JP5224721B2 - Video projection system - Google Patents

Description

  The present invention relates to a video projection system that projects a video on a dome screen using a plurality of projectors and is applied to a planetarium or the like.

When projecting an image on a dome screen, a multi-projection system that obtains a high-resolution image using a plurality of projectors is often used as a planetarium device.
This is extremely useful because it has the feature that it can project high-resolution images that cannot be projected by a single projector. In the dome image device that projects the image on the entire dome screen, the multi-projection method is easier to obtain than the monocular method using a fisheye lens, and the aberration of the peripheral image peculiar to the fisheye lens, that is, the deterioration of the image quality near the horizon or the peripheral reduction. Since it is difficult to cause problems such as insufficient brightness near the horizon due to light, the multi-projection method is preferably used for image projection in medium to large dome theaters. Further, when an optical planetarium is used in combination with this planetarium facility, a multi-projection method that can be installed around the dome without installing the projector at the center is extremely useful.

  In a multi-projection video system, video is projected from a plurality of projectors onto a common screen surface. At this time, the areas to be projected are set separately for each, and the boundary line is set so that the images projected from the respective projectors are connected to one to form an image of the entire dome screen. In order for the images projected from the projectors to be connected and viewed as a single image, it is natural to correct various image distortions caused by differences in the projection positions and variations in characteristics, making it perfect at the joints on the screen surface. Of course you have to project to match. The alignment in the multi-projection method is proposed in Patent Documents 1 and 2.

Since such distortion correction requires complicated calculation, it is almost the case that a video transmission device or a video deformation device having a distortion correction function is used, and a computer is used for actual distortion correction calculation.
However, in reality, slight errors in projection cannot be made completely zero, so in order to make the joints between the projection planes inconspicuous, the images on the multiple projection planes sandwiching the joints have a certain width in the vicinity of the joints. In addition, the processing is performed so that the joints are made inconspicuous by applying gradation to each. This is called edge blending.
Edge blending is roughly divided into electronic blending and optical blending.

Electronic blending is a screen projected from each projector, and the area that does not need to be projected because it overlaps the projection surface of another projector is either masked in advance on the video itself, or electrically masked on the video signal. It is a method to do.
On the other hand, in optical blending, an optical mask having a predetermined shape of a transmission window is installed in front of a projection lens of a projector, and unnecessary portions of each projection screen are shielded from light. Since the projection lens is not focused on the light-shielding plate, the shadow of the optical mask results in a certain degree of blur, resulting in a gradation effect (Patent Document 3).

In electronic blending, there is a technique that can freely set boundaries with software, make full use of computer calculations, and can automate blending itself by imaging with a camera (Patent Document 1).
However, in liquid crystal type, DLP type, and laser projection type projectors that are often used as projectors, it is difficult in principle to make perfect black because the light irradiated in advance is controlled for each pixel by an optical element. In other words, as long as the light source of the projector is lit, the portion that should be completely black appears to be slightly black, and the contrast of the image is lowered. In particular, in the planetarium, when projecting together with a star, a frame due to this black floating phenomenon is seen, which affects the way the star is seen. In particular, in a region where a large number of projection planes overlap on the screen, the black float is added by the amount corresponding to the projection plane, and unnaturalness is exaggerated more and more. It is difficult to solve this problem only by the electronic blending method.
Patent 3497805 JP 2006-337682 A Japanese Patent Laid-Open No. 5-19346

Therefore, optical blending is used. In optical blending, an extra portion of light is physically cut off, which can reduce the problem of an electronic background frame. In addition, a method of obtaining a good connection by cutting off excess light while leaving a certain margin with an optical mask and correcting the lightness in the vicinity of the joint uniformly by electronic correction.
By the way, in recent planetariums, composite planetariums using both an optical stellar projector and a digital image projected from a projector have become more effective.
This is because a star can be projected more beautifully when projected from an optical stellar projector, and the projector is good at complicated effects. However, when using an optical stellar projector and a projector in combination, when projecting a star from the optical stellar projector and projecting constellation lines or constellation pictures from the projector, the black floating of the projector will look like a starry sky. It has a great impact and must be minimized. Also, constellation pictures and constellation lines need not be projected so brightly because of the balance of brightness with stars.

  Therefore, in order to coexist with the starry sky, the projector is used with the light amount of the projector being reduced, but in this case, it is most ideal to project with the aperture stopped down. In addition to the fact that many commercially available projectors have such a diaphragm mechanism built in, the contrast of bright and dark areas can be improved by narrowing the diaphragm. Because it becomes. However, if you do not want to use it with stars, or if you want to emphasize the brightness and gorgeousness of the image, the amount of light is insufficient when the aperture is stopped, so there are many cases where you want to project with the aperture open. Therefore, it is desirable that the aperture can be opened or closed depending on the production scene.

However, in optical blending, for example, when the numerical aperture of the projection lens of a projector changes, the depth of focus changes and the gradation effect changes. When the iris is closed, the depth of focus is deepened, so that the shadow of the optical mask is clear, and when it is opened, it becomes blurred.
However, in the image correction method as described above, such a change in the characteristics of the optical mask due to the change in the numerical aperture of the diaphragm is not taken into consideration. There was a problem that unevenness would occur if changed. Therefore, the conventional multi-projection system has a problem in that it is difficult to project with the aperture being opened or closed, and it is difficult to achieve both bright and gorgeous effects and the compatibility with the starry sky.

  From the above background, conventionally, in order to project a high-quality image on the dome screen, the method of installing a projector using a fisheye lens in the center of the dome has a problem that the resolution near the horizon and the image near the horizon deteriorate. there were. Further, in the method of projecting on a plurality of screens, although the positional relationship between the plurality of screens can be accurately maintained by a method using a reference mark or the like, there is a technical problem in edge blending. In other words, the majority of projectors such as the liquid crystal system and the DLP system cannot express black completely, and there is a black floating phenomenon. In software edge blending that masks unnecessary parts of the image, the background black floating phenomenon occurs. Waking up has the problem that the projection frame frame of each projector looks unnatural, and this black float is particularly noticeable in the part where the projection surfaces of a plurality of projectors overlap.

On the other hand, when an optical mask is used, the above problem is less likely to occur. However, with correction using only a physical mask, it is extremely difficult to accurately adjust the shape of the mask at the dividing boundary line, and accurate adjustment is possible. However, there is a problem that the joint becomes unnatural due to uneven light. The method of using edge blending with an optical mask and software can solve the above-mentioned problems, but it can be used when screening with the aperture open to show a bright image clearly, and when projecting an image with the starry sky. When displaying dark images, it is desirable to change the aperture of the projector, but since the shadow pattern by the optical mask changes depending on the numerical aperture, the change pattern of the dividing boundary line changes, so the numerical aperture of the aperture is changed. If it is changed, there is a problem that unevenness in the amount of light occurs remarkably at the dividing boundary line.
The present invention has been made to solve the above problems.

