JPH05120439A - Lattice point calculating method and texture mapping method - Google Patents

Abstract

PURPOSE:To reduce the work amount of the extraction of a coordinate value by calculating the coordinate value on a projected figure on which a three- dimensional space necessary for the setting of the parameter of a projection conversion is perspective-projected, from the coordinate value of a specific point. CONSTITUTION:An area surrounded by lattices are respectively defined as a mapping area 416, and the mapping of a texture picture indicated by a figure (b) is operated. That is, the parameter of the projection conversion of the plane of a figure (c) into the projected figure of a figure (a) is detected from the coordinate values of a point a11 (411), point a41 (412), point a14 (413), point a44 (414) of the figure (a), and point b11 (421), point b41 (422), point b14 (423), and point b44 (424) of the figure (c), and the coordinate value of a point bij (425), so that the coordinate value of a point aij (415) can be decided. The texture mapping matched with the structure of the perspective projection can be obtained by using the parameter of the above mentioned projection conversion.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for calculating position information of grid points found in images, photographs and the like, and a texture mapping method.

[0002]

2. Description of the Related Art To extract position information of an arbitrary point in an image, a point whose position moves according to input information is overwritten on the image, and a moving point is input when the moving point overlaps the point to be extracted. The position information of this point is extracted by the information. To extract the coordinate value of an arbitrary point in a digital image on a computer, usually, by displaying the target image on the display, superimposing the mouse cursor on it, and moving the mouse, Move the mouse cursor,
The coordinate value input to the mouse when it coincides with the point to be extracted is the coordinate value of that point. Hereinafter, the extraction of the coordinate value based on the above principle will be referred to as “direct extraction with a mouse or the like”. However, the input device of the computer includes a trackball, a light pen, a touch panel, a digitizer, and the like in addition to the mouse, and the same can be done with these, so this operation is not limited to the mouse.

Fumiko Ito et al. Texture simulation system considering simplicity and reality, IPSJ 39th (second half of 1989) national convention (hereinafter, known example 1
In the texture mapping that synthesizes a texture image with a perspective view that is a perspective projection of a three-dimensional space, in order to synthesize a texture image that matches the composition of the perspective projection in the projection diagram, The coordinate values of the corresponding quadrilateral vertices in the two images of the projected view are directly extracted with a mouse or the like, the parameters of the projective transformation between the two images are calculated from these coordinate values, and the texture image is obtained by this projective transformation. There is a description of the method of converting and synthesizing.

Further, in the above-mentioned known example 1, in the texture mapping for synthesizing the second image with the first image, the contour of the area to be texture-mapped on the first image is determined by direct extraction with a mouse or the like. There is a description of a method of limiting the range in which texture mapping is performed.

[0005]

The conventional method has the following problems.

The problem to be solved by directly extracting all the necessary points with a mouse or the like when extracting the position information of an arbitrary point in the image will be described below (Problem 1). For example, even if there is regularity between the position information of multiple points to be extracted, this method repeats the same procedure for all the required number of points. There is a problem that the amount becomes huge. Further, in that case, since the position information is extracted by an independent procedure, the regularity of the original position information is ignored.

Further, in a texture mapping for synthesizing a texture image with respect to a perspective view which is a perspective projection of a three-dimensional space, the coordinate values of the vertices of the corresponding quadrangle in the two images of the texture image and the perspective view are set by a mouse or the like. A method of extracting directly by using the coordinate values, calculating parameters of projective transformation between two images from these coordinate values, and converting and synthesizing a texture image by this projective transformation will be described below. .. For example, when the same texture image is successively mapped to a plurality of regions on a projection view on the same plane in a three-dimensional space, there is regularity between coordinate values of a plurality of points to be extracted. Even in such a case, this method repeats the same procedure for all the necessary points, and when the number is large, the amount of work becomes enormous. Further, in that case, since the coordinate values are extracted by an independent procedure, the regularity of the original coordinate values is ignored.

