KR20090002698A - Apparatus and method for calibration of projected image - Google Patents

Abstract

An apparatus and a method for correcting a projected image are provided to correct a projected image by reflecting a geometrical characteristic of a screen on a calculated projection surface, there more accuately modeling the geometrical characteristic of the projection surface. An apparatus(100) for correcting a projected image comprises an encoding unit(110), a projection unit(120), a photographing unit(130), a preprocessor(140), a detection unit(150), a decoding unit(160), a matching block(170), a correction value calculation unit(180), a storage unit(190) and a correction unit(195). The projection unit projects an image for correction and a series of coding images having a different color pattern to a projection surface. The correction value calculation unit calculates a geometrical characteristic based on the projected correction image and the first and second photographing images in which a series of coding images is photographed. The correction unit corrects an input image to be projected on the projection surface by reflecting the calculated geometrical characteristic of the projection surface.

Description

Apparatus and method for calibration of projected image}

The present invention relates to image correction, and more particularly, to a method and apparatus for correcting a projection image.

In recent years, as the spread of portable projectors has been expanded, in order to project image frames to arbitrary projection surfaces such as walls and curtains, as well as screens that are widely used for projectors, distortion of image frames due to characteristics of the projection surfaces may be corrected. There is an increasing demand for technology that can be used.

In general, a display system using a projector displays an image frame by projecting the image frame onto a flat screen. However, when the screen has its own color, the image frame projected on the screen has a distorted color as compared with the actual image frame. In addition, when the screen is not a perfect plane, geometric distortion due to the curvature of the screen may occur in the projected image frame.

Therefore, in order to correct the distortion of the projected image frame, color and geometric characteristics of the screen must be modeled. According to the prior art for modeling the characteristics of the screen, before projecting the image frame, the operation of comparing the projected pattern image and the photographed pattern image by projecting a predetermined pattern image on the screen and shooting the projected pattern image with a camera This was preceded. A function representing the characteristics of the screen can be obtained by comparing the projected pattern with the photographed pattern. When the inverse function of the obtained function is applied to the image frame to be projected, the user can correct the distortion due to the characteristics of the screen. You will be able to see

However, the conventional technology has a problem that it is difficult to accurately model the geometrical characteristics of the screen when the screen has a pattern that cancels the pattern image. For example, if the pattern image of the vertical stripe is projected on the wall having the pattern of the vertical stripe, it is difficult to distinguish between the vertical stripe of the wall and the vertical stripe of the pattern image when the projected image is taken. As a result, it is difficult to accurately model the wall geometry.

Therefore, a technique for more accurately modeling the geometrical characteristics of the screen is required.

It is an object of the present invention to model the geometrical characteristics of the projection surface more accurately.

However, the objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned will be clearly understood by those skilled in the art from the following description.

In order to achieve the above object, an image correction apparatus according to an exemplary embodiment of the present invention includes a projection unit for projecting a correction image and a series of coded images having different color patterns onto a projection surface, the projected correction image, and a series of coded images. A correction value calculator which calculates geometrical characteristics of the projection surface based on the first photographed image and a series of second photographed images, and an input to project onto the projection surface by reflecting the calculated geometrical characteristics of the projection surface It includes a correction unit for correcting the image.

In order to achieve the above object, an image correction method according to an embodiment of the present invention includes projecting a correction image and a series of coded images having different color patterns onto a projection surface, and projecting the projected correction image and the series of coded images. Calculating geometric characteristics of the projection surface based on the first photographed image and a series of second photographed images, and correcting the input image to project on the projection surface by reflecting the calculated geometric characteristics of the projection surface It includes a step.

As described above, according to the apparatus and method for correcting the projection image according to the present invention, there is an effect that can accurately model the geometrical characteristics of the projection surface.

Specific details of other embodiments are included in the detailed description and drawings, and the advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings.

However, the present invention is not limited to the embodiments disclosed below, but may be embodied in various different forms, and the present embodiments merely make the disclosure of the present invention complete and common knowledge in the technical field to which the present invention belongs. It is provided to fully inform the person having the scope of the invention, which is defined only by the scope of the claims. Like reference numerals refer to like elements throughout.

Hereinafter, the present invention will be described with reference to the drawings for a block diagram or a processing flowchart for explaining an image correction method and apparatus according to embodiments of the present invention. At this point, it will be understood that each block of the flowchart illustrations and combinations of flowchart illustrations may be performed by computer program instructions. Since these computer program instructions may be mounted on a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, those instructions performed through the processor of the computer or other programmable data processing equipment are described in the flowchart block (s). It will create a means to perform the functions.

