KR101491334B1 - Apparatus and method for detecting color chart in image - Google Patents

Abstract

The present invention relates to a technique for detecting a color chart from an input image and extracting color values for respective color patches for the detected color chart. A color patch detecting apparatus comprises: a detecting unit which detects primary corner points for each of a detailed color patches by using an algorithm for detecting corner points from a color chart image; a selecting unit which selects primary reference corner points of a reference color patch corresponding to the primary corner points; and a calculating unit which performs a primary calculation on a relation between a first reference matrix for coordinates of the primary reference corner points and a first target matrix for coordinates of the primary corner points to extract coordinates of the detailed color patches.

Description

TECHNICAL FIELD [0001] The present invention relates to a color chart detecting method and apparatus,

To an apparatus and method for detecting a color chart in an input image and extracting a color value of each color patch for the detected color chart.

In a typical digital imaging apparatus, an optical image sensor receives light input through an optical device such as a lens, a diaphragm, and a shutter, converts the light into a predetermined electrical signal, and generates a digital image.

In the process of generating a digital image, the color value is determined by various causes such as characteristics of an image sensor, characteristics of an optical part such as a lens, illumination conditions, and a preprocessing process of a camera.

It is possible to cause mutual influences in determining the color values, and problems may arise due to the presence of elements of different color channels in each color channel. In addition, the reproducibility with respect to color may be lowered, which may result in deterioration of image quality.

In order to solve such a problem, there is a need for a technique of comparing and correcting a reference color value and a color value in an actually obtained image to perform an accurate color analysis.

Therefore, there is a need for a technique of detecting a color chart in an input image and extracting color values of each color patch for the detected color chart.

According to one aspect, a detection unit for detecting a first corner point of each of the detail color patches using an algorithm for corner point detection in a color chart image; A selector for selecting a first reference corner point of a reference color patch corresponding to the first corner point; And generating a first matrix for extracting the coordinates of the detail color patch by performing a first operation between a first reference matrix for the coordinates of the first reference corner point and a first object matrix for the coordinates of the first corner point There is provided a color patch detecting apparatus including a calculating section.

According to one embodiment, the algorithm may be at least one of an algorithm using eigenvalues and a corner detection algorithm.

According to another embodiment, the selector classifies the corner type of the first corner point, and the first corner type includes a first corner type at the upper left corner of the detail color patch, a second corner type at the upper right corner of the detail color patch, An edge type, a lower right third corner type of the detail color patch, and a lower left fourth corner type of the color patch.

According to another embodiment, the edge type may be determined by comparing the intensity value of the pixel position in a predetermined direction with respect to the first corner point.

According to one embodiment, the first operation may be performed using a method of least squares.

According to another embodiment, the first reference corner point may be selected as a point having a high degree of similarity through a distance comparison with the first corner point.

According to yet another embodiment, the distance comparison may comprise a first distance between the first corner point and the first reference corner point, and a second distance between the color value of the first corner point and the color value of the first reference corner point It is possible to determine that the degree of similarity is large.

The color value of the corner point may be a color value at any point in the color patch considering the color value or corner type at the corner point position.

According to another embodiment, the detecting unit may convert the coordinates of the first reference corner point to each second reference corner point using the first matrix, and the selector may correspond to each of the secondary reference corner points And the calculation unit performs a second calculation between the second reference matrix for the coordinates of the second reference corner point and the second object matrix for the coordinates of the second corner point in the color chart image To generate a second matrix for extracting the coordinates of the detail color patches.

According to another embodiment, the second operation may be performed using a method of least squares.

According to an exemplary embodiment, the secondary corner point may be selected as a point having a high degree of similarity through distance comparison with the secondary reference corner point.

According to another embodiment, the distance comparison is made between a third distance between the second corner point and the second reference corner point and between a color value of the second corner point and a color value of the second reference corner point When the sum of the fourth distances is minimum, it can be determined that the degree of similarity is large.

