JP2007096622A - Correction data creation method, correction data creation apparatus and image projection system - Google Patents

Abstract

A technique for realizing a seamless image in an area with a low luminance level is provided.
The projected images projected from the image projecting devices 104A to 104D are divided into an effective image area for displaying image data and an invalid image area for not displaying image data, and a boundary line between these effective image areas. The area surrounded by the border of the invalid image area is set as a correction unit image area, and correction data to be applied to a low luminance level area is created for each correction unit image area, and the brightness level calculated separately is relatively high. Combine with the correction data to apply to the region.
[Selection] Figure 1

Description

  The present invention relates to a technique for creating correction data for correcting a pixel value of an image in an image projection system that projects a desired image by connecting a plurality of images on a screen.

  In an image display device that projects images on a screen using a plurality of image projection devices, the brightness of the images projected from the respective image projection devices, It is necessary to adjust the hue.

  As a technique for performing such adjustment, Patent Document 1 arranges images projected from a plurality of image projection apparatuses so that adjacent images have overlapping portions, and corrects the overlapping portions for each pixel. A technique for performing a blending process to realize a seamless image is disclosed.

WO99 / 31877 pamphlet

  However, in the conventional technology, in the range where the luminance level is high, the adjustment range of the color and luminance is large, so that it is possible to add subtle adjustments for each pixel, and it is easy to realize a seamless image, while the luminance level is low. In the range, since the adjustment range of the color and brightness is small, it is not possible to add subtle adjustments for each pixel, and there is a drawback that correction unevenness between pixels appears remarkably by adjusting for each pixel.

  Therefore, an object of the present invention is to provide a technique for realizing a seamless image even in a low luminance level range.

  In order to solve the above problems, in the present invention, correction data is created by a different method in a range where the luminance level is low and other ranges.

  For example, in the present invention, the image data output from the image reproduction unit can be projected from a plurality of image projection units, and a plurality of images projected from the plurality of image projection units can be connected to project a desired image. In the image projection system according to the present invention, there is provided a correction data creation method for creating correction data for correcting pixel values of the image data based on photographing data obtained by photographing an image projected from the image projection unit. Correction data for pixel values below one luminance level is calculated for each correction unit image area, correction data for pixel values above a second luminance level higher than the first luminance level is calculated for each pixel, Combine the correction data.

  As described above, according to the present invention, correction data is created in units of predetermined planes in a range with a low luminance level, so that correction unevenness in a range with a low luminance level can be reduced.

  FIG. 1 is a schematic configuration diagram of an image projection system 100 according to an embodiment of the present invention.

  The image projection system 100 includes an image reproduction device 101A to 101D, an image switching device 102, an image correction device 103, an image projection device 104A to 104D, a light shielding member 105A to 105D, a screen 106, an imaging device 107, And an arithmetic processing unit 108.

  The image reproducing devices 101A to 101D read image data stored in an external storage device (not shown) and output the image data to an image switching device 102 described later.

  The image data here may be either a still image or a moving image, and examples of the external storage device include a hard disk, a CD-ROM (Compact Disk Read Only Memory), and a DVD-ROM (Digital Versatile Disk Read Only Memory). A magnetic disk or the like can be used.

  In the present embodiment, 8-bit image data is handled, and at least (R, G, B) = (0, 0, 0) black image data and (R, G, B) = (255, 255, 255) and the so-called checkered coordinate calculation image data shown in FIG. 6 are stored in the external storage devices of the respective image reproducing devices 101A to 101D.

  The image switching device 102 is a switching device that controls whether or not the image data sent from the image reproducing devices 101A to 101D is output to the image projection devices 104A to 104D via the image correction device 103 described later.

  Here, in the present embodiment, the output of the image data sent from the image reproduction device 101A is turned on or off by the image switching device 102, thereby turning on or off the projection of the image from the image projection device 104A. Have been able to. Similarly, by turning on or off the output of the image data sent from the image reproduction device 101B, it is possible to turn on or off the projection of the image from the image projection device 104B. By turning on or off the output of the received image data, the projection of the image from the image projection device 104C can be turned on or off, and further, the output of the image data sent from the image reproduction device 101D can be turned on. Alternatively, the image projection from the image projection device 104D can be turned on or off by turning it off.

  Note that the control of the image switching device 102 is performed by an arithmetic processing device 108 described later.

  The image correction device 103 performs pixel correction such as geometric correction such as position and shape correction and color and luminance correction on the image data sent from the image reproduction devices 101A to 101D via the image switching device 102. Make corrections.

