JP2005099150A - Image correction data calculation method of image display apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image correction data calculation method of an image display apparatus in which an image is observed in an optimum quality according to an observation position. <P>SOLUTION: A test pattern is projected on a screen of the image display apparatus and a test pattern photographing data is acquired by photographing the projected test pattern by a photographing means from a predetermined photographing position and the test pattern photographing data is transformed to a characteristics data obtained when the test pattern is photographed from a predetermined observation position different from the predetermined photographing position based on a transformation function calculated beforehand and the image correction data for correcting display characteristics of the image display apparatus based on the characteristics data is calculated. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image correction data calculation method for an image display device.

  As conventional image display devices, for example, those shown in FIGS. 38, 39 and 40 are known.

  The image display device shown in FIG. 38 is called a multi-screen display in which a plurality of (in this case, four) image display devices 1 are combined up and down to form one screen. The apparatus 1 includes a projector 3 disposed in a rectangular parallelepiped housing 2 and a planar screen 4 disposed on the front surface of the housing 2 so as to display a projection image thereof. In FIG. 38, one image display device 1 is shown separately from the other three image display devices 1.

  The image display apparatus shown in FIG. 39 combines a plurality of projectors 3 (up to nine in this case), three each in the vertical direction, and partially exceeds the projected image from the adjacent projector 3 on the flat common screen 4. This is called a multi-seamless display that seamlessly wraps to form a single screen. A plurality of projectors 3 are arranged in a rectangular parallelepiped casing 2 and the entire surface of the casing 2 is displayed. A screen 4 is arranged on the screen.

  The image display apparatus shown in FIG. 40 combines a plurality of projectors 3 (here, 10 projectors) up and down to display projected images from the projectors 3 on a common arch-shaped screen 4. This is called an arch screen display that constitutes one screen. In addition, various types are known such as a so-called dome screen display that displays a projection image from a plurality of projectors on a common dome-shaped screen to form a single screen.

  This type of image display device is widely used in, for example, event venues and showrooms because the depth can be shortened and the brightness of the display image can be increased as compared with a display that forms a large screen with a single projector.

  By the way, in the image display device using a plurality of projectors as described above, in particular, luminance unevenness and color unevenness (in-plane unevenness) in the display screen by each projector, and luminance unevenness between display screens by different projectors and If there is uneven color (unevenness between surfaces), the image quality deteriorates, which causes a problem in observing the image.

  The main factors of this variation include variations in color filters within each projector, variations in lamps that are projection light sources, changes over time, etc. The change with time etc. are mentioned.

  Conventionally, as a method of correcting the in-plane unevenness and the inter-surface unevenness, a data converter for converting digitally converted video signal data into video signal data corrected for brightness unevenness and a data converter corresponding to each projector And an image correction device for controlling the conversion processing content in the projector. First, each projector projects and displays a video of a predetermined luminance level on the screen, and the screen is photographed by a television camera to detect the luminance distribution. After correcting the luminance unevenness in the plane based on the luminance distribution, an operation for correcting the luminance unevenness between the planes so as to eliminate the luminance difference between the projectors at the luminance level is performed by a plurality of luminance levels (for example, 100%). , 75%, 50%, 25%), and it is known that brightness unevenness and color unevenness are corrected at all brightness levels. For example, see Patent Document 1).

  As another correction method, first, R, G, B colors are projected and displayed on a screen by a plurality of projectors, and the light amount data of each color is detected by photographing the screen with a video camera. By correcting the in-plane unevenness for each projector based on the light intensity data, and then adjusting the white balance between the projectors by controlling the drive voltage of each projector, and finally adjusting the γ characteristics, An apparatus that corrects unevenness is also known (see, for example, Patent Document 2).

Japanese Patent No. 3287007 JP 7-333760 A

  However, as disclosed in Patent Documents 1 and 2 above, when the image displayed on the screen is captured by the camera, the diffused light on the screen varies depending on the angle, and the light receiving sensitivity of the camera is also an optical characteristic of the photographing lens. Therefore, as the shooting angle becomes wider, the sensitivity differs between the central portion and the peripheral portion. In addition, the brightness difference between the screen central portion and the peripheral portion of the image displayed on the screen increases as the projection angle by the projector increases.

  For this reason, since the image viewed from the camera is observed as different brightness at the center and the periphery of the screen, if correction processing is performed in this state, a good image is obtained when observing from the camera position. Although it can be observed, it may be observed as an unsightly image when observed at other positions. Hereinafter, this point will be described in more detail.

  For example, as shown in FIG. 41A, when an image projected on the screen 4 by the projector 3 is taken by the camera 11, the light rays from the projector 3 spread radially, so Incident angle increases. Here, among the light rays diffused by the screen 4, the light beam directed to the camera 11 is denoted by a, the light beam directed in the normal direction of the screen 4 is denoted by b, and the light beam in the direction in which the light goes straight at the same emission angle as the incident angle. When the code c is used and the screen 4 has a characteristic in which the intensities of the rays a, b, and c are in the relationship of a <b <c, the luminance distribution of the image photographed by the camera 11 is shown in FIG. As a result, the luminance decreases and becomes darker toward the periphery.

  Further, as shown in FIG. 42A, even when the screen 4 having the characteristic that the intensity of the light beam b is highest is used, the luminance distribution of the image taken by the camera 11 is as shown in FIG. As shown in the figure, the brightness decreases as the peripheral portion becomes darker.

  Further, as shown in FIG. 43 (a), even if the screen 4 has a characteristic in which all the intensities of the light rays a directed to the camera 11 are equal, the peripheral sensitivity of the camera 11 depends on the optical characteristics of the taking lens. If it has a characteristic of lowering, the luminance distribution of the image taken by the camera 11 is as shown in FIG. 43B, and the luminance is also lowered and darker in the peripheral part.

  For this reason, if the image correction is performed so that the luminance, color, γ characteristics, and the like are uniform in the screen based on the image data captured by the camera 11, even if the image viewed through the camera 11 is uniformly corrected, the observation is performed. An image viewed by a person is not necessarily optimal. For example, in the case of FIG. 42 (a), if the brightness of the light beam a toward the camera 11 is corrected so as to be constant, after the correction, as shown in FIGS. 44 (a) and (b), as a result, the screen 4 When the image is observed from a position other than the camera, the luminance difference between the central portion and the peripheral portion is generated.

  Therefore, by applying this correction method to a multi-screen display as shown in FIG. 38, for example, as shown in FIG. 45 (a), images displayed on the screens 4 of the two horizontal image display devices 1 are displayed. When each of the images is taken by the camera 11 and the in-plane unevenness and the inter-surface unevenness are corrected, the corrected luminance distribution becomes non-uniform as shown in FIG.

  In addition, when applied to a multi-seamless display as shown in FIG. 39, for example, as shown in FIG. 46A, an image displayed on the screen 4 by the three horizontal projectors 3 is taken by one camera 11. Then, when the in-plane unevenness and the inter-surface unevenness are corrected, the corrected luminance distribution becomes non-uniform because the luminance of the peripheral portion becomes high as shown in FIG.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide an image correction data calculation method for an image display device capable of observing an image having an optimum image quality in accordance with an observation position.

  The invention of the image correction data calculation method for an image display device according to claim 1 that achieves the above object comprises a first step of projecting a test pattern on a screen of the image display device, and the projected test pattern is predetermined by an imaging means. A second step of obtaining test pattern photographing data by photographing from the photographing position, and the test pattern photographing data is photographed from a predetermined observation position different from the photographing position based on a conversion function calculated in advance. And a fourth step of calculating image correction data for correcting display characteristics of the image display device based on the characteristic data. It is.

