Computer graphics processing and selective visual display system – Computer graphics processing – Attributes
Reexamination Certificate
2000-06-02
2004-01-27
Luu, Matthew (Department: 2672)
Computer graphics processing and selective visual display system
Computer graphics processing
Attributes
C345S611000
Reexamination Certificate
active
06683617
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to an antialiasing method and an image processing apparatus using the same, and more particularly to an antialiasing method capable of a high-quality image display without causing any significant reduction in processing speed and any increase in apparatus scale.
2. Description of the Related Arts
In recent image processing systems such as game machines making use of computer graphics, it is desired to enhance the virtual realities through high-quality and high-speed displays of graphics having three-dimensional coordinates, on a two-dimensional screen such as a CRT composed of a multiplicity of display elements (pixels).
In such image processing systems, graphics to be displayed in the three-dimensional space are regarded as accumulations of polygons (polygonal surfaces), so that use is made of attribute data (hereinafter, referred to as polygon data) possessed by the polygons to create and display final two-dimensional images.
The polygon data is typically polygon vertex data including three-dimensional coordinate values of each vertex in a predetermined three-dimensional space, vertex colors, texture mapping coordinate addresses, vertex transparencies, and vertex normal vectors, etc.
FIG. 14
illustrates a schematic configuration of the image processing system. Description will hereinbelow be made of a flow of typical processing effected to display images on the screen. In
FIG. 14
, a CPU
1
provides a control of entire functions of the image processing apparatus and performs a management of game sequences for example. A RAM
7
stores therein programs, data and display lists, etc, for use in the CPU
1
.
The polygon data is fed to a geometry computing unit
2
in which it is subjected to processings such as coordinate transformation and perspective transformation, for the conversion into two-dimensional data corresponding to the display screen. The thus converted polygon data is stored in a polygon buffer
8
.
A hidden-surface processing unit
3
then executes a hidden-surface removal (hereinafter, referred to as hidden-surface processing) by use of the polygon data. In cases where a plurality of polygons are superposed, one upon another, at the same site on the two-dimensional screen, it is necessary to judge which polygons are actually visible from the user's eyes. The hidden-surface processing makes such judgment to determine the range of each polygon to be displayed.
In response to the results of the hidden-surface processing, a rendering unit
4
then converts the polygon data into pixel based data. This processing is called rendering. For the data converted on a pixel-by-pixel basis (hereinafter referred to simply as pixel data), a texture unit
5
and a shading unit
6
execute texture mapping and shading processes, respectively. A texture buffer
9
stores therein texture data and provides data corresponding to the pixel texture mapping coordinate addresses, allowing texture mapping to be effected on a pixel-by-pixel basis.
Through these steps, determination of the pixel based data such as display colors is complete of each polygon. Using such data, a blending unit
10
defines finally the colors to be displayed on the screen, on a pixel-by-pixel basis. More specifically, in cases where for example a plurality of polygons share a single pixel (i.e., where the pixel includes plural pieces of valid pixel data) with pixels being translucent, the color data is produced from a values possessed by the respective pixels, to determine the final pixel display color.
The image data created on &agr; pixel basis is written to the frame buffer
11
. The frame buffer
11
feeds the data as frame images to a display unit for final display of the images.
Although images are displayed on the CRT or the like through the above processing steps, such digitized images represent the images in the form of the aggregate of pixels (the set of rectangles), intrinsically suffering from the problem of aliasing. This state is shown in
FIGS. 15A
to
15
C.
FIGS. 15A
to
15
C illustrate by way of example the case where a graphic of
FIG. 15A
is desired to represent. Even in the case of internally possessing information on a theoretically correct geometry as shown in
FIG. 15A
, once it is represented as the aggregate of pixels, diagonal lines such as the triangular polygon edges for example may result in a stepwise display as indicated in FIG.
15
B. This is due to the fact that a single color is used to represent the entire pixel within which the color may internally (theoretically) vary. This is a problem intrinsically difficult to solve as long as the digital images are represented as the aggregates of pixels.
In particular, in cases where without taking any measures thereagainst, the color possessed internally by a representative point (e.g., a pixel center) within a pixel is regarded as the pixel color for processing as in the prior art, the above stepwise display will become remarkable.
Such a problem could however be solved to a substantially (visually) negligible extent by combining plural pieces of color data originally internally possessed by a pixel to determine the display color of that pixel. The processing therefor is called antialiasing. The graphic of
FIG. 15C
is an example of the display image obtained when subjected to the antialiasing processing.
It is one key function for the creation of highly realistic, high-quality images to execute such antialiasing processing.
This means that the antialiasing processing is indispensable for any image processing systems such as game machines since it ensures high-quality images in a pseudo-manner without using any high-resolution monitor.
For the determination of one pixel color in the antialiasing processing, however, it is necessary to generate information on a plurality of colors associated with the pixel and information on positions of those colors within the pixel and to make a certain calculation by use of them. For this reason, the antialiasing processing has hitherto been regarded as generally heavy processing, inevitably bringing about any increase in the processing time and system scale for the acquisition of high-quality images.
In order to alleviate such a drawback, a variety of measures have been adopted, although all the measures are not free from any deficiencies. The translucent edge technique commonly used in, e.g., personal computers is a color synthesizing technique in which transparency is determined depending on the proportions of the area occupied by polygons in a pixel within which polygon edges lie.
Although this technique is advantages in terms of the processing speed and system scale, it is problematic from the viewpoint of quality since the picture may often break or background colors may remain left. This arises from the fact that all the polygons are treated as translucent at their edge portions, so that the edge portions are not to be subjected to correct hidden-surface processing in the hidden-surface processing unit
3
described above.
An alternative conventional technique is an over-sampling method. In this technique, processing is, made at a higher resolution than a display resolution on the screen and averaging is made immediately before the display, to obtain the display resolution. More specifically, this is a method in which the entire process steps up to the frame buffer
11
described above are executed on the basis of a smaller rectangle (subpixel) than the pixel size to be displayed.
This method is advantageous in that with its simple concept, the above-described standard processing steps need not substantially be changed and that a fairly high-quality is expected, although the amount of data to be processed increases in proportion to the number of subpixels, resulting in an increase in processing time and in system dimensions.
Some other methods than the above-described techniques have also been conceived, but none of them have presented well-balanced results in terms of the processing speed
Morioka Seisuke
Naoi Junichi
Dickstein , Shapiro, Morin & Oshinsky, LLP
Luu Matthew
Sega Enterprises Ltd.
LandOfFree
Antialiasing method and image processing apparatus using same does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Antialiasing method and image processing apparatus using same, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Antialiasing method and image processing apparatus using same will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3229979