Correction value calculation method, program, and projection device

Abstract

If there are black color inconsistencies throughout the projected image, correct for these color inconsistencies. [Solution] The correction value calculation method includes: calculating a first color unevenness correction value for each of a plurality of first correction points DP included in the first part PT1 of a first image PI1 having a first part PT1 that overlaps on the projection surface SC with a second image PI2 projected from a first projection device 10A onto the projection surface SC by a second projection device 10B, and a second part PT2 that does not overlap on the projection surface SC with a second image PI2 projected from a first projection device 10A onto the projection surface SC by a second projection device 10B; and calculating a brightness correction value for the second part PT2 based on the first color unevenness correction value to reduce the difference in black brightness between the first part PT1 and the second part PT2 on the projection surface SC.

Description

Technical Field

[0001] The present invention relates to a correction value calculation method, a program, and a projection device.

Background Art

[0002] Patent Document 1 discloses a black level area setting method for setting a black level area, which is an adjustment target area of black level, on a projection image when projection images are projected from a plurality of projection devices such that a part of each projection image overlaps. In this black level area setting method, a plurality of position-adjustable guides are arranged on the projection image, and a black level area is set based on the arrangement of the guides. Further, in this black level area setting method, the brightness and chromaticity of black are adjusted in the black level area.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the black level area setting method according to Patent Document 1, since the brightness and chromaticity of black are adjusted for each black level area set based on the arrangement of guides, for example, when there is color unevenness of black over the entire projection image, the correction of the color unevenness cannot be performed.

Means for Solving the Problems

[0005] A method for calculating a correction value according to an aspect of the present invention includes: calculating a first color unevenness correction value for each of a plurality of first correction points included in the first portion of a first image, which has a first portion that overlaps on the projection surface with a second image projected from a first projection device onto the projection surface by a second projection device, and a second portion that does not overlap on the projection surface with the second image, based on a target gradation value; and calculating a brightness correction value for the second portion based on the first color unevenness correction value for reducing the difference in black brightness between the first portion and the second portion on the projection surface.

[0006] A program according to an aspect of the present invention causes a computer to perform the following actions for each of a plurality of first correction points included in the first portion of a first image, which has a first portion that overlaps on the projection surface with a second image projected on the projection surface by a second projection device and a second portion that does not overlap on the projection surface with the second image and a first portion that is projected on the projection surface by the first projection device and a second portion that does not overlap on the projection surface: calculate a first color unevenness correction value for each of a plurality of first correction points included in the first portion of the first image, based on a target gradation value; and calculate a brightness correction value for the second portion, based on the first color unevenness correction value, for performing brightness correction to reduce the difference in black brightness between the first portion and the second portion on the projection surface.

[0007] A projection device according to an aspect of the present invention includes at least one image processing circuit that performs the following: calculating a first color unevenness correction value for each of a plurality of first correction points included in the first portion of a first image, which is projected onto a projection surface and has a first portion that overlaps on the projection surface with a second image projected onto the projection surface by another projection device, and a second portion that is projected onto the projection surface and does not overlap on the projection surface with the second image, based on a target gradation value; and calculating a brightness correction value for the second portion, based on the first color unevenness correction value, for performing brightness correction to reduce the difference in black brightness between the first portion and the second portion on the projection surface. [Brief explanation of the drawing]

[0008] [Figure 1] A diagram showing the overall configuration of projection system 1. [Figure 2] Block diagram of projection device 10A. [Figure 3] Functional block diagram of the correction unit 136. [Figure 4] A flowchart showing an example of operation of the projection device 10A according to the first embodiment. [Figure 5] A diagram showing an example of a grid point LP. [Figure 6] A diagram showing an example of a grid point LP. [Figure 7] A diagram showing an example of a grid point LP. [Figure 8] A flowchart showing the substeps SS4[1] to SS4[5] that make up step S4. [Figure 9] A diagram showing an example of the correction region of an LC brightness correction circuit. [Figure 10] An explanatory diagram showing the processing procedures for the brightness correction circuit LC and the color uniformity correction circuit UC. [Figure 11] A diagram showing an example of the calculation process for the color uniformity correction value S0, brightness correction value C0, ideal output value Ai, and color uniformity correction value S1 in substep SS4[1]. [Figure 12] A diagram showing an example of the calculation process for the color uniformity correction value S0, brightness correction value C0, ideal output value Ai, and color uniformity correction value S1 in substep SS4[2]. [Figure 13] A diagram showing an example of the calculation process for the color uniformity correction value S0, brightness correction value C0, ideal output value Ai, and color uniformity correction value S1 in substep SS4[2]. [Figure 14] A diagram illustrating the processing steps in substep SS4[3]. [Figure 15] A diagram showing an example of the calculation process for the color uniformity correction value S0, brightness correction value C0, ideal output value Ai, and color uniformity correction value S1 in substep SS4[3]. [Figure 16]A diagram showing an example of the calculation process for the color unevenness correction value S0, brightness correction value C0, ideal output value Ai, and color unevenness correction value S1 in substep SS4[4]. [Figure 17] A diagram showing an example of the calculation process for the color unevenness correction value S0, brightness correction value C0, ideal output value Ai, and color unevenness correction value S1 in substep SS4[5]. [Figure 18] A flowchart showing substeps SS4[1]_1 to SS4[1]_7 that constitute substep SS4[1]. [Figure 19] A diagram showing an example of target grayscale values ​​(r,g,b) in a grid point dynamic programming (DP). [Figure 20] This figure shows an example of the target grayscale values ​​(r,g,b) at grid point DP and the grayscale value t at grid point NP adjacent to the boundary between the superimposed region DR and the non-superimposed region NR. [Figure 21] This figure shows an example of a method for determining the target grayscale value t for a grid point NP whose target grayscale value t is undetermined. [Figure 22] This figure shows an example of a method for determining the target grayscale value t for a grid point NP whose target grayscale value t is undetermined. [Figure 23] A diagram showing an example of how the target grayscale value t is calculated. [Figure 24] A diagram showing an example of how the target grayscale value t is calculated. [Figure 25] A diagram showing an example of how the target grayscale value t is calculated. [Figure 26] A diagram illustrating an example of smoothing. [Figure 27] A diagram illustrating an example of smoothing. [Figure 28] A diagram illustrating an example of the smoothing process for the target grayscale value t. [Figure 29] A diagram illustrating an example of the smoothing process for the target grayscale value t. [Figure 30] A diagram illustrating an example of the smoothing process for the target grayscale value t. [Figure 31] A figure showing a three-dimensional representation of the target values ​​of each grid point NP before smoothing. [Figure 32] A diagram showing a three-dimensional representation of the target values ​​of each grid point NP after smoothing is complete. [Figure 33] An explanatory diagram of the projected image PI_A when the projection device 10 is arranged in a 2x2 configuration. [Figure 34] This figure shows the grid points LP included in the non-overlapping region NR, the two-projection overlapping region DR[2], and the four-projection overlapping region DR[4] in the projection image PI1. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. However, in each drawing, the dimensions and scale of each part have been appropriately changed from those of the actual parts. Furthermore, the embodiments described below are preferred specific examples of the present invention and are subject to various technically preferred limitations, but the scope of the present invention is not limited to these embodiments unless otherwise stated in the following description.

[0010] 1: First Embodiment The projection system 1 according to the first embodiment will be described below with reference to Figures 1 to 32.

[0011] 1-1: Configuration of the First Embodiment 1-1-1: Overall structure Figure 1 shows the overall configuration of projection system 1. Projection system 1 comprises projection device 10A, projection device 10B, projection device 10C, and image supply device 20. Projection device 10A is an example of the "first projection device". Projection device 10B is an example of the "second projection device".

[0012] Projection devices 10A, 10B, 10C, and image supply device 20 are connected to each other via a communication line LN, enabling them to communicate with one another.

[0013] Projectors 10A, 10B, and 10C project various images or videos onto the projection surface SC. For example, among projectors 10A, 10B, and 10C, projector 10A acts as the master projector for projectors 10B and 10C. Projectors 10B and 10C act as slave projectors for projector 10B. Specifically, projector 10A transmits various control signals to projectors 10B and 10C. As a result, projector 10A controls projectors 10B and 10C. The above control signals include various correction values, which will be described later.

[0014] Furthermore, the image supply device 20 supplies various images to the projection devices 10A, 10B, and 10C. Each of the projection devices 10A, 10B, and 10C projects the images supplied by the image supply device 20 onto the projection surface SC. Alternatively, the image supply device 20 may supply images only to the projection device 10A, and the projection device 10A may supply the images projected by each projection device 10 to the projection devices 10B and 10C, respectively. In this embodiment, when projection devices 10A through 10C are not distinguished, they are referred to as projection device 10. Alternatively, in the projection system 1, projection devices 10A, 10B, and 10C may each read the image to be projected from their respective storage devices 14 and project it onto the projection surface SC. Alternatively, projection device 10A may read the image to be projected from its storage device 14 and supply the image to be projected by each projection device 10 to projection devices 10B and 10C, respectively. In this case, the projection system 1 does not necessarily have to include an image supply device 20.

[0015] In the example shown in Figure 1, projection device 10A projects projection image PI1 onto projection surface SC. Projection image PI1 is an example of a "first image". Projection device 10B projects projection image PI2 onto projection surface SC. Projection image PI2 is an example of a "second image". Projection device 10C projects projection image PI3 onto projection surface SC. Projection images PI1, PI2, and PI3 are projected onto projection surface SC, with parts of them overlapping each other, so that a single projection image PI_A is displayed on projection surface SC as a whole.

