JP2001169211A - Image projection apparatus and distortion correction method thereof - Google Patents

Abstract

(57) [Summary] [Problem] For curved screens as well as flat screens,
Achieve accurate image distortion correction with an inexpensive camera. A light spot of a laser pointer (5) arranged near an ideal viewpoint and controllable in angle is projected on a curved screen (2). On the other hand, from the CPU 11 to the graphics board 12
The measurement image (point image) generated in step (1) is drawn in the frame memory of the image transformation circuit 13 and projected from the projector 3 onto the screen 2. A projection image including a light spot on the screen and a point image is captured by the camera 4 and the captured image is captured by the capture board 14.
Take in. The CPU 11 measures the positions of the light spot on the input image of the board 14 and the point image. When the two points match while moving the point image, the pixel coordinates of the point image on the frame memory are determined. Replace with the coordinates on the input image. The conversion value of each pixel coordinate is set in the coordinate conversion parameter memory of the image transformation circuit 13.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video projector for projecting and displaying an image on a screen, and more particularly to a function of correcting a distortion of a projected image.

[0002]

2. Description of the Related Art As a conventional method for correcting image distortion of a video projection apparatus, for example, Japanese Patent Application Laid-Open No. 8-88860 (Cited Example 1).
And Japanese Patent Application Laid-Open No. 8-331610 (Cited Example 2) are known. In the cited example 1, the image projected on the screen is photographed using a camera to recognize the image distortion. In the cited example 2, for example, distortion is corrected by using a flat screen that looks like a trapezoid when the rectangle is distorted.

[0003]

The image distortion correction method according to the above cited reference 1 requires an optically precise camera in order to reduce the distortion, since the image pickup by the camera tends to cause distortion particularly in the peripheral portion. It is expensive. Alternatively, it was necessary to accurately center the photographed image and display the enlarged image.

The method for correcting image distortion according to the above cited reference 2 has difficulty in applying to a screen other than a flat surface such as a curved surface. The distortion of the projected image on the curved screen will be described with reference to FIGS. The projector here is a rectangle (4: 3) with respect to a flat screen
Is a general projection. In (a), the projector 3 projects the curved screen 2 from the viewpoint different from the observer.

An image obtained by projecting a rectangular original image on the curved screen 2 is distorted due to a difference in the distance between the projector and the screen, and appears to the observer as an image 21 having a curved shape as shown in FIG. Image 2 distorted in this curved shape
It is necessary to correct 1 so that the observer looks rectangular. This distortion shape changes greatly depending on the shape and curvature of the screen and the position of the projector and the observer.

FIG. 17C shows a composite image of a video 21-1 and a video 21-2 projected from two projectors on one curved screen. The original image is a rectangle, and the enlarged ideal projected image 21-4 is indicated by a dotted line. In this case, an arbitrary point 21-5 in an area 21-3 where the actual projected images 21-1 and 21-2 overlap each other is a video 2 as a dotted line unless the corrections from both projectors match.
1-4 cannot be obtained. For this purpose, it is necessary to have mutually common positional information, and it is necessary to accurately measure the positional relationship between the screen point and the coordinates of each projector.

SUMMARY OF THE INVENTION An object of the present invention is to overcome the above-mentioned problems of the prior art, to realize an inexpensive camera without the need for an expensive camera, and to accurately correct image distortion even in a screen shape such as a curved surface. It is an object of the present invention to provide a video projection device and an image distortion correction method thereof. Another object of the present invention is to provide an image distortion correction method that can be applied to a composite screen of a plurality of projectors.

[0008]

In order to achieve the above object, means for projecting a light spot on a screen, means for changing the projection direction of the light spot projecting means, and the relative position between the projector image and the light spot By providing a means for measuring the light spot and identifying the relative positions of the light spot and the projected image of the projector, replacing the coordinates of the projector image at the position where both coincide with the light spot coordinates, the image on the screen including the curved surface The distortion can be corrected.

That is, the present invention relates to a method for correcting distortion of a video projection apparatus in which a display image drawn in an image memory is projected on a flat or curved screen by a projector, wherein the light projection apparatus is arranged at or near an ideal viewpoint of the screen. A light point projected while changing the projection position of the screen, and a display image for measurement projected on the screen from the projector are captured and captured by a camera, and the light spot and the display image for measurement are captured on the input image. Measuring the relative position, when it is determined that the light spot and the measurement display image approached within a predetermined distance while moving the relative position, the pixel coordinates of the measurement display image on the image memory, A conversion parameter to be replaced by the coordinates of the light spot on the input image is set.