In the multi-projection system using an optical mask, the applicant of the present application pays attention to the projection light quantity distribution by the optical system, that is, the change in the optical profile, which occurs when the numerical aperture is changed by controlling the aperture of the projector, and this change is observed. We propose a method of canceling and correcting the brightness distribution of video.
That is, first, the projection light quantity distribution to be projected by each projector, that is, the target profile, is calculated based on the preset division projection setting, and compared with the optical profile that is the projection light quantity distribution of the projector affected by the optical mask. By calculating, the correction amount of each part of the video is determined, and this is used as a correction profile. By correcting the brightness of the video based on this correction profile, a video having a brightness matching the target profile is obtained.

  An object of the present invention is to provide an image projection system for projecting an image output from a computer, a video device, or the like onto a screen, and superimposing screens projected by a plurality of projectors on the screen to form one image. It is an object of the present invention to provide a high-definition video projection system capable of projecting both bright images and dark images in harmony with the starry sky, even if the number changes, hardly causing uneven brightness near the joints of the projection surfaces.

In order to achieve the above object, claim 1 of the present invention projects an image output from image generation means of a computer or a video apparatus onto a screen, and displays a screen projected by a plurality of projectors having variable numerical apertures. An image projection system that forms a single image by superimposing on a screen, sets the numerical aperture of the projector to a predetermined value, projects the image onto the screen, and shoots the projected image with a camera to obtain a correction map. A projection image acquisition means capable of acquiring data of a projection image projected on the screen, and a projection correction map is created by calculation using the correction map according to the numerical aperture at the time of projection of an actual projector, and the correction at the time of projection Image correction data calculating means for calculating image correction data at the numerical aperture at the time of projection by a map, and using the image correction data It characterized that you correct the brightness of an image.
According to a second aspect of the present invention, an image output from image generation means of a computer or a video device is projected on a screen, and a plurality of screens projected by a projector capable of changing the numerical aperture are superimposed on the screen. An image projection system for forming an image of the image, having a camera capable of photographing the image projected on the screen and photographing a test pattern projected on the screen with a numerical aperture set to a predetermined value for each projector Then, the captured image acquisition means for obtaining the captured image data of the projected images of the respective projectors obtained by photographing, and the numerical aperture of the diaphragm of each projector is set to a value different from that at the time of the test pattern photographing to project the image. A predetermined calculation is performed based on the imaging data, the numerical aperture of each projector at the time of shooting, and the numerical aperture at the time of projection. Characterized by comprising the image correction data calculating means for calculating correction data for correcting the luminance of an image, and image correction projection means for projecting performing correction of the luminance of the image by the image correction data.
According to a third aspect of the present invention, the image output from the image generation means of the computer or video apparatus is projected onto the screen, and the image projected by the plurality of projectors is superimposed on the screen to form one image. The system includes a camera capable of capturing an image projected on the screen, and each of the plurality of projectors has an aperture that can be opened and closed by an external command, and the numerical aperture of each aperture is set for each projector. The test pattern projected on the screen is set to a predetermined value, and the captured image acquisition means for obtaining the captured image data of the projected image of each projector, and the numerical aperture of the diaphragm of each projector, When projecting a video with a value different from that at the time of shooting the test pattern, it is obtained from the imaging data of the captured image acquisition means Correction that corrects the luminance of each position of the image by a predetermined calculation based on the preset target profile or target blend mask that shows the optical profile to be projected and the amount of light that each projector should project according to the position on the screen and image correction data calculating means for obtaining the data, characterized by comprising a video correction projection means for projecting performing correction of the luminance of the image by the correction data.
The optical profile can be obtained by projecting a test pattern onto a screen with a projector affected by a light shielding mask and capturing a projected image with a camera. For example, after obtaining an optical profile in the case of multiple different numerical apertures (open aperture and minimum aperture, etc.) and obtaining an optical profile at the actual numerical aperture during use, or after obtaining an optical profile at a predetermined numerical aperture The optical profile at a numerical aperture different from that at the time of photographing is estimated by numerical means, and correction calculation is performed based on this. Then, the corrected image is projected.
According to a fourth aspect of the present invention, in the first, second, or third aspect, the screen is a dome-shaped screen having a curved surface.
According to a fifth aspect of the present invention, in the invention according to the second, third, or fourth aspect, the photographed image acquisition means obtains a plurality of imaging data at a plurality of values with known numerical apertures of the projector, and the video correction data The data used by the calculation means is the plurality of imaging data and the known numerical aperture.
According to a sixth aspect of the present invention, in the invention according to the second, third, fourth, or fifth aspect, the front surface of the projector has a light shielding mask for shielding light around the screen.
According to a seventh aspect of the present invention, in the invention according to the sixth aspect, the predetermined calculation is a function of a relationship between a position and a transmitted light amount when the aperture of the diaphragm is shielded from light by an edge of the light shielding mask. It is characterized by using a converted gradation function.
According to an eighth aspect of the present invention, in the invention according to the seventh aspect, the optical gradation area formed by the light shielding mask and the overlapping area of the projection surface are at least partially overlapped.
A ninth aspect of the present invention is the camera according to any one of the second to eighth aspects, wherein the camera is mounted on a stellar projector, and the direction of the camera is set by angular position control of a rotation axis of the stellar projector. It has the means.

According to the above-described configuration, the image that is automatically adjusted and projected using the plurality of projectors without causing unevenness without performing complicated and costly artificial adjustment work. The image can be projected with uniform brightness.
When the projector's black level is not low enough, when it is used together with an optical mask for the purpose of removing excess light, the optimal shape is displayed when determining the shape of the transmission window of the optical mask. Regardless of this, the transmission window can be easily created.
In addition, since the camera used for imaging can be used together with the camera used for image distortion correction, the position and brightness and color of the image can be shared with a camera that is mounted on a stellar projector and performs high-precision alignment. Can be corrected without additional hardware. This makes it possible to provide a high-quality dome image easily, stably and at a low cost.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1A is a perspective view for explaining the appearance of a video projection system according to the present invention.
At the center of the dome screen 1, an optical stellar projector 2 with a video camera 6 is installed. In this embodiment, the optical stellar projector 2 does not use the stellar projection function, and is used as a mounting means for the video camera 6 described later.
A plurality of video projectors 3 are installed around the dome screen 1, and the images projected from the video projectors 3 overlap each other on the dome screen, and are fixed at respective positions and posture angles so as to cover the entire surface of the dome screen 1 Has been. In the figure, only two of them and the screen projected therefrom are shown.