These computer program instructions may be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular manner, and thus the computer usable or computer readable memory. It is also possible for the instructions stored in to produce an article of manufacture containing instruction means for performing the functions described in the flowchart block (s).

Computer program instructions may also be mounted on a computer or other programmable data processing equipment, such that a series of operating steps are performed on the computer or other programmable data processing equipment to create a computer-implemented process to create a computer or other programmable data. Instructions for performing the processing equipment may also provide steps for performing the functions described in the flowchart block (s).

In addition, each block may represent a portion of a module, segment, or code that includes one or more executable instructions for executing a specified logical function (s). It should also be noted that in some alternative implementations, the functions noted in the blocks may occur out of order. For example, the two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending on the corresponding function.

1 is a diagram illustrating a configuration of an apparatus (hereinafter, referred to as an image correcting apparatus) for correcting a projected image according to an exemplary embodiment. As shown, the image correcting apparatus 100 includes an encoding unit 110, a projection unit 120, a photographing unit 130, a preprocessor 140, a detector 150, a decoder 160, a matching unit ( 170, a correction value calculator 180, a storage 190, and a correction unit 195.

The encoder 110 may generate a plurality of coded images having different color patterns by coding the correction image. Here, the image for correction (hereinafter, I ₀ ) refers to an image used to correct distortion that may occur in the input image displayed on the projection surface due to the color or geometrical shape of the projection surface on which the input image is to be projected. . As illustrated in FIG. 2, the correction image I ₀ may include a plurality of white patches spaced at predetermined intervals, and the interval between the patches may be black. It can be seen that the correction image I ₀ illustrated in FIG. 2 includes 30 × 30 patches each of width × length. In addition, the center coordinates of each patch constituting the correction image I ₀ may be stored in the first data structure DS0.

On the other hand, coding image _{(I i; i = 1,2,} ..., m) is a picture which has the same geometry and the correction image (I _0), to be assigned to each patch constituting the correction image (I ₀₎ Color information for generating an identifier may be included.

Referring back to FIG. 1, the encoding unit 110 codes each patch constituting the correction image I ₀ to any one of a plurality of colors, thereby encoding a plurality of coded images I _i (i = 1, 2, ..., m), and then based on the color combinations of the patches at the same position in the plurality of coded images I _i (i = 1, 2, ..., m), the correction image I ₀ You can create an identifier for the patch. Here, for example, eight colors may be used to code the correction image I _{0. In the} following description, black, red (R), green (G), yellow (Y), and blue ( B), magenta (M), cyan (C), and white (W) will be described as an example.

Of the eight colors, black may be used as a background color of the coded image I _i (i = 1, 2, ..., m). In other words, black may be used as a color that fills the gap between the patches constituting the coded image I _i (i = 1, 2, ..., m). 3, numbers from 0 to 6 may be sequentially assigned to the remaining colors except for black, and numbers from 0 to 6 may be represented by binary numbers, respectively.

As such, when seven colors except black are used, in order to assign a unique identifier to the n patches constituting the correction image I ₀ , a total of m coded images I _i ; i = 1, 2, ... m) is required. Here, m may be determined as a minimum integer that satisfies Equation 1 below.

For example, when the correction image I ₀ includes 900 patches, since the minimum value of the integer m satisfying 7 ^m ≥ 900 is 4, a unique identifier is assigned to the 900 patches constituting the correction image I ₀ . It can be seen that a total of four coded images I ₁ , I ₂ , I ₃ , and I ₄ are required for allocation.

On the other hand, when the patches constituting the coded image are ordered from left to right to top to bottom, the color of the x-th patch in each of the coded images I ₁ , I ₂ , I ₃ , and I ₄ is lower. In Equation 2, it may be determined as a color corresponding to a ₀ , a ₁ , ..., a _m-1 , respectively. That is, the color of the x-th patch in the first coded image I ₁ is a color corresponding to a ₀ , and the color of the x-th patch in a second coded image I ₂ is a color corresponding to a ₁ , The color of the x-th patch in the third coded image I ₃ is determined to be a color corresponding to a ₂ , and the color of the x-th patch in the fourth coded image I ₄ is determined to be a color corresponding to a ₃ . Can be.