According to an embodiment of the present invention, there is further provided an extraction unit for generating a first coordinate value for the color chart image by calculating a matrix for an arbitrary reference coordinate in the order of the first matrix and the second matrix, can do.

According to another aspect, there is provided a color patch unit for detecting a detailed color patch in a color chart image and extracting a color value of the detailed color patch; A color conversion matrix generator for generating a first color conversion matrix by performing a third operation on a color value of the reference color patch and a color value of the detailed color patch; And a color correction application unit for correcting the color value of the color patch by an operation of multiplying the color conversion matrix by the first color conversion matrix.

According to one embodiment, the color patch section comprises: a detector for detecting a first corner point of each of the detail color patches using an algorithm for corner point detection in a color chart image; A selector for selecting a first reference corner point of a reference color patch corresponding to the first corner point; And generating a first matrix for extracting the coordinates of the detail color patch by performing a first operation between a first reference matrix for the coordinates of the first reference corner point and a first object matrix for the coordinates of the first corner point And a calculation unit.

According to another embodiment, the first operation may be performed using a method of least squares.

According to one embodiment, the third operation may be performed using a method of least squares.

According to another embodiment, the primary reference corner point may be selected as a point of high similarity through distance comparison with the primary corner point.

According to one embodiment, the distance comparison may be based on a first distance between the first corner point and the first reference corner point, and a second distance between the color value of the first corner point and the color value of the first reference corner point. When the sum of the two distances is minimum, it can be determined that the degree of similarity is large.

According to another embodiment, the apparatus may further include a color chart detector for detecting the input image as a color chart image.

According to another aspect, a method includes detecting a first corner point of each of a detailed color patch using an algorithm for edge point detection in a color chart image, Selecting a first reference corner point of a reference color patch corresponding to the first corner point; And a calculation unit for performing a first calculation between a first reference matrix for the coordinates of the first reference corner point and a first object matrix for the coordinates of the first corner point to extract coordinates of the detail color patch, A color patch detection method is provided.

According to one embodiment, the first operation may be performed using a method of least squares.

According to another embodiment, the first reference corner point may be selected as a point having a high degree of similarity through a distance comparison with the first corner point.

According to yet another embodiment, the distance comparison may comprise a first distance between the first corner point and the first reference corner point, and a second distance between the color value of the first corner point and the color value of the first reference corner point It is possible to determine that the degree of similarity is large.

According to another aspect, there is provided a computer-readable recording medium having recorded thereon a program for performing the manufacturing method.

1 is a block diagram showing a configuration of a color patch detecting apparatus according to an embodiment.
2 is a flowchart showing a procedure of a color patch detection method according to an embodiment.
Fig. 3 is an exemplary view for explaining the classification of edge types according to one embodiment. Fig.
FIG. 4 is a flowchart illustrating in detail a procedure for selecting reference corner points in FIG. 2 according to one embodiment.
5 is an exemplary diagram illustrating extracting color values of each color patch, according to one embodiment.
6 is a block diagram illustrating a configuration of an image color correction apparatus according to an exemplary embodiment.
7 is a flowchart illustrating a procedure of an image color correction method according to an exemplary embodiment.

In the following, some embodiments will be described in detail with reference to the accompanying drawings. However, it is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements.

Although the terms used in the following description have selected the general terms that are widely used in the present invention while considering the functions of the present invention, they may vary depending on the intention or custom of the artisan, the emergence of new technology, and the like.

Also, in certain cases, there may be terms chosen arbitrarily by the applicant for the sake of understanding and / or convenience of explanation, and in this case the meaning of the detailed description in the corresponding description section. Therefore, the term used in the following description should be understood based on the meaning of the term, not the name of a simple term, and the contents throughout the specification.

1 is a block diagram showing a configuration of a color patch detecting apparatus according to an embodiment.

The color patch detecting apparatus 100 may include a detecting unit 110, a sorting unit 120, a calculating unit 130, and an extracting unit 140.