The video projected by the video projector 3 is supplied from the video generation PC 4 as an analog RGB signal or DVI signal. The video generation PC 4 is connected to the control PC 5 via a communication line, and can transmit and receive various commands and data for video generation.
It is assumed that the position and orientation angle of each projector 3 are known, and the coordinates on the projection screen of each projector 3 and the horizontal coordinates on the dome screen 1 are in a state where mutual conversion is possible by coordinate conversion. The coordinate transformation on the projector screen from the horizontal coordinate is referred to as projection transformation, and the inverse transformation is referred to as projection inverse transformation.
For example, it is possible to perform projection conversion and projection reverse conversion with high accuracy using the reference mark means provided in the star projector 2. FIG. 1B shows a display example of the horizontal coordinate of the dome screen.

FIG. 2 is a perspective view of the structure of a stellar projector installed at the center of the dome screen.
A base 11 is installed on a base 10 so as to be rotatable around an azimuth axis, and a gantry fork 12 is attached thereon. The base 11 and the gantry fork 12 can be arbitrarily controlled for rotation by an azimuth axis servo motor (not shown). The gantry fork 12 is provided with a horizontally mounted latitude rotating shaft 13 for adjusting the latitude, and a cylindrical daily base 14 is installed around the latitude rotating shaft 13 so as to be rotatable. ing. The rotational movement of the daily base 14 around the latitude axis can also be arbitrarily controlled by a latitude axis servo motor (not shown). In addition, the northern sky star sphere 15 and the southern sky star sphere 16 having a function of projecting a star can be arbitrarily rotated around the diurnal axis installed on the diurnal base 14 at right angles to the latitude axis. However, this diurnal motion is not used in this embodiment.
A video camera 6 is attached to the diurnal base 14 as projection image acquisition means in a direction orthogonal to both the latitude axis and the diurnal axis. The video camera is a camera capable of electronic photography such as a CCD camera, and the imaging result is obtained in the form of a video signal or a digital video signal by LVDS, etc., though not shown, via a latitudinal slip ring or an azimuth axis slip ring. Video signals can be sent to the outside. Also, the video transmission method can be wireless.

The main point of the camera is set away from the latitude axis by COFS.
The orientation of the video camera 6 mounted on the stellar projector 2 can be arbitrarily oriented by controlling the rotation of the latitude axis and the azimuth axis. In this case, the azimuth angle AZM of the optical axis of the camera can be set by rotation control of the azimuth axis, and the elevation angle ALT of the optical axis can be set by rotation control of the latitude axis.
For example, the elevation angle is 0 degree when the northern star is 15 up and the diurnal axis is upright, and the elevation angle is 90 or -90 degrees when the diurnal axis is horizontal. The north-south horizon is a straight line connecting the south horizon and the north horizon when the stellar projector is used for stellar projection, and the elevation angle of the video camera 6 is an angle formed by the north-south horizon and the camera optical axis.
Many of the video cameras 6 have a function of automatically changing the gain and the aperture, but in this embodiment, both the gain and the aperture are always used in a fixed state. In addition, the physical luminance and the luminance obtained from the imaging data (data obtained from each pixel of the CCD image sensor of the video camera) are often not linearly related (camera gamma characteristics). When the pixel of the video camera is read, the luminance data corrected by the gamma characteristic function and linear with the luminance as an actual physical quantity is obtained.

This gamma function can be acquired based on the luminance data obtained from the imaging result and the illuminance at that time, by taking a picture with a video camera while changing the illuminance of the subject in advance. Can be known and easily guessed, and is not the essence of the present invention, so a specific procedure is omitted. Thereafter, all brightness data obtained from the camera is treated as the brightness after this gamma correction unless otherwise specified.
In the present invention, it is not essential to mount a video camera on the stellar projector as the camera imaging means. However, in one use of the present invention, the planetarium, a stellar projector is often placed in the center of the dome screen. By mounting a video camera on this stellar projector, it is possible to take an image of the entire dome without blind spots at an ideal position based on the dome's center without moving the stellar projector, and to control the direction of the camera Since it is realized by the motion function of the stellar projector, it is not necessary to prepare a special drive mechanism separately. Further, the video camera mounted on the stellar projector 2 is advantageous because it can be used in combination with other devices that automatically and highly accurately correct image distortion using a reference mark. In this embodiment, a video camera mounted on a stellar projector will be described for the reasons described above.
In the case where there is no stellar projector, the direction of the video camera can be arbitrarily controlled by other means, and such an embodiment is also included in the scope of the present invention.

FIG. 3 is a perspective view showing an embodiment of a projector used in the present invention.
The projector 3 is a commercially available video projector using a liquid crystal or a DLP element, and has a function of projecting a video signal input from the video input cable 21 (video signal from the video generation PC). The projection lens 22 has a built-in movable aperture (not shown in the figure) that can change the numerical aperture electrically. With the control signal (for example, RS-232C command) input from the control signal cable 23, the numerical aperture of the movable diaphragm can be controlled within the range (for example, F3.2-F8) set by the specifications unique to the projector. . When the aperture is opened, the projected image becomes brighter but the contrast ratio is lower. When the aperture is closed, the projected image becomes dark, but the contrast ratio increases.

  The projector 3 is mounted on a base 24, and a glass substrate 25 having sufficient transparency is installed upright through the base 24 in front of the projection lens 22, and the projection image is a glass plate of the glass substrate 25. It is designed to pass through. Although the transmittance of glass is extremely high, a ghost may occur in the projected image when there is surface reflection, and thus an antireflection film is provided on one side or both sides. A light shielding mask 27 that does not transmit light is attached to the surface of the glass plate. The light shielding mask 27 is made of a thin aluminum plate whose surface is painted black. The light-shielding mask 27 is provided with a projection window 28 having a predetermined shape so that the light projected by the projector can pass halfway. The projection window 28 allows most of the image projected from the projector 3 to pass through, but blocks the light in the peripheral portion, and has the effect of masking the peripheral portion of the image projected on the dome screen in a dimly dark manner. The shape of the projection window will be described later.

Next, the angle of view of the video camera will be described.
As for the screen resolution of the video camera, the width is CW pixels, the height is CH pixels, and the pitch of the CCD elements is CP.
If the focal length of the camera is F, the horizontal field angle QW and the vertical field angle QH are QW = atan (CW * CP / (2 * F) * 2) (1)
QH = atan (CH * CP / (2 * F) * 2) (2)
It becomes.
As a result, when the dome radius of curvature R, the azimuth angle of the camera mounted on the stellar projector is determined as AZM, and the altitude is determined as ALT, the coordinates on the dome screen reflected in any pixel (CX, CY) on the video camera screen are transformed. Can be calculated by Since a specific calculation method is known, it will be omitted.