Further, in the texture mapping for synthesizing the second image with the first image, the contour of the area to be texture-mapped on the first image is determined by direct extraction with a mouse or the like, and the texture mapping is performed. The subject of the method of limiting the range to be performed is described below (subject 3). In this method, there is no limitation on the range of the second image, so there is a problem in that it is necessary to prepare an image in which any pixel can be texture-mapped as the second image. In addition, when limiting the range of the texture mapping of the first image, in a method of designating the contour of the area, when it is desired to change whether or not to perform the texture mapping for each pixel of the first image, There is a problem that designation of the area is very complicated and a huge amount of work is required.

A first object of the present invention is to obtain all the lattice points to be obtained when the position information on the projection diagram is obtained with respect to the perspective projection of the lattice points on a certain plane in the three-dimensional space. It is an object of the present invention to provide a method of extracting a smaller number of points with a mouse or the like instead of extracting with a mouse or the like, and calculating position information for the remaining points so as to match the composition of perspective projection.

A second object of the present invention is to perform texture mapping in which a texture image is projectively transformed and combined with respect to a projection view obtained by perspectively projecting a three-dimensional space. In the three-dimensional space before projection, the texture mapping is performed on the same plane. When mapping the same texture image to a plurality of areas on a certain projection map one after another, it is necessary to set the parameters for projective transformation for each area. When obtaining the above coordinate values, instead of extracting four points for every region on the projection diagram with a mouse or the like, a smaller number of points are extracted with a mouse or the like, and the remaining points are perspective-viewed. It is to provide a method of calculating coordinate values so as to match a projection composition. Further, although the coordinate values on the projection drawing that match the composition of the perspective projection are calculated by this texture mapping, it is to provide a texture mapping method that makes the composition in which the texture images are too artificially arranged in a more natural way. .. Furthermore, the calculated coordinate value is due to an error in direct extraction with a mouse or the like, distortion of the image, and the reason that the grid points are not exactly on the same plane in the original space. Although it is possible that the true position is not shown on the projection view, it is to correct this and provide a texture mapping method that can realize more accurate mapping. A third object of the present invention is to determine whether or not texture mapping is performed for each pixel of at least one of the first image and the second image in the texture mapping for synthesizing the second image with the first image. The object of the present invention is to provide a texture mapping method that can easily control the texture.

[0011]

In order to achieve the first object, a lattice structure on a plane in a three-dimensional space, positional information on the plane of four lattice points and a projected view of the four points are provided. The parameter of the projective transformation from the plane to the projection drawing is calculated from the position information in (3), and the position information of the points projected on the projection drawing of all the grid points on the plane is calculated.

Further, in order to achieve the second object,
The coordinate values on the projection diagram of the four points corresponding to each area, which are necessary for setting the parameters of the projective transformation for each area, and the coordinate values on the plane of the four points, It is calculated from the coordinate values of the four points on the plane and on the projected view. Also, after calculating the coordinate values required for setting the parameters of the projective transformation, the disturbance is given to the coordinate values and disturbed, and the resulting coordinate values are used for setting the parameters of the projective transformation. is there. Also, after calculating the coordinate values necessary for setting the parameters of projective transformation,
A true position is searched for around the coordinate value, and the coordinate value obtained as a result is used for setting a parameter for projective transformation.

Further, in order to achieve the above-mentioned third object, data holding information describing whether or not texture mapping is performed for each pixel of at least one of the first image and the second image is held. Then, by referring to it, it is determined whether or not to perform texture mapping.

[0014]

From the lattice structure on the plane in the three-dimensional space, the positional information on the planes of the four lattice points, and the positional information on the projected views of the four points, the parameters of the projective transformation from the plane to the projected plan The operation of the method of calculating and calculating the position information of the points projected on the projection view of all the grid points on the plane will be described below.

Consider a case in which a grid point is present on a plane in a three-dimensional space, there is a perspective projection projection of the grid point, and it is desired to obtain position information on the projection view of the grid point.