The color patch detecting apparatus 100 can perform a process of finding a corner point of a detailed color patch in a color chart area extracted at a specific resolution.

According to an embodiment, the color patch detecting apparatus 100 may calculate coordinate values of an input image by calculating coordinate values of a matrix H and a reference color patch. That is, the matrix H is a matrix for calculating a coordinate value of the input image corresponding to a certain coordinate value of the reference color patch.

According to an embodiment, it is possible to know which point corresponds to the first image by selecting any point in the reference coordinate system and performing an operation with the matrix H.

According to one embodiment, the matrix H is expressed by Equation (1).

The matrix H can be obtained through the condition (affine homography) of h ₃₁ = h ₃₂ = 0 and h ₃₃ = 1 in the above equation (1).

The position of each corner point of the reference color patch can be known in advance. It is possible to capture each corner of the reference color patch in units of pixels by taking a photograph on a color chart drawing dimension, an actual measurement, or an image.

According to one embodiment, a color chart in the form of a square-shaped detailed color patch array as well as a Macbeth chart can be applied to the present invention. In the case of a Macbeth chart, there are a total of 24 patches of detail color, so 96 edge point positions can be set as reference positions.

The matrix H can be obtained by knowing the pair of the reference corner point and the corner point of the input image corresponding to the reference point.

According to one embodiment, since affine homography has six unknowns, it is possible to obtain values of h ₁₁ to h ₂₃ by knowing at least three pairs of corresponding points. The unknowns can be obtained by a least-squares method.

According to an embodiment, the color patch detecting apparatus 100 may obtain a corner coordinate transformation matrix through a first process (a process of obtaining a first matrix) and a second process (a process of obtaining a second matrix). The first and second processes will be described later in the detailed description of FIG.

The detection unit 110 can detect the first corner points of each of the detailed color patches using an algorithm for corner point detection in the color chart image.

According to one embodiment, the color chart image may be related to the input image input to the color patch detecting apparatus 100. [ The color chart image may be at least one of a gray scale image of the image obtained as a result of the color chart detection, a specific channel image among the color channels, and the color channel value difference image.

According to one embodiment, a particular one of the color channels may be a channel for R, a channel for G, and a channel for B in an RGB color coordinate system.

According to an exemplary embodiment, the color channel difference image may be an image for R - G, an image for R - B, and an image for G - B.

Typically, the color space of an image that is the basis of image processing is expressed in various spaces such as RGB, CMYK, HS series, CIE and Y series according to the viewpoint of the color space such as the viewpoint of color mixing, . It is obvious to those skilled in the art that the conversion between color spaces expressed in a heterogeneous form can be converted by a simple mathematical conversion formula. Therefore, in the following description of the present invention, It can be said that it is applicable as it is.

According to one embodiment, the algorithm may be at least one of an algorithm using an eigenvalue and a corner detection algorithm.

According to another embodiment, it is possible to set such that only strong edge points can be detected through internal parameter setting of the algorithm.

The selector 120 may select a first reference corner point of a reference color patch corresponding to the first corner point.

The sorting unit 120 may classify the corner type of the first corner point. Wherein the first edge type includes a first corner type of a left upper corner of the detail color patch, a second corner type of a right upper corner of the detail color patch, a third corner type of a lower right corner of the detail color patch, And a corner type.

The corner type may be determined by comparing intensity values of pixel positions in a predetermined direction with respect to the primary corner point.

The calculation unit 130 performs a first calculation between the first reference matrix for the coordinates of the first reference corner point and the first object matrix for the coordinates of the first corner point to extract coordinates of the detailed color patches 1 < / RTI > matrix.

The first calculation may be performed using a method of least squares.

The primary reference corner point may be selected as a point having a high degree of similarity through a distance comparison with the primary corner point.