  The invention according to claim 2 is the image correction data calculation method of the image display device according to claim 1, wherein the conversion function used in the third step is a plurality of measurements in the x and y directions orthogonal to the screen. A point is set and a test pattern is projected. The test pattern is measured from the reference position by the first measuring means to obtain the display characteristics of all the measuring points, and measured from the position different from the reference position by the second measuring means. And calculating the display characteristics of each measurement point based on the display characteristics of all measurement points obtained from the reference position and the display characteristics of each measurement point obtained from a position different from the reference position. To do.

  The invention according to claim 3 is the image correction data calculation method for the image display device according to claim 2, wherein the reference position is the predetermined photographing position or a position equivalent thereto. .

  The invention according to claim 4 is the image correction data calculation method of the image display device according to claim 2 or 3, wherein the position different from the reference position is a position directly facing each measurement point. Is.

  The invention according to claim 5 is the image correction data calculation method of the image display device according to claim 2 or 3, wherein the position different from the reference position is a position separated from the screen by a predetermined distance. The second measuring means is sequentially directed to each measurement point to measure the display characteristics.

  According to a sixth aspect of the present invention, in the image correction data calculation method of the image display device according to the second or third aspect, the position different from the reference position is parallel to an x direction or a y direction spaced apart from the screen by a predetermined distance. It is a linear position, and the display characteristic of each measurement point is measured by rotating the second measuring means along the linear position and rotating it in the y direction or the x direction.

  The invention according to claim 7 is the image correction data calculation method of the image display device according to any one of claims 2 to 6, wherein the second step uses a camera having an image sensor as the photographing means. As the first measuring means and the second measuring means, the camera or a camera having an image sensor having substantially the same light receiving characteristics as the first measuring means is used, and when the display characteristics of each measurement point are measured by the camera as the second measuring means. In addition, each measurement point is sequentially photographed at a substantially central portion of the image sensor.

  According to an eighth aspect of the present invention, in the image correction data calculation method of the image display device according to the fourth aspect, in the second step, a camera having an image sensor is used as the photographing means, and the first measuring means is the above-described one. Using a camera or a camera having an image sensor with almost the same light receiving characteristics as the second measuring means, a measuring element such as a colorimeter or a photoelectric sensor is used as the second measuring means, and the measuring element is sequentially opposed to each measurement point. The display characteristic is measured.

  According to a ninth aspect of the present invention, in the image correction data calculation method of the image display device according to the first aspect, in the second step, a plurality of photographing positions are set as the predetermined photographing positions, and each of the photographing positions is set. The screens on which the test patterns are projected by the photographing means are partially overlapped and photographed to obtain the test pattern photographing data, and the conversion function used in the third step has x and y directions orthogonal to the screen. A plurality of measurement points are set to project a test pattern, and the test patterns are photographed by partially overlapping each other by the photographing means from the plurality of photographing positions to obtain display characteristics of all the measurement points, A table of each measurement point measured by the second measuring means from a position different from the plurality of shooting positions. Seeking properties, it is characterized in that calculated on the basis of the display characteristics of each measurement point obtained from the display characteristics and the reference position different from the position of all the measurement points obtained from the reference position.

  According to a tenth aspect of the present invention, in the image correction data calculation method of the image display device according to the first aspect, in the fourth step, the image correction is performed so that the display characteristics are optimal when the screen is observed from infinity. It is characterized by calculating data.

  According to an eleventh aspect of the present invention, in the image correction data calculation method for an image display device according to any one of the first to tenth aspects, the display characteristics include brightness, hue, and γ of each color of R, G, and B. It is at least one of characteristic, white balance, and offset.

  According to the present invention, a test pattern projected and displayed on a screen is photographed from a predetermined photographing position to obtain test pattern photographing data, and the test pattern photographing data is different from the photographing position based on a previously calculated conversion function. Since the image correction data is calculated based on the converted characteristic data from the observation position, the image correction data is calculated based on the converted characteristic data. Image correction data that can be observed can be calculated.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
1 to 22 show a first embodiment of the present invention, FIG. 1 is a schematic plan view of a multi-screen display for explaining the operation, and FIG. 2 is a diagram showing how the multi-screen display is used. 3 is a block diagram showing a configuration of the image correction apparatus shown in FIG. 2, FIG. 4 is a flowchart for explaining an outline of an image correction data acquisition method, and FIG. 5 is a diagram showing a schematic configuration for acquiring image correction data. FIG. 6 is a diagram showing a configuration when acquiring image correction data, FIG. 7 is a diagram showing an example of block division of a screen display area when calculating a conversion function, and FIG. 8 is a test pattern when calculating the conversion function. FIG. 9 is a diagram for explaining an example of the imaging method, FIG. 9 is a diagram showing two other examples of measurement points in the block when calculating the conversion function, and FIG. 10 is a measurement obtained by the imaging method shown in FIG. FIG. 11 is a diagram showing an example of luminance measurement results at points, FIG. 11 is a diagram showing a conversion function f (x) in the x direction calculated from the measurement results of FIG. 10, and FIG. FIG. 13 shows a function f (r, x, y), FIG. 13 shows a conversion function at an arbitrary screen position (p, q) obtained from the conversion function f (r, x, y), and FIG. 14 shows a conversion function. FIG. 15 is a diagram for explaining another example of a test pattern for calculating a conversion function, and FIG. 16 is a diagram illustrating a case where the sensitivity characteristics of the camera are not uniform due to the imaging method shown in FIG. FIG. 17 is a diagram illustrating an example of block division of the screen display area when acquiring characteristic data, FIG. 18 is a diagram illustrating an example of characteristic data, and FIG. 19 is a calculation of image correction data. To explain a specific example of the method FIG. 20 is a diagram showing an example of correction characteristic data in the correction data calculation process, FIG. 21 is a diagram for explaining the correction effect according to the present embodiment, and FIG. 22 is a diagram for explaining the correction result. It is.

  In the present embodiment, in the multi-screen display as shown in FIG. 38, each image displayed on the screen 4 as shown in FIG. Image correction data for correcting in-plane unevenness and inter-surface unevenness of the image display device 1 is calculated.

  As shown in FIG. 2, each image display device 1 includes an image correction device 15 that corrects the video signal output from the video output device 16 such as a personal computer or a video player as described above and supplies it to the projector 3. Is provided.

  Each image correction device 15 includes an image correction data storage unit 22, a video signal correction processing unit 23, a control unit (CPU) 24, a communication unit 25, a video signal input unit 26, and a video as shown in FIG. A signal output unit 27 is provided. The image correction data storage unit 22 stores the image correction data of the image display device 1 through the control unit 24 and the communication unit 25 so as to be able to read / write data with an arithmetic processing unit to be described later.

  In this way, the input video signal input from the video output device 16 via the video signal input unit 26 is real-time based on the image correction data stored in the image correction data storage unit 22 in the video signal correction processing unit 23. Then, the corrected video signal is supplied from the video signal output unit 27 to the corresponding projector 3 and projected and displayed on the screen 4.