For example, the patches located 400th in each coded image I ₁ , I ₂ , I ₃ , and I ₄ may be determined as R, G, G, and G, respectively. Because 400 = 399 + 1 = 0 × 7 ⁰ + 1 × 7 ¹ + 1 × 7 ² + 1 × 7 ³ +1, where a ₀ to a ₄ are [0,1,1,1 ] And the color corresponding to the number 0 in FIG. 3 is red (R) and the color corresponding to the number 1 is green (G). As another example, the 300th patches among the patches constituting each of the coded images I ₁ , I ₂ , I ₃ , and I ₄ may be determined by C, R, W, and R, respectively. Because 300 = 299 + 1 = 5 × 7 ⁰ + 0 × 7 ¹ + 6 × 7 ² + 0 × 7 ³ +1, where a ₀ to a ₄ are [5,0,6,0 3, the number 5 is cyan (C), the color corresponding to the number 0 is red (R), and the color corresponding to the number 6 is white (W).

Based on the above Equations 1 and 2, the generated coded images I ₁ , I ₂ , I ₃ , and I ₄ are illustrated in FIG. 4. As shown in FIG. 4, the color combinations of patches existing at the same positions in the four coded images I ₁ , I ₂ , I ₃ , and I ₄ are used to determine the coded image among the patches constituting the correction image I ₀ . An identifier of a patch existing at a location corresponding to the patch may be assigned. For example, if the color combination of the 400th-positioned patches in each coded image (I ₁ , I ₂ , I ₃ , I ₄ ) is (R, G, G, G), the color combination is a correction image (I _0). ) Can be assigned as the identifier of the 400th patch. If the color combination of the 300th-positioned patches in each coded image (I ₁ , I ₂ , I ₃ , I ₄ ) is (C, R, W, R), the color combination is determined in the correction image (I ₀ ). Can be assigned to the identifier of the 300th patch.

The projector 120 may project the corrected image I ₀ and the four coded images I ₁ , I ₂ , I ₃ , and I ₄ generated by the encoder 110 in order on the projection surface.

The photographing unit 130 may photograph the correction image I ₀ and the coding images I ₁ , I ₂ , I ₃ , and I _{4 that} are sequentially projected onto the projection surface. In the following description, a corrected image photographed through the photographing unit 130 will be referred to as a first photographed image, and a coded image photographed through the photographing unit 130 will be referred to as a second photographed image. The first captured image O ₀ and the second captured images O ₁ , O ₂ , O ₃ , and O _{4 are} shown in FIGS. 5 and 6, respectively.

The preprocessor 140 may perform preprocessing on the first photographed image O ₀ and the second photographed images O ₁ , O ₂ , O ₃ , and O ₄ , respectively. First, the preprocessing process for the first captured image O ₀ will be described, and then the preprocessing process for the second captured images O ₁ , O ₂ , O ₃ , and O ₄ will be described.

The preprocessor 140 may generate a binary image B ₀ consisting of black and white by separating all patches from the first photographed image O ₀ . To this end, the pre-processing unit 140 is the first captured image (O ₀₎ set by the threshold value of the brightness necessary to remove the background and a patch in the following, first the brightness of all the pixels constituting the photographed image (O ₀₎ the threshold In comparison with the value, a binary image B ₀ may be generated by specifying a value of a pixel having a brightness greater than or equal to a threshold as white and a value of a pixel having a brightness less than or equal to a threshold as black. For example, the preprocessor 140 may generate a binary image B ₀ by applying a thresholding algorithm to the first captured image O ₀ .

Thereafter, the preprocessor 140 may filter the binary image so that the boundary of each patch becomes clear in the binary image B ₀ . To this end, the preprocessor 140 may perform template-based image processing. Here, the template-based image processing determines a value of a pixel corresponding to the first pixel in an output image based on a value of the first pixel in the original image and a value of the plurality of pixels surrounding the first pixel. Image processing method. In this case, the first pixel and the plurality of pixels surrounding the first pixel are called templates, and the template may include N × N pixels. FIG. 7 illustrates the types of templates, a template E ₄ including four pixels except for a center pixel among N × N pixels, and a template E ₈ including eight pixels except for a center pixel. A template E ₂₄ is shown that contains 24 pixels excluding the, and center pixels.

When the binary image B ₀ is filtered by applying the template-based image processing described above, an image having a clear boundary between a background and a patch may be obtained as compared to the previous binary image. 8 illustrates a binary image B ₀ and a binary image F ₀ obtained as a result of the filtering. Referring to FIG. 8, it can be seen that the binary image F ₀ obtained as a result of the filtering has a clearer boundary between the background and the patch than the binary image B ₀ before filtering. In the following description, the binary image F ₀ obtained as a result of the filtering will be referred to as a filtered image.