The number of grid points is n, and their positions on the plane and on the projected view are (xi yi) and (x'i), respectively.
y'i) (here (i = 1 ... n)).

Here, (xi yi) (i = 1 ... n)
Is known.

Further, it is assumed that (x'i y'i) is known for each of the four points.

Here, known points are i = 1, 2, 3,
4 is set, but this setting does not impair generality.

The known lattice structure means that (x
i yi) (i = 1 ... n) is known.

Since (xi yi) (i = 1 ... n) is on the same plane, and (x'i y'i) (i = 1 ... n) is also on the same plane, the formula 1 is satisfied for all i. There exists a constant matrix A (Equation 2) that holds. Here, the homogeneous coordinate system is used.

[0022]

[Equation 1]

[0023]

[Equation 2]

Equation 3 is derived from Equations 1 and 2.

[0025]

[Equation 3]

Expression 4 is obtained by expressing Expression 3 in normal coordinates.

[0027]

[Equation 4]

From Equation 4, Equation 5 and Equation 6 are derived.

[0029]

[Equation 5]

[0030]

[Equation 6]

Here, when i = 1, 2, 3, and 4 are summarized, the formulas 7 and 8 are obtained.

[0032]

[Equation 7]

[0033]

[Equation 8]

Here, if the matrix T1 is a regular matrix, the inverse matrix T2 is obtained (Equation 9).

[0035]

[Equation 9]

For i = 1, 2, 3, 4, (xi
Since yi) and (x'i y'i) are known, each element of the matrix A is determined by the equation (10).

[0037]

[Equation 10]

According to the equations (1) and (2), the position information on the projection view of all the lattice points can be obtained. According to the method described above, the position information of the projection points on the projection view of all the grid points on the plane can be calculated from the position information of the smaller number of grid points, so that the workload is reduced. In addition, position information that matches the composition of perspective projection is obtained.

Further, the coordinate values on the projection diagram of the four points corresponding to each area, which are necessary for setting the parameters of the projective transformation for each area, are set to the coordinate values on the plane of the four points and the area. The operation of the method of calculating from a set of coordinate values on a plane and a projected view of four points that do not depend will be described below.

Similarly to the explanation of the operation of the above-mentioned grid point calculation method, (xi yi) is a point on the plane, and (x'i y'i)
Is a point on the projection. Further, here, four points which have known coordinate values and do not depend on the area are set as i = 1, 2, 3, 4, and points where (x′i y′i) is unknown are set to i = 5 or more. Allocate so that

Here again, there is a matrix A (Equation 2) that establishes Equation 1. For equations 5 and 6, i = 1, 2,
Summarizing about 3 and 4, the following equations 7 and 8 are obtained.
Here, when the inverse matrix T2 of the matrix T1 is obtained, each element of the matrix A is determined as in Expression 10.

According to Equations 1 and 2, given (xi yi), the corresponding (x'i y'i) is calculated.

According to the above-mentioned procedure, if four points are associated in the plane and the projection drawing, when the mapping destination is a plurality of areas on the same plane, the projective transformation parameter is set for each area. Therefore, when the coordinate value of a certain point is needed, if the point is assigned as a point where the above i is 5 or more, the coordinate value on the projection diagram can be easily obtained.

Further, after calculating the coordinate values necessary for setting the parameters of the projective transformation, the coordinate values are disturbed by disturbance, and the resulting coordinate values are set for the parameter setting of the projective transformation. The operation of the method used will be described below.

The influence of disturbance is exerted on the coordinate values (x'i y'i) of the points on the projection drawing calculated as described above.

The function that returns a random number is R (). x '
The values obtained by applying R () to i and y′i are respectively defined as x ″ i and y ″ i, which are used for setting the parameters of the projective transformation.

According to the above procedure, the tendency depending on the method of calculating coordinate values can be alleviated.