Next, a system connection configuration will be described with reference to FIG. 1A.
Screens to be projected by a plurality of video projectors are respectively generated by the video generation PC 4 assigned to each projector. The control PC 5 is connected to all the video generation PCs 4 by communication lines, and from the control PC 5 it is possible to generate arbitrary graphics and video by the video generation PC 4 by a predetermined command from the communication port. For example, vector graphics such as points, straight lines, circles, or moving images. In addition, the opening / closing of the movable diaphragm that can change the numerical aperture of the projector 3 can be controlled by the diaphragm control line from the image generation PC 4, and the numerical apertures of the diaphragms of all projectors can be controlled collectively from the control PC. It has become.
The control PC 5 is also connected to the star projector 2 via a communication line, outputs a pulse train corresponding to a predetermined movement amount from the servo I / F, and controls the rotation angle position and rotation angular velocity of the star projector via the servo control device. It can be done. The stellar projector 2 has a built-in video camera 6 as described above, and a video signal picked up by the video camera 6 can be taken into the control PC 5 by a frame grabber of the control PC 5. . The frame grabber stores luminance information of each pixel of the acquired video, and the luminance of an arbitrary pixel can be accessed and updated for each synchronization signal of the video camera.

FIG. 4 is a diagram showing an example of a layout for projecting an image on the entire dome screen using a plurality of projectors. This figure is an expression method called a dome master. A square inscribed circle is the horizon, and the center corresponds to the zenith. That is, the coordinates on the dome master can be easily converted into altitude and azimuth (horizontal coordinates) on the dome screen. The whole sky is divided into 6 planes S1 to S6 and projected by individual projectors.
However, since it is extremely difficult to perfectly match the boundary lines of the projection planes of the projectors, overlapping areas with a certain width are provided between the projection planes, and they overlap each other with gradation. It is necessary to project so that This is called edge blending.

FIG. 5A is a diagram showing the overlapping area of the projection screen of each projector on the dome master.
The overlapping region is provided with a certain width at the boundary line of each projection plane. Here, it is assumed that a white image (all white image) is projected at a predetermined luminance (target luminance) on the entire dome screen by a plurality of projectors. The area occupied by each projection plane is projected only by the corresponding projector. In the overlapping area, projection light from two or more projectors adjacent to each other across the boundary line is synthesized. At this time, the luminance of the overlapping area must match the occupied area. For this purpose, it is necessary to reduce the amount of light projected from each projector at a constant ratio in the overlapping region. FIG. 5B shows the luminance of the projector on the S1 plane and the S2 plane at the boundary line between the S1 plane and the S2 plane. By crossing and changing the light amounts of S1 and S2 in the overlapping region, the sum of both becomes equal to the luminance of the occupied region. Under such conditions, both the occupied area and the overlapping area can be seen with uniform brightness.

The amount of light that should be projected by each projector when performing divided projection according to the position on the screen is referred to as a target profile. The target profile can be calculated each time based on data such as the position of the dividing boundary line and the width of the boundary line. However, in practice, it is better to calculate the target profile once over the entire surface of the dome screen to obtain two-dimensional data. It is good because the amount of calculation is reduced. This is expressed on the dome master and expressed on the projection screen. In this embodiment, the one on the projection screen is used. The target profile is synonymous with a blend mask, and can be created by transforming a dome master type blend mask into a projection screen using projection transformation or inverse projection transformation. A target profile is prepared for each projection plane.
For example, a dome master type blend mask is created with a monochrome 256 gradation image having a predetermined resolution, for example, 2000 PIXEL. The brightness of each pixel corresponds to the brightness at that position. For example, if the brightness of a pixel is 0, the brightness is zero, and the position is not projected on the screen. When all the pixels of the dome master type target blend mask of all the projection planes are added, all become 1 (255 in 256 gradations).

FIG. 6 is an example showing the dome master type blend mask of the S1 surface in the example of FIG. FIG. 7 also shows a dome master type blend mask on each projection plane.
Here, before explaining the procedure of the specific embodiment, the basic characteristics of the optical profile output in a projector equipped with a light-shielding mask will be explained, and the luminance of the actual video will be corrected to match the target profile The basic principle to be used and the correction profile used for correction will be described.

FIG. 8 is a diagram for explaining a luminance distribution of an image projected from a video projector to which a light shielding mask is attached. Only one of the multiple units used is shown.
A diaphragm 29 is built in the principal point of the projection lens 22 of the video projector 3 and is set to a predetermined numerical aperture. It is assumed that a light shielding mask 27 is installed in front of the video projector 3 and the projection light projected from the video projector 3 covers a range that sufficiently covers the projection window of the light shielding mask 27. Here, when a video of uniform brightness, for example, an image of all white, is projected from the video projector 3, the light around the dome screen 1 is vignetted and blocked by the light shielding mask 27. The shadow is greatly blurred because it is out of focus.
When the video projector 3 itself has no projection unevenness, the optical profile represents the luminance distribution on the screen when the projector projects an all-white screen without any luminance correction, with the maximum value being “1”. I will call it. The optical profile has a substantially constant value in the entire transmission region that is not affected by the shadow of the light shielding mask.

On the other hand, an area where no image is reflected by the shadow of the light shielding mask 27 is referred to as a total light shielding area. An optical gradation region in which the brightness changes continuously with a certain width is formed between the total transmission region and the total light shielding region. The non-projection area is also an area occupied by another projection plane.
As is apparent from this figure, the pattern of the target profile is lower than the optical profile. However, the gradation area of the optical profile and the overlapping area of the target profile partially overlap.
Since the optical profile represents the maximum luminance that can be physically projected, the target profile must be less than or equal to the optical profile at all positions on the projection screen. In this figure, the optical profile and the target profile are related by a one-dimensional function. However, since a two-dimensional image is actually handled, any profile is expressed by a two-dimensional function.

The trouble is that changing the numerical aperture of the video projector changes the curve of the optical profile.
FIG. 9 is a diagram for explaining a change in the optical profile when the numerical aperture of the diaphragm of the video projector is changed. The position of the light shielding mask is also shown. FIG. 9A shows an example in which the numerical aperture is F8 and FIG. 9B is F3.2. The optical profiles change almost symmetrically across the edge position of the light-shielding mask. However, in the case of F8, the depth of field becomes deeper and the change is abrupt because the aperture is narrower. I understand.
When projecting an image based on such a relationship, in order to make the luminance distribution of the image coincide with the target profile, the image itself may be corrected in advance or the image signal may be corrected. The target profile for correcting the luminance value, that is, the one indicating the correction amount of the luminance value from the ratio between the target profile and the optical profile is called a correction profile, and is shown in FIG. However, since the correction can only be dimmed, the correction profile must always be 1 or less at all positions on the screen.
FIGS. 9A and 9B both satisfy this condition, and show how the correction profile changes depending on the difference in numerical aperture.