One form of the measurement image is a point image.
Alternatively, it is a display range in which no image is displayed. For the predetermined in-approach, the optimal approach is the closest approach or display limit (display limit of one pixel or one dot) between the light spot and the display image for measurement.

Further, in the present invention, a projected image of a screen is fetched by utilizing an enlargement function of a camera, and the predetermined within-approach is determined on the enlarged input image. According to this, accurate distortion correction can be performed using an inexpensive camera.

Further, according to the present invention, when a composite image is projected on a screen by a plurality of projectors, the light spot is projected onto an overlapping area of the projection area of each projector, and the conversion parameter based on the light spot is overlapped with the projector. Set each time. According to this, it is possible to obtain a high-quality composite image having a smooth overlapping portion.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. FIG. 1 shows a configuration of a video projection device according to an embodiment of the present invention. Video projection control device 1
Has a main processing unit CPU11 and is constituted by a personal computer (PC). The graphics board 12, the image transformation circuit 13, the capture board 14, and the camera platform control circuit 15
The system bus 16 is connected to the CPU 11.
The graphics board 12 has a VGA format (64
An RGB analog signal of (0 pixel × 480 pixels) is output. The RGB analog output is input to the image transformation circuit 13. The image transformation circuit 13 converts the coordinates of the input image as described later, and outputs the input image as RGB analog signals in VGA format.

The projector 3 receives the RGB analog signals of the image transformation circuit and outputs an image to the curved screen 2. Generally, a curved screen is effective in enhancing the sense of reality. The camera 4 is an ITV camera capable of capturing a video signal. It is installed in a place where the entire surface of the screen 2 can be photographed, and a projected image of the screen 2 is photographed. The video signal of the camera 4 is taken into the video projection control device 1 via the capture board 14.

In this embodiment, a laser pointer 5 is provided for measuring the position of the screen 2 in accordance with the viewpoint of a person who views the image on the screen 2. The laser pointer 5 is mounted on the electric pan head 6. The head 6 is a head control circuit 15
Can be swung up, down, left, and right by the control from, and the light spot of the laser pointer 5 can scan the screen 2 in the horizontal and vertical directions. As will be described later, the light spot of the laser pointer 5 is displayed at a known point on the screen 2 and coordinate transformation is performed so that the position of the light spot captured by the camera 4 and the corresponding position of the distorted image are matched. In addition, accurate image distortion correction can be realized.

FIG. 2 shows the configuration of the image transformation circuit. The image transformation circuit 13 performs coordinate transformation for distortion image correction and outputs the result to the projector 3. The CPU 131 of the image transformation circuit 13
It is connected to the system bus 16 of the control device 1 via a bus I / F 132 and an internal bus 137. A / D134
Converts an RGB analog signal input to a digital signal.
The frame memory 135 is a memory for two surfaces of 640 pixels × 480 pixels and has a double buffer configuration. When one frame buffer is reading image data from the A / D 134, the other frame buffer transfers an image to the D / A 136. Output data. The D / A 136 converts and outputs the digital data into RGB analog signals in VGA format.

The coordinate conversion parameter memory 133 stores the VG
A For each pixel of all pixels (640 pixels x 480 pixels),
Has coordinate data memory. For example, input coordinates (1
When it is desired to convert the pixel of (0, 15) into a pixel of output coordinates (30, 35), (10, 15) is written to an address corresponding to the output coordinates (30, 35) of the coordinate conversion parameter memory.

The CPU 131 refers to the coordinate conversion parameter memory 133 and instructs an address to the frame memory 135 to output the coordinates to the memory of the coordinates corresponding to the pixel.
For example, at the output timing of the output coordinates (30, 35), the CPU 131 starts from the address corresponding to (30, 35) in the coordinate conversion parameter memory 133 and (10, 15).
Read the coordinate data of. Then, an address corresponding to the coordinates of (10, 15) is given to the frame memory 135,
The frame memory 135 outputs the pixel of (10, 15).

Next, the operation of the image projection apparatus according to the above embodiment for correcting image distortion will be described. FIG. 3 shows an overall flow controlled by the CPU 11. In the coordinate transformation parameter generation (step A), the reference position is indicated by the laser pointer 5 (light spot), the image for measurement is output from the projector 3, and the image is captured by the camera 4. A coordinate conversion parameter for distortion correction is obtained while performing position measurement using the image for use.