  In the present embodiment, with respect to geometric correction, the images projected from the image projection apparatuses 104A to 104D are geometrically connected on the screen 106 by performing correction by a known method, and the pixels With respect to value correction, settings are made such that the image data input from the image reproducing apparatuses 101A to 101D is passed through without any correction or input / output is equivalent. These controls can be performed by the arithmetic processing unit 108.

  The image projection devices 104A to 104D project the image data output from the image correction device 103 from the respective projection units 104a onto the screen 106. Specifically, a liquid crystal projector, a DLP (registered trademark) projector, or the like is used. Can be used.

  Regarding the adjacent image projection devices 104A to 104D, at least a part of the adjacent projection areas on the screen 106 is overlapped with each other, and the images projected by all the image projection devices 104A to 104D are connected. Thus, a desired image can be projected on the screen 106.

  The light shielding members 105A to 105D are provided around the projection unit 104a of each of the image projection apparatuses 104A to 104D and on the optical axis from the projection unit 104a toward the screen 106.

  The projection light projected from the image projection devices 104A to 104D is converted into image data output from the image reproduction devices 101A to 101D as shown in FIG. 2 due to structural factors of the image projection devices 104A to 104D. The corresponding effective image area 111 and the invalid image area 112 that is located around the effective image area 111 and is not necessary for the image data output from the image reproducing apparatuses 101A to 101D (not corresponding to the image data). And a first leaked light 113 (see FIG. 2A) projected at a position distant from the effective image area 111 and a second leaked light 114 (see FIG. 2B). It is common.

  Here, the light shielding members 105 </ b> A to 105 </ b> D are configured to prevent the first leaked light 113 and the second leaked light 114 projected at positions away from the effective image area 111 from being projected on the screen 106. It is installed so as to block the light around the image area 111.

  Strictly speaking, the second leaked light 114 cannot be distinguished from the invalid image area 112, but the area shielded by the light shielding members 105A to 105D is separated from the effective image area 111 by a predetermined interval. By doing so, an area that is a projection area that is not necessary for the image data and remains around the effective image area 111 becomes the invalid image area 112.

  In this regard, if the light shielding area by the light shielding members 105 </ b> A to 105 </ b> D is too close to the effective image area 111, the effective image area 111 is optically blended by the light shielding members 105 </ b> A to 105 </ b> D. The peripheral part may become dark. Therefore, in the present embodiment, a light shielding region by the light shielding members 105A to 105D is provided to be separated from the effective image region 111 by a predetermined distance, and the invalid image region 112 between the light shielding region and the effective image region 111 is formed by a method described later. The pixel value is removed by correcting it.

  If the first leakage light 113 and the second leakage light 114 are not projected on the screen 106 due to the structure of the image projection devices 104A to 104D, it is not necessary to provide such light shielding members 105A to 105D. .

  Specifically, as shown in FIG. 1, the light shielding members 105 </ b> A to 105 </ b> D have a rectangular shape having a rectangular through hole for allowing the effective image area 111 and the invalid image area 112 to pass through at the center. It is comprised by the plate-shaped member.

  In addition, about this rectangular through-hole, it is desirable to be able to change the magnitude | size and shape by sliding the plate-shaped member around it mutually.

  In the present embodiment, although not shown, the light shielding members 105A to 105D are fixed to an installation base for fixing the positions of the image projection apparatuses 104A to 104D, but for example, the image projection apparatuses 104A to 104D, etc. The light shielding members 105 </ b> A to 105 </ b> D may be fixed to each other.

  For the screen 106, either front projection or rear projection can be used, and any shape such as a plane, cylinder, hemisphere, etc. can be used, but in this embodiment, A rectangular projection is used for rear projection.

  The imaging device 107 captures an image projected on the screen 106 and transmits the captured image data to the arithmetic processing unit 108 described later.

  Here, the imaging apparatus 107 can be realized by using a digital camera, and can use an apparatus that can transmit image data to the arithmetic processing unit 108 via an information transmission unit such as an IEEE1394 interface or a USB interface. To do.

  Further, the imaging device 107 is installed so that the entire projection area projected on the screen 106 from all the image projection devices 104A to 104D can be photographed.

  As will be described later, the arithmetic processing unit 108 extracts the effective image area and the invalid image area from the shooting data shot by the imaging device 107, and calculates the coordinates for each pixel of the shooting data shot by the imaging device 107. , Processing for dividing the area of the image data captured by the imaging device 107, processing for calculating the average luminance for each divided area, processing for calculating the offset value between the divided area and the reference luminance, and calculation of correction data The image switching device 102 and the image correction device 103 are controlled.