  The image correction data storage unit 22 is configured with a non-volatile memory (for example, a flash memory), or is backed up with a battery so that the data is not lost even when the power is turned off using a volatile memory (SDRAM). However, since the input video signal is corrected in real time with reference to the image correction data in the video signal correction processing unit 23, preferably, a non-volatile memory and a volatile memory are used. The image correction data is always stored in a non-volatile memory with a relatively slow processing speed, and the image correction data is downloaded from the non-volatile memory to a volatile memory with a relatively high processing speed when the power is turned on. .

  Next, a method of calculating image correction data stored in the image correction data storage unit 22 of each image correction device 15 will be described.

  In the present embodiment, as shown in an overall schematic flowchart in FIG. 4, first, a test pattern is projected on the screen 4 in each image display device 1 (step S1), and the entire screen on which the test pattern is displayed is displayed. Shooting is performed from a predetermined shooting position (in this embodiment, a position separated from the center of the screen by a predetermined distance) by a shooting means, for example, a camera or a video camera having an imaging element such as a CCD (hereinafter simply referred to as a camera) (step) S2). Next, the test pattern photographing data obtained by photographing is converted into characteristic data obtained when the test pattern is photographed from almost infinity based on a conversion function calculated in advance (step S3). Thereafter, image correction data for correcting in-plane unevenness of each image display device 1 and unevenness between surfaces of the image display devices 1 is calculated based on the characteristic data of all the image display devices 1 (step S4). The calculated image correction data is stored in the image correction data storage unit 22 of the corresponding image correction device 15 (step S5).

  For this reason, as shown schematically in FIG. 5, a common arithmetic processing unit 31 is provided for all the image display devices 1 to be combined. The arithmetic processing unit 31 includes a test pattern sending unit 32, a conversion function storage unit 33, a characteristic data calculation unit 34, a characteristic data storage unit 35, and an image correction data calculation unit 36. The test pattern sending unit 32 is connected to each image correction device. The image correction data calculation unit 36 is connected to the communication unit 25 of each image correction device 15 via a hub (HUB) 37, and the characteristic data calculation unit 34 includes As will be described later, test pattern photographing data from the camera is input. This arithmetic processing unit 31 is constituted by a personal computer (PC), for example.

  In this way, when acquiring the image correction data of each image display device 1, a through mode command signal is given from the arithmetic processing unit 31 to the image correction device 15 as shown in FIG. A test pattern signal is sent from the test pattern sending unit 32 to the video signal input unit 26 of the image correction device 15, and image correction data is acquired according to the flowchart shown in FIG.

  That is, in each image display device 1, the test pattern signal from the arithmetic processing device 31 is supplied to the projector 3 in the through mode without being corrected by the video signal correction processing unit 23, and is projected and displayed on the entire surface of the screen 4. The test pattern is photographed from the reference position by the camera 41 while the outside light is shielded by the hood 40 in the dark room, and the test pattern photographing data is taken into the characteristic data calculation unit 34 of the arithmetic processing unit 31. When the hood 40 is used, the inner surface is formed of a material that does not reflect light so that the image on the screen 4 is not reflected by the hood 40.

  The characteristic data calculation unit 34 captures the input test pattern imaging data based on the conversion function corresponding to the image display device 1 stored in the conversion function storage unit 33 when the test pattern is imaged from almost infinity. The obtained characteristic data is converted and stored in the characteristic data storage unit 35. After that, when the characteristic data of all the image display apparatuses 1 are stored in the characteristic data storage unit 35, the image correction data calculation unit 36 based on the characteristic data and the in-plane unevenness of each image display apparatus 1 and the image display apparatus 1 Image correction data for correcting the inter-surface unevenness is calculated, and the calculated image correction data is stored in the image correction data storage unit 22 of the corresponding image correction device 15 via the hub 37.

  Hereinafter, an example of a method for acquiring a conversion function stored in the conversion function storage unit 33 of the arithmetic processing device 31 corresponding to each image display device 1 will be described.

  In the present embodiment, as described above, the in-plane unevenness and the inter-surface unevenness of the image display device 1 are corrected so that the image displayed on the screen 4 has an optimum image quality when observed from almost infinity. . For this reason, in each image display device 1, a plurality of measurement points are set on the screen 4 as shown in FIG. Here, the display area on the screen 4 is set to 25 blocks in total with x = 0, 1, 2, 3, 4 and y = 0, 1, 2, 3, 4 at almost equal intervals on the orthogonal x and y coordinates. The display area is divided into grids, a predetermined area at the center of each block is set as a measurement point, and the display characteristics are measured.

In the following, description will be made by taking the part indicated by the one-dot chain line in the x direction of FIG. Note that display characteristics to be measured include luminance (cd / m ² ) and color (x, y, z value, L, a, b value, etc.). Only listed. The final correction target includes the luminance, hue, γ characteristic, white balance, offset, and the like of each of the R, G, and B colors. The luminance of R will be described as a representative example.

  First, as shown in FIG. 8, the entire display of the screen 4 on which the test pattern is projected by the camera used when acquiring the image correction data or the camera 51 having the same light receiving characteristic as the first measurement means. An image is taken from the reference position, and the luminance of light rays (indicated by solid lines) incident on the reference position from each measurement point is measured. Here, the reference position is a position that is separated from the center (x, y) = (2, 2) of the screen 4 by a predetermined distance, and a predetermined pattern for photographing the test pattern projected on the screen 4 when acquiring the image correction data. The position is the same as or equivalent to the shooting position.

  Also, using the camera 51 as the second measuring means, moving it in the x and y planes parallel to the screen 4 including the reference position, and sequentially photographing the test pattern from the front of each block indicated by the broken line, The brightness of light rays (indicated by broken lines) in the screen normal direction at each measurement point is measured from the photographing data of the central portion of the photographing element. Note that the luminance in the normal direction at the center measurement point of the screen 4 can also be obtained from the captured data at the reference position.

  Here, as the second measuring means for measuring the luminance in the screen normal direction at each measurement point, another camera having the same light receiving characteristic as the camera 51 photographed at the reference position can be used. If they are equal, a measuring element such as a colorimeter or a photoelectric sensor can be used. In addition, the measurement point of each block on the screen 4 is not limited to a predetermined area at the center of the block as shown in FIG. 7, but may be a local area at the center of the block as shown in FIG. As shown in FIG. 9B, the average luminance can be measured for the entire area in the block.

  The luminance of each measurement point at the reference position and the luminance in the screen normal direction at each measurement point are, for example, signal values r = 0 (black) represented by 256 steps (8 bits) as r test patterns. Up to 255 (red), that is, the luminance is stepped from 0% to 100%, for example, 0% (r = 0), 25% (r = 64), 50% (r = 128), 75% (r = 192) ) And 100% (r = 255), and the test pattern of each luminance is measured in order of 5 steps.

  In this way, when each R luminance test pattern is photographed at a position facing the reference position and each measurement point, luminance data as shown in FIG. 10 is obtained for a test pattern having a luminance of 100%, for example. In FIG. 10, ◯ indicates the luminance of each measurement point at the reference position, and X indicates the luminance at the directly facing position of each measurement point.

  As is clear from FIG. 10, the brightness at each measurement point measured at the reference position differs depending on the angle of the diffused light from the screen 4 and the light receiving sensitivity of the camera 51 also varies depending on the angle due to the optical characteristics of the photographing lens. As the shooting angle is wider, the sensitivity is different between the central portion and the peripheral portion. Also, the test pattern displayed on the screen 4 is the central portion of the screen as the projection angle by the projector is wider. Since the difference in luminance between the peripheral portion and the peripheral portion increases, the luminance decreases toward the periphery of the screen 4.