Referring back to FIG. 1, the preprocessing unit 140 is configured such that the second photographed images O ₁ , O ₂ , O ₃ , and O ₄ obtained through the photographing unit 130 have a uniform brightness distribution. 2 Brightness of the captured images O ₁ , O ₂ , O ₃ , and O ₄ can be corrected. To this end, the preprocessor 140 may first convert a signal format of the first captured image O ₀ from a RGB signal format to a signal format including a brightness component, for example, a YUV signal format. Then, the pre-processing unit 140 is the first captured image to determine the brightness of the (O _0), the first captured image (O ₀₎ of pixels are all, for the same brightness (for example, constituting the first captured image O ₀ The average brightness or the maximum brightness of the first photographed image O ₀ ). Thereafter, the preprocessor 140 corrects the brightness by applying the pixel-specific correction values to the corresponding pixels of the second photographed images O ₁ , O ₂ , O ₃ , and O ₄ , respectively, thereby adjusting the brightness of the second photographed images. The distribution can be made uniform.

FIG. 9 is a diagram illustrating second captured images CO ₁ , CO ₂ , CO ₃ , and CO ₄ whose brightness is corrected by the preprocessor 140. When comparing the second photographed images O ₁ , O ₂ , O ₃ , and O ₄ of FIG. 6 with the second photographed images CO ₁ , CO ₂ , CO ₃ , and CO _{4 of} FIG. 9, FIG. 6. The second photographed images O ₁ , O ₂ , O ₃ , and O ₄ illustrated in FIG. 9 have lower brightness of the outer portion than the brightness of the central portion, whereas the second captured images shown in FIG. It can be seen that CO ₁ , CO ₂ , CO ₃ , and CO ₄ ) have a uniform distribution of brightness.

Referring back to FIG. 1, the detector 150 may detect edges of patches constituting the filtered image F ₀ . The detector 150 may use arithmetic operators T ₅ and T ₉ to detect edges of each patch in the filtered image F ₀ . In this case, the arithmetic operators T ₅ and T ₉ may be determined as shown in Equation 3 below, and the arithmetic operators T ₅ and T ₉ may be set to 1 as shown in FIG. 10.

Referring back to FIG. 1, the detector 150 may detect an edge of each patch in the filtered first captured image F ₀ and then calculate a center point of each patch. In order to calculate the center coordinates of each patch, the detector 150 may track pixels corresponding to an edge of each patch. The pixel tracking tracks a pixel having a non-zero value among neighboring pixels present in left, right, up, down, top, bottom, right, and bottom directions with respect to a starting point, as shown in FIG. 11. It can proceed in a manner. In this case, the starting point may be set to a pixel having a non-zero value found when the pixels of the filtered image F ₀ are searched in the order of left to right to top to bottom. According to the above-described method, when pixels corresponding to the edge of each patch are detected, the center coordinates of the patch may be calculated based on the coordinates of the detected pixels. As such, the result of detecting the information and the center coordinate of the pixels corresponding to the boundary of each patch is shown in FIG. 12, and the detection result may be stored in the second data structure DS1.

Referring back to FIG. 1, the decoding unit 160 may extract a combination of colors of patches having the same position in the second captured images CO ₁ , CO ₂ , CO ₃ , and CO ₄ whose brightness is corrected. . Here, the extracted color combination corresponds to an identifier of each patch constituting the first captured image O ₀ .

In order to extract the color combination, the decoding unit 160 _first , the luminance difference between the R-channel image and the G-channel image for each of the second photographed images CO ₁ , CO ₂ , CO ₃ , and CO ₄ whose brightness is corrected. (ΔRG), a luminance difference ΔRB between the R-channel image and the B-channel image, and a luminance difference image indicating the luminance difference ΔGB between the G-channel image and the B-channel image.

Next, the decoding unit 160 may refer to the mapping table 200 illustrated in FIG. 13 to color the pixels constituting the corrected second captured images CO ₁ , CO ₂ , CO ₃ , and CO ₄ . You can check it. For example, when the luminance difference between each channel is calculated for a pixel, if the value of (Δ RG, ΔRB, ΔGB) is (1, 1, 0), the pixel may be identified as having a red color (R). have.

As described above, when the colors of the pixels constituting the corrected second captured images CO _i (i = 1, 2, 3, 4) are confirmed, the decoder 160 may correct the second captured images CO _i. i = 1,2,3,4) to separate the patches and generate binary images including the white patches. Next, the decoding unit 160 corresponds to each of the corrected second photographed images CO _i ; i = 1, 2, 3, and 4 by replacing the pixels corresponding to the white patches of the binary images with the colors identified above. Octal images OO _i ; i = 1, 2, 3, and 4 may be generated as shown in FIG. 14.