Further, after calculating the coordinate values necessary for setting the parameters of the projective transformation, the true position is searched for around the coordinate values, and the coordinate values obtained as a result are set for the parameter setting of the projective transformation. The operation of the method used for this will be described below.

A true position is searched for in the vicinity of the point (x'i y'i) on the projection drawing calculated as described above. When the true position is a feature point that can be extracted by image processing on the projection diagram, create a suitable spatial filter to extract it and try to operate it while changing the position near this point, and perform template matching. The coordinate value of the position having a high correlation value is set as the true value by the operation.

According to the above procedure, more accurate texture mapping can be realized.

Further, the data in which the information as to whether or not the texture mapping is performed is described for each pixel of at least one of the first image and the second image is held, and the texture mapping is performed by referring to the data. The operation of the method for determining whether or not to perform is described below.

The image has a pixel value that indicates for each pixel what color or density it represents.

In addition to this data, whether the pixel is a pixel of a texture image and may be mapped to another image, and whether the pixel is mapped from another image and the pixel value is rewritten. Whether or not a pixel is a good pixel will have data described for each pixel. By referring to this data, it is determined whether or not texture mapping is performed.

Although it has been described above that this data is held for each pixel, it is also possible that this data is held in a compressed form such as run length data.
It takes a description of the function determined by the position of the pixel,
It is possible to extend it to have it as data. It is also possible to make a description such that the description of the above function also functions as a pixel value.

By the above-mentioned procedure, the data in which the information as to whether or not the texture mapping is performed is stored for each pixel of at least one of the first image and the second image is held, and with reference to it, By determining whether or not to perform texture mapping for each pixel, whether or not to perform texture mapping is controlled for each pixel of either one of the first image and the second image. It is also possible to control whether or not to perform texture mapping for each pixel of both the first image and the second image.

[0056]

[Embodiment] Regarding one embodiment of the grid point calculation method of the present invention,
This will be described in detail with reference to FIGS. 1 to 3.

Lattice points (point b11121, point b21125, point b31 1 on the plane 130 as shown in FIG. 1B).
22, point b12 123, point b22 126, point b32
124) is placed in a three-dimensional space and is perspectively projected to obtain the image shown in FIG.
It looks like (a) (point a11 111, point a2
1 115, point a31 112, point a12 113, point a2
2 116, point a32 114). In projection, the point bij
Corresponds to the point aij (where i = 1, 2, 3,
j = 1, 2).

At this time, it is assumed that all the coordinate values of the point bij and the coordinate values of the points a11, a31, a12, and a32 are obtained.

Here, the coordinate value of each point is represented by a point a11 (x'1 y'1) and a point a21 (x'5 y ').
5), point a31 (x'2y'2), point a12 (x'3
y'3), point a22 (x'6 y'6), point a32
(X'4 y'4) Point b11 (x1 y1), point b21 (x5 y5), point b31 (x2 y2), point b12 (x3 y3), point b
22 (x6 y6) and point b32 (x4 y4), the coordinate values of the points a21 and a22 can be calculated by the procedure described in the action.

In the above-mentioned embodiment, the description has been made assuming that the lattice is composed of 6 points as shown in FIG. 1, but this does not limit the present embodiment to the lattice structure as shown in FIG. Any grid may be used. For example, even when there are grids that are separated as shown in FIG. 2, if the grids are on the same plane, the procedure similar to the above can be performed, so that such expansion is hindered. is not.

Further, in the above-mentioned embodiment, the known aij
The points of 4 were set as shown in Fig. 1, but of course the other 4
You may choose a point. However, it is necessary to select so that the inverse matrix T2 described in the operation exists.

Further, in the above embodiment, the point bi on the plane side
The coordinate space for defining the coordinate value of j may be any one as long as it constitutes a linear space. For example, as shown in FIG. 3, whether the coordinate axis (1) 301 is taken or the coordinate axis (2) 302 is taken, it does not affect the coordinate values on the projection drawing side calculated using the coordinate axis (1) 301. According to this, for example, in the case of FIG. 1B, instead of the information of the coordinate values of all the points bij, “two congruent squares are arranged side by side with one side in common, and the upper left corner of the left square Point is point b11
, The lower left point is the point b12, the upper right point of the right square is the point b31, and the lower right point is the point b32. " That is, it is sufficient if the description can be determined as a similar figure and the points correspond to each other. As described above, the coordinate values of all grid points on the projection view that match the composition of the perspective projection are calculated.

