JP2856465B2 - Image generation processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an image generation processing method, and in particular, to a three-dimensional shape, color, reflectance, light source position, and light intensity of an object set as numerical data. The present invention relates to an image generation processing method for generating a two-dimensional image based on colors, viewpoint positions, and the like (these are collectively referred to as scene data).

[Conventional technology]

Conventionally, in this type of method, the luminance of all pixels in an image is set to 1
Since each pixel was calculated using a complicated processing method, it took a long time to generate an image. One way to solve this problem is to detect a set of pixels that do not require brightness calculation by a complex processing method without accurately calculating the brightness of all pixels in the image one pixel at a time, and The luminance can be calculated in a short time by calculating the luminance in a simple process by making effective use of the luminance of the pixel whose luminance is already calculated in the vicinity of the pixel or the information obtained in the process of calculating the luminance of the pixel. There is a generation method (Akimoto, Mase: “Image Generation Device”, Japanese Patent Application No. 60-42020, March 4, 1985, and Akimoto, Mase: “Pixel-selective ray tracing method” Transactions of the Institute of Electronics, Communication and Communication Engineers) , J69-D, No. 12, December 1986).

[Problems to be solved by the invention]

However, in this method, a small object, an elongated object, or a sharp projection of the object in the image may disappear or be interrupted from the generated image. This is because these small objects are buried between the coarsely sampled pixels, and the brightness of the pixel where the object is to be displayed is determined by interpolating the brightness of the pixels outside the object. Because.

 The present invention aims to solve this point.

[Means for solving the problem]

In the present invention, a high-quality image is obtained by detecting some of the pixels to which erroneous luminance may be given and recalculating the luminance of those pixels by a method capable of calculating more accurate luminance. Is generated at high speed.

[Action]

When the luminance is corrected by the recalculation, the correction result is reflected on neighboring pixels based on the corrected result.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of the present invention, and FIG. 2 shows a part of an image being generated by the image generation processing method of the present invention.

In FIG. 1, reference numeral 101 denotes an image memory for recording the luminance of each pixel of an image being generated, a method for calculating the luminance, and information obtained in the luminance calculation process, and reference numeral 102 denotes a pixel based on scene data. A luminance calculation unit for calculating an accurate luminance, 103 is a luminance estimation unit for estimating the luminance of a certain pixel from information of pixels in the vicinity of the pixel by simple processing, and 104 is a flat region detection for detecting a region where the luminance change is flat. , 105 is an image generation control unit that controls the entire image generation process, 106 is an erroneous luminance detection unit that detects pixels that may have erroneous luminance, and 107 is a unit that has erroneous luminance. A luminance recalculation control unit 108 for performing control for accurately recalculating the luminance of a certain pixel is a display for displaying the generated image.

The method of calculating the brightness in the brightness calculation unit 102 is obvious in the field of computer graphics, and a description thereof will be omitted. For example, the luminance estimating unit 103 obtains the average of the luminance of several pixels near a certain pixel, and obtains the luminance of the pixel at high speed. For example, the flat area detection unit 104 compares the brightness of the pixels at the four corners of the small square area, and determines that the area is not a flat area if the brightness is similar, and is not a flat area if the brightness is significantly different. , Flat areas are detected. The erroneous luminance detection unit 106 examines information of two adjacent pixels and a method of calculating the luminance of the pixels, and the two pixels are not similar and the luminance of both or one of the pixels is estimated. If so, the pixel whose luminance is estimated is detected as a pixel that may have incorrect luminance. Here, “not similar” means that, for example, the object displayed by the two pixels is different. Alternatively, the luminance of the two pixels is significantly different. The luminance recalculation control unit 107 recalculates the correct luminance of a pixel that may have an erroneous luminance, and redetermines the flatness of an area that may not be a flat area due to the recalculation of luminance. Then, if necessary, the luminance calculation unit 102, the flat area detection unit 104, and the image memory 101 are controlled so as to recalculate the luminance of other pixels.

Next, the operation of FIG. 1 will be described. Prior to the image generation operation, scene data to be displayed is provided to the luminance calculator 102. For simplicity of description, the image generation operation will be specifically described in order using an example of the image shown in FIG. FIG. 2 is an explanatory diagram showing a state of image generation in the case of the present invention.

In the figure, circles indicate pixels calculated by luminance calculation, 4
The corners represent the pixels calculated by the luminance estimation, the squares represent the pixels for which the luminance has been recalculated, and the oblique lines represent the color of the “object 1”.

Process 1) Under the control of the image generation control unit 105, the luminance of the pixels spaced apart by D (pixels 1 to 6 in FIG. .
It is assumed that “object 1” as shown by diagonal lines in the drawing exists, and the brightness of “object 1” and “object 2” are greatly different.

Process 2) The flat area detection unit 104 outputs the pixels at the four corners of the area forming the square (for example, the pixels 1 to 4 and the pixels 3 to 6 shown in FIG. 2B or the pixels shown in FIG. 2C). 3,7,4,8) similarity is determined.

Process 3) According to the control of the image generation control unit 105, if the region is a flat region, the luminance estimating unit 103 estimates the luminance of the center pixel of the region (the pixel 7 shown in FIG. 2B), and the above-mentioned process 2) Proceed to. If the area is not a flat area, the luminance calculation unit 102 calculates the center pixel of the area (pixel 8 shown in FIG.
The luminance of the pixel 12 shown in FIG.
Proceed to.