[0016] Specifically, projected image PI1 includes parts PT1 and PT2. Projected image PI2 includes parts PT3, PT4, and PT5. Projected image PI3 includes parts PT6 and PT7. Part PT1 of projected image PI1 and part PT3 of projected image PI2 are superimposed onto the projection surface SC. Additionally, part PT5 of projected image PI2 and part PT6 of projected image PI3 are superimposed onto the projection surface SC. Note that Part PT1 is an example of "Part 1," and Part PT2 is an example of "Part 2."

[0017] As a result, the projection surface SC region RL1 will have part PT2 of projection image PI1 and part PT3 of projection image PI2 projected onto it. The projection surface SC region RL2 will have only part PT1 of projection image PI1 projected onto it. The projection surface SC region RL3 will have only part PT4 of projection image PI2 projected onto it. The projection surface SC region RL4 will have part PT5 of projection image PI2 and part PT6 of projection image PI3 projected onto it. The projection surface SC region RL5 will have part PT7 of projection image PI3 projected onto it.

[0018] Of the multiple regions RL on the projection surface SC, regions RL1 and RL4 are superimposed regions DR. On the other hand, regions RL2, RL3, and RL5 are non-superimposed regions NR.

[0019] 1-1-2: Projection device configuration Figure 2 is a block diagram of projection device 10A. Projection devices 10B and 10C may have the same configuration as projection device 10A. Alternatively, projection devices 10B and 10C may have the configurations essential for projection devices, but may not include at least one of the following: imaging device 12, imaging control unit 132, image analysis unit 133, correction value calculation unit 134, image acquisition unit 135, and correction unit 136. The projection device 10A comprises a projector 11, an imaging device 12, a processing device 13, a storage device 14, and a communication device 15. Each element of the projection device 10A is interconnected by one or more buses for communicating information. Furthermore, each element of the projection device 10A consists of one or more devices, and some elements of the projection device 10A may be omitted.

[0020] The projector 11 is a device that projects various projection images PI onto a projection surface SC such as a screen or wall. The projector 11 projects various projection images PI under the control of the processing device 13. The projector 11 includes, for example, a light source, a projection lens, a dichroic mirror, a prism, and a liquid crystal panel. Light from the light source is modulated using the liquid crystal panel, and the modulated light is projected onto the projection surface SC via the projection lens. The light source, projection lens, dichroic mirror, and prism are examples of projection optical systems.

[0021] The imaging device 12 is a device that captures the projected image PI projected onto the projection surface SC. The imaging device 12 captures various images under the control of the processing device 13. The imaging device 12 is, for example, an image sensor. The imaging device 12 is an example of a "sensor".

[0022] The processing unit 13 is a processor that controls the entire projection device 10A, and is composed of, for example, one or more chips. The processing unit 13 is composed of a central processing unit (CPU) that includes, for example, interfaces with peripheral devices, an arithmetic unit, and registers. Some or all of the functions of the processing unit 13 may be implemented by hardware such as a DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), and FPGA (Field Programmable Gate Array). The processing unit 13 may also include a SoC (System on Chip). The processing unit 13 executes various processes in parallel or sequentially. The processing unit 13 or the projection device 10A is an example of a "computer". The processing unit 13 is an example of an "image processing circuit".

[0023] The storage device 14 is a recording medium readable by the processing unit 13 and stores multiple programs, including the control program PR1 executed by the processing unit 13. The storage device 14 also stores a measurement pattern image projected from the projector 11 during the correction described later. Hereafter, the measurement pattern image may be referred to as a measurement pattern. The storage device 14 may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), and RAM (Random Access Memory). The storage device 14 may also be called a register, cache, main memory, or primary memory.

[0024] The communication device 15 is hardware acting as a transmitting and receiving device for communicating with other devices. The communication device 15 is also called, for example, a network device, network controller, network card, or communication module. The communication device 15 may be equipped with a connector for wired connection and an interface circuit corresponding to the connector. The communication device 15 may also be equipped with a wireless communication interface. Examples of connectors and interface circuits for wired connection include those conforming to wired LAN (Local Area Network), IEEE1394, and USB (Universal Serial Bus). Examples of wireless communication interfaces include those conforming to wireless LAN or Bluetooth®.

[0025] The processing unit 13 functions as a projection control unit 131, an imaging control unit 132, an image analysis unit 133, a correction value calculation unit 134, an image acquisition unit 135, a correction unit 136, and a communication control unit 137 by reading and executing the control program PR1 from the storage device 14. The control program PR1 may also be transmitted via a communication network from another device, such as a server that manages the projection device 10A.

[0026] The projection control unit 131 causes the projector 11 to project the above-mentioned measurement pattern onto the projection surface SC. The projection control unit 131 also causes the projector 11 to project the image acquired by the image acquisition unit 135 (described later) from the image supply device 20 onto the projection surface SC.

[0027] The imaging control unit 132 causes the imaging device 12 to image the reflected light of the measurement pattern projected onto the projection surface SC.

[0028] The image analysis unit 133 analyzes the reflected light of the measurement pattern captured by the imaging device 12 and calculates a measured value indicating the color of the measurement pattern in the captured image.

[0029] The correction value calculation unit 134 calculates a correction value to be set in the correction unit 136, which will be described later, based on the measured values ​​calculated by the image analysis unit 133. A specific example of how the correction value calculation unit 134 calculates the correction value will be described later in the explanation of the operation of this embodiment, with reference to Figures 4 to 32.

[0030] The following describes the measurement pattern for correcting the black color projected by the projection control unit 131 in this embodiment.

[0031] A typical projection device with color adjustment functionality divides the color range from the lowest black level to the highest white level into roughly equal parts, projects colored light with the corresponding color at each division point, and calculates a correction value based on the image captured of the projected colored light on the projection surface. In this process, the projection device estimates the intermediate color levels between the first colored light level and the second colored light level based on interpolation calculations such as spline interpolation, and calculates a correction value based on the results of this estimation.

[0032] Alternatively, a projection device with a standard color adjustment function, as another example, projected colored light with a base of intermediate gray tones and calculated correction values ​​based on the image captured from the projected colored light on the projection surface. In this case, for colors other than gray, the properties of additive color mixing were used to estimate them based on interpolation calculations, and correction values ​​were calculated based on the results of these estimations.

[0033] However, the color of low-gradation light, close to black (gradation 0), is a color light that is created by adding gradually modulated RGB light to color light that has black as its entirety (gradation 0). As a result, the change in chromaticity is large, and conventional measurement methods have the problem of large estimation errors. Therefore, in this embodiment, color light having the color of each gradation, which is divided more finely than gradations above the predetermined gradation, is projected as a measurement pattern from gradation 0 to a predetermined gradation. This reduces the estimation error in the low-gradation range close to black compared to when measurements are performed with the same fineness regardless of the gradation. For example, the range up to 146 gradations in the first of seven divisions of the gradation width from gradation 0 to gradation 1023 may be divided evenly. The fineness of the division can be two or more divisions. Furthermore, it is preferable to use a measurement pattern of gray included between gradation 0 and a predetermined gradation, and multiple measurement patterns in which one of the R, G, and B color components is changed based on the gray measurement pattern. Compared to dividing the low-tone range equally, the accuracy of estimating the change in chromaticity in the low-tone range is improved.

[0034] In the above explanation, as an example, we described the case where the liquid crystal panel of the projector 11 has three panels: a panel corresponding to the R component, a panel corresponding to the G component, and a panel corresponding to the B component. However, the same effect is achieved even when the liquid crystal panel has only one panel.

[0035] Returning to Figure 2, the image acquisition unit 135 acquires the image to be projected from the image supply device 20.

[0036] The correction unit 136 corrects the image acquired by the image acquisition unit 135 using the correction value calculated by the correction value calculation unit 134.

[0037] Figure 3 is a functional block diagram of the correction unit 136. The correction unit 136 includes a brightness correction circuit LC and a color uniformity correction circuit UC.

[0038] The brightness correction circuit LC corrects the brightness of the image acquired by the image acquisition unit 135 using the correction value calculated by the correction value calculation unit 134. This correction includes correction for so-called "black level floating." "Black level floating" refers to the difference in brightness between the superimposed area DR and the non-superimposed area NR on the projection surface SC when the projection device 10 attempts to display a black image with a grayscale of 0 on the projection surface SC by projecting the projection image PI onto the projection surface SC. Therefore, correcting "black level floating" is a brightness correction that reduces the difference in brightness between the superimposed area DR and the non-superimposed area NR.

[0039] The color unevenness correction circuit UC corrects the color unevenness of the image acquired by the image acquisition unit 135 using the correction value calculated by the correction value calculation unit 134. Details of the brightness correction circuit LC and the color unevenness correction circuit UC will be described later.

[0040] In Figure 2, the projection control unit 131 projects the image corrected by the correction unit 136 as the projected image PI onto the projection surface SC using the projector 11.

[0041] The communication control unit 137 causes the communication device 15 to send and receive various types of information with external devices. This information includes correction values ​​that are transmitted from projection device 10A to projection device 10B and projection device 10C, respectively.

[0042] 1-2: Operation of the First Embodiment Figure 4 is a flowchart showing an example of the operation of the projection device 10A according to the first embodiment.

