a display device.
[0261] In the conventional pulse-width modulation method, the
higher the bit-level the larger the number of repetitions. For
example, if the images are represented by 8-bit data, the same
image is presented 128 times for the highest bit-level while for
the lowest bit level, only one presentation is made. Because the
image presented are binary coded holograms (b/w), noises are
emphasized even more intensely, in other words, inhomogeniety in
intensity (gradations) becomes more noticeable. As described above,
holograms prepared as N-bit level digital images are used to
generate N layers of bit images for each bit-level, and when
presenting the same bit image, images prepared by different image
processing procedures, including error diffusion processing, are
presented so as to vary the locations of appearance of the
speckles. By adopting this procedure, the higher the number of
presentations of the same bit image, higher the probability of
mutual cancellation of speckles, thereby enabling to reduce
inhomogeniety in the background and produce clearer images.
[0262] [Embodiment 6]
[0263] Embodiment 6 relates to a method and apparatus for producing
a plurality of digital images for displaying moving picture by
extracting only the high bit-level digital images representing the
attributes of a display object, and assigning such high bit-level
images to a plurality of screens in the moving pictures.
[0264] The method is comprised by the steps of: computing fringe
patterns for a plurality of display objects produced by the
reference beam and light from each display object; converting the
fringe patterns to digital images; separating bit-arrays of each
pixel in digitized images into pixel arrays for each bit-level;
extracting those pixels having bit-levels higher than a
pre-determined bit-level from the decomposed pixel arrays;
distributing those pixels extracted to a plurality of screens of
moving pictures, thereby producing dynamic digital images comprised
by pixel arrays of high bit-levels; and displaying produced dynamic
digital images. These steps can be recorded in a recording medium
in the form of programs to be read by a computer to produce
holographic fringe patterns produced by light emitted by an object
and a reference beam and to display digital images of the fringe
patterns.
[0265] When extracting the images of higher ranking bit-levels, the
level of the high bit-levels to be extracted may be varied to suit
the attributes of a display object.
[0266] Also, the apparatus to execute the method is comprised by: a
fringe pattern computation section; a digital image generation
section for converting the fringe patterns to digital images and
arranging bit-arrays of pixels into pixel arrays of different
bit-levels; a moving picture generation section for selecting
images of display object having higher bit-levels and generating
moving picture arrays for display; and a display section for
successively displaying moving pictures.
[0267] The moving picture generation section may alter the
bit-level according to the attributes of a display object, when
selecting pixel images of higher ranking bit-levels.
[0268] The method and apparatus of the present embodiment will be
exemplified in a simulation to demonstrate that the above approach,
of selecting only higher ranking bit levels in the digital images,
produces images of excellent visual qualities.
[0269] The simulation studies showed that, when the fringe patterns
generated by computer generated holography method are decomposed
into digital images for different bit-levels and reproduced as
pixel arrays for each bit-level, pixels that make a real
contribution in reproducing the images were high-level bits
only.
[0270] For example, a hologram produced by an image shown in file
611 in FIG. 31A produces a fringe pattern shown in file 612 in FIG.
31B. When the fringe pattern in file 612 is converted to an 8-bit
digital image, and the bit-arrays for each pixel are decomposed
into pixel arrays for each bit-level, the images appear as shown in
file 613 in FIG. 31C. In other words, pixel arrays 6131.about.6138
shown in file 613 correspond, respectively, to
[0271] 2.sup.0.about.2.sup.7 bit-levels of the digitized images of
the display object. Images reproduced using only these pixel arrays
are shown in file 614 in FIG. 31C. It means that images 6141 to
6148 are produced from pixel arrays 6131 to 6138. Reproduced images
for digital image display can be computed by weighting the images
shown in file 614 according to the bit-levels. When the pixel
arrays 6141.about.6148 are added with suitable weighting values to
correspond with the respective bit-levels (2.sup.0.about.2.sup.7),
then an image shown in file 616 in FIG. 31E is obtained.
