JPS6366682A - Image producing and processing system - Google Patents

Abstract

PURPOSE:To correct image data at a high speed by scanning a beam tracking tree through a relevant route and using successively space information held for each node to produce the image data. CONSTITUTION:An image producing and processing part 34 tracks beams with object data 3 set as an input and the image data. At the same time, the space information outputs obtained at each intersecting point of beams is turned into data where a visual point is defined as a root for each beam together with each light spot defined as a node in a tree form. Thus a beam tracking tree 37 is obtained and held. An image correction processing part 38 is started after designation of the color, etc. of the body of the data 3 is changed. Then the part 38 extracts the space information held by each node according to the structure of the tree 37 set for each beam. Based on this space information and the data 3, data 5 is produced. Thus calculation processing is omitted for tracking of beams in case correction is given to designation of colors, etc. Then the image data can be produce at a high speed with connection input.

Description

[Detailed Description of the Invention] [Summary] This is a processing method for speeding up image modification processing using image generation processing using a ray tracing method.

To generate an image, the intersection of a ray and an object is traced, and the spatial information generated for each intersection is held as ray tracing tree data linked in a tree with the intersection as a node.
When specifications such as attributes are changed, image data is generated using only the data of the ray tracing tree without performing ray tracing.

This method eliminates the need for spatial information calculation processing, which takes a relatively long processing time, for images in which only the color or the like has been changed, and images can be generated at high speed.

C. Industrial Application Field] The present invention relates to image processing in a computer, and particularly to a method for generating an image with a partially changed designation.

In image generation processing, ray tracing is widely used as a process for obtaining high-quality image data.

[Conventional technology]

FIG. 3 is a block diagram showing an example of the configuration of the computer.

The processing device 1 receives the object data 3 specified from the input device 2, generates image data 5 through processing by the image quality generation processing section 4, and outputs this to the display device 6 to display the specified image. .

Object data 3 includes coordinate data specifying the geometric shape and arrangement of each basic object constituting the object, the color of each basic object,
Attributes (surface brightness, reflectance, transparency, refractive index, etc.)
, and, if necessary, mapping designation (picture designation), viewpoint position, location designation, etc.

In order to generate image data by the ray tracing method, the image generation processing unit 4 generates a ray of light that is emitted from a viewpoint 9 and passes through a point 1o on a screen 11, which is assumed as an image generation surface, as shown in FIG. 4, for example. 12, and for this ray 12, objects 13a, 13 input as object data 3
Calculate the intersection with surfaces such as b and 13c.

For example, if the light ray first intersects the object 13a, the intersection point 14 is determined by calculation, the normal vector 15 to the surface of the object at this point is determined, and the inclination of the normal to the light source 16 and this intersection point are calculated. The brightness of this point is calculated from the color and other attribute data, and a brightness value that contributes to the brightness of the point 1o on the screen is determined from the brightness.

Also, the vectors of the reflected ray 17 and, if necessary, the refracting ray 8 that transmits through the object 13a at the intersection 14 are calculated, and the intersection of the traveling light of these rays with the object is determined. That is, for example, for a reflected ray 17, the initial object 1
Find the intersection point 2o with 3b.

Once the intersection point 20 has been found, the normal vector, reflected ray vector, refracted ray vector, brightness, etc. are found, as in the case of the intersection 14, and the brightness value at which this brightness contributes to the point 10 on the screen is determined. , added to the previously determined brightness value.

In this way, the destinations of reflected and refracted rays are tracked sequentially,
When all the rays no longer intersect with objects in their destinations, the tracking of one ray 12 starting from the viewpoint is completed.

Through the above tracking process, point 1 on the screen 11
Data such as brightness values and colors are obtained as image data to be added to 0, and by performing such processing on each point taken at an appropriate density on the screen 11, image data representing a specified object can be obtained. Complete.