[0043] In step S1, the projection device 10A calculates brightness correction values ​​and color uniformity correction values ​​for gradations other than black. Furthermore, the projection device 10A uses these correction values ​​to adjust the brightness and color of the projected image PI from projection devices 10A to 10C so that they are uniform among the projection devices 10.

[0044] Here, "black" refers to the color contained within the first gradation range, which includes gradation 0, when the gradation range from the smallest gradation value (gradation 0) to the largest gradation value (gradation 1023) is divided into N sections. Here, N is an integer greater than or equal to 3. For the sake of explanation, N may be set to 7 below. In the case of N=7, "black" refers to the color contained within the gradations between gradation 0 and gradation 146.

[0045] Furthermore, the brightness correction described above is a correction that reduces the difference in brightness between the superimposed region DR and the non-superimposed region NR for gradations other than black. Also, the method used for the brightness and color unevenness correction described above may be a conventional method. For example, in Figure 1, the method may be one in which any region RL from regions RL1 to RL5 is designated as the target region, and the brightness and color unevenness are corrected by comparing the image value shown by the image value obtained by imaging the target region with the image value obtained by imaging the other regions RL.

[0046] In this case, the adjustment point when the processing unit 13 corrects the color unevenness of the projected image PI is, for example, the grid points LP of the projected image PI, which are 11 rows x 21 columns. The number of gradations is based on the gradations at the boundaries when the gradation width from gradation 0 to gradation 1023 is divided into 7 equal parts.

[0047] In step S1, the processing unit 13 determines whether each of the grid points LP used as the color unevenness correction circuit UC is included in the superposition region DR or in the non-superposition region NR. For example, before step S1, the processing unit 13 projects a full white image as a projection image only from projector 11 and captures the full white image with the imaging device 12 of projection device 10A. Next, it projects a full white image as a projection image only from projection device 10B and captures the full white image with the imaging device 12 of projection device 10A. From these imaging results, the processing unit 13 detects the position of the right edge of the full white image projected from projection device 10A and the position of the left edge of the full white image projected from projection device 10B, and determines that the region from the position of the right edge to the position of the left edge is the superposition region DR in the coordinate system of the captured image. Then, based on the correspondence between the coordinate systems of the projection device 10A, the projection device 10B, and the captured image, which were generated before step S1, the superposition region DR in the captured image coordinate system is converted into the superposition region DR in the coordinate systems of the projection device 10A and the projection device 10B. Based on the captured image of the grid points LP, the processing device 13 determines whether each of the grid points LP belongs to the superposition region DR in the coordinate systems of the projection device 10A and the projection device 10B. The correspondence can be calculated, for example, by a well-known calibration technique using Gray code. The coordinate system of the projection device 10A is the two-dimensional coordinate system of the liquid crystal panel. The coordinate system of the projection device 10B is the same. The same also applies to the projection devices 10B and 10C. It should be noted that the above process is merely an example, and the method for determining whether each grid point LP is included in the superposition region DR or the non-superposition region NR can be modified as appropriate.

[0048] Figures 5 to 7 show examples of grid points LP. More specifically, Figure 5 is an example of a grid point LP corresponding to the projected image PI1 projected from projection device 10A. Figure 6 is an example of a grid point LP corresponding to the projected image PI2 projected from projection device 10B. Figure 7 is an example of a grid point LP corresponding to the projected image PI3 projected from projection device 10C.

[0049] In Figures 5 to 7, grid point LP includes grid point DP, grid point NP, and grid point PP. In these figures, hatched circles indicate grid point DP included in the superimposed region DR. White circles indicate grid point NP included in the non-superimposed region NR. Circles with dotted outlines indicate grid point PP where it could not be determined whether it was included in the superimposed region DR or the non-superimposed region NR. For each grid point LP shown in Figures 5 to 7, there are measured values ​​calculated by the image analysis unit 133. Note that lattice point DP is an example of a "first correction point," and lattice point NP is an example of a "second correction point."

[0050] In step S2 of Figure 4, the projection device 10A projects the black correction measurement pattern described above. Specifically, the processing device 13 in the projection device 10A functions as a projection control unit 131. The processing device 13 reads the measurement pattern from the storage device 14 and causes the projector 11 to sequentially project the measurement pattern onto the projection surface SC.

[0051] In step S3, the processing unit 13 in the projection device 10A functions as an image capture control unit 132. The processing unit 13 causes the imaging device 12 to image each measurement pattern projected onto the projection surface SC. The processing unit 13 also functions as an image analysis unit 133. The processing unit 13 analyzes the measurement patterns captured by the imaging device 12 and calculates a measurement value indicating the color of the measurement pattern in the captured image.

[0052] First, the processing unit 13 calculates the correspondence between the gradation values ​​(r, g, b) of the color light projected by the projector 11 at each point of the grid LP and the measured values ​​(R, G, B) indicating the color of the color light in the captured image, calculated by analyzing the color light as a measurement pattern captured by the imaging device 12, using interpolation. The interpolation calculation should be appropriate depending on the type and number of measurement patterns used. For example, the processing unit 13 may perform curve interpolation using a spline curve. Alternatively, the processing unit 13 may perform parabolic interpolation. Alternatively, the processing unit 13 may perform cubic curve interpolation. One example of such cubic curve interpolation is cubic interpolation.

[0053] The image analysis unit 133 converts the RGB values ​​into XYZ values ​​by multiplying a matrix whose components are the RGB values ​​representing the color of the colored light in the estimated captured image by a transformation matrix specific to the imaging device 12. As a result, the image analysis unit 133 can estimate the output values ​​XYZ for any grayscale.

[0054] Furthermore, the image analysis unit 133 calculates the brightness component Y and the chromaticity component u'v' by converting the estimated XYZ values ​​into Yu'v' values ​​using the following equation 1.

number

[0055] Returning to Figure 4, in step S4, the processing unit 13 in the projection device 10A functions as a correction value calculation unit 134. The processing unit 13 calculates correction values ​​to be set for the brightness correction circuit LC and the color unevenness correction circuit UC, respectively, so that the brightness of the corrected black becomes the target brightness as seen from the imaging device 12, the color unevenness of the corrected black is suppressed, and the chromaticity is uniform as seen from the imaging device 12.

[0056] Figure 8 is a flowchart showing the substeps SS4[1] to SS4[5] that constitute step S4.

[0057] In substep SS4[1], the processing unit 13 calculates ideal output values ​​for all grid points LP in the superimposed region DR and the non-superimposed region NR. Specifically, the processing unit 13 calculates ideal output values ​​to correct color unevenness in the superimposed region DR as correction values ​​for grid points DP included in the superimposed region DR. The processing unit 13 also calculates ideal values ​​to correct color unevenness in the non-superimposed region NR and to match the brightness of the non-superimposed region NR to the brightness of the superimposed region DR as correction values ​​for grid points LP included in the non-superimposed region NR. After these processes, the processing unit 13 decomposes the ideal output values ​​for grid points DP included in the superimposed region DR and grid points NP included in the non-superimposed region NR into correction values ​​to be set in the brightness correction circuit LC and correction values ​​to be set in the color unevenness correction circuit UC. The ideal output values ​​are an example of "target gradation values". The method for calculating the "ideal output value" mentioned above will be explained later in section "1-3: Addendum (Method for Calculating the Ideal Output Value)".

[0058] Here, we will explain the brightness correction circuit LC and the color uniformity correction circuit UC.

[0059] Figure 9 shows an example of the correction region of the brightness correction circuit LC. In Figure 9, the superimposed region DR and the non-superimposed region NR are the correction regions that the brightness correction circuit LC targets. When the brightness correction circuit LC performs correction on black, it functions as a circuit that corrects so-called "black floating". The brightness correction circuit LC can be set to any shape as the correction region, as shown in Figure 9. In addition, the brightness correction circuit LC can be set to a brightness correction value C0, which is a uniform adjustment amount for each of the superimposed region DR and the non-superimposed region NR as exemplified in Figure 9.

[0060] On the other hand, the color unevenness correction circuit UC can set a correction value for each of the discrete grid points LP exemplified in Figures 5 to 7. Furthermore, the color unevenness correction circuit UC can set a color unevenness correction value S as an adjustment value for each discrete gradation. Here, as an example, we set color unevenness correction values ​​S for eight gradations. Specifically, the color unevenness correction value S at gradation 0 is S0, the color unevenness correction value at gradation 146 is S1, ... the color unevenness correction value at gradation 1023 is S7. In addition, the color unevenness correction values ​​S for gradations between gradation 0 and gradation 146...1023 are determined by linear interpolation.

[0061] Figure 10 is an explanatory diagram showing the processing procedure of the brightness correction circuit LC and the color unevenness correction circuit UC. As shown in Figure 10, when a gradation value is input to the brightness correction circuit LC, the brightness correction value C0 is added to the input value. In the case of gradation 0, the brightness correction value output from the brightness correction circuit LC is C0. When the brightness correction value C0 is output from the brightness correction circuit LC to the color unevenness correction circuit UC, the color unevenness correction circuit UC performs linear interpolation between the point where the x-coordinate is gradation 0 and the y-coordinate is the color unevenness correction value S0 and the point where the x-coordinate is gradation 146 and the y-coordinate is the color unevenness correction value S1, thereby obtaining the following equation 2. According to the following equation 2, the output value A of the color unevenness correction circuit UC is u The following is calculated. Note that the color unevenness correction value S1 has already been calculated in step S1 of Figure 4.

number

[0062] At the grid point LP, the output value A can be obtained by using equation 2 above. u The following is calculated. For pixels between grid points LP, linearly interpolated values ​​are calculated.