In writing the coordinate conversion parameters (step B), the obtained coordinate conversion parameters (here, the coordinate values to be converted) are written in the coordinate conversion parameter memory 133. In the video projection (step C), the display image generated by the CPU 11 on the graphics board 12 is drawn and deformed by the image deformation circuit 13 and projected by the projector 3.

FIG. 4 shows a procedure for generating coordinate transformation parameters (step A). First, before measurement, an identity parameter is written to the coordinate conversion parameter memory 133 (A11). The identity parameter is a parameter that makes the output of the image transformation circuit 13 the same as the input. For example, the pixel at the input coordinates (10, 15) is directly output coordinate (1).
(0, 15) are parameters for operating the image transformation circuit 13 so as to be pixels. Note that there is a method of directly inputting the output signal of the graphics board 12 to the projector 3 without writing the identity parameter.

Next, the reference position of the screen 2 is indicated by the laser pointer 5 (A12). The reference position is a reference point of the screen 2, generally, the center of the screen 2. The laser pointer 5 is arranged so as to face an ideal viewpoint (for example, in the case of a spherical screen, the line of sight passes through the center of the spherical surface) so that the light spot of the laser pointer 5 hits this reference position.

Next, a position measurement is performed by projecting a measurement image for position measurement from the projector 3 (A13). FIG. 5 shows a screen state at this time. On the curved screen 2, the projection area 21 of the projector 3 becomes a display area distorted from the original rectangle. This projection area 2
1 includes a light spot 211 (black point) of the laser pointer 5 and a point image 212 (white point) which is a measurement image from the projector 3.
Is displayed.

Therefore, the reference position indicated by the light spot of the laser pointer 5 is measured as to which pixel of the pixel coordinates of the projector 3 (ie, a coordinate conversion parameter) as described later. When the measurement of one designated position is completed, the angle of the laser pointer 5 is moved to the next measured position until all the measured positions are completed (A14) (A15). Movement of the measurement position is instructed by the operation angle of the camera platform 6 from the camera platform control circuit 15, and scanning is performed in a matrix at equal angular intervals vertically and horizontally. When the direct measurement is not performed for all the display pixels of the projector 3, interpolation is performed using a spline function or the like so that the data of the measurement points in the vertical and horizontal directions are smoothly connected (A16).

FIG. 6 shows a detailed procedure of position measurement. This process is mainly performed by the CPU 11. First, the point image 21 is placed at the image center position (x, y) = (320, 240) of the graphic board 12.
Draw 2. The background is black, and the dot image 212 is white. The data drawn on the graphics board 12 is input to the image transformation circuit 13 as an analog signal. As described above, since the identity parameter is written in the image transformation circuit 13, the same image as the input is output and projected from the projector 3 (A131-1).

Next, the projected image on the screen 2 is
Ingest through. The input image on the capture board 14 has a point 211 indicated by the laser pointer 5 and a point image 212 projected by the projector 3 drawn on a black background. Image recognition is performed on this input image, and two points 211 and
12 is measured, and this is set to L (A131-
2).

FIG. 7 shows an image diagram of a drawn image in the frame memory and an image input by the camera. (A) is written in the graphics board 12 and the frame memory 13
5 is a point image 2 at a pixel at coordinates (x, y).
12 is drawn. The adjacent pixels shown in black are coordinates (x−1, y), (x + 1, y), (x, y−1), (x, y + 1), respectively.
Is a candidate point when searching for a position near the light spot 211.

(B) is an input image on the capture board 14 obtained by capturing the projected image of the screen 2 by the camera 4;
A light point 211 and a point image 212 of coordinates (x, y) are drawn,
The distance L between the two points is calculated. Further, the four x-point images 212-1 to 212-4 are respectively adjacent pixels (x−1,
These are virtual points corresponding to (y), (x + 1, y), (x, y−1), and (x, y + 1). In this manner, adjacent pixels on the frame memory are drawn on the screen in a distorted manner like each x point.

Next, after clearing the image on the graphics board 12, the point image 212 is drawn at (x-1, y) (A
132-1). Measure the distance between the two points as before. This is set to M (A132-2). For the point 121 of the laser pointer 5, this time (x-1, y) from the previous (x, y) L
If M is close (A132-3), this is set to a new (x,
y), L, and the processing from step A132-1 is L ≦
It repeats until it becomes M (A132-4).