An embodiment of the texture mapping method of the present invention will be described in detail with reference to FIGS. 4 and 5.

Lattice points (point b11 421, point b41 422, point b14 42) on the plane as shown in FIG.
A point bij 425 including the third point b44 424) is placed in the three-dimensional space, and is perspectively projected to look like FIG. 4A. A point bij 425 on the plane on the projection view
Let points aij 415 respectively correspond to (here, i = 1, 2, 3, 4 j = 1, 2, 3, 4).

Here, let us consider mapping the texture image as shown in FIG. 4B by setting the regions surrounded by the lattice as the mapping regions 416, respectively.

Here, all the coordinate values of the point bij and the point a
It is assumed that the coordinate values of 11 411, point a41412, point a14 413, and point a44 414 have been obtained.

The coordinate values of the respective points are point a11 (x'1 y'1) and point a41 (x'2 y ').
2), point a14 (x'3y'3), point a44 (x'4)
y′4) If a point b11 (x1 y1), a point b41 (x2 y2), a point b14 (x3 y3), and a point b44 (x4 y4) are used, the coordinate value of bij is converted from the coordinate value of bij by the procedure described in the action. The value can be calculated.

In the above-mentioned embodiment, the lattice is 16 as shown in FIG.
Although the description has been given as a lattice of dots, this does not limit the present embodiment to the lattice structure shown in FIG.
Any set of mapping regions may be used. For example, even when there are mapping regions that are separated as shown in FIG. 5, the same procedure as above can be performed if the mapping regions are on the same plane. It does not prevent the expansion.

Also, in the above embodiment, the known aij
Although the four points are set at the four end points as shown in FIG. 4, other four points may of course be selected. However, it is necessary to select so that the inverse matrix T2 described in the operation exists.

Furthermore, in the above embodiment, the point bi on the plane side
The coordinate space for defining the coordinate value of j may be any one as long as it constitutes a linear space. According to this
For example, in the case of FIG. 4C, instead of the information of the coordinate values of all the points bij, “a total of 16 square lattices are arranged in the vertical direction and 4 in the horizontal direction, and the point b11 corresponding to the point a11 is displayed. Is the upper left point, the point b14 corresponding to the point a14 is the lower left point, the point b41 corresponding to the point a41 is the upper right point, and the point a4
The point b44 corresponding to 4 is the lower right point. " That is, it is sufficient if the description can be determined as similar figures and the points correspond to each other.

From the coordinate values of the points aij obtained as described above, the coordinate values of the four points necessary for setting the parameters of the projective transformation can be obtained for each mapping area. Can be set.

By using the projective transformation parameters, texture mapping that matches the composition of perspective projection can be realized.

Further, in the present embodiment, there is a change in that the calculated coordinate values are not used as they are for setting the parameters of the projective transformation, but the values obtained after performing the following processing are used. It is possible.

One is a process of giving a random number to disturb the value.

Let Rmax be the upper limit of the disturbance with respect to the coordinate value, and R () be a function that generates a random number in the range from -1 to 1 each time it is called. The calculated coordinate value is (x ′
iy′i), (x′i + RmaxR () y ′
i + RmaxR () is used to set the parameters of the projective transformation.

Further, the other is a process of searching for a true position around the calculated coordinate value to obtain an accurate value.

In order to detect the intersecting line segments, a matrix as shown in FIG. 6 is used, and the product of the matrix component and the corresponding pixel value is calculated around the calculated coordinate values to obtain the correlation. The true position is obtained by detecting the high point, and the coordinate value of this position is used for the parameter setting of the projective transformation.