[0063] Figure 11 shows the color uniformity correction value S0, brightness correction value C0, and ideal output value A in substep SS4[1]. iand a diagram showing an example of the calculation status of the color unevenness correction value S1. In FIG. 11, for simplicity of explanation, the lattice points DP included in the overlapping region DR are arranged in one vertical row and three horizontal rows, and the lattice points NP included in the non-overlapping region NR are arranged in three vertical rows and three horizontal rows. As shown in FIG. 11, at the stage when the sub-step SS4[1] is completed, for the ideal output value A i and the color unevenness correction value S1, values have been calculated at all lattice points DP and lattice points NP. On the other hand, for the color unevenness correction value S0 and the brightness correction value C0, values have not been calculated at any lattice points DP and lattice points NP. Note that the numerical values shown in FIG. 11 are integers, but this is for convenience of explanation. The values corresponding to the lattice points DP and lattice points NP may be decimal numbers. The same applies to the following drawings.

[0064] In the sub-step SS4[2] of FIG. 8, the processing device 13 calculates the color unevenness correction value S0 and the brightness correction value C0 at the lattice points DP included in the overlapping region DR. Specifically, the processing device 13 divides the ideal output value A i at the lattice points DP included in the overlapping region DR into the color unevenness correction value S0 and the brightness correction value C0.

[0065] The processing device 13 pre-determines the brightness correction value C0 = C Lap of the overlapping region DR, and calculates the color unevenness correction value S0 = S0 u at gradation 0 so that the final output value, the output value A i = A Lap becomes the ideal output value A Lap = A Lap . The brightness correction value C0 = C i = A Lap , and the color unevenness correction value S1 = S1 Lap at gradation 146 are already known. Substituting these values into Equation 2 results in Equation 3. By transforming Equation 3 into Equation 4, the value of the color unevenness correction value S0 = S0 Lap is calculated. In FIG. 11, the ideal output value A i included in the overlapping region DR is the above-mentioned ideal output value A i = A Lap . Note that the color unevenness correction value S0 = S0 for the superimposed region DR Lap This is an example of the "first color unevenness correction value".

number

[0066] Note that the brightness correction value C0 of the superimposed region DR is set to C Lap Set = 0, and set the color unevenness correction value S0 to S0 Lap =A Lap That's fine.

[0067] Figures 12 and 13 show the color uniformity correction value S0, brightness correction value C0, and ideal output value A in substep SS4[2]. i This figure shows an example of the calculation status of the color unevenness correction value S1. Note that Figure 12 is C Lap This is the figure when =20. Figure 13 is C Lap This figure shows the case where = 0. As shown in Figures 12 and 13, when substep SS4[2] is completed, the ideal output value A i Regarding the color unevenness correction value S1, values ​​have been calculated for all grid points DP and NP. Furthermore, regarding the color unevenness correction value S0 and brightness correction value C0, values ​​have been calculated for grid points DP included in the superimposed region DR. As described above, at the substep SS4[2] stage, the brightness correction value C0 and the color unevenness correction value S0 are determined for each grid point DP included in the superimposed region DR.

[0068] In substep SS4[3] of Figure 8, the processing unit 13 calculates a color unevenness correction value S0 and a brightness correction value C0 for grid points NP included in the non-overlapping region NR that are adjacent to grid points DP included in the overlapping region DR, with the boundary between the non-overlapping region NR and the overlapping region DR in between.

[0069] When a grid point DP and a grid point NP are adjacent, the processing unit 13 calculates a brightness correction value C0 for the grid point NP such that the color unevenness correction value S0 for level 0 at the grid point NP is the same as the color unevenness correction value S0 for level 0 at the grid point DP.

[0070] Figure 14 is an explanatory diagram of the processing in substep SS4[3]. The projected image PI1 shown in Figure 14 corresponds to the projected image PI1 shown in Figure 5.

[0071] In Figure 14, the grid point NP enclosed by the dotted line is a grid point NP adjacent to the superimposed region DR, with the boundary between the non-superimposed region NR and the superimposed region DR in between. The grid point enclosed by the dotted line is an example of a "third correction point". In substep SS4[3], the processing unit 13 calculates the correction value at the grid point NP. The grid point DP enclosed by the solid line is a grid point DP in the superimposed region DR adjacent to the grid point NP.

[0072] When the processing unit 13 calculates the brightness correction value C0 at the grid point NP0 shown in Figure 14, the color unevenness correction value S0 at the grid point NP0 is equal to the color unevenness correction value S0 at the grid point DP0. Lap The brightness correction value C0 at grid point NP0 should be the same as C NonLap The following calculation is performed. Grid points NP0 and DP0 are located in the same row of the grid and are adjacent in the row direction. Even if a grid point PP exists between grid points NP0 and DP0 whose ownership cannot be determined (either the superimposed region DR or the non-superimposed region NR), the grid points DP0 and NP0 that are closest to each other in either the row or column direction, across the boundary between the non-superimposed region NR and the superimposed region DR, are treated as adjacent grid points. Ideal output value A i =A NonLap Color unevenness correction value S1 = S1 in 146 gradations NonLap , and color unevenness correction value S0 = S0 Lap The values ​​of are known, and substituting these values ​​into Equation 2 gives Equation 5. By transforming Equation 5 into Equation 6, we get the brightness correction value C0 = C NonLap The value is calculated.

number

[0073] Figure 15 shows the color uniformity correction value S0, brightness correction value C0, and ideal output value A in substep SS4[3]. i This figure shows an example of the calculation status of the color unevenness correction value S1. In Figure 15, the ideal output value A of a lattice point NP adjacent to lattice point DP, which is included in the non-superposition region NR. i However, the ideal output value A mentioned above i =A NonLap Therefore, the color unevenness correction value S1 for a grid point NP adjacent to grid point DP among the grid points NP included in the non-overlapping region NR is the above color unevenness correction value S1 = S1. NonLap Therefore, the color unevenness correction value S0 at the grid point DP included in the superimposed region DR is S0 = S0. Lap Therefore, using these values ​​and equation 6, the brightness correction value C0 = C for the grid point NP adjacent to grid point DP is obtained. NonLap This is calculated. As shown in Figure 15, at the stage when substep SS4[3] is completed, the ideal output value A i Regarding the color unevenness correction value S1, values ​​have been calculated for all grid points DP and NP. Furthermore, regarding the color unevenness correction value S0 and the brightness correction value C0, values ​​have been calculated for grid points DP included in the superimposed region DR, and for grid points NP adjacent to grid point DP among the grid points NP included in the non-superimposed region NR.

[0074] In substep SS4[4] of Figure 8, the processing unit 13 calculates the brightness correction value C0=C for the grid points NP adjacent to the superimposed region DR, which is on either side of the boundary between the non-superimposed region NR and the superimposed region DR, as calculated in substep SS4[3]. NonLap Therefore, the brightness correction value C0 = C in the entire non-overlapping region NR. NonLapThis is determined uniformly. Furthermore, the brightness correction value C0 = C for grid points NP adjacent to the superimposed region DR. NonLap This is an example of a "provisional brightness correction value". As an example, the processing unit 13 calculates the brightness correction value C0 = C for grid points NP adjacent to the superimposed region DR, across the boundary between the non-superimposed region NR and the superimposed region DR, which was calculated in substep SS4[3]. NonLap The average value is calculated, and this average value is used as the brightness correction value C0 = C for all grid points NP included in the non-overlapping region NR. NonLap As described above with reference to Figure 9, the brightness correction circuit LC sets a brightness correction value C0, which is a uniform adjustment amount for both the superimposed region DR and the non-superimposed region NR.

[0075] Figure 16 shows the color uniformity correction value S0, brightness correction value C0, and ideal output value A in substep SS4[4]. i This figure shows an example of the calculation status of the color unevenness correction value S1. In Figure 15, the brightness correction value C0 = C was set at the grid point NP adjacent to the superimposed region DR, with the boundary between the non-superimposed region NR and the superimposed region DR in between. NonLap The average value of is the brightness correction value C0 = C for all grid points NP in Figure 16. NonLap It will be set as follows. As shown in Figure 16, when substep SS4[4] is completed, the brightness correction value C0 and the ideal output value A are i Regarding the color unevenness correction value S1, values ​​have been calculated for all grid points DP and NP. Furthermore, regarding the color unevenness correction value S0, values ​​have been calculated for grid points DP included in the superimposed region DR, and for grid points NP adjacent to grid point DP among the grid points NP included in the non-superimposed region NR.

[0076] In substep SS4[5] of Figure 8, the processing unit 13 calculates the brightness correction value C0 = C at the grid point NP of the entire non-overlapping region NR, which was calculated in substep SS4[4]. NonLap Therefore, color unevenness correction value S 0= S0 NonLap’The following is calculated. Note that the processing in substep SS4[5] is performed in substep SS4[3] where the brightness correction value C0=C NonLap The calculation is performed for grid points NP adjacent to grid points DP included in the superimposed region DR, across the boundary between the non-superimposed region NR and the superimposed region DR. This is because the brightness correction value C0 = C at the grid point NP. NonLap This is because it changed after processing in substep SS4[4].