Wherein the distance comparison is performed such that a sum of a first distance between the primary corner point and the primary reference corner point and a second distance between a color value of the primary corner point and a color value of the primary reference corner point is at least , It can be determined that the degree of similarity is large.

The color value of the corner point may be a color value at any point in the color patch considering the color value or corner type at the corner point position.

According to one embodiment, the detection unit 110 may transform the coordinates of the primary reference corner point to the secondary reference corner point using the first matrix, respectively.

The selector 120 may select secondary corner points corresponding to the secondary reference corner points from the color chart image.

The calculation unit 130 performs a second calculation between the second reference matrix for the coordinates of the secondary reference corner point and the second object matrix for the coordinates of the secondary corner point to extract the coordinates of the detailed color patch A second matrix can be generated.

The second operation may be performed by using a method of least squares.

The secondary corner point may be selected to have a high degree of similarity through distance comparison with the secondary reference corner point.

Wherein the distance comparison is performed such that a sum of a third distance between the secondary corner point and the secondary reference corner point and a fourth distance between the color value of the secondary corner point and the color value of the secondary reference corner point is at least , It can be determined that the degree of similarity is large.

The extracting unit 140 may calculate a first coordinate value for the color chart image and a color value by operating the matrix for arbitrary reference coordinates and the matrix H, The matrix H is obtained by sequentially calculating the first matrix H1 and the second matrix H2.

2 is a flowchart showing a procedure of a color patch detection method according to an embodiment.

The color patch detection method can obtain a corner coordinate transformation matrix through a first step and a second step.

According to one embodiment, the first step is as follows. The color patch detecting apparatus 100 may detect a first corner point (210). The color patch detecting apparatus 100 may select the first corner point corresponding to the first corner point (220). The first matrix, which is the first color chart corner coordinate transformation matrix H1, can be obtained 230 using the pair of corresponding points. The first matrix H1 may be obtained by least squares.

Step 210 is a step of detecting the first corner point.

According to one embodiment, the detection unit 110 may detect the first corner points of each of the detail color patches using an algorithm for edge point detection in the color chart image.

Step 220 is to find the primary reference corner point corresponding to the primary corner point.

According to one embodiment, the corner type can be determined for each primary corner point in the input image. A set of points with the same corner type for each of the primary corner points can be found at the reference corner points. The corner points at which the similarity expression becomes minimum among the set elements can be determined as corresponding points.

The edge type can be determined by comparing the intensity value of the pixel position in a predetermined direction with respect to the primary corner point.

The similarity equation is shown in Equation (2).

는 두 점의 좌표값 사이 거리이다. 상기 좌표값 사이 거리는 유클리드(Euclidean) 또는 맨하탄거리(Manhattan distance)를 이용하여 구할 수 있다. 상기 (x_r , y_r)은 레퍼런스 색상 패치 좌표이다. 상기 (x_i , y_i)는 입력 영상의 좌표이다.

Is the distance between the coordinate values of the two points. The distance between the coordinate values can be obtained using Euclidean or Manhattan distance. (X _r , y _r ) is a reference color patch coordinate. (X _i , y _i ) is the coordinates of the input image.

는 두 컬러값 사이의 거리이다. 상기 컬러값 사이의 거리는 유클리드(Euclidean) 또는 다른 색차 메트릭(color difference metric)을 이용하여 구할 수 있다. 상기 color_r은 레퍼런스 색상 패치 좌표에 대한 컬러값이다. 상기 color_i는 입력 영상 좌표에 대한 컬러값이다.

Is the distance between the two color values. The distance between the color values can be obtained using Euclidean or another color difference metric. The color _r is a color value for the reference color patch coordinates. The color _i is a color value for the input image coordinates.

Step 230 is a step of obtaining a first-order matrix using a pair of corresponding points. The primary matrix is the primary color chart edge coordinate transformation matrix H1.

According to one embodiment, the first order matrix may be obtained by least squares.