FIG. 8 is an explanatory diagram of step A132.
In the same image as FIG. 7, the point image 212 is represented by coordinates (x−1, y).
, And the distance M between the point image 212-1 and the light spot 211 is calculated. If the distance M is smaller than the distance L, the point image 212 is approaching the light spot 211. Therefore, the point image 212 is updated to x−1⇒x, M⇒L, and the point image 212 is further moved to the coordinates (x−1, y) (to the left of the screen), and the above processing is repeated. In the example shown in the figure, since the light spot 211 is on the + side of x of the first display pixel point image 212, the first calculated M
Has already become larger than L, and the process shifts to step A133.

In step A133, the point image 212 is converted to (x
(1, +1), and the same processing as step A132 is repeated until (x, y) and L are updated, until L ≦ M. As shown in FIG. 9, in this example, the point image 212-2 is repeated until the point image 212-2 substantially coincides with x of the light spot 211.

Similarly, in step A134, (x, y-
From 1), in step A135, the processing from (x, y + 1) is repeated. In this example, as shown in FIG. 10, the processing in the negative y direction (upward on the screen) is stopped immediately because the processing goes away from the light spot 211. On the other hand, when going in the positive y direction as shown in FIG. 11, the point image 212 eventually comes closest to the light spot 211. At the time of the closest approach, the coordinates (X, Y) of the light spot 211 in the input image with respect to the coordinates (x, y) of the point image 212 in the frame memory 135 are stored in the coordinate conversion parameter memory 133. Then, when the pixel at the coordinates (x, y) of the frame memory 135 is output, if the coordinates are converted so as to be output as the pixel at the coordinates (X, Y), image correction that matches the light spot on the screen can be realized.

Here, the reason why the repetition is performed from the beginning of A132 in A33, A134 and A135 in FIG. 6 is as follows. When the screen 2 is not a plane, even if the point image 212 approaches the light spot 211 in the x direction or the y direction, the distance between the two points may not always be reduced. This can be easily understood by considering, for example, a distance error in a map representing the earth in Mercator projection. However, when one side changes two-dimensionally, such as a cylindrical surface, the distance between the two points is always shortened when approaching the target point from that direction, so that repetition from the beginning is not necessary.

In the above process, the coordinates (X, Y) of the light spot and the coordinates (x, y) of the pixel on the frame memory are moved while moving the reference position by the light spot over the entire screen.
And the pixel coordinates of the frame memory between the light points are interpolated, and the light point coordinates corresponding to the coordinates of all the pixels of the frame memory are stored in the coordinate conversion parameter memory 133. According to this, if the above-described coordinate conversion parameters are obtained for the arrangement of the screen and the projector to be used, simple and accurate distortion correction can be realized for the projection distortion of a screen having an arbitrary shape such as a curved surface. it can.

In this embodiment, the camera platform is used to control the projection direction of the light spot. However, similar effects can be obtained by using a movable mirror instead. In addition, the rotation can be performed manually. In this case, the number of rotations of the laser pointer in the vertical and horizontal directions with respect to the reference position of the laser pointer is input to the control device body 1. In this embodiment, the geometric deformation parameters of the entire image are detected. However, the present method may be applied only when measuring the outer shape of a projected image in which the deformation can be easily recognized by human eyes.

Next, a modification of this embodiment will be described. In the above embodiment, when the resolution of the camera 4 is low, FIG.
Even if the light spot 121 on the screen 2 and the projected image point 122 are slightly shifted as in (a), they are regarded as the same point in image recognition on the input image 140 as in (b). Therefore, the accuracy of position measurement is improved by using the enlargement function of the camera.

FIG. 13 shows a procedure of a position measuring method using the enlargement function of the camera. In this example, A131 to A of FIG.
A procedure A136 is added after the measurement procedure 135. That is, after determining the closest measurement point between the light spot 211 and the display pixel 212 based on the camera image before enlargement, it is checked whether the camera 4 can be enlarged (A136-1). If enlargement is possible, the measurement point regarded as the same point in step A135 is enlarged and photographed (A136-2). The light spot 211 and the display pixel 21 are displayed on the magnified input image.
2, the processing from step A132 is repeated based on the distance L and the coordinates (x, y) of the display pixel 212. When the maximum magnification of the camera has been reached, the processing ends. According to this, even if an inexpensive camera with low optical accuracy is used,
Accurate distortion correction becomes possible.

Next, a position measuring method according to another embodiment of the present invention will be described. FIG. 14 shows a procedure of position measurement in the y direction. First, the upper and lower ranges (left and right in the x direction) of the display image are set by the number of dots (B131). For example, y0 =
The height yh from 0, y0 is the total number of dots (yh0 = 48
0). Then, the laser pointer 5 is pointed to a point to be measured on the screen 2, the light spot 211 is enlarged and photographed by the camera 4, and the coordinates (x, y) of the laser pointer 5 in the input image are read (B132).