Figure 5 shows the flow of the above tracking process for each ray,
After setting the ray in processing step 20, processing step 2
In step 1, the point of intersection with the object is calculated, and in step 22, based on the calculation result, it is determined whether or not the ray intersects with the object.

If there is an intersection, the first intersection is selected in process step 23, the normal vector at that intersection is calculated, and the brightness value is determined in process step 24 as described above.

In process step 25, reflected and refracted ray vectors are calculated and stacked, for example. In processing step 26, one of the ray vectors is taken out from the stack and the processing of processing step 21 is performed on that ray, and in this way, all the ray vectors in the stack are processed and the tracing processing for one ray is completed. do.

[Problem that the invention seeks to solve]

Image generation using the ray tracing method, which is obtained as described above, is widely used because it can obtain high-quality images. However, as expected from the ray tracing process described above, this image generation requires a relatively long processing time.

In image creation work, it is often necessary to repeatedly change the color of the object, the color of the light source, the pattern of the object, etc. while looking at the image displayed using image data obtained through ray tracing processing.

Conventionally, in the case of such a correction, the above-mentioned ray tracing process is executed again to generate new image data, so there has been a problem that a long processing time is required for each correction, regardless of the extent of the correction.

[Means for solving problems]

FIG. 1 is a block diagram showing the configuration of the present invention.

In the figure, 31 is a processing device, 3 is object data, 34 is an image generation processing unit, 5 is image data, 37 is a ray tracing tree,
38 is an image correction processing section.

[For production]

The image generation processing unit 34 receives the object data 3 as input, performs ray tracing processing in the conventional manner, outputs image data 5, and generates spatial information (intersection point coordinates and normal vector information) obtained at each intersection of the rays. ) is data linked in a tree shape with the viewpoint as the root for each ray and each intersection as a node, and is held as a ray tracing tree 37.

When the image correction processing unit 38 is started after changing the specification of the object color etc. of the object data 3, the image correction processing unit 38 sequentially applies the data to each node along the tree-like structure for each ray of the ray tracing tree 37. The retained spatial information is extracted and the image data 5 is generated using this data and the object data 3.

As described above, in the case of corrections such as color specification, the intersection point calculation and normal line calculation in the ray tracing process can be omitted, so that image data of the same quality as that obtained by the ray tracing process can be obtained at high speed.

〔Example〕

FIG. 2 shows an example of the processing flow of the image generation processing section 34 and the image correction processing section 38 shown in FIG. 1.

FIG. 2 (al indicates the processing of the image generation processing unit 34; when one ray for ray tracing processing is set in processing step 20, an intersection point is calculated in processing step 21, and processing step 2
In step 2, the presence or absence of an intersection is determined based on the calculation result.

If there is an intersection, the first intersection point is selected in process step 23, the normal at that intersection is calculated, and the process step 24
The brightness value is determined as described above.

In the next processing step 40, the spatial information consisting of intersection coordinates and normal vectors obtained up to the preprocessing step is
8 are linked as data of the relevant node to a ray tracing tree 37 having a tree-like data structure in which the viewpoint is the root and each intersection point corresponds to a node.

That is, the ray tracing tree 37 conceptually has the configuration shown in FIG.
To describe the data corresponding to FIG. 4, starting from the route 50 corresponding to the viewpoint, information on the intersection point coordinates and the normal vector of the intersection point 14 is held in the node 51 corresponding to the first intersection point 14. .

Next, each node 52, 53 corresponding to the intersection 20 of the reflected ray 17 and the intersection 21 of the refracted ray 18 at the intersection 14 is set as a node linked to the node 51, and the processing result of the processing step 23 regarding the equal intersection is as follows. The data is held at the node in processing step 40, and in this way, tree-structured data is constructed that branches and grows in accordance with the reflection and refraction of the light rays.

In addition to spatial information, each node also includes processing step 24.
If a pointer to the color specification used in the calculation of the brightness of the intersection point is included, it will be convenient for use in the image modification process described later.