  For example, although not shown, the arithmetic processing unit 108 is realized by a computer system including a CPU (Central Processing Unit), a main memory, an auxiliary storage device, an interface, and a system bus that couples them. be able to.

  Processing when creating correction data in the video projection system 100 according to the present embodiment configured as described above will be described below.

  In the following description, 8-bit image data will be described as an example, but the present invention is not limited to such a mode.

  FIG. 3 is a flowchart showing the flow of processing when creating correction data in the video projection system 100.

  First, a first image photographing process is performed (S201).

  Here, black image data with pixel values (0, 0, 0) is output from the image reproducing devices 101A to 101D, and the image switching device 102 switches the image data to be output to each of the image projection devices 104A to 104D. Each image projection device 104 </ b> A to 104 </ b> D independently projects the black image data on 106, and each black image is captured by the image capturing device 107.

  That is, when shooting the image data projected from the image projection device 104A, the projection light of the other image projection devices 104B, 104C and 104D is prevented from being projected onto the screen 106, and the image data projected from the image projection device 104B is shot. When projecting, the projection light of the other image projection devices 104A, 104C and 104D is prevented from being projected onto the screen 106, and when photographing the image data projected from the image projection device 104C, the other image projection devices 104A, 104B and The projection light of 104D is not projected on the screen 106, and the projection light of the other image projection devices 104A, 104B, and 104C is not projected on the screen 106 when shooting the image data projected from the image projection device 104D. Yes.

  For example, when using the image projection devices 104A to 104D that can open and close the projection unit 104a with a mechanical shutter, the mechanical shutters of the image projection devices 104A to 104D other than those that perform shooting are used. By closing, projection light can be prevented from being projected onto the screen 106. In addition, by turning off the power of the image projection apparatuses 104A to 104D other than the one that performs photographing, or covering the projection unit 104a of the image projection apparatuses 104A to 104D other than the one that performs photographing with a shielding plate. It is also possible to prevent projection light from being projected onto the screen 106.

  Further, when photographing with the photographing apparatus 107, the aperture and exposure time of the photographing apparatus 107 are adjusted so that the effective image area 111 and the invalid image area 112 shown in FIG. 2 are photographed. For example, both the effective image area 111 and the invalid image area 112 can be photographed by releasing the aperture and extending the exposure time.

  An example of shooting data shot in this way is shown in FIG. FIG. 4A shows photographing data obtained by photographing an image projected from the image projecting device 104A by the image photographing device 107. Similarly, FIGS. 4B, 4C, and 4D are as follows. These are the shooting data obtained by shooting the images projected from the image projection devices 104B, 104C, and 104D by the image shooting device 107, respectively.

  Here, in these shooting data, an area including both the effective image areas 111A to 111D and the invalid image areas 112A to 112D is shot as an area having a predetermined brightness (the actual shooting data includes The boundary lines between the effective image areas 111A to 111D and the invalid image areas 112A to 112D are not clearly captured).

  In the present embodiment, the rear projection screen 106 is used, and the portions other than the effective image areas 111A to 111D and the invalid image areas 112A to 112D are not projected, and thus are photographed in black.

  The photographing data photographed in this way is sent from the imaging device 107 to the arithmetic processing device 108 and stored as first image data in the arithmetic processing device 108.

  Next, a second image shooting process is performed (S202).

  Here, white image data with pixel values (255, 255, 255) is output from the image reproduction apparatuses 101A to 101D, and the image data to be output to the image projection apparatuses 104A to 104D is switched by the image switching apparatus 102, and the screens are switched. Each image projection device 104 </ b> A to 104 </ b> D individually captures white image data on the image 106, and the image capturing device 107 captures the image.

  In this regard, during the first image shooting process (S201), the projection light of the image projection apparatuses 104A to 104D that are not the shooting target is not projected on the screen 106. In the second image shooting process (S202) using a high-level image, it is possible to sufficiently eliminate the influence of the image projection devices 104A to 104D that are not the shooting target by setting the threshold value. Therefore, the above-described black image data is projected from the image projecting devices 104A to 104D that are not to be photographed. Note that, similarly to the first image shooting process (S201), the projection light of the image projection apparatuses 104A to 104D that are not the shooting target may not be projected on the screen 106.