  As described above, when the brightness of each measurement point at the reference position and the brightness in the screen normal direction at each measurement point are measured for the test patterns of each brightness of R, the test pattern of each brightness is measured at the reference position of each measurement point. A conversion coefficient f for converting the test pattern photographing data obtained by photographing from the reference position into characteristic data obtained by photographing the test pattern from almost infinity by dividing the luminance by the luminance in the normal direction of the screen. At the same time, a conversion function f (x) in the x direction as shown in FIG. 11 is obtained by approximately interpolating the conversion coefficient f between measurement points by the least square method or the like. FIG. 11 shows the conversion function f (x) in the case of FIG. Similarly, a conversion function f (y) in the y direction is obtained.

  In this way, when the conversion function f (x, y) is obtained for the test patterns of each luminance of R, as shown in FIG. 12, 5 × 5 = 25 (points) on the x and y planes, and further 5 levels Since a coefficient of 25 × 5 = 125 (points) is obtained with the luminance (r) of R, a conversion function f (r, x, y) for the total luminance of R is obtained by approximating interpolation between the luminances by the least square method or the like. Is stored in the conversion function storage unit 33 of the arithmetic processing unit 31. As a result, the conversion function at an arbitrary position (p, q) on the screen 4 can be expressed as shown in FIG. 13, and the image is taken from the reference position in the process of calculating image correction data using this conversion function. The test pattern photographing data obtained in this way can be converted into characteristic data obtained when the test pattern is photographed from almost infinity.

  Similarly, a conversion function for all luminances of G and B is obtained and stored in the conversion function storage unit 33. Similarly, for other image display devices 1 constituting the multi-screen display, a conversion function f (color) (r, x, y) of all luminances of R, G, and B is obtained and converted by the arithmetic processing unit 31. Store in the function storage unit 33.

  FIG. 14 shows a flowchart of the above-described conversion function acquisition method, the details of which are as described above, and a description thereof will be omitted here.

  If the above conversion function is calculated once and stored in the conversion function storage unit 33 of the arithmetic processing unit 31, it is not necessary to calculate again thereafter. For example, the conversion function is performed once when the image display device 1 is installed, or The data can be calculated at the time of shipment from the factory, and the data can be recorded on a portable recording medium such as a CD-R and read from the recording medium, or stored in a database connectable to a network and taken in via the network. .

  In the above description, the luminance of each measurement point at the reference position is measured by the camera used when acquiring the image correction data, or the camera 51 having the same light receiving characteristic, and the screen normal at each measurement point. The luminance in the direction is measured by using a camera 51 that is photographed at the reference position, another camera having a light receiving characteristic equal to the camera 51, or a measuring element such as a colorimeter or a photoelectric sensor having the same light receiving characteristic as the camera 51. Therefore, the measurement luminance at the reference position at the measurement point (x, y) = (2, 2) in the center of the screen 4 and the measurement luminance in the screen normal direction coincide with each other. Therefore, the luminance in the normal direction at the measurement point in the center of the screen 4 can be obtained from the photographing data at the reference position without being measured independently.

  However, if the light reception characteristics of the camera that measures the brightness at each measurement point from the reference position and the camera or brightness measurement device that measures the brightness in the screen normal direction directly opposite each measurement point are different, It is necessary to measure the luminance in the normal direction at the same as the luminance measurement in the normal direction at other measurement points. In this case, for example, as shown in FIG. 15, the measurement brightness at the center measurement point is different from L1 and L2, so that the entire screen is measured at the reference position so that the brightness at the center measurement point is the same. What is necessary is just to normalize with the brightness | luminance of a center measurement point by multiplying the measured value obtained by imaging | photography with L2 / L1.

  Note that when image correction data is obtained by photographing a test pattern with a single camera as in the prior art, nonuniform sensitivity characteristics within the surface of an image sensor such as a CCD are reflected in the image correction data. It will be. For example, if the sensitivity characteristic of a part of the image sensor is low, the image correction data is calculated so that the part is brightened, and therefore, that part of the input image is corrected brightly.

  On the other hand, as described above, the camera 51 that measures the luminance of each measurement point by photographing the entire screen at the reference position is used to photograph the luminance in the normal direction by sequentially facing each measurement point. When measuring from the image data of the central part of the element, since the measurement is always performed at substantially the same point of the image sensor in the camera 51, even if the sensitivity characteristics in the plane of the image sensor are not uniform, It will not be affected. Moreover, when acquiring the image correction data, the captured image is always converted according to the conversion function calculated based on the result measured at the center portion of the image sensor, so that the sensitivity characteristics in the plane of the image sensor are obtained by the conversion. There is an advantage that non-uniformity is corrected at the same time.

  FIGS. 16A to 16C show such a state. FIG. 16 (a) shows the measurement brightness (◯ mark) at each measurement point at the reference position and the measurement brightness (× mark) at the directly-facing position of each measurement point. The measured brightness at the point-to-point position is shown uniformly. Here, when the entire screen is imaged from the reference position, due to in-plane variation of the image sensor, the sensitivity of the part of the image sensor corresponding to the second measurement point from the left in the figure is lower than the other parts. The case where the measured brightness | luminance has shifted | deviated below from the curve of a figure is shown.

  In this case, as shown in FIG. 16 (b), the value of the conversion function f (x) in the x direction is increased only on the contrary, so that the same value from the reference position is obtained when acquiring the image correction data. The test pattern photographed by the camera 51 is converted by this function, so that it is corrected to a uniform image g (x) as shown in FIG. 16C, and the in-plane variation of the image sensor is corrected. .

  Next, a specific example of the image correction data calculation method according to the present embodiment will be described in more detail.

  In each image display device 1, first, as a through mode, a test pattern of only R is applied from, for example, a signal value r = 0 (black) to r = 255 (bright red) represented by 256 steps (8 bits), that is, the luminance is 0. From 100% to 100% and projected sequentially on the screen 4, and each time, the entire screen is photographed by the camera 41 from the predetermined photographing position corresponding to the reference position when the conversion function is obtained, and the test is performed. The pattern photographing data is taken into the characteristic data calculation unit 34 of the arithmetic processing unit 31.

  Here, the gradation level of the test pattern can be, for example, 0, 16, 32, 48,..., 240, 255, every 16 gradations, or the actual input video γ coefficient. Is 0.45, the gradation of black video is emphasized, and white video tends to be relatively rough. For example, 0, 4, 8, 12, 16, 24, 32, 40, 48 64, 96, 128, 160, 192, 224, 255, etc., the black image can be fine and the white image can be made slightly rough. Here, for convenience of explanation, the luminance of the test pattern is set to 0% (r = 0), 25% (r = 64), 50% (r = 128), 75% (as in the case of obtaining the conversion function). It is assumed that r = 192) and 100% (r = 255) are sequentially changed in 5 steps every 64 gradations.

  As shown in FIG. 17, the characteristic data calculation unit 34 divides the display area of the screen 4 into a plurality of blocks at almost equal intervals in a grid pattern, and stores the shooting data of the test pattern for each gradation in a conversion function. Based on the conversion function f (r, x, y) at the corresponding position stored in the unit 33, the image data is converted into luminance characteristic data obtained when photographing from approximately infinity. That is, when the luminance at the (p, q) point of the test pattern photographed from the reference position by the camera 41 is L, the luminance at this point is converted to L ′ = f (L, p, q).