Thereafter, the decoder 160 determines the actual color of each patch constituting the octal images OO _i (i = 1, 2, 3, 4) with reference to the information stored in the second data structure DS1. Can be. For example, the decoder 160 may determine the color of the highest frequency among the pixels included in a patch as the actual color of the patch with reference to the second data structure DS1. Thereafter, the decoder 160 may extract a color combination of the patches existing at the same position in the octal images OO _i (i = 1, 2, 3, 4). The extracted color combination corresponds to an identifier of a patch constituting the first captured image O ₀ .

Referring back to FIG. 1, the matching unit 170 may include the first data structure DS0, the second data structure DS1, the coding images I _I , I ₂ , I _{3 , and} I ₄ , and the octal images ( OO _i : i = 1, 2, 3, and 4 may be provided to match a patch having the same identifier in the correction image I ₀ and the first captured image O ₀ . In this case, the matching relationship between patches having the same identifier may be stored in the third data structure DS2 = DS0 + DS1.

The correction value calculator 180 may calculate a correction value indicating a geometric mapping relationship between the pixels of the correction image I ₀ and the pixels of the first photographed image O ₀ with reference to the third data structure DS2. .

Specifically, when the coordinate of the pixel of the correction image (I ₀ ) is referred to as (x _i , y _i ) and the coordinate of the pixel of the first photographed image is referred to as (x _o , y _o ), mapping between the pixels in the two images is performed. The relationship may be modeled as in Equation 4. In Equation 4, a, b, c, d, e, f, g, and h are model constants.

와 제1 촬영 영상을 구성하는 패치 중 상기 4개의 패치에 대응하는 패치들의 중심좌표

간의 관계는 [수학식 5]와 같이 나타낼 수 있다. Meanwhile, the center coordinates of four patches adjacent to each other among the patches constituting the correction image.

And central coordinates of patches corresponding to the four patches among the patches constituting the first captured image.

The relationship can be expressed as shown in [Equation 5].

However, Equation 5 may be expressed in a matrix form as shown in Equation 6 below.

In Equation 6, T and B may be obtained from the third data structure DS2. Therefore, A may be calculated through Equation 7 below. Each element of A is a correction value representing a geometric mapping relationship between the pixels of the correction image I ₀ and the pixels of the first photographed image O ₀ , and performs geometric correction on the input image to be projected on the projection surface. It can be used to

The storage unit 190 may store the correction value calculated by the correction value calculator 180 in the form of a look up table (LUT). The storage unit 190 may be implemented as at least one of a nonvolatile memory such as a flash memory, a volatile memory, and a storage medium such as a hard disk.

The correction unit 195 may perform geometric correction by referring to the lookup table when an input image to be projected on the projection surface is provided. The corrected input image may be projected onto the projection surface through the projection unit 120. In addition, the correction unit 195 may also perform color correction on the input image. Since the technology for color correction is a known technique, a detailed description thereof will be omitted.

Next, FIG. 15 is a flowchart illustrating an image calibrating method according to an exemplary embodiment.

First, the encoding unit 110 may generate a plurality of coded images I ₁ , I ₂ , I ₃ , and I ₄ having the same geometrical structure as the correction image I ₀ and having different color patterns (S310). ). The number of coded images may be determined by the number of patches constituting the correction image I ₀ . The coded image may be generated using eight colors including, for example, black and white. In this case, black may be used as a background color of the coded image, and each patch constituting the coded image may be coded in any one of seven colors except for black.

When a plurality of coded images are generated, the encoding unit 110 configures an image I ₀ for correcting color combinations of patches having the same position in the plurality of coded images I ₁ , I ₂ , I ₃ , and I ₄ . It can be assigned to the identifier of each patch (S320). Before generating the coded image, the center coordinates of each patch constituting the correction image I ₀ may be stored in the first data structure DS0.

The correction image I ₀ and the plurality of coded images I ₁ , I ₂ , I ₃ , and I ₄ may be sequentially projected onto the projection surface through the projection unit 120 (S330).

The photographing unit 130 may photograph a correction image and a plurality of coded images that are sequentially projected onto the projection surface. As a result, the photographing unit 130 is obtained by photographing the first photographed image O ₀ and the plurality of coded images I ₁ , I ₂ , I ₃ , and I ₄ obtained by photographing the correction image I ₀ , respectively. A plurality of second photographed images O ₁ , O ₂ , O ₃ , and O ₄ may be provided (S340).