  Further, when photographing with the photographing apparatus 107, the effective image areas 111A to 111D shown in FIG. 2 are clearly photographed, and the invalid image areas 112A to 112D are not clearly photographed, that is, a threshold value. With this setting, the aperture and exposure time of the photographing apparatus 107 are adjusted so that the effective image areas 111A to 111D can be cut out. For example, only the effective image areas 111 </ b> A to 111 </ b> D can be clearly photographed by reducing the aperture and shortening the exposure time.

  FIG. 5 shows an example of the shooting data thus shot. FIG. 5A shows photographing data obtained by photographing an image projected from the image projecting device 104A by the image photographing device 107. Similarly, FIGS. 5B, 5C, and 5D are as follows. These are the shooting data obtained by shooting the images projected from the image projection devices 104B, 104C, and 104D by the image shooting device 107, respectively.

  Further, the shooting data shot in this way is sent from the imaging device 107 to the arithmetic processing unit 108 and stored in the arithmetic processing unit 108 as second image data.

  Next, a third image photographing process is performed (S203).

  Here, the image reproducing devices 101A to 101D output coordinate calculation image data for calculating the coordinates at which the pixels of the images projected from the image projecting devices 104A to 104D are displayed on the screen 106. The image switching device 102 switches the coordinate calculation image data to be output to each of the image projection devices 104A to 104D so that each of the image projection devices 104A to 104D independently projects the coordinate calculation image data on the screen 106. Are respectively photographed by the image photographing device 107.

  This coordinate calculation method can be performed by a known method, and is not limited to the method of this embodiment described here.

  In the present embodiment, as shown in FIG. 6, square regions 120 a with a high luminance level such as white and square regions 120 b with a low luminance level such as black are alternately arranged vertically and horizontally. The coordinate calculation image data 120 including the so-called checkered pattern and the reference marking pattern 120c used as the reference coordinate at the time of coordinate calculation is used. However, the present invention is not limited to this.

  Further, such shooting of the coordinate calculation image data 120 may be performed in the same manner as the shooting in the second image shooting process (S202).

  In other words, the image projection devices 104A to 104D that are not subject to photography project the above-described black image and perform photography, and by reducing the aperture and shortening the exposure time, a so-called checkered pattern is obtained. Can be taken clearly.

  FIG. 7 shows an example of shooting data thus shot. FIG. 7A shows shooting data obtained by shooting the image data 120 for coordinate calculation projected from the image projection device 104A with the image shooting device 107. Similarly, FIG. 7B, FIG. 7C, and FIG. (D) is shooting data obtained by shooting the image data 120 for coordinate calculation projected from the image projection devices 104B, 104C, and 104D by the image shooting device 107, respectively.

  Further, the photographing data photographed in this way is sent from the imaging device 107 to the arithmetic processing device 108 and stored in the arithmetic processing device 108 as third image data.

  Next, a fourth image photographing process is performed (S204).

  Here, black image data having pixel values (0, 0, 0) is output from the image reproducing devices 101A to 101D, and the black image data is output to all the image projection devices 104A to 104D by the image switching device 102. Then, the black images projected from all the image projection devices 104 </ b> A to 104 </ b> D on the screen 106 are photographed by the image photographing device 107.

  Also in this case, as in the case of the first image capturing process (S201), the aperture and exposure of the image capturing apparatus 107 so that the effective image area 111 and the invalid image area 112 shown in FIG. Adjust the time. For example, the effective image area 111 and the invalid image area 112 can be photographed by releasing the aperture and extending the exposure time.

  As shown in FIG. 8, in the shooting data shot in this way, the effective image areas 111A to 111A projected from all the image projection apparatuses 104A to 104D in the imaging area 130 imaged by the image pickup apparatus 107. 111D and invalid image areas 112A to 112D partially overlap each other.

  Further, the shooting data shot in this way is sent from the imaging device 107 to the arithmetic processing unit 108 and stored in the arithmetic processing unit 108 as fourth image data.

  Next, extraction processing of the effective image areas 111A to 111D is performed (S205).

  This is because each of the second image data (white image shooting data individually projected from the image projection devices 104A to 104D) obtained in the second image shooting processing (S202) in the arithmetic processing unit 108 is obtained. This is processing for extracting the effective image areas 111A to 111D.

  Specifically, the effective pixel regions 111A to 111D can be distinguished from other regions by performing binarization processing on each of the second image data using a predetermined threshold value.

  Note that the invalid image areas 112A to 112D may be photographed around the effective image areas 111A to 111D as the second image data, but by using an appropriate threshold value when binarization is performed. The effective image areas 111A to 111D can be extracted.