[0077] Ideal output value A i =A NonLap Color unevenness correction value S1 = S1 in 146 gradations NonLap , and brightness correction value C0=C NonLap The values ​​of are known, and substituting these values ​​into Equation 2 gives Equation 7. By transforming Equation 7 into Equation 8, we obtain the color unevenness correction value S0 = S0 NonLap’ The value is calculated. Note that the color unevenness correction value S0 = S0 for non-overlapping region NR NonLap’ This is an example of a "second color unevenness correction value".

number

[0078] Figure 17 shows the color uniformity correction value S0, brightness correction value C0, and ideal output value A in substep SS4[5]. i This figure shows an example of the calculation status of the color unevenness correction value S1. In Figure 17, the ideal output value A of a lattice point NP adjacent to lattice point DP, which is included in the non-superposition region NR. i However, the ideal output value A mentioned above i =A NonLap Therefore, the color unevenness correction value S1 for a grid point NP adjacent to grid point DP among the grid points NP included in the non-overlapping region NR is the above color unevenness correction value S1 = S1. NonLap Therefore, the brightness correction value C0 at the grid point NP included in the non-overlapping region NR is given by the above C=S0 NonLap Therefore, using these values ​​and equation 8, the color unevenness correction value S is obtained. 0= S0 NonLap’ This is calculated. As shown in Figure 17, when substep SS4[5] is completed, the color uniformity correction value S0, the brightness correction value C0, and the ideal output value A are i Furthermore, regarding the color unevenness correction value S1, the value has been calculated for all grid points DP and NP.

[0079] In step S5 of Figure 4, the processing unit 13 provided in the projection device 10A functions as a correction value calculation unit 134. The processing unit 13 sets the brightness correction value C0 calculated in step S4 to the brightness correction circuit LC and the color unevenness correction value S0 to the color unevenness correction circuit UC.

[0080] In Figure 2, when the brightness correction circuit LC receives the grayscale value of each pixel in the image acquired by the image acquisition unit 135, it calculates a grayscale value to output to the color unevenness correction circuit UC using the brightness correction value C0. When the color unevenness correction circuit UC receives the grayscale value calculated by the color unevenness correction circuit UC, it outputs the color unevenness correction circuit A based on the color unevenness correction value S0. u Output. Output value A u This is the grayscale value and is input to the projection control unit 131. The projection control unit 131 outputs value A u This is converted into a control signal to drive the LCD panel and output to the projector 11. As a result, the projector 11 projects a projected image PI onto the projection surface SC, with the black color unevenness and "black floating" corrected.

[0081] 1-3: Appendix (Method for calculating ideal output values) Figure 18 is a flowchart of substeps SS4[1]_1 to SS4[1]_7 that constitute substep SS4[1]. In the flowchart shown in Figure 18, the processing unit 13 first sets the ideal output value A of the superimposed region DR. i After calculating the target values ​​for brightness and chromaticity of the non-overlapping region NR, the processing unit 13 then calculates the ideal output value A in the non-overlapping region NR based on the calculated target values. i The following is calculated. Here, the "target value for brightness" is the target value for correcting the "black level issue" mentioned above. The "target value for chromaticity" is the target value for correcting the "color unevenness" mentioned above. For the sake of simplicity, the following explanation assumes that the ideal output value A at grid point LP included in the projected image PI2 is used. i Only the calculation method will be explained. However, the ideal output value A at grid point LP included in projected images PI1 and PI3 will not be explained. i The calculation method is basically the same.

[0082] In substep SS4[1]_1, the processing unit 13 assumes that the target brightness and chromaticity in the superimposed region DR have already been determined by a known method, and sets an ideal output value A at the grid point DP included in the superimposed region DR. i The processing unit 13 calculates the ideal output value A. i Known methods may be used to calculate this.

[0083] Specifically, the processing unit 13 predetermines the target gradation value for the superimposed region DR. For each grid point DP of the superimposed region DR shown in Figures 5 to 7, the processing unit 13 calculates the brightness component Y as a target value from the determined gradation value. This "brightness component Y as a target value" is the brightness component Y used to correct the "black level lift" mentioned above.

[0084] The processing unit 13 calculates a target brightness component Y for each grid point DP. Specifically, even if brightness unevenness originally exists within the superimposed region DR, the processing unit 13 does not uniformly adjust the brightness unevenness within the superimposed region DR, but rather calculates a target brightness component Y for each grid point DP. If the processing unit 13 were to uniformly adjust the brightness unevenness, it would be necessary to lower the brightness at each grid point DP, which would reduce the number of gradation steps that can be corrected at each grid point DP. The reason the processing unit 13 pre-determines a target brightness component Y for each grid point DP is to ensure that the minimum correction amount is sufficient to uniformly remove color unevenness at each grid point DP as described above.

[0085] Furthermore, the processing unit 13 predetermines the chromaticity component u'v' as the target chromaticity value for the superimposed region DR. This "target chromaticity component u'v'" is the chromaticity component u'v' used as the target value for correcting the color unevenness mentioned above. The processing unit 13 determines the same target chromaticity component u'v' for all superimposed regions DR based on the average chromaticity among the three projectors 10 (projector 10A, projector 10B, and projector 10C), the average chromaticity in each projector 10, the chromaticity design value of each projector 10 at the time of product shipment, etc. In this case, after correction, all superimposed regions DR will have the same chromaticity. Alternatively, the processing unit 13 determines the chromaticity component u'v' as the same target value within each superimposed region DR based on the above parameters. In this case, after correction, the chromaticity will be the same within each superimposed region DR. As a result, color unevenness in the superimposed DR region is corrected.

[0086] As described above, the processing unit 13 converts the target value Yu'v' value, which consists of the brightness component Y and the chromaticity component u'v', at each grid point DP within the superimposed region DR, into an RGB value. Furthermore, the processing unit 13 inversely calculates the target gradation value (r,g,b) to be output from the correction unit 136 to the projector 11 using the converted RGB value. The processing unit 13 then determines this to be the ideal output value A that should be output from the correction unit 136 to the projector 11. i Therefore, in this section, the ideal output value Ai represents the target grayscale values ​​(r, g, b).

[0087] Figure 19 shows an example of the target grayscale value (r,g,b) at the grid point DP calculated by the processing unit 13. In Figure 19, an example of the value of one of the r, g, and b components included in the grayscale value (r,g,b) is shown. Figure 19 corresponds to the projected image PI2 shown in Figure 6. Furthermore, each rectangle shown in Figure 19 represents a portion PT of the projected image PI, including each grid point LP shown in Figure 6. Region RL1 includes multiple portions PT11. Region RL3 includes multiple portions PT31. Region RL4 includes multiple portions PT41. The numerical values ​​inside the rectangle above represent the values ​​of one of the r, g, and b components included in the target grayscale value (r, g, b). For the sake of simplicity, the values ​​in Figure 19 are shown as integers, but they may actually be decimals. The same applies to the values ​​shown below.

[0088] Furthermore, among the rectangles shown in Figure 19, those with double borders correspond to lattice point DP. As described above, lattice point DP is lattice point LP included in the superposition region DR. Rectangles with single borders correspond to lattice point NP. As described above, lattice point NP is lattice point LP included in the non-superposition region NR. Rectangles with double borders, one of which is a dotted line, correspond to lattice point PP. As described above, lattice point PP is lattice point LP that could not be determined to belong to either the superposition region DR or the non-superposition region NR.

[0089] In substep SS4[1]_2 of Figure 18, the processing unit 13 determines the target gradation values ​​(r,g,b) at the grid point NP such that the brightness and color are matched between the grid point DP belonging to the superimposed region DR and the grid point NP belonging to the non-superimposed region NR, which are adjacent to each other across the boundary between the superimposed region DR and the non-superimposed region NR.

[0090] The target grayscale value (r,g,b) at the grid point NP is calculated based on the target grayscale value (r,g,b) at the grid point DP. The specific calculation method is described below.

[0091] In Figure 19, lattice point DP1 is an example of lattice point DP. Similarly, lattice point NP1 is an example of lattice point NP. Lattice points DP1 and NP1 are adjacent to each other, separated by the boundary between the superposition region DR and the non-superposition region NR.

[0092] In Figure 19, the target gradation values ​​(r,g,b) for grid point DP1 have already been calculated. Based on these gradation values ​​(r,g,b), the processing unit 13 estimates the RGB values ​​as measured values ​​by the imaging device 12. The processing unit 13 then converts the estimated RGB values ​​as measured values ​​into XYZ values. Furthermore, the processing unit 13 applies the above equation 1 to the converted XYZ values ​​to further convert the converted XYZ values ​​into Yu'v' values.

[0093] As a result, the Yu'v' value of grid point DP1 is calculated, and the processing unit 13 sets this Yu'v' value as the target Yu'v' value for grid point NP1. Furthermore, the processing unit 13 calculates the target grayscale values ​​(r, g, b) for grid point NP1 based on this target Yu'v value.

[0094] In Figure 19, region RL3 includes multiple intermediate portions CP between the first adjacent portion GP1 and the second adjacent portion GP2.

[0095] For the sake of simplicity, the following explanation will focus on the brightness target value among the target values ​​for grid point NP1. Since the method for calculating the chromaticity target value is the same as that for the brightness target value, the explanation regarding the brightness target value also applies to the chromaticity target value.

[0096] When the brightness of grid point NP1 reaches the target value, the processing unit 13 calculates a grayscale value t as the target grayscale value (r,g,b) for grid point NP such that the r, g, and b components are (r,g,b)=(t,t,t). Specifically, the processing unit 13 calculates a grayscale value t such that when gray light is projected from the projector 11, the brightness at grid point DP1 and grid point NP1 are the same.