Next, while changing the image display range from the CPU 11 via the graphics board 12, the image transformation circuit 13, and the projector 3, the display range of the light spot 211 is searched as follows. First, from y0, height yh / 2 (0
Displays range of 240 dots) (B133), the coordinates (x ₀ of the light spot 211 to the display range, y ₀₎ is determined whether the display (B 134). If the light range 211 is included in the display range, the process shifts to step B137, and the y height is set to yh / 2 (240 dots) without changing y0.
It returns to B133. Then, the new display range is from y0 to the height yh / 2 (0 to 240 dots). On the other hand, when the light spot 211 is not included in the display range,
The upper part is set as y0 = y0 + yh / 2 (B135), the y height is set as yh = yh / 2 (B137), and the process returns to step B133. However, when the updated y0 exceeds the display limit (y0 <1 dot), the process ends (B136).

FIG. 15 is an explanatory diagram of this example. As shown in (a), the light spot 211 is included in the first upper half display range. Next, in (b), y0 = 0 to yh ′ / 2 (= y
h / 4) is the display range (0 to 120 dots), and does not include the light spot 211 here. Therefore, y0 = y
h / 4 (below y of the previous display range), y height is yh '= yh /
2 (= yh / 4), a new display range (120 to 18)
(0 dot), the light spot 211 is included.

Thus, the y coordinate of the display range narrowed down to just before the display limit, that is, the y coordinate in the frame memory, corresponds to the y coordinate (y ₀ ) of the light spot 211.
Similarly, the x coordinate of the display range narrowed down in the x direction is obtained, and the coordinates of the pixel of the frame memory corresponding to the coordinates (x ₀ , y ₀ ) of the light spot 211 are obtained. According to this measurement method, the display of the measurement pixel 212 and the calculation of the distance from the light spot 211 are not required, so that the measurement processing can be sped up.

Next, as still another embodiment of the present invention, an application example in a system for projecting from a plurality of projectors will be described. FIG. 16 shows a composite image by two projectors. A projection area 20-1 from the left projector and a projection area 20-from the right projector on the screen 2.
2 partially overlaps. When the left and right portions of the rectangle are partially overlapped and displayed by the left and right projectors, the correction is performed so that an ideal projected image 20-4 is obtained.

The luminance of the composite image is adjusted so that there is no difference between the luminance of the composite image and the non-overlapping region so that the image of the overlapping region 20-3 looks smooth. That is, the luminance from the left projector decreases as going to the right, and the luminance from the right projector decreases as going to the left, and the sum of the left and right luminance values becomes the original luminance value. For this reason, in the overlapping area, the left and right projectors need to project images at the same position, and it is necessary to mutually have position information for distortion correction.

In the present embodiment, one light spot 201 is displayed in the overlapping area 20-3 by the laser pointer 5, and first, position measurement for displaying only the image of the left projector is performed.
A coordinate conversion parameter of a pixel corresponding to the light spot 201 is obtained. This position measurement may be performed by any of the above-described embodiments. Next, a coordinate conversion parameter of a pixel corresponding to the light spot 201 is obtained by position measurement displaying only the image of the right projector. In this case, the brightness of the pixel at the conversion destination based on the coordinate conversion parameter from the left is reduced as described above, and is added to the brightness of the same conversion destination based on the coordinate conversion parameter from the right and displayed.

According to this embodiment, when displaying a composite image by a plurality of projectors on a screen such as a curved surface,
Since the luminance is added by correcting the distortion of the image position in the overlapping area, a smooth video without discomfort can be displayed.

[0046]

According to the present invention, a light spot projected from an ideal viewpoint on a screen and a light spot projected on a frame memory are drawn,
The position deviation from the display image for measurement projected on the screen from the projector is measured on the input image captured from the camera, the display image is moved, and the coordinates of the light spot when both coincide are stored in the frame memory of the display image. , The image distortion can be easily and accurately corrected not only for a flat screen but also for a curved screen.

Further, since the displacement is measured by using the enlargement function of the camera, the distortion can be corrected with high accuracy even with an inexpensive camera.

Further, when the video projector is composed of a plurality of projectors, distortion is corrected by the respective projections at the same position of the overlapping area on the screen, so that a high-quality composite image can be obtained.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an image projection device according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of an image transformation circuit according to one embodiment.