  Further, as preprocessing for performing binarization, it is desirable to remove noise and the like by filtering the second image data.

  The second image data that has been binarized in this way is stored in the arithmetic processing unit 108 as effective image area extraction data.

  Next, an invalid image region extraction process is performed (S206).

  This is because each of the first image data (black image shooting data individually projected from the image projection devices 104A to 104D) obtained in the first image shooting processing (S201) in the arithmetic processing unit 108 is obtained. This is processing for extracting invalid image areas 112A to 112D.

  Specifically, after performing binarization processing using a predetermined threshold that can extract the effective image regions 111A to 111D and the invalid image regions 112A to 112D from each of the first image data, the effective image By subtracting the effective image area extraction data calculated in the area extraction process (S205) from the corresponding pixel position, the invalid image areas 112A to 112D can be distinguished from other areas. In addition, when subtracting, it carries out by what was projected from the same image projection apparatus 104A-104D.

  Further, as preprocessing for performing binarization, it is desirable to remove noise and the like by filtering the first image data.

  In this way, the first image data that has been binarized is stored in the arithmetic processing unit 108 as invalid image area extraction data.

  Next, coordinates for each pixel of the photographing data are calculated (S207).

  This is because each of the third image data obtained by the third image shooting process (S203) in the arithmetic processing unit 108 (shooting data of the coordinate calculation images individually projected from the image projection devices 104A to 104D). From this, the image data representing the pixel position of each pixel of the imaging data of the imaging device 107 with specific coordinates is generated.

  Specifically, the effective image area image data obtained in the effective image area extraction process (S205) is added to the third image data obtained in the third image shooting process (S203), thereby obtaining the third image data. Image data obtained by extracting a portion corresponding to the effective image area from the image data is generated.

  Then, edge detection processing is performed on the image data obtained by extracting the portion corresponding to the effective image area from the third image data, and the vertical and horizontal grid lines 120d and 120e are detected as shown in FIG. The coordinates of the intersection of the grid lines 120d and 120e are obtained.

  Next, the coordinates of the reference marking pattern 120c are obtained from the image data obtained by extracting the portion corresponding to the effective image area from the third image data, and set as the reference coordinates.

  It is determined how many grids the intersection point of the previously obtained grid lines 120d and 120e is at a distance from the reference coordinate in the vertical and horizontal directions, and set as a relative coordinate.

  The reference coordinates and relative coordinates thus obtained are used as fourth image data (shooting data obtained by projecting a black image from all the image projection devices 104A to 104D) obtained in the fourth image shooting process (S204). ), The pixel position of each pixel of the shooting data can be specified on the coordinates.

  Note that the pixel positions of the pixels other than the intersections of the grid lines 120d and 120e can be specified by interpolation with the distance from the coordinates of the intersections.

  The pixel position of each pixel calculated in this manner is stored in the arithmetic processing unit 108 as coordinate data.

  Next, a correction unit image area separation process is performed (S208).

  Here, as will be described later, the correction unit image region is an image region that is a unit for generating correction data when generating correction data for a region having a low luminance level in the present invention. Specifically, these are the areas delimited by the boundary lines of the effective image areas 111A to 111D and the boundary lines of the invalid image areas 112A to 112D.

  In the present embodiment, “1” serving as an identifier is input to each of the valid image areas 111A to 111D and the invalid image areas 112A to 112D in the image areas projected by the image projection apparatuses 104A to 104D. Although the division is performed by setting the code, it is not limited to such a mode.

  That is, the valid image areas 111A to 111D extracted by the effective image area extraction process (S205) and the invalid image areas 112A to 112D extracted by the invalid image area extraction process (S206) are identifiers at different bit positions. A value in which “1” is input is set, and a value of “0” is input to the other areas, and all are added to perform separation.

  Specifically, a value of “1” is set in the invalid image area 112A, a value of “10000” is set in the valid image area 111A, and a value of “0” is set in the other areas, and the invalid image area 112B is set. Is set to “10”, the effective image area 111B is set to “100000”, the other areas are set to “0”, and the invalid image area 112C is set to “100”. A value of “1000000” is set in the effective image area 111B, a value of “0” is set in the other areas, a value of “1000” is set in the invalid image area 112D, and “10000000” is set in the effective image area 111B. By setting the value of “0” in other areas and adding all the values, it becomes possible to delimit which area the corresponding pixel belongs to.