According to one embodiment, the second step is as follows. The color patch detecting apparatus 100 may detect a second corner point (240). The color patch detection apparatus 100 may select secondary edge points corresponding to secondary reference edge points (250). A second matrix, which is a second color chart corner coordinate transformation matrix (H2), can be obtained (260) using a pair of corresponding points. The second matrix H2 can be obtained by a least squares method.

Step 240 is a step of detecting a second corner point.

According to one embodiment, the color chart image input to the color patch detecting apparatus 100 for secondary edge point detection may be a gray scale image image of the image obtained as a result of color chart detection, And a difference image of the color channel value.

According to one embodiment, the secondary corner point may be detected in a manner similar to the primary corner point. In addition, a larger number of corner points can be obtained through internal parameter adjustment of the corner point detection algorithm.

Step 250 is a step of detecting a secondary corner point corresponding to the secondary reference corner point.

According to an exemplary embodiment, the primary color chart corner coordinate transformation matrix H1 may be used to coordinate the primary reference corner point to a secondary reference corner point.

According to an embodiment, a set of corner points in the input image having the same corner type for each of the secondary corner points can be found at the secondary reference corner point.

The point at which the similarity expression is minimized among the elements of the set can be determined as a corresponding point. The similarity equation is shown in Equation (3).

는 두 점의 좌표값 사이 거리이다. 상기 좌표값 사이 거리는 유클리드(Euclidean) 또는 맨하탄거리(Manhattan distance)를 이용하여 구할 수 있다.

는 2차 레퍼런스 모서리 점에 대한 좌표값이다.

는 2차 모서리 점에 대한 좌표값이다.

Is the distance between the coordinate values of the two points. The distance between the coordinate values can be obtained using Euclidean or Manhattan distance.

Is the coordinate value for the second reference corner point.

Is the coordinate value for the second corner point.

는 두 컬러값 사이의 거리이다. 상기 컬러값 사이의 거리는 유클리드(Euclidean) 또는 다른 색차 메트릭(color difference metric)을 이용하여 구할 수 있다. 상기 color_r은 레퍼런스 색상 패치 좌표에 대한 컬러값이다. 상기 color_i는 입력 영상 좌표에 대한 컬러값이다.

Is the distance between the two color values. The distance between the color values can be obtained using Euclidean or another color difference metric. The color _r is a color value for the reference color patch coordinates. The color _i is a color value for the input image coordinates.

Step 260 is a step of obtaining two matrices using a pair of corresponding points. The second matrix is a second color chart edge coordinate transformation matrix (H2).

According to one embodiment, the quadratic matrix can be found by least squares.

Step 270 is a step of extracting a color value of the color chart image. The extracting unit 140 can generate a first coordinate value for the color chart image and extract a color value by operating a matrix and a matrix H for a certain reference coordinate. The matrix H is obtained by sequentially calculating the first matrix H1 and the second matrix H2.

The matrix H is expressed by Equation (4).

Fig. 3 is an exemplary view for explaining the classification of edge types according to one embodiment. Fig.

According to one embodiment, the selector 120 may select a first reference corner point of a reference color patch corresponding to a first corner point.

According to another embodiment, the selector 120 can select secondary corner points corresponding to each of the secondary reference corner points from the color chart image.

According to one embodiment, the sorting unit 120 classifies a corner type of a first corner point, and the first corner type is a first corner type of a left upper corner of the detail color patch, a right corner of the detail color patch Two corner types, a lower right third corner type of the detail color patch, and a lower left fourth corner type of the color patch.

The corner type may be determined by comparing intensity values of pixel positions in a predetermined direction with respect to the primary corner point.

3, only one of the brightness values of positions 1 to 4 may exist in the color patch. The remaining three places can belong to the background color. Accordingly, it is possible to classify the edge type by determining which position has the highest brightness value. In the case of FIG. 3, if it is determined that the lower left portion of the color patch is the brightest, it may correspond to the fourth edge type.