The matrix used in the above embodiment is an example of a matrix for detecting intersecting line segments,
Other materials may be used as long as they can achieve the same effect. Another embodiment of the texture mapping method of the present invention will be described in detail below with reference to FIGS. 7 to 9.

An example of a texture mapping method capable of controlling whether or not to map the image 2 on the image 1 for each pixel of the image 2 will be described with reference to FIG. Consider that the image 2 (FIG. 7B) is texture-mapped to the image 1 (FIG. 7A) at the same size. Regarding the image 2, in addition to the pixel value data (FIG. 7B), the pixel 701 indicating a portion that may be mapped is 1 and the pixel 702 indicating a portion that should not be mapped is 0 ( FIG. 7D is created in advance, and at the time of texture mapping, only the pixels whose data is 1 are displayed in the image 1 (FIG. 7D).
When the texture mapping is performed in (a)), as shown in FIG. 7C, a mapping result in which it is controlled whether or not each pixel of the image 2 is mapped is obtained. At this time, how to write data such as 0 or 1 is not limited to the above. Further, the texture mapping method is not limited at all, and it may be a composition of equal size or a composition of performing projective transformation.

An embodiment of a texture mapping method capable of controlling whether or not to perform mapping for each pixel of the image 1 when the image 2 is texture-mapped on the image 1 will be described with reference to FIG. Consider that the image 2 (FIG. 8B) is texture-mapped to the image 1 (FIG. 8A) at the same size. Regarding the image 1, in addition to the pixel value data (FIG. 8A), the pixel 801 representing the part that may be mapped is 1 and the pixel 802 representing the part that should not be mapped is 0 ( FIG. 8D) is created in advance, and when texture mapping is performed, if texture mapping is performed from the image 2 only on the pixels for which this data is 1, one of the images of the image 1 is obtained as shown in FIG. 8C. A mapping result that controls whether or not to perform mapping for each pixel is obtained. At this time, how to write data such as 0 or 1 is not limited to the above. Further, the texture mapping method is not limited at all, and it may be a composition of equal size or a composition of performing projective transformation.

FIG. 9 shows an embodiment of a texture mapping method capable of controlling whether or not to perform the image mapping of the image 2 on the image 1 for each pixel of both the images 1 and 2. It will be explained using. Consider that the image 2 (FIG. 9B) is texture-mapped to the image 1 (FIG. 9A) at the same size. Regarding the image 1, in addition to the pixel value data (FIG. 9A), the pixel 901 indicating a portion that may be mapped is 1 and the pixel 902 indicating a portion that should not be mapped is 0 ( 9 (d)) is created in advance, and in addition to the pixel value data (FIG. 9 (b)) in the image 2, the pixel 903 representing a portion that may be mapped is 1, and mapping is not performed. Data (FIG. 9 (e)) in which the pixel 904 representing a prohibited portion is 0 is created in advance, and these data are set to 1 when texture mapping is performed.
If texture mapping is performed from image 2 to image 1 only when
It is possible to obtain a mapping result that controls whether or not mapping is performed for each pixel in both images of the image 2 and the image 2. At this time,
How to write data such as 0 or 1 is not limited to the above.
Further, the texture mapping method is not limited at all, and it may be a composition of equal size or a composition of performing projective transformation.

Further, although it has been described above that this data is held for each pixel, it is also possible to hold this data in a compressed form such as run length data. This makes it possible to have equivalent information with a smaller capacity.

It is also possible to extend it by describing a function determined by the position of a pixel and having it as data. For example, if the horizontal coordinate of the image is i,
The vertical coordinate is j, the upper left corner of the image is the origin,
It is assumed that i and j take positive directions in the rightward and downward directions, respectively. At this time, for example, j <i + 100 (i, j)
It is also possible to determine whether or not to perform texture mapping on the condition that the pixel of is subjected to texture mapping. This does not need to have data for each pixel, but it can be implemented as a modification of the above-described embodiment by having information that can determine whether or not to perform texture mapping for each pixel as data. As a result, information can be held with a smaller capacity.