After saving the data, the reflected and refracted ray vectors are calculated in a processing step 24, e.g. stacked, and in a processing step 25 one of the ray vectors in the stack is retrieved and the processing from processing step 21 is performed on that ray. In this way, all the ray vectors in the stack are processed and the tracing process for one ray is completed. At the same time, one piece of tree-like data corresponding to the ray is generated in the ray tracing tree 37.

If you want to evaluate the display of the image generated as described above and modify a part of it, first modify the required part of the object data 3 that is input and held, and then generate image data from this. .

If this correction is to the geometric configuration of the object, etc., the same ray tracing process as described above by the image generation processing unit 34 must be re-executed on new data.

However, if there is no change in the geometrical configuration specifications due to corrections such as color specification or designation, the image correction processing section 38 can be activated to generate image data by the processing described below.

The image correction processing unit 38 performs the processing step 41 of FIG.
When one ray for ray tracing processing is set in step 42, data corresponding to the ray in the ray tracing tree 37 is selected in processing step 42.

Hereinafter, the tree-like data will be traced from the root and processing will proceed sequentially for each node, processing step 43.
First, the spatial information held in the node corresponding to the first intersection point is extracted, and in processing step 44, the brightness value is determined using the coordinates of the intersection point and the normal vector in the same manner as described above.

In processing step 45, it is determined whether all nodes have been scanned, and if there is an unprocessed node, the process returns to processing step 43 and the brightness value is calculated using the spatial information held in the next node, and in this way all nodes are scanned. Finish scanning.

As described above, one point of image data is obtained which is the same as when the ray tracing processing for this l ray is executed as described above, but in this case, processing steps 21, 24.
25, the calculation of intersection points, normal vectors, reflected light vectors, and refracted light vectors is not required.

Therefore, according to the above correction process, 60
Calculation processing, which accounts for ~70%, is omitted, and image data can be generated quickly by corrected input.

[Effects of the Invention] As is clear from the above description, according to the present invention, when some data specifying an object is modified in image generation processing using the ray tracing method, image data can be modified at high speed. It has significant industrial effects.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of the present invention, FIG. 2 is a flowchart of processing in an embodiment of the present invention, FIG. 3 is a block diagram of a conventional configuration example, FIG. 4 is an explanatory diagram of ray tracing processing, FIG. 5 is a flow chart of conventional processing, and FIG. 6 is an explanatory diagram of a ray tracing tree. In the figure, 1.31 is a processing device, 2 is an input device, 3 is object data, 4.34 is an image generation processing unit, 5 is image data, 6 is a display device, 9 is a viewpoint, 12 is a light beam, 13a, 13b , 13c is an object, 14.20.21
is the intersection point, 20-26.40-45 is the processing step, 37
38 is a ray tracing tree, 38 is an image correction processing unit, 50 is a root, and 51 to 53 are nodes showing the configuration of the present invention. Example block diagram Figure 3 Illustration of ray tracing processing Figure 4 Flowchart of conventional processing Figure 5 Illustration of ray tracing tree

Claims (3)

[Claims] (1) In a device (31) that generates image data using a ray tracing method based on object data, for each ray that is set as emitted from a viewpoint in order to generate ray tracing data, the ray and the Reflection and refraction of light rays Spatial information obtained for a predetermined intersection between a ray and the object is configured as tree-like data in which the viewpoint is used as a root, the intersection is used as a node, and the nodes are linked in the order in which the ray travels. means (34, 37) for maintaining a ray tracing tree; and
An image generation process characterized by comprising means (38) for scanning the ray tracing tree (37) from the root and sequentially using spatial information held for each node to generate image data. method. (2) The image generation processing method according to claim 1, wherein the spatial information is information indicating the intersection position coordinates and the normal vector to the object regarding the intersection. (3) The image generation processing method according to claim 1 or 2, wherein the ray tracing tree holds, for each node, a pointer to color designation and attribute designation data for the intersection of the objects. .