FIG. 4 is a flowchart illustrating in detail a procedure for selecting reference corner points in FIG. 2 according to one embodiment.

Step 410 is a step of determining a corner type for a corner point of the color patch detecting apparatus 100. [

According to one embodiment, the sorting unit 120 classifies the types of primary edge points, wherein the primary edge type is a first upper edge type of the detail color patch, a second upper edge corner of the detail color patch Type, a lower right third corner type of the detail color patch, and a left lower fourth corner type of the color patch.

The corner type may be determined by comparing brightness values of pixel positions in a predetermined direction with respect to the primary corner point.

Step 420 is a step for finding a point having the same corner type with respect to the corner point of the color patch.

According to one embodiment, the color chart image may have a plurality of color patches. The selector 120 can find points with the same corner type at the reference corner points.

Step 430 is a step of determining the degree of similarity and selecting a reference corner point corresponding to the corner point.

According to an exemplary embodiment, the first-order reference corner point can be selected as a point having a high degree of similarity through the distance comparison with the first-order corner point.

The distance comparison can determine that the degree of similarity is large when the sum of the first distance and the second distance becomes minimum. The first distance is the distance between the primary corner point and the primary reference corner point. The second distance is the distance between the color value of the primary corner point and the color value of the primary reference corner point.

5 is an exemplary diagram illustrating extracting color values of each color patch, according to one embodiment.

510 is an exemplary diagram showing reference edges for a reference color patch. 520 is an exemplary view showing a reference edge for a color patch of an input image.

According to one embodiment, the coordinates of the detail color patch can be extracted by performing a first operation between the first reference matrix for the coordinates of the first reference corner point and the first object matrix for the coordinates of the first corner point.

According to another embodiment, the coordinates of the detailed color patch can be extracted by performing a second operation between the second reference matrix for the coordinates of the second reference corner point and the second object matrix for the coordinates of the second corner point.

The extracting unit 140 can generate a first coordinate value for the color chart image and extract a color value by operating a matrix and a matrix H for a certain reference coordinate. The matrix H is obtained by sequentially calculating the first matrix H1 and the second matrix H2.

Equation (5) for generating the first coordinate value is as follows.

The [reference coordinates] are matrices for arbitrary reference coordinates. The [first coordinate value] is a matrix obtained by calculating a matrix for the arbitrary reference coordinate and a matrix (H).

6 is a block diagram illustrating a configuration of an image color correction apparatus according to an exemplary embodiment.

The image color correction apparatus 600 may include a color patch unit 610, a color conversion matrix generator 620, a color correction application unit 630, and a color chart detector 640.

The color patch unit 610 may detect a color patch in a color chart image of the input image and extract a color value of the color patch.

The description of the color patch unit 610 is the same as that described in FIGS. 1 to 5.

The color conversion matrix generation unit 620 may generate a first color conversion matrix by performing a third operation on the color values of the reference color patch and the color values of the detailed color patch.

The third operation may be performed by using a method of least squares.

When the color values of the detail color patches are obtained from the input image, the color conversion matrix generation unit can generate the first color conversion matrix.

According to one embodiment, the color conversion matrix C, which is the first color conversion matrix, is expressed by Equation (6).

은 레퍼런스 컬러차트의 각 세부 색상 패치의 컬러값인 평균값이다.

Is the average value of the color values of each detail color patch of the reference color chart.

은 입력 영상에서 검출된 세부 색상 패치의 컬러값이다.

Is the color value of the detail color patch detected in the input image.

Although the color space RGB has been described with respect to the color conversion matrix, it may correspond to other color spaces such as YCbCr and YUV.

According to one embodiment, since a total of 24 color values are known in the case of a Macbeth chart, the color conversion matrix C can be obtained through a least square method.

The color correction applying section 630 can correct the color patches by multiplying the first color conversion matrix with the color values of the color patches.