  For example, the correction unit areas 141 to 144 shown in FIG. 9 are non-overlapping areas including only invalid image areas, and have values of “1”, “10”, “100”, and “1000”, respectively.

  The correction unit areas 145 to 148 shown in FIG. 10 are double overlapping areas of invalid image areas, and have values of “11”, “101”, “1010”, and “1100”, respectively.

  The correction unit areas 149 to 152 shown in FIG. 11 are non-overlapping areas that are only effective image areas, and have values of “10000”, “100000”, “1000000”, and “10000000”, respectively.

  The correction unit areas 153 to 161 shown in FIG. 12 are double overlapping areas of an effective image area and an invalid image area, which are “10010”, “11000”, “10100”, “1000001”, “1001000”, The values are “10000100”, “10000010”, “101000”, and “100001”.

  The correction unit areas 162 to 168 shown in FIG. 13 are triple overlapping areas of a double invalid image area and a single effective image area, which are “11010”, “11100”, “1001001”, “1001010”, The values are “10000101”, “101100”, and “101001”.

  The correction unit regions 169 to 172 shown in FIG. 14 are quadruple overlapping regions of a triple invalid image region and a single effective image region, and are “11110”, “1001011”, “10000111”, and “101101”, respectively. It is a value.

  The correction unit areas 173 to 176 shown in FIG. 15 are double overlapping areas of the effective image areas, and have values of “110000”, “1010000”, “11000000”, and “10100000”, respectively.

  The correction unit areas 177 to 180 shown in FIG. 16 are triple overlapping areas of a single invalid image area and a double effective image area, and values of “111000”, “1011000”, “11000010”, and “10100100”, respectively. It has become.

  The correction unit areas 181 to 185 shown in FIG. 17 are quadruple overlapping areas of a double invalid image area and a double valid image area, which are “111100”, “10010110”, “10101010”, and “11000011”, respectively. , “10100101”.

  The correction unit areas 186 to 188 shown in FIG. 18 are quadruple overlapping areas of a single invalid image area and a triple effective image area, and have values of “10110100”, “11010010”, and “11100001”, respectively. .

  The correction unit area 189 shown in FIG. 19 is a quadruple overlapping area of only the effective image area, and has a value of “11110000”.

  As described above, the correction unit image areas surrounded by the boundary lines of these areas can be divided by associating the invalid image areas 112A to 112D and the valid image areas 111A to 111D with values having different numbers of bits. it can.

  Next, the average luminance of the black image for each correction unit image area is calculated (S209).

  This is because the fourth image data (black image shooting data projected from all the image projection devices 104A to 104D) shot in the fourth image shooting process (S204) and the correction unit image area separation process ( This is a process of calculating the average luminance for each correction unit image area by associating the correction unit image areas divided in S208).

  The average luminance is obtained by calculating an average value of pixel values of pixels located in the correction unit image area.

  As a pre-process for calculating such average luminance, it is desirable to filter the fourth image data to remove noise and the like.

  Next, an offset value in the black image for each correction unit image area is calculated (S210).

  This compares the average luminance for each correction unit image area calculated in the average luminance calculation process (S209) of the black image for each correction unit image area, and aligns the average luminance of the black image for all correction unit image areas. In this process, a target value is determined, and the difference between the target value and the average luminance for each correction unit image area is used as an offset value for each correction unit area.

  Here, as the target value, it is general to use the highest brightness among the average brightness for each correction unit image area.

  In addition, when the highest brightness among the average brightness for each correction unit image area is set as the target value, the brightness of the black image, which is the lowest brightness level, is increased, and the contrast of the entire image is reduced. Therefore, it is possible to set an upper limit for this target value. When such a condition is provided, there may occur a region where the average luminance for each correction unit image region becomes higher than the target value. At this time, the offset value of that region is treated as “0”.

  Further, since the correction unit image areas 141 to 148 including only the invalid image areas 112A to 112D cannot be corrected by the image correction apparatus 103, the offset value calculation in the black image for each correction unit image area is calculated. It excludes from the object of a process (S210).

  Next, correction data generation processing is performed (S211).

  This is a process of generating correction data by combining the offset value calculated by the offset value calculation process (S210) in the black image for each correction unit image area with correction data at a luminance level other than the black image calculated separately. It is.