[0097] Figure 20 shows an example of the target grayscale value (r,g,b) at grid point DP and the target grayscale value t at grid point NP adjacent to the boundary between the superimposed region DR and the non-superimposed region NR. Furthermore, without correction, the non-overlapping region NR will be darker than the overlapping region DR. Therefore, in order to make the brightness component Y the same for both the overlapping region DR and the non-overlapping region NR after correction, it is necessary to make the target tonal value t at grid point NP larger than the target tonal value (r,g,b) at grid point DP.

[0098] In substep SS4[1]_3 of Figure 18, the processing unit 13 calculates further target grayscale values ​​t for grid point NP by averaging the target grayscale values ​​t of grid point NP calculated in substep SS4[1]_2 with the target grayscale values ​​t of adjacent grid point NP.

[0099] Specifically, for a grid point NP whose target grayscale value t is undetermined, the processing unit 13 calculates the average value of the target grayscale values ​​t of at least one of the adjacent grid points NP (up, down, left, or right) whose target grayscale value t is determined. The processing unit 13 then sets the calculated average value as the target grayscale value t for the grid point NP whose target grayscale value t is undetermined.

[0100] Figures 21 and 22 show examples of methods for determining the target grayscale value t for a grid point NP whose target grayscale value t is undetermined.

[0101] In Figure 21, grid point NP2, whose target tone value t is undetermined, is located at the edge of the non-overlapping region NR. In this case, there are five grid points NP adjacent to grid point NP2: grid points NP3 to NP7. Of these grid points NP3 to NP7, grid point NP3 is set to a target tone value t=52. Grid point NP4 is set to a target tone value t=51. On the other hand, grid points NP5 to NP7 do not have a target tone value t set. Therefore, the processing unit 13 sets the average value of the target tone value t=52 of grid point NP3 and the target tone value t=51 of grid point NP4 as the target tone value t of grid point NP2.

[0102] In Figure 22, grid point NP8, whose target tone value t is undetermined, is located inside the non-overlapping region NR. In this case, there are eight grid points NP adjacent to grid point NP8: grid points NP9 to NP16. Of these eight grid points NP, grid point NP9 is set to a target tone value t=52. Grid point NP10 is set to a target tone value t=51. Grid point NP11 is set to a target tone value t=52. On the other hand, grid points NP12 to NP16 do not have a target tone value t set. Therefore, the processing unit 13 sets the average value of the target tone value t=52 of grid point NP9, the target tone value t=51 of grid point NP10, and the target tone value t=52 of grid point NP11 as the target tone value t of grid point NP8.

[0103] In substep SS4[1]_4 of Figure 18, the processing unit 13 determines whether the target grayscale value t has been calculated for all grid points NP included in the non-overlapping region NR. If the target grayscale value t has been calculated for all grid points NP in the non-overlapping region NR ("YES" in substep SS4[1]_4), the processing unit 13 executes the process in substep SS4[1]_5. On the other hand, if the target grayscale value t has not been calculated for all grid points NP in the non-overlapping region NR ("NO" in substep SS4[1]_4), the processing unit 13 executes the process in substep SS4[1]_4.

[0104] As a result, the processing unit 13 sequentially extends the grid points NP used to calculate the target grayscale value t into the interior of the non-overlapping region NR.

[0105] Figures 23 to 25 show examples of the calculation status of the target tone value t. More specifically, Figure 23 shows the calculation status of the target tone value t for grid points NP adjacent to the boundary between the superimposed region DR and the non-superimposed region NR, in the direction inward of the non-superimposed region NR, compared to the state in Figure 22. Figure 24 shows the calculation status of the target tone value t for grid points NP adjacent to the direction inward of the non-superimposed region NR, for grid points NP for which the target tone value t has been newly calculated in Figure 23. Figure 25 shows the state in which the target tone value t has been calculated for all grid points NP.

[0106] In substep SS4[1]_5 of Figure 18, the processing unit 13 repeatedly smooths the target grayscale values ​​t of the calculated target grayscale values ​​t of grid points NP other than those adjacent to the boundary between the superimposed region DR and the non-superimposed region NR. Specifically, for grid points NP other than those adjacent to the boundary, the processing unit 13 smooths using the target grayscale values ​​t of valid grid points NP among the grid points NP adjacent to it in the upper, lower, left, and right directions.

[0107] As shown in Figure 25, when the processing unit 13 calculates the target tone value t of a grid point NP, it sequentially calculates the target tone value t of adjacent grid points NP to the right, starting from the first superimposed region DR, region RL1, inside the non-superimposed region NR. In parallel with this, the processing unit 13 sequentially calculates the target tone value t of adjacent grid points NP to the left, starting from the second superimposed region DR, region RL4, inside the non-superimposed region NR. As a result, in Figure 25, as an example, a large difference occurs between the target tone value t=45 of grid point NP17 and the target tone value t=41 of grid point NP18, which is adjacent to grid point NP17. In other words, there is a step between the target tone value t=45 of grid point NP17 and the target tone value t=41 of grid point NP18.

[0108] The processing unit 13 eliminates the step difference in the interpolated target grayscale value t and performs the smoothing described above so that the target grayscale value t of the grid point NP included in the non-overlapping region NR smoothly connects from region RL1, which is the first superimposed region DR, to region RL4, which is the second superimposed region DR. As a result, as described later, the distribution of the target gradation value t in region RL3, which is a non-overlapping region NR, becomes a continuous or stepped distribution in the direction from region RL1 to region RL4.

[0109] Figures 26 and 27 show examples of smoothing. Figure 26 corresponds to Figure 21. Figure 27 corresponds to Figure 22.

[0110] In Figure 26, the processing unit 13 smooths the target gradation value t=51 at grid point NP2 using the target gradation values ​​t=52 at grid point NP3, t=51 at grid point NP4, t=51 at grid point NP5, t=50 at grid point NP6, and t=50 at grid point NP7.

[0111] In Figure 27, the processing unit 13 smooths the target gradation value t=51 at grid point NP8 using the target gradation values ​​t=52 at grid point NP9, NP10, NP11, NP12, NP13, NP14, NP15, and NP16.

[0112] In substep SS4[1]_6 of Figure 18, the processing unit 13 determines whether the range of change in the target grayscale value t of the grid point NP is below a threshold before and after smoothing. More specifically, the processing unit 13 determines whether the sum of the ranges of change in the target grayscale value t of all grid points NP is below a threshold. If the sum of the ranges of change in the target grayscale value t of all grid points NP is below a threshold ("YES" in substep SS4[1]_6), the processing unit 13 executes the process in substep SS4[1]_7. On the other hand, if the sum of the ranges of change in the target grayscale value t of all grid points NP exceeds a threshold ("NO" in substep SS4[1]_6), the processing unit 13 executes the process in substep SS4[1]_5.

[0113] In other words, the processing unit 13 repeats the smoothing process until the sum of the changes in the target grayscale value t for all grid points NP falls below a threshold.

[0114] Figures 28 to 30 show examples of the smoothing process of the target tone value t. More specifically, Figure 28 shows an example of the target tone value t after the first smoothing process. Figure 29 shows an example of the target tone value t after the second smoothing process. Figure 30 shows an example of the target tone value t after the 18th smoothing process. Note that smoothing is considered complete after the 18th smoothing process shown in Figure 30.

[0115] As is clear from comparing Figures 28 to 30, the more times the smoothing process is performed, the smaller the difference in target grayscale values ​​t between adjacent grid points NP becomes overall.

[0116] Figure 31 shows a three-dimensional representation of the target grayscale value t for each grid point NP before smoothing. Figure 32 shows a three-dimensional representation of the target grayscale value t for each grid point NP after smoothing is complete. In both Figures 31 and 32, the x-axis indicates the column position of the grid point NP, the y-axis indicates the row position of the grid point NP, and the z-axis indicates the target grayscale value t. As is clear from comparing Figure 31 and Figure 32, after smoothing is complete, the steps in the target grayscale value t are eliminated compared to before smoothing.

[0117] In substep SS4[1]_7 of Figure 18, the processing unit 13 functions as a correction value calculation unit 134. The processing unit 13 calculates the target grayscale values ​​(r,g,b) at the grid point NP included in the non-overlapping region NR.

[0118] As described above, the processing unit 13 determines the target gradation value (t,t,t) of brightness at the grid point NP after smoothing is completed by processing up to substep SS4[6]. Based on the gradation value (t,t,t), the processing unit 13 calculates the target gradation value (r,g,b) at the grid point NP. The processing unit 13 may use a known method as the method for calculating the target gradation value (r,g,b).

[0119] Specifically, the processing unit 13 calculates the brightness component from the target grayscale value (t,t,t). The processing unit 13 sets the calculated brightness component as the brightness component Y of the target value.

[0120] Furthermore, the processing unit 13 determines the chromaticity component u'v' of the target value in the same manner as the brightness component Y of the target value at the grid point NP, as described above.

[0121] The processing unit 13 converts the Yu'v' value, which is the target value at each grid point NP in the non-overlapping region NR, into RGB values. Furthermore, the processing unit 13 inversely calculates the target gradation values ​​(r,g,b) that are output from the correction unit 136 to the projector 11 using the converted RGB values.

[0122] The processing unit 13 outputs the target grayscale values ​​(r, g, b) at each grid point DP and each grid point NP to an ideal output value A. i Let's assume that.