Processing 4) For an uneven area, the erroneous luminance detection unit 106
Compares the pixel at the center of the region with the pixels at the four corners, and detects a pixel whose luminance has been calculated by luminance estimation and which is not similar to the central pixel (pixel 7 shown in FIG. 2 (c)). I do. Then, under the control of the brightness recalculation control unit 107, the brightness calculation unit 102 recalculates the correct brightness of the pixel.

Process 5) When the luminance of a pixel and the information held by the pixel change due to the luminance recalculation, the flatness of a region including the pixel in the four corners may also change. Therefore, the brightness recalculation control unit 107
Extracts a region where the flatness may change with a change in the brightness of the pixel or the information held by the pixel (region centering on pixels 9, 10, and 11 shown in FIG. 2C), The flatness is re-determined by the flat area detection unit 104, and the luminance calculation unit 102 is controlled to recalculate the luminance of the central pixel of the non-flat area (pixels 9, 10, and 11 shown in FIG. 2C). This process is repeated until there are no more pixels whose luminance is to be recalculated. During this time, the processing steps as shown in FIG.

Process 6) The above processes 2) to 5) are repeated until the luminance of all pixels is determined. FIG. 2E shows a finally generated image.

The reason why the pixel having the possibility of having the wrong luminance can be detected by the determination method in the above process 4) is as follows.
It is determined as follows. FIG. 3 is an explanatory diagram showing a state of image generation when the present invention is not used. Fig. 3 (a)
(B), (c), (d) and (e) are FIGS. 2 (a) and (b) respectively.
(C), (d), and (e) are shown. The luminance of the pixel 7 shown in FIG. FIG. 3E shows an image generation result when the present invention is not used. FIG. 3C shows an example in which the tip of a sharp projection is missing due to the presence of the pixel 13 corresponding to the pixel 7 in FIG. Is shown. As can be seen from FIG. 3 (e), in the vicinity where the protrusion is missing, the luminance of one of the two adjacent pixels is obtained by the luminance estimation even though the luminance of the two adjacent pixels is different. (The pixel shown in FIG. 3 (e))
13,14). By the way, there should be a boundary (edge) of an object in the vicinity where the luminance of a pixel changes greatly. However, since the brightness of the pixel 13 in FIG. 3 (e) has been obtained by estimation, it is not certain whether this pixel is inside or outside the boundary of the object. That is, it can be considered that the luminance of the pixel calculated by the luminance estimation may be incorrect.

〔The invention's effect〕

According to the present invention, the above-described erroneous luminance detection is performed in the middle of the image generation processing, and the luminance of a pixel that may be erroneous is corrected earlier to generate an image with less deterioration in image quality. Can be. FIGS. 2 (e) and 3 (e) show the effect of preventing image quality deterioration according to the present invention.
It can be confirmed by comparing with.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is an explanatory diagram illustrating a state of image generation in the case of the present invention, and FIG. 3 is a diagram illustrating a state of image generation when the present invention is not used. FIG. In the figure, 101 is an image memory, 102 is a brightness calculation unit, 103 is a brightness estimation unit, 104 is a flat area detection unit, 105 is an image generation control unit, 106 is an erroneous brightness detection unit, and 107 is a brightness recalculation control unit. .

────────────────────────────────────────────────── ─── Continuation of the front page (56) References Akimoto et al. “Improved Pixel Selected Ray-tracing by Using Ray-Object Intersection on Tree”, Proc. Of the 1989 IEICE Spring Conference, IEICE. , March 28, 1989, p. 7-377 to 7-378. Akimoto et al. “Pixel-selective ray tracing method”, IEICE Transactions on '86 / 12, IEICE, December 1986, Vol. J69-D, no. 12, p. 1943-1952 (58) Field surveyed (Int. Cl. ⁶ , DB name) G06T 15/50 JICST Science and Technology Literature Database

Claims (1)

(57) [Claims] 1. An image generation processing method for calculating the brightness of a pixel based on image creation data to generate an image, comprising the steps of: (1) roughly extracting pixels in an image to be generated at predetermined intervals and extracting the pixels; Step (2): Collecting a plurality of pixels for which the luminance has been calculated to form a set of pixels, and calculating the luminance of the pixels and / or pixels which are information obtained in the process of calculating the luminance. By comparing the information possessed by, the similarity of the pixels in the set is checked, and if they are similar, the area spatially surrounded by the pixels in the set is determined to be a flat area, and if they are not similar, Determines the region spatially surrounded by the pixels in the set as a non-flat region, and Step (3) calculates the brightness of the central pixel of the flat region by a brightness estimation process if the flat region is determined. To step (2) Ri, when it is determined that the non-flat areas,
The luminance of the center pixel of the non-flat area is accurately calculated by the luminance calculation process, and the process proceeds to step (4). Step (4) The center pixel of the non-flat area and each of the pixels whose luminance has already been calculated in the non-flat area Examine the similarity with the pixel, and among the pixels whose luminance has already been calculated in the non-flat area, the luminance of which is calculated by the luminance estimation process, and there is a pixel that is not similar to the center pixel. In this case, the correct luminance of the pixel is calculated by a luminance calculation process (step (5)).
To step (2) if there is no such pixel
Returning to step (5), for a pixel whose exact luminance has been recalculated by the luminance calculation processing, the similarity between the information of the pixel before the luminance recalculation and the information after the recalculation is checked. Returns to step (2), and if not similar, re-determines whether the area including the pixel is a flat area or a non-flat area using the information of the pixel after the recalculation of the luminance, If it is determined that the area is an area, the process returns to step (2). If it is determined that the area is a non-flat area, steps (4) and (5) are repeated for the non-flat area. ) To step (5)
An image generation processing method which repeats until luminance of all pixels in an image to be generated is calculated.