The color correction application unit can correct the color value of the desired area in the input image using the color conversion matrix. (7) "

The color chart detector 640 can detect a color chart area with respect to the input image. The color chart detector 640 may upscale or downscale the input image to generate pyramid images S1 to Sn. The color chart detector may apply a video classifier to the pyramid images to detect color chart areas.

The image classifier may be an image classifier learned using various learning techniques such as a Viola-Jones AdaBoost using positive data and negative data.

The positive data may be a normalized color chart image. The color chart may be a Macbeth chart or another type of color chart.

The negative data may be an image not including a color chart. It may also be a color chart image that does not satisfy the normalization condition.

According to one embodiment, the resolution of the window for detection by the color chart detector may be low. When the detection is completed, the color chart detector can utilize the resolution of the color chart area in the original input image as it is, or resize it to an appropriate level of resolution.

7 is a flowchart illustrating a procedure of an image color correction method according to an exemplary embodiment.

Step 710 is a step of detecting a color chart from the image input to the image color correcting apparatus 600.

According to one embodiment, the image color correction apparatus 600 may detect a normalized color chart image. The color chart may use a Macbeth chart. Other types of color charts may also be included.

Step 720 is a step of detecting a color patch in the color chart image of the input image and extracting a color value of the color patch.

According to one embodiment, the color patch unit 610 may perform a process of finding a corner point of the detailed color patch in the color chart image.

According to one embodiment, the color patch section 610 may detect the first corner points of each of the detail color patches using an algorithm for edge points in the color chart image.

The color patch unit 610 may select a first reference corner point of a reference color patch corresponding to the first corner point.

The coordinates of the detailed color patch can be extracted by performing a first operation between the first reference matrix for the coordinates of the first reference corner point and the first object matrix for the coordinates of the first corner point.

The first calculation may be performed using a method of least squares.

According to another embodiment, the color patch unit 610 may transform the coordinates of the primary reference corner point to secondary reference corner points, respectively, using a first matrix.

The color patch section 610 may select secondary corner points corresponding to the respective secondary reference corner points from the color chart image.

The color patch unit 610 performs a second operation between the second reference matrix for the coordinates of the second reference corner point and the second object matrix for the coordinates of the second corner point to extract the coordinates of the detailed color patch .

The color patch unit 610 may calculate a first coordinate value for the color chart image and a color value by operating a matrix H and a matrix for arbitrary reference coordinates. The matrix H is obtained by sequentially calculating the first matrix H1 and the second matrix H2.

Step 730 is a step in which the color conversion matrix generation unit 620 performs a third operation on the color value of the reference color patch and the color value of the detailed color patch to generate the first color conversion matrix.

The third operation may be performed using a method of least squares.

Step 740 is a step in which the color correction applying section 630 corrects the color value of the color patch through an operation of multiplying the first color conversion matrix by the color conversion matrix.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA) A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Includes hardware devices that are specially configured to store and execute program instructions such as magneto-tical media and ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

Claims (24)