[0123] 2: Variant The above configurations can be modified in various ways. Specific examples of modifications are given below. The configurations exemplified below and the configurations shown in the above embodiments can be merged as appropriate, within the bounds of mutual consistency. In the modified configurations exemplified below, for elements whose operation and function are equivalent to those in the embodiments, the reference numerals used in the above description will be reused, and detailed explanations of each will be omitted as appropriate.

[0124] 2-1: Variation 1 In the above embodiment, the projection system 1 comprises three projection devices 10, namely projection devices 10A to 10C, and calculates correction values ​​for grid points DP included in the superimposed region DR and grid points NP included in the non-superimposed region NR when these three projection devices 10 are aligned in one direction.

[0125] However, the projection system 1 may include other numbers of projection devices 10. Furthermore, the images projected from these multiple projection devices 10 are not limited to being arranged in a single direction. For example, there may be two or more projection devices 10, and the images may be arranged vertically, horizontally, or in a matrix such as 2 rows and 2 columns or 3 rows and 3 columns.

[0126] Figure 33 is an explanatory diagram of the projected image PI_A when the projection device 10 is arranged in a 2x2 grid. In the example shown in Figure 33, projection device 10A projects projected image PI1 onto the projection surface SC. Projection device 10B projects projected image PI2 onto the projection surface SC. Projection device 10C projects projected image PI3 onto the projection surface SC. Projection device 10D projects projected image PI4 onto the projection surface SC. Projected images PI1, PI2, PI3, and PI34 are projected onto the projection surface SC, with parts of them overlapping each other, so that a single projected image PI_A is displayed on the projection surface SC as a whole.

[0127] Specifically, projected image PI1 includes parts PT51, PT52, PT53, and PT54. Projected image PI2 includes parts PT55, PT56, PT57, and PT58. Projected image PI3 includes parts PT59, PT60, PT61, and PT62. Projected image PI4 includes parts PT63, PT64, PT65, and PT66. On the projection surface SC, the portion PT53 of projection image PI1 and the portion PT57 of projection image PI2 are superimposed. Also, on the projection surface SC, the portion PT52 of projection image PI1 and the portion PT60 of projection image PI3 are superimposed. Also, on the projection surface SC, the portion PT61 of projection image PI3 and the portion PT65 of projection image PI4 are superimposed. Also, on the projection surface SC, the portion PT56 of projection image PI2 and the portion PT64 of projection image PI4 are superimposed. Also, on the projection surface SC, the portion PT54 of projection image PI1, the portion PT58 of projection image PI2, the portion PT62 of projection image PI3, and the portion PT66 of projection image PI4 are superimposed.

[0128] As a result, region RL51 of projected image PI_A includes only part PT51 of projected image PI1. Region RL52 of projected image PI_A includes part PT53 of projected image PI1 and part PT57 of projected image PI2. Region RL53 of projected image PI_A includes only part PT55 of projected image PI2. Region RL54 of projected image PI_A includes part PT52 of projected image PI1 and part PT60 of projected image PI3. Region RL55 of projected image PI_A includes part PT54 of projected image PI1, part PT58 of projected image PI2, part PT62 of projected image PI3, and part PT66 of projected image PI4. Region RL56 of projected image PI_A includes part PT56 of projected image PI2 and part PT64 of projected image PI3. Region RL57 of projected image PI_A includes only portion PT59 of projected image PI3. Region RL58 of projected image PI_A includes portion PT61 of projected image PI3 and portion PT65 of projected image PI4. Region RL59 of projected image PI_A includes only portion PT63 of projected image PI4.

[0129] In the projected image PI_A, regions RL51, RL53, RL57, and RL59 are non-overlapping regions NR. Regions RL52, RL54, RL56, and RL58 are two-projection overlapping regions DR[2]. Region RL55 is a four-projection overlapping region DR[4].

[0130] Figure 34 shows the grid point LP contained in the non-overlapping region NR, the two-projection overlapping region DR[2], and the four-projection overlapping region DR[4] in the projected image PI1. The non-overlapping region NR contains the grid point NP. The two-projection overlapping region DR[2] contains the grid point DP[2]. The four-projection overlapping region DR[4] contains the grid point DP[4]. In Figure 34, the region RL81 enclosed by the dotted line contains lattice point DP[2] and adjacent lattice point NP. The region RL82 enclosed by the dashed line contains lattice point NP and adjacent lattice point DP[2]. The region RL83 enclosed by the dashed line contains lattice point DP[4] and adjacent lattice point DP[2]. The region RL84 enclosed by the solid line contains lattice point DP[2] and adjacent lattice point DP[4]. Note that "adjacent" here includes cases where points are adjacent across a lattice point PP.

[0131] In this modified example, the processing apparatus 13 performs the same processing as substeps SS4[1] to SS4[5] shown in Figure 18. Below, we will mainly explain the differences between substeps SS4[1] to SS4[5] in this modified example and substeps SS4[1] to SS4[5] in the above embodiment.

[0132] In general terms, in substep SS4[1] of this modified example, the processing device 13 performs the same processing as up to substep SS4[1] in the above embodiment to obtain an ideal output value A for all grid points LP. i The following is calculated. Next, in substeps SS4[2] to SS4[5] of the above embodiment, the processing unit 13 performs a process in which each of the "non-overlapping area NR" and "overlapping area DR" in the above embodiment is replaced with each of the "two-unit overlapping area DR[2]" and "four-unit overlapping area DR[4]". Finally, in substeps SS4[3] to SS4[5] of the above embodiment, the processing unit 13 performs a process in which each of the "non-overlapping area NR" and "overlapping area DR" in the above embodiment is replaced with "non-overlapping area NR" and "two-unit overlapping area DR[2]".

[0133] Specifically, in this modified example, the processing unit 13 performs the processing on the superimposed region DR in substep SS4[2] of the above embodiment on the four superimposed region DR[2]. In other words, the processing unit 13 processes the ideal output value A of the four superimposed region DR[4]. iThis is broken down into a brightness correction value C0 and a color uniformity correction value S0.

[0134] Furthermore, in this modified example, the processing device 13, in the same process as substep SS4[3] in the above embodiment, calculates a brightness correction value C0 for grid points DP[2] included in region RL83, which is the boundary between the two-unit superimposed region DR[2] and the four-unit superimposed region DR[4], such that the color unevenness correction value S0 for gradation 0 is the same as the color unevenness correction value S0 for grid points DP[4] included in the four-unit superimposed region DR[4].

[0135] Furthermore, in this modified example, the processing device 13 calculates the average value of the brightness correction value C0 of the grid points DP[2] included in region RL83 in the same process as substep SS4[4] in the above embodiment, and sets this average value as the brightness correction value C0 that is uniformly set for the grid points DP[2] included in regions RL82 and RL83. Furthermore, in this modified example, the processing device 13 performs the same processing as in substep SS4[5] in the above embodiment for calculating the color unevenness correction value S0 at the grid point NP included in the non-overlapping region NR, for the grid point DP[2] included in the two-unit overlapping region DR[2].

[0136] Subsequently, the processing unit 13 calculates the brightness correction value C0 and the color unevenness correction value S0 at the grid point NP included in the non-overlapping region NR using the same procedure as substeps SS4[3] to SS4[5] in the above embodiment.

[0137] 2-2: Variation 2 In substep SS4[4] of the above embodiment, the processing device 13 sets the brightness correction value C0 = C for the grid point NP adjacent to the superimposed region DR, across the boundary between the non-superimposed region NR and the superimposed region DR. NonLap The average value is calculated, and this average value is used as the brightness correction value C0 = C for all grid points NP included in the non-overlapping region NR. NonLapHowever, the processing unit 13 may calculate the median, mode, maximum, and minimum values ​​instead of the mean. Generally, the processing unit 13 statistically processes the provisional brightness correction values ​​of the grid points NP adjacent to the superimposed region DR across the boundary between the non-superimposed region NR and the superimposed region DR, thereby obtaining a brightness correction value C0 = C for all grid points NP included in the non-superimposed region NR. NonLap Calculate.

[0138] 2-3: Variation 3 The above-mentioned projection devices 10A to 10C were equipped with a projector 11 that had a liquid crystal panel. However, projection devices 10A to 10C may also be projection devices equipped with a DMD (Digital Micro-mirror Device) instead of a liquid crystal panel.

[0139] 3: Summary of this disclosure A summary of this disclosure is provided below.

[0140] (Note 1) A correction value calculation method comprising: calculating a first color unevenness correction value for each of a plurality of first correction points included in the first portion of a first image having a first portion that overlaps on the projection surface with a second image projected from a first projection device and projected on the projection surface by a second projection device, and a second portion that does not overlap on the projection surface with the second image, based on a target gradation value; and calculating a brightness correction value for the second portion based on the first color unevenness correction value for performing brightness correction to reduce the difference in black brightness between the first portion and the second portion on the projection surface.

[0141] The correction value calculation method described in Appendix 1 involves first calculating a first color unevenness correction value to correct color unevenness in the first image, and then calculating a brightness correction value based on the first color unevenness correction value to reduce the difference in brightness of black between the first and second parts. As a result, brightness correction is performed according to the color unevenness correction value, so if there is color unevenness in black throughout the entire projected image, that color unevenness can be corrected.