FIG. 3 is an overall flowchart of an image distortion correction method according to an embodiment of the present invention.

FIG. 4 is a flowchart of a coordinate conversion parameter generation method according to one embodiment.

FIG. 5 is an explanatory diagram of position measurement.

FIG. 6 is a flowchart showing a detailed procedure of position measurement.

FIG. 7 is an explanatory diagram of position measurement.

FIG. 8 is an explanatory diagram of a process of position measurement.

FIG. 9 is an explanatory diagram of a position measurement process.

FIG. 10 is an explanatory diagram of a process of position measurement.

FIG. 11 is an explanatory diagram of a process of position measurement.

FIG. 12 is an explanatory diagram showing the necessity of enlarged photographing.

FIG. 13 is a flowchart of position measurement using an input image of enlarged photographing.

FIG. 14 is a flowchart of position measurement according to another embodiment.

FIG. 15 is an explanatory diagram of position measurement according to another embodiment.

FIG. 16 is an explanatory diagram of image distortion using a plurality of projectors.

FIG. 17 is an explanatory diagram of a problem in the conventional technique.

[Explanation of symbols]

1 ... image display control device (PC), 2 ... screen, 3 ...
Projector, 4 ... Camera, 5 ... Laser pointer, 6 ...
Head, 11 ... CPU, 12 ... Graphics board, 1
3 image transformation circuit, 14 capture board, 15 pan head control circuit, 131 CPU, 132 coordinate conversion parameter memory, 135 frame memory, 211 light spot,
212: Point image (measurement image).

Claims (10)

[Claims] 1. A distortion correcting method for a video projector, wherein a display image drawn on an image memory is projected on a flat or curved screen by a projector, wherein a projection device arranged at an ideal viewpoint of the screen or in the vicinity thereof is used to adjust the screen. A camera captures and captures a light spot projected while changing the projection position and a measurement display image projected on the screen from the projector, and measures a relative position between the light spot and the measurement display image on the input image. When it is determined that the light spot and the measurement display image have approached within a predetermined distance while moving the relative position, the pixel coordinates of the measurement display image on the image memory are displayed on the input image. Setting a conversion parameter to be replaced by the coordinates of the light spot in (1). 2. The display image for measurement according to claim 1, wherein the display image for measurement is a point image in which an arbitrary pixel in an image memory is drawn, and a distance between the point image and the light spot is obtained. And correcting the display of the point image to approach the predetermined area. 3. The measurement display image according to claim 1, wherein the display image for measurement represents a display range of an image memory, and the display range is sequentially reduced so that the light spot is included in the display range so as to approach the predetermined inside. A distortion correction method for a video projection device, comprising: 4. The distortion correcting method according to claim 1, wherein the predetermined in-approach is such that the closest approach or display limit between the light spot and the display image for measurement is an optimum value. 5. The distortion correcting method according to claim 1, wherein the predetermined in-approach is determined on a magnified input image of the camera. 6. The image according to claim 1, wherein the conversion parameter of a pixel which is not replaced by the coordinates of the light spot in the pixel coordinates of the image memory is obtained by interpolation between light spots. A distortion correction method for a projection device. 7. The projector according to claim 1, wherein when projecting a composite image by a plurality of projectors onto the screen, the light spot is projected on an overlapping area of a projection area of each projector. A distortion correction method for a video projection apparatus, wherein the conversion parameter is set for each overlapping projector. 8. A video projection apparatus comprising: a flat or curved screen; a projector for projecting a projection image on the screen; and an image control device for rendering a display image to be the projection image in a frame memory and outputting the display image to the projector. Placed at or near the ideal viewpoint of the screen,
A projection device for projecting a light spot on an arbitrary projection position on a screen, a device for changing the projection position of the light spot, and a camera for photographing an image of the screen, further comprising an input image captured from the camera An image storage device for storing image data and image distortion correction means for measuring a relative position of the display image for measurement by the projector and the light spot on the input image to correct distortion of the projected image. An image projection device, comprising: 9. The image distortion correction unit according to claim 8, wherein the image distortion correction unit calculates pixel coordinates on the frame memory when the light spot and the measurement display image match on the input image by using the coordinates of the light spot. An image projection device comprising a coordinate conversion parameter memory for replacement. 10. The projector according to claim 9, wherein the frame memory is configured as a double buffer, and an input buffer for drawing the display image and the drawn display image are replaced with the coordinates of the coordinate conversion parameter memory to the projector. A video projection device, wherein an output buffer for outputting is changed every frame.