  Note that FIG. 17 shows a case where the display area is divided into a total of 25 blocks of x = 5 and y = 5, as in the case of obtaining the conversion function. What is necessary is just to set. For example, when the projector 3 is SVGA, it can be divided into a total of 480,000 blocks of a maximum of 800 horizontal and 600 vertical, and in the case of SXGA (1280 × 1024), 16 × 16 is set. One block can be divided into a total of 5,120 blocks of 80 in the horizontal direction and 64 in the vertical direction.

FIG. 18 shows an example of R characteristic data obtained by the above processing, and FIG. 18 (b) is a graph of FIG. 18 (a). In FIG. 18, R1 indicates the luminance (cd / m ² ) of the first block, R2 indicates the luminance of the second block, R3 indicates the luminance of the third block, and the luminance of the other blocks is illustrated. It is omitted.

  The characteristic data for G and B is calculated in the same manner, and the characteristic data for each color of R, G, and B is stored in the characteristic data storage unit 35 of the arithmetic processing unit 31. In this manner, the characteristic data of each color of R, G, and B of the N (in this case, four) image display apparatuses 1 constituting the multi-screen display is calculated and stored in the characteristic data storage unit 35.

  Thereafter, based on the characteristic data stored in the characteristic data storage unit 35, the image correction data calculation unit 36 calculates image correction data for correcting the in-plane unevenness and the inter-surface unevenness, and the calculated image correction data is used as the hub. The image correction data is stored in the image correction data storage unit 22 of the corresponding image correction device 15 via the control unit 37.

  That is, as shown in the flowchart of FIG. 19, first, the characteristic data of the N image display devices 1 stored in the characteristic data storage unit 36 is taken into the image correction data calculation unit 36 (step S11). Thereafter, the image correction data calculation unit 36 compares the characteristic data for each of the R, G, and B colors of the k (k = 1 to N) -th image display device 1 to obtain the lowest luminance in photographing a 100% luminance image. A value W (k) and a maximum luminance value B (k) in photographing a 0% luminance image are extracted (step S12). Next, the upper limit reference data Wmin that is the minimum value of W (k) (k = 1 to N) and the lower limit reference data that is the maximum value of B (k) (k = 1 to N). Bmax is extracted (step S13).

  That is, in step S12, the minimum values R-White (k), G-White (k), and B-White (k) at 100% luminance of each color in the k (k = 1 to 4) -th image display device 1 are displayed. And maximum values R-Black (k), G-Black (k), and B-Black (k) at 0% luminance are extracted. In step S13, R-White (k), G-White (k) are extracted. ), B-White (k) (k = 1 to N), the upper limit reference data Wmin (R-Whitemin, G-Whitemin, B-Whitemin) and R-Black (k), G The lower limit reference data Bmax (R-Blackmax, G-Blackmax, B-Blackmax) which is the maximum value among -Black (k) and B-Black (k) (k = 1 to N) is extracted.

  Thereafter, for all the blocks of each image display device 1, the image correction data is set such that 100% luminance of each color becomes the upper limit reference data Wmin, 0% luminance becomes the lower limit reference data Bmax, and a predetermined γ characteristic curve is obtained. And the image correction data is set in the corresponding image correction device 15 of the image display device 1 (step S14).

  That is, in step S14, first, other characteristic data is obtained using the above-mentioned R-Whitemin, G-Whitemin, B-Whitemin and R-Blackmax, G-Blackmax, B-Blackmax as the upper limit reference data and the lower limit reference data, respectively. The correction coefficient is calculated so as to match the reference data, and the luminance of each block is corrected with the correction coefficient. That is, for all the blocks, the characteristic data is corrected so that the luminance of the 0% luminance image and the luminance of the 100% luminance image match the respective reference data.

  FIG. 20 shows an example of R correction characteristic data obtained by the above correction process of characteristic data, and FIG. 20B is a graph of FIG. 20A. In FIG. 20, as in the case of FIG. 18, the first to third blocks are shown, and the other blocks are not shown.

  In this way, as shown in FIG. 21A, for example, luminance distributions having different black levels (gradation 0) and maximum luminance levels (gradation 255) in the horizontal direction or vertical direction of each image display device 1 are obtained. Even if they are present, they can be made equal as shown in FIG. FIG. 21 shows the black level and the maximum luminance level of two image display devices.

  However, as is apparent from FIG. 20, even if the characteristic data is corrected so that the luminances of gradation 0 and gradation 255 match the extracted reference data for all blocks, the γ coefficient of each block If they are different, the brightness in the halftones does not necessarily match.

  Therefore, in step S14, image correction data that also matches the γ coefficients is calculated for all blocks. Specifically, an average of the γ characteristic curve is calculated for each color of R, G, and B, and a correction coefficient that provides the average γ characteristic curve (predetermined γ characteristic curve) for each block, that is, an image. Correction data is calculated.

  In this way, as shown in FIG. 22, the luminance at each gradation can be matched for all blocks. FIG. 22 shows the R correction result Rave, and FIG. 22B shows a graph of FIG. 22A.

  The image correction data for matching the luminance and γ characteristic curves described above is calculated by a single calculation process, and the image correction data of each image display device 1 obtained thereby is distributed to the corresponding image correction device 15. Thus, the image correction data storage unit 22 stores the block and the image correction data in correspondence with each other using a lookup table (LUT) method.

  Thereafter, the video output device 16 is connected to the video signal input unit 26 of each image correction device 15 as shown in FIG. 2, and the input video signal from the video output device 16 is stored in the image correction data storage unit 22. Based on the stored image correction data, correction processing is performed by the video signal correction processing unit 23, and projection display is performed on the screen 4 by the projector 3, and one image is displayed on the four image display devices 1.

  In the present embodiment, a test pattern projected and displayed on the screen 4 of each image display device 1 is photographed from a predetermined photographing position to obtain test pattern photographing data, and the test pattern photographing data is converted into a previously calculated conversion function. Based on the converted characteristic data, the image correction data for correcting the in-plane unevenness and the inter-surface unevenness is calculated based on the converted characteristic data. Therefore, as shown in FIG. 1, the brightness at each point on the screen 4 can be made uniform when observed from almost infinity, and the same is true for the left and right observers as well as the central observer. It is possible to observe images with uniform image quality, such as brightness and color. In FIG. 1, the right observer is the same in that the left part of the screen is slightly dark, and the left observer is the same in that the right part of the screen appears slightly dark. This is better than the conventional case described, and when observing a large screen from various observation positions with a large number of observers, all observers can see a substantially uniform image in this embodiment. it can.

  In the present embodiment, a test pattern sending unit 32 is provided in the arithmetic processing device 32 so that the test pattern signal is distributed from the arithmetic processing device 32 to the image correction device 15 of each image display device 1 and displayed. However, each image correction device 15 is provided with a test pattern sending unit, and either a test pattern signal from the test pattern sending unit or a video signal from the external video output device 16 is selected by a selector and displayed. It can also be made to do. Further, in the present embodiment, the conversion function is calculated by photographing the screen 4 for each image display device 1, but the conversion function is calculated in the same manner with the entire screen of the multi-screen display as one screen. You can also

(Second Embodiment)
FIGS. 23 to 28 show a second embodiment of the present invention. FIG. 23 is a schematic plan view of an arch screen display for explaining the operation. FIG. 24 is an image of a test pattern for calculating a conversion function. FIG. 25 is a diagram for explaining an example of the method, FIG. 25 is a diagram showing an example of block division of the screen display area when calculating the conversion function, and FIG. FIG. 27 is a diagram illustrating an example of a measurement result, FIG. 27 is a diagram illustrating a conversion function f (x) in the x direction calculated from the measurement result of FIG. 26, and FIG. 28 is a flowchart of a conversion function acquisition method.