  For example, as shown in FIG. 20, the correction data of the black image in the specific pixel specified based on the offset value calculated by the offset value calculation process (S210) in the black image for each correction unit image area is obtained. The correction data at the luminance level other than the black image in this pixel calculated separately shown in the graph 191 is shown in the graph 192. As a process for combining correction data, there is a method using a spline curve or a beige curve. In this embodiment, the offset value when the luminance level of the input image is “0” is calculated at a luminance level other than the black image. The correction data is obtained by interpolating between the correction values of the minimum brightness level (input value “32” in the graph 192) of the valid correction data (those that do not show much correction unevenness). To join. The correction data calculated in this way is shown in a graph 193.

  In this way, the processing for creating correction data in the video projection system 100 is completed.

  Note that the offset value calculated in this way can be changed as appropriate by performing feedback adjustment for each correction unit image region. In such a feedback adjustment process, only the offset value calculated in the offset value calculation process (S210) in the black image for each correction unit image area is sent as correction data from the arithmetic processing unit 108 to the image correction apparatus 103, and the image correction apparatus The process of registering in 103 and calculating a new offset value is repeated until a predetermined evaluation value or number of times is reached.

  The correction data calculated as described above is calculated for each pixel and registered in the image correction apparatus 103, so that the image data output from the image reproduction apparatuses 101A to 101D and selected by the image switching apparatus 103 is The image is corrected via the image correction device 103, projected from the image projection devices 104A to 104D onto the screen 106, and displayed as a seamless image as shown in FIG.

  In the embodiment described above, four image projection devices 104A to 104D and image reproduction devices 101A to 101D are used. However, the number is not limited to this number, and the number of images to be displayed on the screen is divided. These may be provided depending on the situation.

  Further, the numbers of the image projection apparatuses 104A to 104D and the image reproduction apparatuses 101A to 101D do not necessarily match. For example, an image division apparatus that divides image data read into one image reproduction apparatus into predetermined frames. May be used to output image data to the four image projection devices 104A to 104D.

  In the present embodiment, the image correction apparatus 103 and the arithmetic processing apparatus 108 are configured as separate apparatuses. However, the present invention is not limited to such an embodiment. For example, the image correction apparatus 103 and the arithmetic processing apparatus 108 are configured by the arithmetic processing apparatus 108 and a computer system. The image correction apparatus 103 may be configured by an expansion board that can be connected to the computer system, such as a PCI board, and these apparatuses may be provided in the same casing.

  Further, in the present embodiment, in the first image shooting process (S201) and the fourth image shooting process (S204), the black image data output from the image reproducing devices 101A to 101D has a pixel value of (0). , 0, 0) is output, but the present invention is not limited to this mode, and the pixel value of the black image data in this embodiment is the lower 10% of 0-255 (rounded up). Any image data may be used within the range, specifically, within the range of (0, 0, 0) to (25, 25, 25).

  In this embodiment, correction data is created using one black image data with a pixel value of (0, 0, 0). However, correction data is obtained from image data with a plurality of pixel values within a low luminance level range. It is also possible to create correction data in a low luminance level range.

  Furthermore, as a method of combining correction data in a relatively high luminance range and correction data in a low luminance level range, correction data in a low luminance level range is created with at least one luminance, and the correction data and the luminance level are created. It is also possible to interpolate between correction data in a relatively high range, and correction data in a low luminance level range is created with at least one luminance, and has already been calculated for a low luminance level range. It may be replaced with correction data.

  Note that the minimum value of the correction data calculated for each pixel when combining the correction data calculated for each correction unit image area with the correction data calculated for each pixel is the luminance level “32” in the present embodiment. However, the present invention is not limited to such an embodiment, and any pixel value can be adopted in a range where the luminance level is higher than the pixel value when calculating correction data for each correction unit image area. In this regard, normally, for the range of the lower 20% or more of the pixel value, good correction data can be obtained by calculating the correction data in pixel units. Accordingly, by performing interpolation or replacement with correction data created in units of planes (correction unit image areas) for a range lower than a specific pixel value of the lower 20% or less of the pixel value, good correction data as a whole is obtained. be able to.

  In the present embodiment, the image projection system 100 is configured. However, the present invention is not limited to such an aspect. For example, the correction data creation device can be configured by the imaging device 107 and the arithmetic processing device 108. is there.