[0272] On the other hand, the images shown in file 615 in FIG. 31D
are obtained by adding images having the high level bits only, and
the image 6151 is comprised by a sum of images of 2.sup.7 level (or
image 6148 only), 6152 by 2.sup.7 and 2.sup.6 level images (a sum
of images 6148 and 6147); 6153 by a sum of 2.sup.7, 2.sup.6 and
2.sup.5 level images (a sum of images 6148, 6147 and 6146); 6154 by
2.sup.7, 2.sup.6, 2.sup.5 and 2.sup.4 level images (a sum of images
6148, 6147, 6146 and 6145). From these results, it can be observed
that by using only the images with high level bits (in this case,
2.sup.7, 2.sup.6, 2.sup.5, and 2.sup.4), an image equivalent in
quality to the image shown in file 616 can be obtained. In other
words, even if the images with low level bits are discarded, there
is little effect on the visual quality of the final image
produced.
[0273] Therefore, the present invention enables to reproduce
computed hologram by replacing lower bit level images with higher
bit level images without degrading the image quality, thereby
significantly reducing the information content required to produce
a high quality holographic image.
[0274] The present embodiment will be presented with reference to
the drawings.
[0275] FIG. 32 is a flowchart for the method, and FIGS.
33A.about.33H show examples of the images obtained. The images are
represented by 8-bit data, and two items (two images in this case)
are used. For example, an original image shown in file 630 in FIG.
33A is a combination of image items "F" and First, fringe patterns
are generated from the original image (step 601). File 631 in FIG.
33B shows fringe patterns generated by image F only. File 632 in
FIG. 33C shows fringe patterns generated by image G only. File 633
in FIG. 33D shows fringe patterns generated by the original image
630.
[0276] If the dynamic range and resolution of the display device
are sufficient, fringe pattern 633 would acceptably reproduce the
original image. For electronic displays with relatively inferior
resolution and display capability for intensity levels, such a
fringe pattern 633 would lose fine features (high frequency
components) in the reproduction. On the other hand, if the fringes
are widely spaced as in fringe patterns 631, 632, electronic
display device will be acceptable because there are fewer high
frequency components.
[0277] Next, computed fringe patterns 631, 632 are converted to
digital images, and pixel arrays for each bit-level are obtained
(step 602). Image 634 in FIG. 33E and image 635 in FIG. 33F are
pixel array images for each bit-level.
[0278] Next, from the pixel array images, only those images with
higher bit-levels (for example, top 4 levels) are extraced (step
603), to produce image array for dynamic representation of the sets
of pixel array images (step 604).
[0279] The pixel array images are serially displayed while giving
weighting (step 605). Methods of weighting may include a technique
of including relative intensity values corresponding to the
bit-levels or adjusting the presentation time intervals for the
same image according to weighting factors. The resulting image thus
produced is shown in file 637 in FIG. 33H.
[0280] The steps in FIG. 32 correspond to those in FIG. 12 as
follows. Step 601 in FIG. 32 corresponds to step 22 in FIG. 12;
step 602 to step 23; steps 603, 604 to step 24; step 605 to step
25. In FIG. 32, a step to correspond with step 21 is omitted.
[0281] Also, in the present invention, when extracting images with
higher bit-levels in step 603, it is possible to change the order
of the high bit-levels to be extracted.
[0282] For example, one approach is to use images of 2.sup.7 level
only for fringe pattern 631, and use 2.sup.6 to 2.sup.0 for fringe
pattern 632. With this method, no special weighting would be
necessary for the pixel array so that normal digital display
processing can be used and normal digital image display apparatus
can be used.
[0283] Other example would include the use of 2.sup.7 level only
for fringe pattern 631, and 27 to 26 levels for fringe pattern
632.
[0284] In this case, because individual weighting operations are
necessary for the pixel array images, it is necessary to control
intensity and presentation interval, but high quality images are
produced.
[0285] Next, structure and operation of the embodiment will be
explained. FIG. 34 is a block diagram of an example of the
apparatus for executing the method of the present invention.
[0286] The apparatus is comprised by: a fringe pattern computation
section 641; a digital image generation section 642; a moving
picture generation section 643; a display section 644; and an image
storage section 645. The operation of the apparatus will be
explained in the following.