A detector for detecting a first corner point of each of the detailed color patches using an algorithm for corner point detection in a color chart image;
A selector for selecting a first reference corner point of a reference color patch corresponding to the first corner point; And
Calculating a first matrix between a first reference matrix for the coordinates of the first reference corner point and a first object matrix for the coordinates of the first corner point to extract coordinates of the detail color patch; part
And a color patch detector. The method according to claim 1,
The algorithm comprises:
An algorithm using an eigenvalue, and a corner detection algorithm. The method according to claim 1,
The sorting unit,
The edge type of the first corner point is classified into a first corner type of a left upper corner of the detail color patch, a second corner type of a right upper corner of the detail color patch, 3 corner type and a lower left fourth corner type of the color patch. The method of claim 3,
Wherein the edge type is determined by comparing an intensity value of a pixel position in a predetermined direction with respect to the first corner point. The method according to claim 1,
The first operation may include:
A color patch detection apparatus for performing a color patch detection using a method of least squares. The method according to claim 1,
The primary reference corner point may be,
And a point having a large degree of similarity is selected through a distance comparison with the first corner point. The method according to claim 6,
The distance comparison may include:
A first distance between the primary corner point and the primary reference corner point and
And determines that the degree of similarity is large when the sum of the color value of the primary corner point and the second distance between the color values of the primary reference corner point becomes minimum. The method according to claim 1,
Wherein the detecting unit converts the coordinates of the primary reference corner point to secondary reference corner points using the first matrix,
Wherein the selector selects the secondary corner point corresponding to each of the secondary reference corner points from the color chart image,
Wherein the calculation unit performs a second operation between a second reference matrix for the coordinates of the second reference corner point and a second object matrix for the coordinates of the second corner point to extract a second matrix Generating
Color patch detection device. 9. The method of claim 8,
Wherein the second operation comprises:
A color patch detection apparatus for performing a color patch detection using a method of least squares. 9. The method of claim 8,
The second corner point may be,
And the second reference corner point is compared with a point having a larger degree of similarity through distance comparison. 11. The method of claim 10,
The distance comparison may include:
A third distance between the second corner point and the second reference corner point and
When the sum of the color value of the second corner point and the fourth distance between the color values of the second reference corner point is the smallest, the degree of similarity is determined to be large. 9. The method of claim 8,
A first coordinate value for the color chart image is calculated by sequentially computing a matrix for an arbitrary reference coordinate by the first matrix and the second matrix,
Further comprising: A color patch unit for detecting a detailed color patch in a color chart image and extracting a color value of the detailed color patch;
A color conversion matrix generator for generating a first color conversion matrix by performing a third operation on a color value of the reference color patch and a color value of the detailed color patch; And
A color correction unit which corrects the color value of the color patch by an operation of multiplying the color value of the color patch by the first color conversion matrix,
And an image correction unit. 14. The method of claim 13,
The color patch unit includes:
A detector for detecting a first corner point of each of the detailed color patches using an algorithm for corner point detection in a color chart image;
A selector for selecting a first reference corner point of a reference color patch corresponding to the first corner point; And
Calculating a first matrix between a first reference matrix for the coordinates of the first reference corner point and a first object matrix for the coordinates of the first corner point to extract coordinates of the detail color patch; part
And an image correction unit. 15. The method of claim 14,
The first operation may include:
An apparatus for correcting image color using a method of least squares. 15. The method of claim 14,
Wherein the third operation comprises:
An apparatus for correcting image color using a method of least squares. 15. The method of claim 14,
The primary reference corner point may be,
And the similarity is selected through a distance comparison with the first corner point. 18. The method of claim 17,
The distance comparison may include:
A first distance between the primary corner point and the primary reference corner point and
And determines that the degree of similarity is large when a sum of a color value of the first corner point and a second distance between the color values of the first reference corner point is minimum. 14. The method of claim 13,
And a color chart detector for detecting an input image as a color chart image. Detecting a first corner point of each of the detail color patches using an algorithm for corner point detection in a color chart image;
Selecting a first reference corner point of a reference color patch corresponding to the first corner point; And
The calculation unit performs a first operation between the first reference matrix for the coordinates of the first reference corner point and the first object matrix for the coordinates of the first corner point to generate a first matrix for extracting the coordinates of the detailed color patches Step
/ RTI > 21. The method of claim 20,
The first operation may include:
A method of detecting a color patch using a method of least squares. 21. The method of claim 20,
The primary reference corner point may be,
Wherein the similarity degree is selected by comparing the distance with the first corner point. 23. The method of claim 22,
The distance comparison may include:
A first distance between the primary corner point and the primary reference corner point and
And determines that the degree of similarity is large if the sum of the color value of the first corner point and the second distance between the color values of the first reference corner point is the smallest. A computer-readable recording medium having recorded thereon a program for performing the color patch detecting method according to any one of claims 20 to 23.

Images