Next, a correction map creation method will be described.
Shooting in two states of high and low numerical aperture of the video projector to obtain the correction map of each state, according to the numerical aperture at the time of actual projection, the correction map at the time of projection by interpolation calculation from the two correction maps It is a procedure to create.
FIG. 10 is a flowchart showing a procedure for obtaining a correction map, and is a flowchart for one of a plurality of projection planes.
The aperture of the video projector can be arbitrarily set in the range of F3.2 to F8, and the projection screen of the video projector and the horizontal coordinate can be converted to each other.
First, the dome master type blend mask of each projection plane is read into the memory unit in advance (step (hereinafter referred to as “S”) 001). Subsequently, the blend mask data is read from the memory unit using the conversion means from the coordinates of all the pixels of the video projector to the horizontal coordinate, and stored in the target profile data memory unit as target profile data on the projector screen. FIG. 11 shows an example in which the target profile of the S1 surface is imaged.

Subsequently, the numerical aperture of the projector is set to open (F3.2) (S002). Thus, camera shooting is performed (camera shooting will be described in detail later), and a correction map 1 is obtained (S003, S004). This is the open correction map.
Next, the numerical aperture is minimized (F8). Camera correction is performed with this numerical aperture to obtain a correction map 2 (S006, S007). This is the minimum aperture correction map 2.

Next, details of the camera photographing procedure will be described.
FIG. 12 illustrates the state of camera shooting.
The image from the video projector can be projected over the projector projection range 30 in the figure, and a figure of any shape and color can be displayed by a command from a control PC (not shown).
The video camera 6 mounted on the stellar projector 2 installed at the center of the dome screen 1 is directed to the azimuth angle AZM and the elevation angle ALT by the rotation angle position control of the stellar projector 2. In this case, the camera imaging range is rectangular, and a part of the dome screen 1 can be imaged as shown.
Here, the video projector 3 uses a white straight line of a predetermined vertical length passing through the center of the camera imaging range 31 as a test pattern 32 so that the video projector 3 sufficiently overflows above and below the camera imaging range 31 without performing any color correction. Project.

FIG. 13 shows the state on the camera screen. This is the camera screen when the camera azimuth AZM is set to 60 degrees and the elevation angle ALT is set to approximately 58 degrees. The test pattern 32 to be projected is shaded. If the test pattern 32 has a certain thickness, the vertical line of pixels passing through the center of the screen overlaps the test pattern on the camera screen. For example, if the resolution of the camera is 640 × 480 pixels, X = 320, Y = 0 to X = 320, Y = 479, which is a total of 480 pixels. From now on, it is assumed that the data of the pixels in one vertical column (acquired pixels) will be used in camera shooting.
For example, two camera screen coordinates (320, 0) and (320, 479) are converted into horizon coordinates with reference to the azimuth and elevation angles of the camera. If these two horizon coordinates are converted into coordinates on the projection screen of the video projector and a white line connecting these two points is drawn on the projector projection screen, for example, with a thickness of about 10 to 20 pixels, A state like 13 is obtained. If the distortion is large, more points on the camera screen may be connected instead of two points. In any case, if the orientation of the camera is known, a straight line passing through the center of the camera screen can be drawn by coordinate conversion. This state is always maintained during camera shooting.
A simple way to find out is to display a figure filled with all white on the full screen of the projector instead of drawing like this. However, if an all white screen is displayed, errors due to scattered light may occur. In the graphic display, it is preferable to keep the acquisition range at the bare part.

FIG. 14 illustrates a range in which the brightness of the entire projection screen of the video projector is photographed by changing the direction of the camera.
Shooting is performed while changing the azimuth angle and elevation angle of the camera so as to sufficiently cover the projection screen 33 of the projector indicated by the dotted line. In this figure, the elevation angle of the camera is divided into three stages, and shooting is performed while moving the azimuth angle at a constant speed at each elevation angle. If the azimuth angle at the center of the screen is 0 degree, the azimuth angle is moved from -WAZM1 to WAZM1 in the first scan. Similarly, the second and third scans are also performed.
FIG. 16 shows the state of scan shooting.
If luminance data is sequentially acquired from the acquired pixels of the camera during this movement, the entire surface of the projection screen can be covered and photographed.

FIG. 15 is a block diagram of a system for obtaining a correction map from this projection image. The procedure and data flow will be described with reference to this figure.
In advance, the aperture of the video projector is set to the first state by the numerical aperture setting unit 41, and the projectors other than the screen to be photographed are turned off.
First, the orientation of the camera is set in the imaging position setting unit 42 by the input device 43. For example, the imaging position setting unit 42 sets the altitude and azimuth angle at the center of the camera screen to a position corresponding to the corner of the projection screen, moves the azimuth angle at a constant speed, and repeats this so as to cover the entire screen. Enter the command.
The servo control device 44 of the stellar projector 2 moves the stellar projector 2 based on this command, and changes the direction of the video camera 6 as described above.
The servo control device 44 holds the rotational angle positions of the azimuth axis and the latitude axis of the stellar projector 2 as needed, and can determine the altitude and azimuth angle at the center of the camera screen from these values. The acquisition pixel setting unit 45 is set with an acquisition pixel on the camera screen, and repeatedly outputs, for example, values of (320, 0) to (320, 479) as needed. The camera pixel coordinate-horizon coordinate conversion unit 46 converts the acquired pixel on the camera screen into the horizontal coordinate by referring to the altitude and the azimuth angle at the center of the video camera screen. The horizon coordinates are further converted into coordinates on the projection screen by the projection conversion unit 47.

On the other hand, based on the acquired pixel coordinates, imaging data of the pixel is read from the video camera 6 through the frame grabber 48. The imaging data is corrected by the gamma function of the video camera by the gamma correction unit 49. The corrected imaging data obtained in this way becomes optical profile data. However, if it remains as it is, it is also a value that stores the physical quantity as it is, so it must be converted. For this purpose, the corrected imaging data is stored in the imaging data memory unit 50. The imaging data memory unit 50 is a two-dimensional array having a number of pixels equal to the number of projection pixels of the projector, and stores the corrected imaging data at coordinates on the projection screen on this array.
In such shooting and writing, the same pixel may be designated a plurality of times, but in that case, it is overwritten or an average value with a value already registered in the imaging data memory unit 50 is obtained. You may write it. Conversely, if the resolution of the video camera is rough or the moving speed of the azimuth is too fast, not all the pixels in the imaging data memory will be filled, and there may be parts where data is not stored. It is necessary to write an approximate value from a nearby value by interpolation calculation.

The optical profile data memory unit 51 is also a two-dimensional array having the same number of pixels as the projection screen of the projector. The optical profile data memory unit 51 obtains a peak value from the imaging data memory unit 50, divides by the peak value, obtains an optical profile, and writes this.
With such processing, when the entire surface of the projection screen is scanned by the video camera, the value is stored in the optical profile data memory unit 51.
As another method, imaging is performed by temporarily retracting the shading mask of the video projector, writing an imaging result in a non-shielding imaging data memory unit (not shown), and then returning the shading mask. The optical profile may be obtained by dividing the imaging data as a result of the calculation for each pixel by the value of the non-shielded imaging data memory unit. In this case, even if there is optical projection unevenness of the video projector, only the influence of a pure light shielding mask can be extracted.