Thereafter, the detector 150 may detect an edge and a center point of each patch in the first photographed image O ₀ (S350). The step S350 may include generating a binary image B ₀ consisting of white and black by separating each patch from the first photographed image O ₀ , and filtering the binary image B ₀ to filter the binary image O. and generating a filtered image (F ₀₎ became boundary is clear between the patch and the background, and a step of detecting the edge of each patch in the filtered image (F ₀₎ as compared to _0), each on the basis of the detected edge information Calculating the center coordinates of the patch. Here, the edge information of each patch and the coordinates of the center point may be stored in the second data structure DS1.

Meanwhile, the decoder 160 may extract the color combinations of the patches at the same position from the plurality of second photographed images O ₁ , O ₂ , O ₃ , and O ₄ (S360). In step S360, the brightness correction values used to uniformize the brightness distribution of the first photographed image O ₀ are applied to the plurality of second photographed images O ₁ , O ₂ , O ₃ , and O ₄ , respectively. Correcting the brightness of the second photographed images O ₁ , O ₂ , O ₃ , and O ₄ , and for each of the plurality of second corrected images CO ₁ , CO ₂ , CO ₃ , and CO ₄ whose brightness is corrected. Generating a plurality of luminance difference images representing luminance differences between channels, and detecting a color of a pixel of the second captured images CO ₁ , CO ₂ , CO ₃ , and CO ₄ from the plurality of luminance differences images. Generating a plurality of binary images including white patches by separating patches from the second photographed images CO ₁ , CO ₂ , CO ₃ , and CO ₄ ; Generating a plurality of octal images OO ₁ , OO ₂ , OO ₃ , and OO ₄ by replacing corresponding pixels with the detected color. The method may include extracting a color combination of the patches having the same position from the plurality of octal images OO ₁ , OO ₂ , OO ₃ , and OO ₄ . The extracted color combination corresponds to an identifier of a patch constituting the first captured image O ₀ .

Correcting the brightness of the second photographed image (O ₁ , O ₂ , O ₃ , O ₄ ) during the process, converts the signal format of the first photographed image (O ₀ ) to a signal format including a brightness component to have a first captured image and (O ₀₎ confirming the brightness of the first photographed image for each pixel, the same brightness (O ₀₎ (for example, the first maximum brightness or the average brightness of the photographed image) Calculating a correction value necessary for correction so that the correction is performed for each pixel, and applying the correction value for each pixel to a corresponding pixel of the second captured images O ₁ , O ₂ , O ₃ , and O ₄ . And correcting the brightness of (O ₁ , O ₂ , O ₃ , O ₄ ).

Next, the matching unit 170 may match patches having the same identifier in the correction image I ₀ and the first photographed image O ₀ (S370). The matching information may be stored in the third data structure DS2.

Subsequently, the correction value calculator 180 refers to the third data structure DS2 and calculates a correction value indicating a geometric mapping relationship between the pixels of the correction image I ₀ and the pixels of the first photographed image O ₀ . Can be calculated (S380). The calculated correction value may be stored in the storage 190 in the form of a lookup table.

Meanwhile, when the input image to be projected on the projection surface is provided, the correction unit 195 may perform geometric correction on the input image with reference to the lookup table (S390). The corrected image may be projected onto the projection surface through the projection unit 120.

Next, an image correction apparatus 400 and a method according to another embodiment of the present invention will be described with reference to FIGS. 16 and 17.

FIG. 16 is a diagram illustrating a configuration of an image correcting apparatus 400 according to another exemplary embodiment. As illustrated, the image calibrating apparatus 400 may include a projection unit 420, a photographing unit 430, a preprocessor 440, a detector 450, a first decoding unit 460-1, and a second decoding unit ( 460-2), a first storage unit 410, a second storage unit 490, a matching unit 470, a correction value calculator 480, and a correction unit 495. In FIG. 16, the projection unit 420, the photographing unit 430, the preprocessor 440, the detector 450, the matching unit 470, the correction value calculator 480, and the correction unit 495 are exemplary embodiments. Since the second decoding unit 460-2 and the second storage unit 490 are the same as the decoding unit 160 and the storage unit 190 according to an exemplary embodiment, a description of overlapping portions will be described. Will be omitted and will be described with reference to the first storage unit 410 and the first decoding unit 460-1.

The first storage unit 410 stores the correction image I ₀ and the coding images I ₁ , I ₂ , I ₃ , and I ₄ necessary to generate an identifier of each patch constituting the correction image I ₀ . Can be stored. The first storage unit 410 may be implemented independently of the second storage unit 490 in hardware, or may be implemented in a merged form with the second storage unit 490.