  In this embodiment, in the arch screen display as shown in FIG. 40, the image displayed on the arch-shaped common screen 4 as shown in FIG. 23 is observed from the center of the last row of the audience seats. Image correction data for correcting in-plane unevenness and inter-surface unevenness of the display screen by each projector 3 so as to obtain an optimum image quality is calculated.

  In such an arch screen display, the screen 4 is relatively large, and if a camera for calculating image correction data is arranged in the center of the last row of the spectator seat, it will be in the way of the spectator. For example, as shown in FIG. In addition, the first shooting position and the second shooting position are set at positions on both sides in front of the audience seat 55, and the cameras 56 and 57 are arranged at these shooting positions, respectively. Shoot the entire screen by partially overlapping the shooting area. The first shooting position and the second shooting position are not limited to the left / right symmetrical positions with respect to the screen 4 and are set to arbitrary positions.

  In the present embodiment, prior to calculating the image correction data, the first shooting position is set as the first reference position, the second shooting position is set as the second reference position, and the first reference position and the second reference position are set. Using two cameras 56 and 57 as the first measuring means, characteristic data obtained when photographing data obtained from these cameras 56 and 57 are photographed from a predetermined observation position in the center of the last row of the spectator seat 55. Get conversion function to convert to.

  Therefore, as in the case of the first embodiment, the display area of the screen 4 is divided into a plurality of blocks, measurement points are set, and the display characteristics of each measurement point are measured. In this embodiment, as shown in FIG. 25, the screen 4 is developed and divided into a total of 30 blocks of 6 in the horizontal direction (x direction) and 5 in the vertical direction (y direction).

  In this way, test patterns having different luminances (r = 0, 64, 128, 192, 255) for each color of R, G, B were sequentially projected on the screen 4, and the test patterns having the respective luminances were displayed. The entire screen is photographed by the two cameras 56 and 57, and the luminance at each measurement point is measured. That is, in FIG. 24, the right portion of the screen 4 is photographed by the camera 56 disposed at the first reference position on the left side, and the left portion of the screen 4 partially overlaps with the camera 57 disposed at the second reference position on the right side. And measure the brightness at each measurement point.

  In addition, a camera 58 as a second measuring means is disposed at a predetermined observation position in the center of the last row of the spectator seat 55 so as to be rotatable in the x and y directions. The brightness of each measurement point is measured in the central area.

  FIG. 26 shows an example of the measurement result in this case, and shows the measurement result in the x direction indicated by the one-dot chain line in FIG. In FIG. 26, ◯ indicates the measurement luminance at each measurement point in the right region of the screen 4 by the camera 56 at the first reference position, and ● indicates the screen 4 by the camera 57 at the second reference position. The measurement brightness at each measurement point in the left side area of FIG. 8 indicates the measurement brightness at each measurement point by the camera 58 at a predetermined position. In addition, the measurement brightness by the cameras 56 and 57 is different at the measurement point in the center. This is due to various characteristics such as the characteristics of the two cameras 56 and 57, the shooting angle, the characteristics of the screen 4, and the projection angle of the projector. It depends on the cause.

  Then, the test pattern photographing data obtained by photographing from the first and second reference positions by dividing the measurement brightness of the camera 58 at each measurement point by the measurement brightness of the camera 56 or the camera 57, and a predetermined observation A conversion coefficient f for conversion to characteristic data obtained when the test pattern is photographed from the position is obtained, and the conversion coefficient f between the measurement points is approximated by a least square method or the like to obtain x as shown in FIG. A conversion function f (x) in the direction is obtained. Similarly, a conversion function f (y) in the y direction is obtained.

  The above processing is performed for each color of R, G, and B with each luminance test pattern to obtain the conversion function f (x, y), and the least square method is used to calculate the luminance between the conversion functions f (x, y). The conversion function f (color) (r, x, y) for the total luminance of each color of R, G, B is obtained by performing approximate interpolation with the above. The conversion function f (color) (r, x, y) obtained in this way is not shown, but the conversion of the arithmetic processing unit similar to that shown in the first embodiment for controlling the projector constituting the arch screen display is performed. Store in function storage.

  FIG. 28 shows a flowchart of the conversion function acquisition method in the second embodiment described above, and the details thereof are as described above, so the description thereof is omitted here.

  Thereafter, when calculating the image correction data, the test pattern photographing data of each luminance of R, G, and B photographed by the cameras 56 and 57 at the first and second photographing positions is used for each color as in the first embodiment. The characteristic data is obtained by conversion with the conversion function of the projector 3, and the display screen of each projector 3 is optimized so as to obtain an optimum image quality when observed from a predetermined observation position at the center of the last row of the spectator seat 55 based on the characteristic data. Image correction data for correcting in-plane unevenness and inter-surface unevenness is obtained and stored in an image correction device corresponding to each projector 3.

  According to the present embodiment, the image correction data calculation cameras 56 and 57 (imaging means) that are used only during calibration and are not normally used are separated from the predetermined observation position in the center of the last row of the spectator seat 55. Therefore, it is possible to perform image correction when observing from a predetermined observation position by disposing at any position that does not disturb the user, so that all the audience seats 55 can enjoy a relatively good image on average. .

  In the present embodiment, the image correction data is calculated with the predetermined observation position as the center of the last row of the spectator seat 55. However, the predetermined observation position is appropriately determined according to the screen shape, the arrangement of the spectator seat, and the like. Can be set. In addition, although the entire screen is photographed by the two cameras 56 and 57, if the entire screen can be photographed by one camera, one reference position (photographing position) is sufficient, and three or more cameras are not used. If the entire screen cannot be shot, three or more reference positions (shooting positions) may be set accordingly. Furthermore, this embodiment can be effectively applied not only to an arch screen display but also to the case of calculating image correction data in a dome screen display.

(Third embodiment)
FIGS. 29 to 34 show a third embodiment of the present invention, FIG. 29 is a schematic plan view of a multi-seamless display for explaining the operation, and FIG. 30 is a test pattern photographing for calculating a conversion function. FIG. 31 is a diagram illustrating an example of a method, FIG. 31 is a diagram illustrating a screen shape of a screen captured by a camera at a reference position, and FIG. 32 is a diagram illustrating a state in which the screen capturing screen of FIG. 33 is a diagram showing an example of the luminance measurement result at the measurement point obtained by the imaging method shown in FIG. 30, and FIG. 34 is a diagram showing the x-direction conversion function f (x) calculated from the measurement result of FIG. It is.

  In the present embodiment, in a multi-seamless display as shown in FIG. 39, a test pattern is projected onto a common flat screen 4 as shown in FIG. 29, and the entire screen on which the test pattern is displayed is projected. Optimum when the camera 61, which is a photographing means permanently installed at a predetermined photographing position on the ceiling, is photographed obliquely and the image displayed on the screen 4 is observed from almost infinity based on the test pattern photographing data. Image correction data for correcting in-plane unevenness and inter-surface unevenness of the display screen by each projector 3 so as to obtain image quality is calculated.

  For this reason, before calculating the image correction data, a conversion function for converting the shooting data obtained from the camera 61 at the predetermined shooting position into characteristic data when observed from almost infinity is acquired.