With reference to the data stored in the optical profile data memory unit 51, the correction profile calculation unit 52 calculates the target profile data in the target profile data memory unit 53, and the correction profile is calculated. The correction profile is written in the correction map 1 of the correction map unit 54.
With the same procedure, the correction map 2 can be obtained by performing the same process with the numerical aperture set to the second state.
In this embodiment, since the angle of view of the video camera is not so wide, a case has been described in which scanning is performed by changing the direction of the video camera. However, more simply, a video camera having a wide angle of view is used for the video camera. The brightness of the entire projection screen may be obtained in a state where is stopped toward the center of the projection screen of the video projector. Further, even when the projection surface of the video projector does not fit on the video camera screen, the same effect may be obtained by reading and synthesizing all the pixels of the video camera screen with the camera direction stationary at a plurality of values. In this case, shooting is completed in a shorter time. However, in this case, since it is used up to the periphery of the camera screen, it is easy to be affected by the peripheral dimming of the video camera and camera lens. desirable.

FIG. 17 is a block diagram for explaining a flow of obtaining a correction map for the numerical aperture to be actually projected by interpolation calculation using the correction map 1 and the correction map 2 obtained as described above.
For example, in a certain pixel, the correction map 1 data with a numerical aperture F3.2 is 0.62, and the correction map 2 data with a numerical aperture F8 is 0.44.
The data when the numerical aperture is F6 when projecting is
A = (0.44-0.62) / (8-3.2) (3)
B = 0.62−A * 3.2 = 0.74 (4)
Data = A * 6 + B = 0.515 (5)
It becomes. This may be performed for all pixels.
Note that the interpolation calculation is merely an approximation calculation, and does not correctly represent the middle of two states that are largely different. In order to increase the accuracy, it is preferable to obtain a correction map and perform an interpolation calculation when the numerical aperture of the projector is larger. Of course, it is also included in the scope of the present invention to obtain an interpolation calculation by obtaining three or more correction maps with three or more numerical apertures.

Next, an embodiment of displaying an image using a correction map will be described.
FIG. 18 is a block diagram showing an embodiment of a video generation PC when a video is corrected and displayed in real time.
The video correction is performed inside the video generation PC 4 that sends a video signal to the video projector 3. Therefore, the data of the correction map 1 and the correction map 2 acquired in advance by the control PC 5 are transferred to the correction map memory unit 61 inside the video generation PC.
The video drawing unit 62 generates an original image to be projected, and may reproduce a video that has been edited and created in advance, or may draw a vector graphic in real time. The pre-edited video is, for example, a video file in AVI format or WMV format, and is created from the whole sky video such as dome master by performing transformation according to each video projector by converting the horizontal coordinate to the projector screen coordinate in advance. Aside, draw this. When drawing a vector figure, the coordinates of the feature points of the figure are specified as horizon coordinates or converted to horizon coordinates by coordinate conversion from other coordinates (for example, equator coordinates). Drawing is performed by converting the coordinates into coordinates on the projector screen.
In any case, the drawing is not performed on the frame buffer 66 that is output to the video projector suddenly, but on the off-screen buffer 63 that does not appear on the screen.
The numerical aperture setting unit 60 sets the numerical aperture of the video projector at the time of screening. The numerical aperture may be input by an input device, or may be set based on a preset sequence. The numerical aperture set here is sent to the video projector 3 and the aperture of the video projector is activated to set the numerical aperture.

On the other hand, this numerical aperture is also sent to the interpolation calculation unit 64, and based on this value, an interpolation calculation is performed between the correction map 1 and the correction map 2, and a correction map corresponding to the numerical aperture at the time of screening is created. . In this figure, the correction map corresponding to the numerical aperture is not provided in an independent form, but the correction value for each pixel obtained by calculating each pixel for each time from the two correction maps is supplied to the video correction unit 65. Although shown as a configuration for sending, of course, a correction map corresponding to the numerical aperture may be provided independently, and correction map data obtained from the two correction maps may be temporarily stored in the memory unit and used.
The video correction unit 65 corrects the screen drawn in the off-screen buffer 63 by performing a calculation based on the correction data, and writes this to the output frame buffer 66.
The correction calculation of the video correction unit 65 is performed by multiplying the pixel data to be projected by the correction data of each pixel obtained from the correction map. However, simple multiplication is not preferable. This is because the luminance projected by the video projector is generally not proportional to the input signal and has a characteristic having a non-linear curve (gamma characteristic of the video projector). An example of the gamma characteristic of the video projector is shown in FIG. The specific characteristics vary depending on the model and settings, so it is necessary to actually measure the characteristics during use or obtain them from the manufacturer's specifications.

Here, the procedure from the drawing output of the off-screen buffer to the output to the output frame buffer will be described.
1. Read the data of each pixel of the off-screen buffer.
2. In the case of a color image, since there are three values of R, G, and B, the luminances of R, G, and B are obtained separately.
3. R, G, and B are converted by a gamma function to obtain gamma correction values R1, G1, and B1 that are actually projected.
4). The gamma correction value is multiplied by the correction value for each pixel. The obtained values are R2, G2, and B2.
5. The correction value is inversely transformed by an inverse function of the gamma function to obtain data luminance. Let R3, G3, B3.
6). R3, G3, and B3 are combined and returned to the original color data, and written to the corresponding pixel in the output frame buffer.
When this process is repeated for each frame period and an image is output, the corrected image having the correct luminance is projected onto the dome screen.
There is also a method of correcting the video to be reproduced in advance instead of correcting in real time.

FIG. 20 is a block diagram for creating corrected video data by correcting video data created in advance.
The original video data 71 is assumed to have been subjected to distortion correction in advance as in FIG. Further, the numerical aperture of the video projector is prepared in advance as a numerical aperture sequence by making a sequence according to the reproduction time. The processing contents are as follows.
1. The correction calculation unit 74 extracts an image from the original video data 71 frame by frame.
2. Then, the time corresponding to the frame is determined, and the numerical aperture at that moment is determined from the numerical aperture control sequence 72.
3. Based on the numerical aperture, correction data at the time of projection is generated by the interpolation calculation unit 73 from the correction maps 1 and 2 of the correction map unit 70.
4). The correction calculation unit 74 corrects the original video data 71 based on the correction data.
5. The corrected video data is written out as a frame, and this is set as corrected video data 75.
If the corrected video data 75 created in this way is reproduced as it is and projected from a video projector, a video with the correct luminance is projected.

Next, a second embodiment will be described.
The basic procedure is the same as in the first embodiment, but a method for obtaining a correction map at an arbitrary numerical aperture based on the result of photographing, which is performed only once instead of multiple times for obtaining a correction map. It is.
First, the basic principle will be described.
The optical gradation of the optical profile is generated when the light shielding mask blocks a substantially circular opening in the projection lens of the video projector.