When the pixel values r, g, b take values from 0 to 255, for example, r <50 and g <5.
It is also possible to determine whether or not to perform texture mapping on the condition that pixels having 0 and b> 200 are subjected to texture mapping. This does not need to have data for each pixel, but it can be implemented as a modification of the above-described embodiment by having information that can determine whether or not to perform texture mapping for each pixel as data. As a result, it is possible to determine whether or not to perform texture mapping according to the area where the pixels exist in the color space.

In addition, when the texture image is projectively transformed and combined by using a plurality of regions on the projection diagram perspectively projecting the three-dimensional space as described above, work is performed by calculating coordinates on the projection diagram. In the embodiment of the texture mapping method in which the amount can be reduced and the texture mapping that matches the composition of perspective projection can be performed, whether or not the texture mapping described above is performed for each pixel of at least one of the projected view and the texture image It is also possible to carry out texture mapping in which data in which the information is described is held, and by referring to it, whether or not to perform texture mapping is determined for each pixel.

In addition, in the execution of the texture mapping, the above-mentioned processing of adding a disturbance to the calculated coordinate value is also added, and a true value is obtained in the vicinity of the calculated coordinate value. It is also possible to add the processing of searching and obtaining.

[0087]

The present invention has the following effects.

(1) According to the grid point calculating method of the present invention, when the position information on the projection diagram is obtained for the perspective projection point of the grid point on a certain plane in the three-dimensional space,
Instead of extracting with a mouse etc. for all the grid points you want to obtain, extract a smaller number of points with a mouse etc., and for the remaining points, calculate the position information so that it matches the composition of the perspective projection. , The workload is reduced. In addition, position information that matches the composition of perspective projection is obtained.

(2) According to the texture mapping method of the present invention, in texture mapping in which a texture image is projectively transformed and combined with a perspective view of a three-dimensional space, a three-dimensional space before projection is used. When mapping the same texture image to a plurality of regions on the projection drawing on the same plane one after another, it is necessary to set the parameters of the projective transformation for each region. When calculating the coordinate values of the points on the projection diagram, instead of extracting four points for each region on the projection diagram with a mouse or the like, a smaller number of points are extracted with a mouse or the like and the remaining points are extracted. With respect to, the workload is reduced by calculating the coordinate values so as to match the composition of the perspective projection. In addition, coordinate values that match the composition of perspective projection are obtained.

Further, after the coordinate values required for setting the parameters of the projective transformation are calculated, the coordinate values are disturbed and disturbed, and the resulting coordinate values are set for the parameter setting of the projective transformation. By using it, the tendency depending on the method of calculating the coordinate values can be eased, and the composition in which the texture images are arranged artificially and neatly can be made more natural.

Furthermore, the calculated coordinate values may include errors in direct extraction with a mouse or the like, image distortion, and the fact that lattice points are not exactly on the same plane in the original space. For some reason, it is possible that the true position is not shown on the projection view, but the true position is searched for around the calculated coordinate values above, and the resulting coordinate values are used to set the parameters for projective transformation. More accurate mapping can be realized by using

(3) Further, according to the texture mapping method of the present invention, in texture mapping for synthesizing the second image with the first image, the pixels of at least one of the first image and the second image Data that describes whether or not to perform texture mapping is stored for each pixel, and whether or not to perform texture mapping is determined by referring to it and determining whether or not to perform texture mapping for each pixel. It is possible to easily control for each pixel of either one of the first image and the second image. Also,
The first image and the second
The control can be easily performed for each pixel of both images.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of an embodiment of a grid point calculation method of the present invention.

FIG. 2 is a diagram for explaining that the lattice structure is not limited in the embodiment of the lattice point calculation method of the present invention.

FIG. 3 is a diagram illustrating that, in the embodiment of the grid point calculation method of the present invention, there are no restrictions on how to take coordinates on a plane other than the condition that the coordinate space is a linear space.