  In acquiring this conversion function, the display area of the screen 4 is divided into a plurality of, for example, 5 × 5 blocks in the x and y directions, and measurement points are set, and the screen 4 has different brightness for each of R, G, and B colors. Are sequentially projected, and the entire screen on which the test patterns of each brightness are displayed is defined as a reference position as shown in FIG. 30, and the camera 61 at that position is used as the first measuring means. And measure the brightness of each measurement point. Further, with respect to each luminance test pattern, the luminance in the normal direction of each measurement point is measured at the center of the image sensor while moving the camera 62 parallel to the screen 4 as the second measuring means. In addition, it can replace with the camera 62 and can also measure a brightness | luminance using measuring elements, such as a colorimeter and a photoelectric sensor.

  In this embodiment, since the entire screen is photographed obliquely from above by the camera 61, the photographing screen becomes indefinite as shown in FIG. 31, but this image is subjected to a separate geometric correction, and is shown in FIG. Convert to rectangle as shown.

  FIG. 33 shows an example of the measurement result in this case, and shows the measurement result in the x direction indicated by the one-dot chain line in FIG. In FIG. 32, ◯ indicates the measurement luminance of each measurement point by the camera 61 at the reference position, and X indicates the measurement luminance in the normal direction of each measurement point by the camera 62.

  Thereafter, the test luminance data obtained by imaging from the reference position is divided into characteristic data in the screen normal direction by dividing the measurement luminance of each measurement point with the camera 62 by the measurement luminance with the camera 61. And a conversion function f (x) in the x direction as shown in FIG. 34 is obtained by approximately interpolating the conversion coefficient f between measurement points by the least square method or the like. Similarly, the conversion function f (y) in the y direction is obtained, and the conversion function f (color) (r, x, B) for the total luminance of each color of R, G, B is obtained in the same manner as in the first embodiment. y), and this conversion function f (color) (r, x, y) is not shown, but is similar to that of the arithmetic processing apparatus similar to that shown in the first embodiment for controlling the projector 3 constituting the multi-seamless display. Store in the conversion function storage.

  Thereafter, when calculating the image correction data, the test pattern shooting data of each luminance of R, G, and B taken by the camera 61 at a predetermined shooting position (reference position) is used for each color as in the first embodiment. The characteristic data is obtained by conversion with the above conversion function, and the in-plane unevenness and the inter-surface unevenness of the display screen by each projector 3 are corrected so as to obtain an optimum image quality when observed from almost infinite distance based on the characteristic data. Image correction data is obtained and stored in the corresponding image correction apparatus.

  According to the present embodiment, the image correction data calculation camera 61 (photographing means) that is used only during calibration and is not normally used is not directly facing toward the center of the screen 4, but instead of facing the viewer. By placing on the ceiling that does not get in the way and shooting from an angle, it is possible to correct the brightness of the entire screen by aligning the brightness of the entire screen when viewed from the front of the screen. Therefore, as in the first embodiment, the center Of course, even the left and right observers can observe images with substantially the same uniform image quality, for example, brightness and color.

(Fourth embodiment)
FIGS. 35 to 37 show a fourth embodiment of the present invention. FIG. 35 is a schematic side view of a multi-seamless display for explaining the operation. FIG. 36 is a test pattern photographing for calculating a conversion function. FIG. 37 is a diagram for explaining an example of the method, and FIG. 37 is a diagram showing an observation state of the screen from a predetermined row.

  In the present embodiment, in a multi-seamless display as shown in FIG. 39, a test pattern is projected onto a common flat screen 4 as shown in FIG. 35, and the entire screen on which the test pattern is displayed is projected. An image is taken obliquely by a camera 65 that is a photographing means permanently installed at a predetermined photographing position on the ceiling, and based on the test pattern photographing data, an image displayed on the screen 4 is displayed on the audience seat at the predetermined observation position. Image correction data for correcting in-plane unevenness and inter-surface unevenness of the display screen by each projector 3 is calculated so as to obtain an optimum image quality when the screen 4 is viewed from below from a predetermined row in the horizontal direction. .

  For this reason, before calculating the image correction data, a conversion function is acquired for converting the shooting data obtained from the camera 65 at the predetermined shooting position into the characteristic data when observed in a predetermined row.

  In acquiring this conversion function, as in the third embodiment, the display area of the screen 4 is divided into a plurality of blocks in the x and y directions, and measurement points are set. A test pattern with different luminance is sequentially projected for each color of B, and the entire screen on which the test pattern with each luminance is displayed is a first in which a camera 65 at a predetermined photographing position is at a reference position as shown in FIG. An image is taken as a measuring means, and the luminance of each measurement point is measured by geometrically correcting the photographing screen into a rectangular shape. Further, for each luminance test pattern, the camera 66, which is the second measuring means, is translated from a predetermined row of the spectator seat in the x direction and rotated in the up and down direction in the y direction, and the center of the image sensor Measure the brightness at each measurement point.

  As a result, as in the above-described embodiment, the measurement brightness of the camera 66 at each measurement point is divided by the measurement brightness of the camera 65 at the reference position, and the measurement points and the brightness are approximated by the least square method or the like. Interpolation is performed to obtain a conversion function f (color) (r, x, y) for all luminances of R, G, B colors, and this conversion function f (color) (r, x, y) is not shown. However, it is stored in the conversion function storage unit of the arithmetic processing unit similar to that shown in the first embodiment for controlling the projector 3 constituting the multi-seamless display.

  Thereafter, when calculating the image correction data, test pattern shooting data of each luminance of R, G, and B taken by the camera 65 at a predetermined shooting position (reference position) is used as in the case of the above-described embodiment. Characteristic data is obtained by conversion with the conversion function of each color, and the optimum data is obtained when the user sits in a predetermined row of the spectator seat and looks up at the screen 4 from below as shown in FIG. Image correction data for correcting the in-plane unevenness and the inter-surface unevenness of the display screen by each projector 3 so as to obtain the image quality is calculated and stored in the image correction device corresponding to each projector 3.

  Therefore, according to the present embodiment, the image correction data calculation camera 65 (imaging means) that is not normally used is placed on the ceiling that does not interfere with the observer, and the screen 4 is photographed at an angle. When looking up from the line, the brightness of the entire screen can be made uniform and corrected to a good image.

  The present invention is not limited to projecting an image using a projector, but can be effectively applied to displaying an image on a CRT, LCD, plasma display, organic EL, LCD display panel, paper display, or the like. it can. Further, the present invention is not limited to a single screen display constituted by a plurality of image display devices, and can be effectively applied to correction of display characteristics in a single large screen image display device.