FIG. 21A illustrates the exit pupil and the light shielding plate of the projection lens from the screen side.
The position P of the light shielding plate 91 with respect to the exit pupil 90 of the lens viewed from the screen surface varies depending on the position of the screen surface.
FIG. 22 illustrates the positional relationship between the exit pupil and the light shielding plate viewed from three points on the dome screen when the light projected from the video projector is shielded from the periphery by the light shielding mask. The light shielding plate 91 creates a shadow on the exit pupil 90, thereby reducing the brightness of the dome screen. Assuming that the light flux distribution through the exit pupil 80 is uniform, the area of the exit pupil viewed from a predetermined position on the dome screen surface is proportional to the luminance of the screen surface at that position.
In this figure, the edge of the light shielding plate 91 is shifted by P from the center of the exit pupil having the radius R and covers the lens surface.
The area of the exposed portion of the exit pupil 90 at this time can be calculated by the equation shown in FIG. 21B, where the fully opened state is “1”.
When R is “1” and this function is graphed, FIG. 23 is obtained. This function is a basic gradation function. This inverse function is an inverse basic gradation function. That is, the principle that an optical gradation curve becomes this function is used.

  The width of the optical gradation is determined by the numerical aperture of the projector aperture, as is apparent from the example of FIG. The fact that it is proportional to the size of the aperture of the diaphragm is apparent from FIG. FIG. 24 shows the optical profile near the edge of the light-shielding mask with the peak optical brightness being 100% for three cases with numerical apertures of F3.2, F5.6, and F8. As can be seen from the figure, the curve shape of the gradation hardly changes, and it can be seen that the gradation simply expands and contracts in inverse proportion to the numerical aperture. In other words, if an optical profile at a certain numerical aperture is obtained, it may be considered that another optical profile can be estimated therefrom.

Therefore, an embodiment will be described in which an optical profile at one numerical aperture is obtained by camera photographing, and an optical profile corresponding to the numerical aperture at the time of projection is estimated, and then a correction map is created.
FIG. 25 shows a second embodiment of the circuit configuration of the video system according to the present invention, and is a block diagram of a circuit for calculating a correction map by the optical profile estimation method.
Since the basic circuit configuration is the same as that of the first embodiment, the description thereof is omitted. First, in the state where the aperture is open (F3.2), for example, the same shooting as the previous time is performed to obtain the optical profile of the video projector. Then, the difference is that the optical profile data obtained here is stored, not the correction map calculated from the optical profile. This is referred to as an optical profile at the time of photographing, and the memory unit is referred to as an optical profile data memory unit 80 at the time of photographing.
At the time of image projection, a projection optical profile is calculated from the imaging optical profile based on the numerical aperture set by the numerical aperture setting unit 41, and stored in the projection optical profile data memory unit 81. This calculation method will be described later.
Subsequent processing is the same as the previous process, the correction profile calculation unit 82 divides the target profile by this projection optical profile to obtain a correction map, and stores it in the correction map unit 83. The video projection method using this correction map is the same as in the previous embodiment. Since the optical profile to be calculated is a relative optical profile having a peak at “1”, the target luminance for target profile calculation is set to a value less than “1”, for example, about 0.8 to 0.9. To do.

Next, a calculation method for obtaining an optical profile at the time of projection using a basic gradation function from the optical profile at the time of shooting will be described.
FIG. 26 illustrates the principle of obtaining the projection optical profile from the relative optical profile. It is assumed that one data value of the relative optical profile is Y1. Here, the X coordinate S on the gradation of the Y1 value can be calculated by putting Y1 into the inverse basic gradation function. If the numerical aperture at the time of shooting is K1, and the numerical aperture at the time of projection is K2, the numerical aperture ratio is K2 / K1. Therefore, Y2 can be calculated by multiplying the S value by the numerical aperture ratio and converting this by the basic gradation function.

Next, an example of calculating Y2 when the numerical aperture at the time of photographing is F3.2, the numerical aperture at the time of projection is F7, and the data Y1 of the pixel having the relative optical profile is 0.61 is shown below.
FIG. 27 is a diagram showing the value of the basic gradation function with X in increments of 0.01.
Row numbers and column numbers are added to help explain the calculation examples. The numerical value is read from this figure and the calculation proceeds.
Although some errors are included for convenience of explanation, approximate values are read from this figure and adopted as they are. However, in practice, interpolation accuracy can be improved by a computer.
The calculated value S by the inverse basic gradation function of 0.61 is X = 0.17, Y = 0.608 in the 38th row of the third column when Y is the closest value to 0.61 in this figure.
Since the numerical aperture ratio is 7 / 3.2 = 2.189, S ′ is 0.17 * 2.189 = 0.372. This is input to the gradation function. That is, when a value closest to the value of X = 0.372 in this figure is searched, the 18th row of the fourth column, X = 0.37, Y = 0.73 can be obtained.

The value of Y becomes the data of the optical profile during projection. This calculation may be performed for all pixels.
In actual calculation, the contents of this figure are read into a computer and the above calculation is performed by interpolation calculation.
In this embodiment, for convenience of explanation, an example is shown in which an optical profile is obtained in a state where the numerical aperture is one, and optical profiles at other numerical apertures are obtained by numerical calculation. However, in practice, a plurality of optical profiles are obtained. In combination with the first embodiment to be used, the accuracy can be further improved.

  In the first and second embodiments described above, an example in which a video camera is used as a projection image acquisition unit has been described. However, in actuality, a digital still camera is used to acquire images continuously or multiple times. A photographed image may be acquired to obtain an optical profile by the same calculation as described above. For example, a fisheye lens can be attached to the camera so that the entire dome can be imaged, and an optical profile can be obtained by capturing images over the entire dome screen. Alternatively, the camera can be controlled in multiple directions. An optical profile may be obtained.

  A system installed in a planetarium facility using a multi-projection system that projects images onto a dome screen using multiple projectors, especially when used in conjunction with a stellar projector, a dark and high-contrast image that harmonizes a bright image with the starry sky. It is a system that can project.