FIG. 4 is an explanatory diagram of an embodiment of a texture mapping method of the present invention.

FIG. 5 is a diagram illustrating that there is no limitation on the way of existence of a mapping area in the embodiment of the texture mapping method of the present invention.

FIG. 6 is a diagram showing an example of a matrix for searching for a true position around a calculated coordinate value in the embodiment of the texture mapping method of the present invention.

FIG. 7 is a method for controlling whether or not to perform texture mapping for each pixel of the second image when texture mapping the second image to the first image in the embodiment of the texture mapping method of the present invention. It is a figure explaining.

FIG. 8 is a method for controlling whether or not texture mapping is performed for each pixel of the first image when texture mapping the second image to the first image in the embodiment of the texture mapping method of the present invention. It is a figure explaining.

FIG. 9 is a diagram illustrating a texture mapping method according to an embodiment of the present invention, wherein when texture mapping a second image to a first image, texture mapping is performed for each pixel of both images of the first image and the second image. It is a figure explaining the method which can control whether to perform.

[Explanation of symbols]

111 ... point a11, 112 ... point a31, 113 ... point a1
2, 114 ... Point a32, 115 ... Point a21, 116 ... Point a22, 121 ... Point b11, 122 ... Point b31, 123
... point b12,124 ... point b32,125 ... point b21,1
26 ... Point b22, 130 ... Plane, 301 ... Coordinate axis (1), 302 ... Coordinate axis (2), 411 ... Point a11, 4
12 ... Point a41, 413 ... Point a14, 414 ... Point a4
4, 415 ... Point aij, 416 ... Mapping area, 42
1 ... Point b11, 422 ... Point b41, 423 ... Point b14, 4
24 ... Point b44, 425 ... Point bij, 701 ... Pixel representing a part that may be mapped, 702 ... Pixel representing a part that should not be mapped, 801 ... Pixel representing a part that may be mapped, 802 ... Pixel representing a prohibited part, 901 ... Pixel representing a part that may be mapped, 902 ... Pixel representing a part that should not be mapped, 903 ... Pixel representing part that may be mapped, 90
4 ... A pixel that represents a portion that should not be mapped.

Claims (8)

[Claims] 1. In a projection view in which a grid point on a plane in a three-dimensional space is perspectively projected, a grid structure on the plane, positional information on the grid point of four points, and position information on the four points. A grid point calculation method for calculating the position information of all grid points on the projection drawing from the position information on the projection drawing. 2. In texture mapping in which a texture image is projectively transformed and combined with respect to a perspective view which is a perspective projection of a three-dimensional space, on a projection view which is on the same plane in the three-dimensional space before projection. When mapping the same texture image to a plurality of areas one after another, the four coordinate values on the projection diagram corresponding to each area, which are necessary for setting the parameters of the projective transformation for each area, A texture mapping method that is calculated from coordinate values on a plane of four points and a set of coordinate values on a plane and a projected view of a set of four points that do not depend on a region. 3. The texture mapping method according to claim 2, wherein the coordinate values obtained by applying disturbance to the calculated coordinate values on the projection diagram of the four points are used to set the parameters of the projective transformation. 4. A search is performed around the calculated coordinate values of the four points on the projection diagram to find a true position on the projection diagram, and the coordinate value of the true position is set as a parameter for projective transformation. The texture mapping method according to claim 2, which is used for 5. In texture mapping for synthesizing a second image with a first image, information on whether or not to perform texture mapping is described for each pixel of at least one of the first image and the second image. A texture mapping method for holding the stored data and determining for each pixel whether or not to perform the texture mapping according to the format of the data. 6. The texture mapping method according to claim 2, wherein the texture mapping is performed for each pixel. 7. The texture mapping method according to claim 3, wherein the texture mapping is performed for each pixel. 8. The texture mapping method according to claim 4, wherein the texture mapping is performed for each pixel.