It is a schematic plan view of the multi-screen display for demonstrating the effect | action of 1st Embodiment. It is a figure which shows the usage condition of a multi-screen display. It is a block diagram which shows the structure of the image correction apparatus shown in FIG. It is a flowchart for demonstrating the outline of the image correction data acquisition method by 1st Embodiment. Similarly, it is a figure which shows schematic structure for acquiring image correction data. Similarly, it is a figure which shows the structure at the time of acquiring image correction data. It is a figure which shows the block division example of the screen display area at the time of calculating the conversion function used in 1st Embodiment. Similarly, it is a figure for demonstrating the imaging | photography method of an example of the test pattern at the time of calculating a conversion function. Similarly, it is a figure which shows two other examples of the measurement point in the block at the time of calculating a conversion function. It is a figure which shows an example of the measurement result of the brightness | luminance in the measurement point obtained by the imaging | photography method shown in FIG. It is a figure which shows the conversion function f (x) of the x direction calculated from the measurement result of FIG. It is a figure which shows the conversion function f (r, x, y) by the test pattern of each brightness | luminance of R. It is a figure which shows the conversion function of the arbitrary screen positions (p, q) obtained from the conversion function shown in FIG. It is a flowchart of the conversion function acquisition method. It is a figure for demonstrating the imaging | photography method of the other example of the test pattern at the time of calculating a conversion function. It is a figure for demonstrating a mode that the nonuniformity of the sensitivity characteristic of a camera is correct | amended with the imaging | photography method shown in FIG. It is a figure which shows the block division example of the screen display area at the time of acquiring characteristic data. It is a figure which shows an example of characteristic data. It is a flowchart for demonstrating the specific example of the correction data calculation method by 1st Embodiment. Similarly, it is a figure which shows an example of the correction characteristic data in the calculation process of correction data. It is a figure for demonstrating the correction effect by 1st Embodiment. Similarly, it is a figure for demonstrating a correction result. It is a schematic plan view of the arch screen display for demonstrating the effect | action of 2nd Embodiment. It is a figure for demonstrating the imaging | photography method of an example of the test pattern at the time of calculating the conversion function by 2nd Embodiment. Similarly, it is a figure which shows the example of a block division | segmentation of the screen display area at the time of calculating a conversion function. It is a figure which shows an example of the measurement result of the brightness | luminance in the measurement point obtained by the imaging | photography method shown in FIG. It is a figure which shows the conversion function f (x) of the x direction calculated from the measurement result of FIG. It is a flowchart of the conversion function acquisition method by 2nd Embodiment. It is a schematic plan view of the multi seamless display for demonstrating the effect | action of 3rd Embodiment. It is a figure for demonstrating an example of the imaging | photography method of the test pattern at the time of calculating the conversion function by 3rd Embodiment. In 3rd Embodiment, it is a figure which shows the screen shape of the screen image | photographed with the camera of a reference position. It is a figure which shows the state which correct | amended the screen imaging | photography screen of FIG. 31 to the rectangular shape. It is a figure which shows an example of the measurement result of the brightness | luminance in the measurement point obtained by the imaging | photography method shown in FIG. It is a figure which shows the conversion function f (x) of the x direction calculated from the measurement result of FIG. It is a schematic side view of the multi seamless display for demonstrating the effect | action of 4th Embodiment. It is a figure for demonstrating an example of the imaging | photography method of the test pattern at the time of calculating the conversion function by 4th Embodiment. It is a figure which shows the observation state of the screen from the predetermined row by 4th Embodiment. 1 is a partially exploded perspective view showing a schematic configuration of a multi-screen display that can implement the present invention. Similarly, it is a perspective view which shows schematic structure of a multi-seamless display. Similarly, it is a perspective view which shows schematic structure of an arch screen display. It is a figure for demonstrating the conventional problem. Similarly, it is a figure for demonstrating a problem. Similarly, it is a figure for demonstrating a problem. Similarly, it is a figure for demonstrating a problem. Similarly, it is a figure for demonstrating a problem. Similarly, it is a figure for demonstrating a problem.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image display apparatus 2 Case 3 Projector 4 Screen 15 Image correction apparatus 16 Video output device 22 Image correction data storage part 23 Video signal correction process part 24 Control part (CPU)
DESCRIPTION OF SYMBOLS 25 Communication part 26 Video signal input part 27 Video signal output part 31 Processing unit 22 Test pattern transmission part 33 Conversion function storage part 34 Characteristic data calculation part 35 Characteristic data storage part 36 Layer correction data calculation part 37 Hub 40 Hood 41, 51, 56, 57, 58, 61, 62, 65, 66 Camera

Claims (11)

A first step of projecting a test pattern onto the screen of the image display device;
A second step of obtaining a test pattern photographing data by photographing the projected test pattern from a predetermined photographing position by photographing means;
A third step of converting the test pattern photographing data into characteristic data obtained when the test pattern is photographed from a predetermined observation position different from the photographing position based on a conversion function calculated in advance;
A fourth step of calculating image correction data for correcting display characteristics of the image display device based on the characteristic data;
An image correction data calculation method for an image display device, comprising:   The conversion function used in the third step projects a test pattern by setting a plurality of measurement points in the orthogonal x and y directions on the screen, and the test pattern is measured from the reference position by the first measurement means. The display characteristics of all measurement points are obtained, the display characteristics of each measurement point are obtained by measuring with the second measuring means from a position different from the reference position, and the display characteristics of all measurement points obtained from the reference position are 2. The image correction data calculation method for an image display device according to claim 1, wherein calculation is performed based on display characteristics of each measurement point obtained from a position different from the reference position.   3. The image correction data calculation method for an image display device according to claim 2, wherein the reference position is the predetermined photographing position or a position equivalent thereto.   4. The image correction data calculation method for an image display device according to claim 2, wherein the position different from the reference position is a position directly facing each measurement point.   3. The position different from the reference position is a position separated from the screen by a predetermined distance, and the display characteristic is measured from the predetermined position by sequentially directing the second measuring means toward each measurement point. Or an image correction data calculation method for the image display device according to 3;   The position different from the reference position is a linear position parallel to the x direction or y direction that is a predetermined distance away from the screen, and rotates in the y direction or x direction while moving the second measuring means along the linear position. 4. The image correction data calculation method for an image display device according to claim 2, wherein the display characteristics at each of the measurement points are measured.   In the second step, a camera having an image sensor is used as the photographing means, and the first measurement means and the second measurement means are the camera or a camera having an image sensor having substantially the same light receiving characteristics as the second measurement means. When measuring the display characteristic of each said measurement point with the camera as a measurement means, each measurement point is image | photographed sequentially in the approximate center part of the image pick-up element. An image correction data calculation method for the image display device according to claim.   In the second step, a camera having an image sensor is used as the photographing means, the camera or a camera having an image sensor having substantially the same light receiving characteristic is used as the first measuring means, and colorimetry is used as the second measuring means. 5. The image correction data of the image display device according to claim 4, wherein a measuring element such as a meter or a photoelectric sensor is used and the measuring element is sequentially opposed to each of the measurement points to measure display characteristics thereof. Calculation method. In the second step, a plurality of shooting positions are set as the predetermined shooting positions, and the screens on which the test patterns are projected from the shooting positions by the shooting means are partially overlapped and shot to test the test pattern shooting. Get the data
The transformation function used in the third step projects a test pattern by setting a plurality of measurement points in the orthogonal x and y directions on the screen, and the test pattern is respectively taken from the plurality of photographing positions by the photographing means. To obtain the display characteristics of all the measurement points by partially overlapping each other, and to obtain the display characteristics of each measurement point by measuring by the second measuring means from positions different from the plurality of shooting positions. 2. The image correction data of the image display device according to claim 1, wherein the image correction data is calculated based on display characteristics of all the measurement points obtained from the above and display characteristics of each measurement point obtained from a position different from the reference position. Calculation method.   2. The image correction data calculation method for an image display device according to claim 1, wherein in the fourth step, the image correction data is calculated so that the display characteristics are optimal when the screen is observed from infinity.   The image display according to claim 1, wherein the display characteristic is at least one of luminance, hue, γ characteristic, white balance, and offset of each of R, G, and B colors. Image correction data calculation method for the apparatus.

Images