It is a perspective view for demonstrating the external appearance of the image | video projection system by this invention. A display example of the horizontal coordinate of the dome screen is shown. It is a perspective view of the structure of the stellar projector installed in the center of the dome dome. It is a perspective view which shows embodiment of the projector used for this invention. It is a figure which shows an example of the layout which projects an image | video on the dome screen whole surface using a some projector. It is the figure which added the duplication area | region of the projection screen of each projector on the dome master, and showed. It is a figure for demonstrating the control method of the brightness | luminance of an overlap area | region. It is a figure which shows the dome master type | mold blend mask of the S1 surface in the example of FIG. 4 with a dome master type | mold. FIG. 5 is a diagram illustrating an example of a projection surface of a dome master type blend mask on surfaces S1 to S6 in the example of FIG. 4 using a dome master type blend mask. It is a figure explaining the luminance distribution of the image | video projected from the projector which attached the light shielding mask. It is a figure for demonstrating the change of the optical profile when changing the numerical aperture of the aperture_diaphragm | restriction of a projector. It is a flowchart which shows the procedure which obtains a correction map. It is a figure which shows an example which imaged the target profile of S1 surface. It is a figure which shows the state of camera imaging | photography. It is the figure which displayed the test pattern on the camera screen. It is a figure for demonstrating the range which changes the direction of a camera and image | photographs the brightness of the projection screen whole surface of a video projector. 1 shows a first embodiment of a circuit configuration of a video projection system according to the present invention, and is a block diagram of a system for obtaining a correction map from a projection image of an optical profile. It is a figure for demonstrating the mode of scan imaging | photography. It is a block diagram for demonstrating the flow which obtains the correction map in the numerical aperture actually projected using the correction map 1 and the correction map 2 by interpolation calculation. It is a block diagram which shows embodiment of video production | generation PC in the case of carrying out correction | amendment to a video in real time and showing it. It is a figure which shows an example of the gamma characteristic of a video projector. It is a block diagram for making correction | amendment to the video data produced beforehand and making corrected video data. It is the figure which showed the exit pupil and light-shielding plate of the projection lens from the screen side. It is a figure which shows the positional relationship of the exit pupil seen from three points on a screen, and the light-shielding plate when the light projected from the video projector is light-shielded by the light-shielding mask. It is the figure which made the relationship with the area of the exposed part of an exit pupil a graph. It is a figure which shows the example of three cases whose numerical aperture is F3.2, F5.6, and F8 about the optical profile near the edge of a light shielding mask. FIG. 5 is a block diagram of a circuit that performs a correction map calculation by an optical profile estimation method according to a second embodiment of a circuit configuration of a video system according to the present invention. It is a figure for demonstrating the principle which calculates | requires the optical profile at the time of projection from a relative optical profile. It is the figure which showed the value of the basic gradation function in X increments of 0.01.

Explanation of symbols

1 Dome screen
2 Stellar projector (optical stellar projector)
3 Projector (video projector)
4 Video generation PC
5 Control PC
6 video camera 10 pedestal 11,24 base 12 mount fork 13 latitudinal axis 14 diurnal base
15 North Star Stellar Sphere
16 Stellar sphere 21 Image input cable 22 Projection lens 23 Control signal cable 24 Base 25 Glass substrate 27 Shading mask 28 Projection window 29 Aperture 30 Projector projection range 31 Camera imaging range 32 Test pattern 33 Projector screen 41 Numerical aperture setting section
42 Image pickup position setting unit 43 Input device 44 Servo control device 45 Acquisition pixel setting unit 46 Camera pixel coordinate-horizontal coordinate conversion unit
47 Projection conversion unit 48 Frame grabber 49 Gamma correction unit 50 Imaging data memory unit 51 Optical profile data memory unit 52 Correction profile calculation unit 53 Target profile data memory unit 54, 61, 83 Correction map (memory) unit 62 Video drawing unit
63 Off-screen buffer 64 Interpolation calculation unit 65 Video correction unit 66 Output frame buffer 80 Optical profile data memory unit during imaging 81 Optical profile data memory unit during projection 82 Correction profile calculation unit

Claims (9)

A video projection system that projects video output from video generation means of a computer or video apparatus onto a screen and forms a single video by superimposing screens projected by a plurality of projectors with variable numerical apertures on the screen Because
Projection image acquisition means capable of setting the numerical aperture of the projector to a predetermined value, projecting an image on the screen, photographing the projection image with a camera to obtain a correction map, and acquiring projection image data projected on the screen; ,
Image correction data calculation means for creating a projection correction map by calculation using the correction map in accordance with the numerical aperture at the time of projection of an actual projector, and calculating image correction data at the numerical aperture at the time of projection by the projection correction map And
Video projection system, characterized that you correct the brightness of the image by using the image correction data. A video projection system that projects video output from video generation means of a computer or video apparatus onto a screen and forms a single video by superimposing screens projected by a plurality of projectors with variable numerical apertures on the screen Because
Each projector has a camera capable of shooting the image projected on the screen, sets the numerical aperture for each projector to a predetermined value, shoots the test pattern projected on the screen, and shoots Photographic image acquisition means for obtaining imaging data of a projected image;
When projecting an image by setting the numerical aperture of the aperture of each projector to a value different from that at the time of shooting the test pattern, based on the imaging data, the numerical aperture of each projector at the time of shooting, and the numerical aperture at the time of projection. Video correction data calculating means for obtaining correction data for correcting the luminance of the video by performing a predetermined calculation;
Video projection system, characterized by comprising a video correction projection means for projecting performing correction of the luminance of the image by the image correction data. An image projection system for projecting an image output from image generation means of a computer or a video device on a screen and superimposing screens projected by a plurality of projectors on the screen to form one image,
Each of the plurality of projectors has a camera that can shoot an image projected on the screen, and has an aperture that can be opened and closed by an external command,
Photographing test patterns projected on the screen by setting the numerical aperture of the aperture to a predetermined value for each projector, photographed image acquisition means for obtaining imaging data of the projected image of each projector,
When projecting an image by setting the numerical aperture of the diaphragm of each projector to a value different from that at the time of shooting the test pattern, the optical profile obtained from the imaging data of the captured image acquisition means and the light amount to be projected by each projector Video correction data calculation means for obtaining correction data for correcting the luminance of each position of the video by a predetermined calculation based on a preset target profile or target blend mask, which is indicated according to the position on the screen;
Video projection system, characterized by comprising a video correction projection means for projecting the corrected data by performing the correction of the luminance of the image.   The video projection system according to claim 1, wherein the screen is a dome-shaped screen having a curved surface. The captured image acquisition means includes:
Obtaining a plurality of imaging data at a plurality of values with a known numerical aperture of the projector;
The video projection system according to claim 2, 3 or 4, wherein the data used by the video correction data calculation means is the plurality of imaging data and the known numerical aperture.   6. The video projection system according to claim 2, further comprising a light shielding mask for shielding light around the screen on a front surface of the projector. The predetermined calculation is:
7. The image projection system according to claim 6, wherein a gradation function is used which functions as a function of a relationship between a position and a transmitted light amount when the aperture of the aperture is shielded by an edge of the light shielding mask in a certain positional relationship.   8. The video projection system according to claim 7, wherein the optical gradation area formed by the light shielding mask and the overlapping area of the projection surface overlap at least partially. The camera
Mounted in star projector, video projection system according to any one of claims 2 to 8, characterized in that it has means for the orientation of the camera to set the angular position control of the rotational shaft of the star projector.