[0142] Furthermore, when arranging multiple projection devices for tiling display, it is conceivable to use both a color unevenness correction circuit and a brightness correction circuit to correct color unevenness and brightness in the superimposed DR area. However, if the brightness of black is corrected first, it becomes impossible to display a black image with a grayscale of 0 in the non-superimposed NR area, and therefore the color unevenness of black cannot be corrected. Conversely, if the color unevenness of black is corrected first, due to the characteristics of the liquid crystal panel, when the brightness is corrected, a different color unevenness may occur than when displaying a black image with a grayscale of 0, and the color unevenness of black may remain even after the brightness is corrected.

[0143] The correction value calculation method described in Appendix 1, having the above configuration, can correct both the unevenness of black tones and the brightness of the entire projected image with high precision.

[0144] (Appendix 2) The correction value calculation method of Appendix 1, further comprising calculating a second color unevenness correction value for each of the plurality of second correction points included in the second part, based on the brightness correction value, for correcting the color unevenness of the first image.

[0145] The correction value calculation method described in Appendix 2, having the above configuration, can correct color unevenness in the non-overlapping region NR.

[0146] (Note 3) The method for calculating the brightness correction value according to Note 2, which includes: calculating a provisional brightness correction value for each of the plurality of third correction points adjacent to the first region from among the plurality of second correction points, based on the first color unevenness correction value of the first correction point adjacent to each of the plurality of third correction points from among the plurality of first correction points; and calculating the brightness correction value for the second region by statistically processing the provisional brightness correction values ​​for the plurality of third correction points.

[0147] The correction value calculation method described in Appendix 3, having the above configuration, can suppress abrupt changes in brightness and chromaticity between the superimposed region DR and the non-superimposed region NR, across the boundary between the superimposed region DR and the non-superimposed region NR.

[0148] (Note 4) The correction value calculation method of Note 3, wherein the calculation of the provisional brightness correction value is performed by calculating the provisional brightness correction value for each of the plurality of third correction points, where the second color unevenness correction value is equal to the first color unevenness correction value of the first correction point adjacent to each of the plurality of third correction points.

[0149] The correction value calculation method described in Appendix 4, having the above configuration, allows for fine-grained correction of the brightness in the projected image PI_A by calculating a provisional brightness correction value for each of the multiple third correction points.

[0150] (Note 5) The correction value calculation method of Note 3, wherein calculating the second color unevenness correction value includes calculating the second color unevenness correction value for each of the plurality of third correction points based on the brightness correction value for the second region.

[0151] The correction value calculation method described in Appendix 5, having the above configuration, allows for fine-grained correction of color unevenness in the projected image PI_A by calculating the second color unevenness correction value for each of the multiple third correction points.

[0152] (Note 6) A program that causes a computer to perform the following actions for each of a plurality of first correction points included in the first portion of a first image, which has a first portion that overlaps on the projection surface with a second image projected by a second projection device onto the projection surface and a second portion that does not overlap on the projection surface with the second image and a first portion that is projected by the first projection device onto the projection surface and a second portion that does not overlap on the projection surface: calculate a first color unevenness correction value for each of a plurality of first correction points included in the first portion of the first image, based on a target gradation value; and calculate a brightness correction value for the second portion, based on the first color unevenness correction value, for performing brightness correction to reduce the difference in black brightness between the first portion and the second portion on the projection surface.

[0153] The program described in Appendix 6 calculates a first color unevenness correction value to correct color unevenness in the first image, and then calculates a brightness correction value to reduce the difference in brightness of black between the first and second parts based on the first color unevenness correction value. As a result, brightness correction is performed according to the color unevenness correction value, so if there is color unevenness in black throughout the entire projected image, that color unevenness can be corrected.

[0154] Furthermore, when arranging multiple projection devices for tiling display, it is conceivable to use both a color unevenness correction circuit and a brightness correction circuit to correct color unevenness and brightness in the superimposed DR area. However, if the brightness of black is corrected first, it becomes impossible to display a black image with a grayscale of 0 in the non-superimposed NR area, and therefore the color unevenness of black cannot be corrected. Conversely, if the color unevenness of black is corrected first, due to the characteristics of the liquid crystal panel, when the brightness is corrected, a different color unevenness may occur than when displaying a black image with a grayscale of 0, and the color unevenness of black may remain even after the brightness is corrected.

[0155] The program described in Appendix 6, having the above configuration, can correct both the unevenness of black tones and the brightness of the entire projected image with high precision.

[0156] (Note 7) A projection device comprising at least one image processing circuit that performs the following: calculating a first color unevenness correction value for each of a plurality of first correction points included in the first portion of a first image, which is projected onto a projection surface and has a first portion on the projection surface that overlaps with a second image projected onto the projection surface by another projection device, and a second portion on the projection surface that does not overlap with the second image, based on a target gradation value; and calculating a brightness correction value for the second portion, based on the first color unevenness correction value, for performing brightness correction to reduce the difference in black brightness between the first portion and the second portion on the projection surface.

[0157] The projection device described in Appendix 7 calculates a first color unevenness correction value to correct color unevenness in the first image, and then calculates a brightness correction value based on the first color unevenness correction value to reduce the difference in brightness of black between the first and second parts. As a result, brightness correction is performed according to the color unevenness correction value, so if there is color unevenness in black throughout the entire projected image, that color unevenness can be corrected.

[0158] Furthermore, when arranging multiple projection devices for tiling display, it is conceivable to use both a color unevenness correction circuit and a brightness correction circuit to correct color unevenness and brightness in the superimposed DR area. However, if the brightness of black is corrected first, it becomes impossible to display a black image with a grayscale of 0 in the non-superimposed NR area, and therefore the color unevenness of black cannot be corrected. Conversely, if the color unevenness of black is corrected first, due to the characteristics of the liquid crystal panel, when the brightness is corrected, a different color unevenness may occur than when displaying a black image with a grayscale of 0, and the color unevenness of black may remain even after the brightness is corrected.

[0159] The projection device described in Appendix 7, having the above configuration, can correct both the unevenness of black tones and the brightness of the entire projected image with high precision. [Explanation of symbols]

[0160] 10...Projection device, 10A...Projection device, 10B...Projection device, 10C...Projection device, 10D...Projection device, 11...Projector, 12...Imaging device, 13...Processing device, 14...Storage device, 15...Communication device, 20...Image supply device, 131...Projection control unit, 132...Imaging control unit, 133...Image analysis unit, 134...Correction value calculation unit, 135...Image acquisition unit, 136...Correction unit, 137...Communication control unit, C...Brightness correction value, CP...Intermediate part, DP...Grid point, DR...Superimposed area, LC...Brightness correction circuit, LN...Communication line, LP...Grid point, NP...Grid point, NR...Non-superimposed area, PI...Projected image, PP...Grid point, PR1...Control program, PT...Part, RL...Area, S...Correction value, S0...Correction value, S1...Correction value, SC...Projection surface, UC...Color unevenness correction circuit, Y...Brightness component

Claims

1. For each of the multiple first correction points included in the first portion of a first image, which has a first portion that overlaps on the projection surface with a second image projected by a second projection device and projected on the projection surface by a second projection device, and a second portion that does not overlap on the projection surface with the second image, a first color unevenness correction value for correcting color unevenness of the first image is calculated based on the target grayscale value. This includes calculating a brightness correction value for the second portion based on the first color unevenness correction value, for which brightness correction is performed to reduce the difference in black brightness between the first portion and the second portion on the projection surface, Method for calculating the correction value.

2. The further step includes calculating a second color unevenness correction value for each of the plurality of second correction points included in the second portion, based on the brightness correction value, in order to correct the color unevenness of the first image. The method for calculating the correction value according to claim 1.

3. Calculating the aforementioned brightness correction value means For each of the multiple third correction points adjacent to the first region among the multiple second correction points, a provisional brightness correction value is calculated based on the first color unevenness correction value of the first correction point adjacent to each of the multiple third correction points among the multiple first correction points. This includes calculating the brightness correction value for the second region by statistically processing the provisional brightness correction values ​​for the plurality of third correction points, The method for calculating the correction value according to claim 2.

4. Calculating the provisional brightness correction value involves calculating the provisional brightness correction value for each of the plurality of third correction points, assuming that the second color unevenness correction value is equal to the first color unevenness correction value of the first correction point adjacent to each of the plurality of third correction points. The method for calculating the correction value according to claim 3.

5. Calculating the second color unevenness correction value includes calculating the second color unevenness correction value for each of the plurality of third correction points based on the brightness correction value for the second region. The method for calculating the correction value according to claim 3.

6. For each of the multiple first correction points included in the first portion of a first image, which has a first portion that overlaps on the projection surface with a second image projected by a second projection device onto the projection surface, and a second portion that does not overlap on the projection surface with the second image, a first color unevenness correction value for correcting color unevenness of the first image is calculated based on the target grayscale value. Based on the first color unevenness correction value, the computer is instructed to calculate a brightness correction value for the second portion to reduce the difference in brightness of black between the first portion and the second portion on the projection surface. program.

7. For each of the multiple first correction points included in the first portion of a first image, which is projected onto a projection surface and has a first portion that overlaps on the projection surface with a second image projected onto the projection surface by another projection device, a first color unevenness correction value for correcting color unevenness in the first image is calculated based on the target grayscale value. Based on the first color unevenness correction value, a brightness correction value is calculated for the second portion to reduce the difference in brightness of black between the first portion and the second portion on the projection surface. Includes at least one image processing circuit that performs the following: Projection device.

Images