The first decoding unit 460-1 may extract a combination of colors of a patch at the same position from the coded images I ₁ , I ₂ , I ₃ , and I ₄ stored in the first storage unit 410. . The extracted color combinations may be allocated as identifiers of the patches constituting the correction image I ₀ .

17 is a flowchart illustrating an image correction method according to another embodiment of the present invention.

First, the first decoder 460-1 may extract color combinations of patches having the same location from the coded images I ₁ , I ₂ , I ₃ , and I ₄ stored in the first storage unit 410. There is (S510). The extracted color combinations correspond to identifiers of respective patches constituting the correction image I ₀ (S520).

Thereafter, when the correction image I ₀ and the coding images I ₁ , I ₂ , I ₃ , and I ₄ are sequentially projected onto the projection surface through the projection unit 420 (S530), the photographing unit 430 may taken of the projected calibration image and the coded picture, and recording the first recording video image photographing a calibration image (I ₀₎ (O ₀₎ and the coded image _{(I 1, I 2, I} 3, I 4) of claim 2 The captured images O ₁ , O ₂ , O ₃ , and O ₄ may be acquired (S540).

Thereafter, the detection unit 450 may detect edges and center coordinates of each patch constituting the first captured image O ₀ and store them in the second data structure DS1 (S550). The detecting may include generating a binary image B ₀ by separating each patch from the first photographed image O ₀ , and generating a filtered image F ₀ by filtering the binary image B ₀ . And detecting edge and center coordinates of each patch in the filtered image F ₀ .

Meanwhile, the second decoding unit 460-2 may detect the color combination of the patch at the same position in the second captured images O ₁ , O ₂ , O ₃ , and O ₄ (S560). The detecting of the color combination may include correcting the brightness of the second captured images O ₁ , O ₂ , O ₃ , and O ₄ , and correcting the brightness of the second images CO ₁ , CO ₂ , and the like. Generating a plurality of octal images (OO ₁ , OO ₂ , OO ₃ , OO ₄ ) corresponding to CO ₃ , CO ₄ ), from the octal images (OO ₁ , OO ₂ , OO ₃ , OO ₄ ) The method may include detecting a color combination of the patches at the same location. In this case, the detected color combination corresponds to an identifier of each patch constituting the first captured image O ₀ .

The matching unit 470 may match a patch having the same identifier in the correction image I ₀ and the first photographed image O ₀ (S570). The matching information may be stored in the third data structure DS2 = DS0 + SD1.

Thereafter, the correction value calculator 480 may calculate, for each pixel, a correction value indicating a geometric mapping relationship between the correction image I ₀ and the first photographed image O ₀ with reference to the third data structure DS2. There is (S580). The correction value may be stored in the second storage unit 490 in the form of a lookup table.

Finally, the correction unit 495 may perform geometric correction on the input image to be projected on the projection surface by referring to the lookup table. The corrected input image may be projected through the projector 420.

Although the apparatus and method for correcting the projection image according to the present invention have been described with reference to the drawings illustrated as above, the present invention is not limited to the embodiments and drawings disclosed herein, and the technical spirit of the present invention. Of course, various modifications may be made by those skilled in the art within the scope.

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

2 is a diagram illustrating an image for correction according to an embodiment of the present invention.

3 illustrates coding and decoding principles according to an embodiment of the present invention.

4 is a diagram illustrating coded images according to an embodiment of the present invention.

5 is a diagram illustrating a first captured image according to an embodiment of the present invention.

6 is a diagram illustrating second captured images according to an exemplary embodiment of the present invention.

7 is a diagram illustrating a template that can be applied to an embodiment of the present invention.

8 illustrates a binary image and a binary image obtained as a result of filtering according to an exemplary embodiment of the present invention.

9 is a diagram illustrating second captured images whose brightness is corrected according to an exemplary embodiment of the present invention.

10 illustrates arithmetic operators T ₅ and T ₉ according to an embodiment of the present invention.

11 is a diagram illustrating a boundary tracking method of a patch according to an embodiment of the present invention.

12 is a diagram illustrating a result of detecting edge and center coordinates of each patch according to an embodiment of the present invention.

FIG. 13 is a diagram illustrating a mapping table required to confirm colors of pixels of a second captured image whose brightness is corrected according to an exemplary embodiment of the present invention.

14 illustrates octal images according to an embodiment of the present invention.

15 is a diagram illustrating an image correction method according to an embodiment of the present invention.