1 is a schematic configuration diagram of an image projection system 100. FIG. Explanatory drawing of the projection image projected from each image projection apparatus 104A-104D. 5 is a flowchart showing a flow of processing when creating correction data in the video projection system. FIG. 3 is a schematic diagram of shooting data obtained by shooting a black image projected from each of the image projection devices 104A to 104D by an image shooting device 107; FIG. 4 is a schematic diagram of shooting data obtained by shooting a white image projected from each of the image projection devices 104A to 104D with the image shooting device 107; Schematic of the image data 120 for coordinate calculation. FIG. 3 is a schematic diagram of shooting data obtained by shooting an image for coordinate calculation projected from each of the image projection devices 104A to 104D with an image shooting device 107; FIG. 3 is a schematic diagram of shooting data obtained by shooting a black image projected from all the image projection apparatuses 104A to 104D by the image shooting apparatus 107; Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining how to divide the correction unit image area. Schematic explaining the correction data combining process. Schematic showing the example of a display after correct | amending in the video projection system 100. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Video projection system 101 Image reproduction apparatus 102 Image switching apparatus 103 Image correction apparatus 104 Image projection apparatus 105 Light-shielding member 106 Screen 107 Imaging apparatus 108 Arithmetic processing apparatus

Claims (14)

In an image projection system in which image data output from an image reproduction unit is projected from a plurality of image projection units, and a plurality of images projected from the plurality of image projection units are connected to project a desired image A correction data creation method for creating correction data for correcting pixel values of the image data based on shooting data obtained by shooting an image projected from the image projection unit,
Correction data for pixel values equal to or lower than the first luminance level is calculated for each correction unit image area, correction data for pixel values equal to or higher than the second luminance level higher than the first luminance level is calculated for each pixel, Combining each correction data,
The correction data creation method characterized by this. The correction data creation method according to claim 1,
The pixel value below the first luminance level is within the lower 10% of the pixel value;
The correction data creation method characterized by this. The correction data creation method according to claim 2,
The correction unit image region is a region divided by a boundary line of an effective image region and a boundary line of an invalid image region among projection regions from the individual image projection units,
The correction data creation method characterized by this. The correction data creation method according to claim 3,
Correction data of pixel values equal to or lower than the first luminance level is to correct the pixel values so that the luminance calculated for each correction unit image region is uniform;
The correction data creation method characterized by this. An image playback unit for outputting image data;
An image correction unit that corrects pixel values of the image data based on correction data;
An image projection unit that projects the corrected image data output from the image correction unit; and a correction data creation device that creates the correction data used in an image projection system,
An imaging device that captures an image projected from the image projection unit;
An arithmetic processing unit that creates the correction data based on the imaging data imaged by the imaging means,
The arithmetic processing unit calculates correction data of pixel values equal to or lower than the first luminance level for each correction unit image area, and correction data of pixel values equal to or higher than the second luminance level higher than the first luminance level is calculated. Calculate for each pixel and calculate the correction data by combining the correction data.
The correction data creation device characterized by this. The correction data creation device according to claim 5,
The pixel value below the first luminance level is within the lower 10% of the pixel value;
The correction data creation device characterized by this. The correction data creation device according to claim 6,
The correction unit image region is a region divided by a boundary line of an effective image region and a boundary line of an invalid image region among projection regions from the individual image projection units,
The correction data creation device characterized by this. The correction data creation device according to claim 7,
The correction data for the pixel value below the first luminance level is to correct the pixel value so that the luminance calculated for each correction unit image region is uniform.
The correction data creation device characterized by this. An image playback unit for outputting image data;
An image correction unit that corrects pixel values of the image data based on correction data;
An image projection unit that projects the corrected image data output from the image correction unit;
An imaging unit that captures an image projected from the image projection unit;
An arithmetic processing unit that creates the correction data based on the imaging data captured by the imaging unit;
The arithmetic processing unit calculates correction data of pixel values equal to or lower than the first luminance level for each correction unit image region, and correction data of pixel values equal to or higher than the second luminance level higher than the first luminance level is calculated. It is calculated for each pixel, and the correction data is calculated by combining the correction data.
An image projection system characterized by The image projection system according to claim 9,
The pixel value below the first luminance level is within the lower 10% of the pixel value;
An image projection system characterized by The arithmetic processing device according to claim 10,
The correction unit image region is a region divided by a boundary line of an effective image region and a boundary line of an invalid image region among projection regions from the individual image projection units,
An image projection system characterized by The image projection system according to claim 11,
The correction data for the pixel value below the first luminance level is to correct the pixel value so that the luminance calculated for each correction unit image region is uniform.
An image projection system characterized by The image projection system according to any one of claims 9 to 12,
Further comprising a shielding member that shields projection light to an area other than the effective image area;
An image projection system characterized by The image projection system according to claim 13,
The shielding member is formed so as to surround the effective image area;
An image projection system characterized by

Images