Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
1999-05-18
2002-04-09
Vo, Cliff N. (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
C345S419000
Reexamination Certificate
active
06369817
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a volume rendering technique which plays an important role when numeric data (volume data) defined in a three-dimensional space is visualized on a display screen. In particular, the present invention relates to a hardware-equipped image synthesis system which makes it possible to perform high speed drawing processing for three-dimensional volume data to be displayed on a display screen. Further, the present invention relates to an image synthesis system constructed by using a small amount of hardware resources.
2. Description of the Related Art
The technique, which is used when a three-dimensional model represented by numeric data (volume data) defined in a three-dimensional space is visualized on a screen of a display unit, is called “volume rendering”. For example, in the medical field, the volume rendering technique is utilized when numeric data (volume data) concerning the inside of human body, which is obtained by using a measuring instrument such as CT scanner and MRI, is displayed as an image on a display. Further, the volume rendering technique is also utilized in a variety of fields.
The volume data for representing the three-dimensional model is expressed by using various methods. However, there is a method in which expression is made by using a set of small cubes (voxels) arranged on a three-dimensional lattice, in the same manner as in the method in which a figure on a two-dimensional plane is expressed by using a set of small square picture elements (pixels) arranged on two-dimensional lattice points.
The technique for expressing the three-dimensional model by means of the voxel expression has the following features. That is, although the amount of data is enormous, the data structure is simple, in which the set operation can be performed extremely easily. The technique is not limited to the expression of a three-dimensional model which is artificial and regular, but it is suitable when an extremely irregular shape existing in the nature is expressed.
One of known rendering techniques based on CG (Computer Graphics) is a technique called “ray casting method” or “ray tracing method”. In the ray casting method, an objective three-dimensional model, which is constructed by voxels, is arranged in an imaginary three-dimensional space defined in a memory space. Numeric data, which represents internal information of the objective three-dimensional model, is given as volume data to each of the voxels. According to the ray casting method, various pieces of internal information concerning the objective three-dimensional model can be freely visualized depending on the purpose. Therefore, this method attracts attention in recent years as a new visualizing technique (volume visualization) for the three-dimensional model represented by those for living bodies.
The volume data as described above includes data called “field data” on the point (voxel) on the three-dimensional lattice. The field data is data, for example, for the color value and the transparency. In the case of the CT scanner, the density of the tissue in the living body is used as the field data. The voxels are colored in accordance with the density to display (visualize) the volume data on the screen. Thus, the situation of the entire data can be made more comprehensible.
The method for visualizing the volume data has been hitherto roughly classified into two kinds of methods. The first method is the surface rendering method in which voxels having equal data in the volume data are connected to one another to form an equivalent surface, and primitives such as surfaces and lines are allotted to the equivalent surface to make display. The second method is the method in which the volume data is directly displayed by using, for example, the ray casting method.
When the surface rendering method is used, an equivalent surface (appreciated as expansion of the two-dimensional contour line) is prepared by connecting equal volume data. Surfaces and patches are applied to the equivalent surface to perform the rendering by means of the surface graphics so that the display is made. That is, this method is also called “slice-by-slice method”. Japanese Patent Publication (a publication of examined application) No. 7-120434 shows that the details of the method are described in the following document (A).
(A) Drebin, R. A., Carpenter, L., and Hanrahan, P., “Volume Rendering”, Computer Graphics, Vol. 22, No. 4, pp. 65-74, 1988.
When the ray casting method is used, the following procedure is adopted. That is, rays (lines of sight) passing through respective cells are generated from a point of sight (pixel) on a surface to be displayed (surface on the display). The volume data is subjected to integration along the rays. Actually, in place of the integration, approximate calculation is performed by means of numerical integration based on the use of values obtained at several sampling points disposed along the rays.
The sampling method includes a method in which sampling is performed at equal intervals along the rays, and a method in which points of intersection between the ray and the cell are used as sampling points to make evaluation. The objective data is the volume data defined on the orthogonal structural lattice in the former case or on the non-structural lattice in the latter case. The Japanese Patent Publication No. 7-120434 also shows that the details of the above are described in the following documents (B) and (C).
(B) Levoy, M., “Display of Surfaces from Volume Data”, IEEE Computer Graphics and Applications, Vol. 8, No. 3, pp. 29-37, 1988.
(C) Garrity, M. P., “Raytracing Irregular Volume Data”, Computer Graphics, Vol. 24, No. 5, pp. 35-40, 1990.
However, in the present circumstances, when it is intended to realize a high degree of display function by means of the conventional volume rendering technique as described above, it is necessary for any of the techniques to use enormous calculation time.
Such a situation results from the following fact. That is, in the case of the two-dimensional surface rendering, exclusive processors have been developed for the purpose of high speed processing, and they are commercially provided to the market as major constitutive elements for the engineering work station. As compared with the two-dimensional surface rendering, no hardware-equipped apparatus capable of high speed processing has been provided for the three-dimensional process, because of the following reasons.
That is, the following fact is firstly pointed out. A large capacity memory is required to accumulate the volume data for the three-dimensional space. Further, in order to perform the operation processing for the volume data in real time, it is necessary to read, at a high speed, the data accumulated in the large capacity memory. However, neither suitable memory nor appropriate interface is available to solve such a problem, resulting in an obstacle to realize hardware equipment.
Secondly, there has been the following cause. That is, in order to realize the three-dimensional visualization, it is necessary to perform the data sampling from the volume data over the entire three-dimensional space. When the sampled volume data is subjected to operation, a large capacity memory is required to accumulate intermediate results of the calculation. Therefore, it is difficult to realize hardware equipment. Such a process has been hitherto realized dominantly by using any software.
Because of the reasons as described above, a lot of time has been hitherto required for the apparatus for visualizing the volume data concerning the three-dimensional space. An inconvenience arises in that when the point of sight or the objective body is moved in the three-dimensional space, it is difficult to synthesize, in real time, an image which reflects the movement.
When the three-dimensional image is displayed by using the conventional volume rendering technique, it is necessary to use enormous volume data concerning the three-dimensional space for expressing the
Ishikawa Masashi
Kashiwade Koji
Mimura Toshiharu
Toeda Kengo
Yamane Daisaku
Frishauf, Holtz Goodman, Langer & Chick, P.C.
Japan Radio Co. Ltd.
Vo Cliff N.
LandOfFree
Image synthesis system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image synthesis system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image synthesis system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2914915