16 is a diagram illustrating a configuration of an image correction device according to another embodiment of the present invention.

17 is a view showing an image correction method according to another embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

100, 400: image correction device 110: coding unit

120, 420: projection unit 130, 430: recording unit

140, 440: preprocessor 150, 450: detector

160: decoding unit 170, 470: matching unit

180, 480: correction value calculation unit 190: storage unit

195 and 495: correction unit 410: first storage unit

460-1: First decoding unit 460-2: Second decoding unit

Claims (24)

A projection unit for projecting a correction image and a series of coded images having different color patterns onto a projection surface; A correction value calculator configured to calculate geometrical characteristics of the projection surface based on the first photographed image and the second photographed image obtained by photographing the projected correction image and the series of coded images; And And a correction unit configured to correct an input image to be projected onto the projection surface by reflecting the calculated geometric characteristics of the projection surface. The method of claim 1, And the correcting image and the series of coded images have the same geometrical structure. The method of claim 1, The image for correction includes a plurality of white patches spaced at predetermined intervals, and the interval is black. The method of claim 1, Each patch constituting the correction image is assigned a first identifier according to a color combination of patches having the same position in the series of coded images. The method of claim 4, wherein And an encoder configured to generate the series of coded images such that the color combinations are unique. The method of claim 4, wherein A preprocessor configured to correct brightness of the series of second captured images based on the brightness of the first captured image; A decoding unit extracting a second identifier of each patch constituting the first captured image based on the second captured image whose brightness is corrected; And And a matching unit matching a patch having the same identifier as the first identifier and the second identifier. The method of claim 6, And the second identifier is a combination of colors of patches having the same position in the second series of captured images of which brightness is corrected. The method of claim 6, And the correction value calculator calculates a correction value for correcting the input image based on at least one of center coordinates and edge information between the matched patches. The method of claim 4, wherein A storage unit which stores the correction image and the series of coded images; A first decoding unit to extract the first identifier; A preprocessor configured to correct brightness of the series of second captured images based on the brightness of the first captured image; A second decoding unit extracting a second identifier of each patch constituting the first captured image based on the second captured image whose brightness is corrected; And And a matching unit which matches the patch with the same first identifier and the second identifier. The method of claim 1, The series of coded images is m, where m is

And n is a total number of patches included in the correction image, and Y is a number of colors of patches in the series of coded images. The method of claim 10, In the series of coded images, the colors of the xth patches are

Is an image correction device determined by the color corresponding to a _i (i = 0, 1, 2, ..., m-1). The method of claim 1, And a photographing unit configured to photograph the projected correction image and a series of coded images. Projecting a correction image and a series of coded images having different color patterns onto a projection surface; Calculating geometric characteristics of the projection surface based on the first photographed image and the second photographed image obtained by photographing the projected correction image and the series of coded images; And Correcting the input image to be projected onto the projection surface by reflecting the calculated geometrical characteristics of the projection surface. The method of claim 13, And the correction image and the series of coded images have the same geometrical structure. The method of claim 13, The image for correction includes a plurality of white patches spaced at predetermined intervals, wherein the interval is black image correction method. The method of claim 13, And assigning a color combination of patches of the same position in the series of coded images to the first identifier of each patch constituting the correction image. The method of claim 16, Generating the series of coded images such that the color combinations are unique. The method of claim 16, Correcting brightness of the series of second captured images based on the brightness of the first captured image; Extracting a second identifier of each patch constituting the first captured image based on the second captured image whose brightness is corrected; And And matching a patch having the same identifier as the first identifier and the second identifier. The method of claim 18, And the second identifier is a combination of colors of patches having the same position in the series of second photographed images whose brightness is corrected. The method of claim 18, Calculating the geometric characteristics of the projection surface, And calculating a correction value for correcting the input image, based on at least one of center coordinates and edge information between the matched patches. The method of claim 16, Storing the correction image and the series of coded images; Extracting the first identifier; Correcting brightness of the series of second captured images based on the brightness of the first captured image; Extracting a second identifier of each patch constituting the first captured image based on the second captured image whose brightness is corrected; And And matching a patch having the same identifier as the first identifier and the second identifier. The method of claim 13, The series of coded images is m, where m is

And n is a total number of patches included in the correction image, and Y is a number of colors of patches in the series of coded images. The method of claim 22, In the series of coded images, the colors of the xth patches are

Is determined by the color corresponding to a _i (i = 0, 1, 2, ..., m-1). The method of claim 13, And photographing the projected correction image and the series of coded images.

Images