[0287] The fringe pattern computation section 641 computes fringe
patterns of the display objects (display images in this case),
interference fringes produced between the display object and the
reference beam (a plane wave radiated from behind the images). The
digital image generation section 642 converts the fringe patterns
generated by the fringe pattern computation section 641 are
digitized according to the capability of the display apparatus 644
to produce images according to the bit-levels of the pixels. For
example, 8-bit data may be used if the display apparatus 644 has a
256-gradations for shading, and 8 layers of pixel array images will
be produced. The moving picture generation section 643 selects only
those pixel images with high bit-levels from a series of pixel
array images to produce moving pictures. The display section 644
displays an array of pixel array images in succession.
[0288] The structures in FIG. 34 correspond to those in FIG. 13 as
follows. The fringe pattern computation section 641, digital image
generation section 642, moving picture generation section 643
correspond to image generation section 7 in FIG. 13; image memory
section 645 to image memory section 4; image display section 644 to
image display section 8. In FIG. 34, a step to correspond to
display object input section 1 is omitted.
[0289] A variation of the present embodiment would be to select a
different number of pixel array images for each item in the pixel
array generation section 643.
[0290] The display apparatus of the present invention is comprised
by: a data reading device for obtaining data from a recording
medium; a hard disk or other similar memory device and the like
which stores and freely retrieves image data such as holograms and
their bit images obtained from the recording medium; buffer
memories or other related devices need for performing various
processing tasks; a display device for displaying information
necessary to perform processing tasks and displaying images such as
digital holograms; and an input device such as keyboard and mouse.
Such devices are controlled by a computer or other similar control
device according to pre-determined algorithms or a sequence of
processing steps such as those illustrated in FIGS. 31A.about.E,
.about.FIG. 34. The application programs to execute such algorithms
and steps can be recorded and distributed in readable memory
devices such as floppy disk, pc card, magneto-optic disk, compact
disk and digital video disk.
[0291] As described above, the present invention enables to reduce
effective data volume by inserting image information into those
images of bit levels which are not necessary.
[0292] It has been demonstrated in Embodiments 1.about.6 above,
that the present method and apparatus for displaying computed
holograms enable to reduce data volume thus allowing to display
many objects simultaneously or to reduce degradation in image
quality caused by superposition of fringes.
[0293] Other variations of the basic method would include a
combination of the technique of displaying a plurality of items by
appropriate sampling techniques presented in Embodiments 1 and 2
with the technique of designing appropriate method of composing
display images presented in Embodiments 3 and 4.
[0294] It would also be possible to incorporate the technique of
reducing the speckle noise presented in Embodiment 5 in those cases
presented in other embodiments when a dynamic frame contains a
repetition of the same image data.
[0295] It should be mentioned that various devices in the display
apparatus presented in FIGS. 2 and 13 and other embodiments are
equivalent to a processing section in its function.
[0296] Holograms may be computed and displayed under the control of
a computer system for executing application programs recorded on
readable recording medium according to the method explained with
reference to FIGS. 1 and 12. Computer system in this context refers
to operating systems and hardwares for peripheral devices. Readable
recording medium includes such common recording devices as floppy
disk, magneto-optic disk, read-only-memory (ROM), and CD-ROM and
internal or external hard disk. It would also be obvious that such
readable recording medium can include dynamic program storage
devices, associated with network communication such as Internet and
telecommunication circuits, as well as volatile memories for short
term data storage in servers and network computers. The term
"application programs" is generic and may consist of instructions
to execute a portion of any function, or of written functions to
operate in combination with programs contained in a computer
system.
[0297] Fields of application of computed holography presented in
Embodiments 1.about.6 will be mentioned briefly.
[0298] The method and apparatus for hologram display relate to a
technology for displaying solid objects represented by computed
holograms, and are applicable to transmission/display/storage of
three-dimensional images of objects. It would be suitable as a
display device for virtual reality images. Specifically,
application fields would include industrial and home-based
applications such as three-dimensional broadcasting at TV
frequencies, museum displays, computer-aided design (CAD) systems,
virtual reality computer games, and medical applications such as
surgery simulations, computer tomography image display, as well as
to other such solid object image display devices as headup display
apparatus for a line-sight-display of solid objects.
* * * * *