JP6039778B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program.

  For objects and backgrounds such as blue in the blue sky and pink of cherry blossoms, there are specific colors that humans image against the objects and background. These colors are called memory colors. Even if a human sees an image that faithfully reproduces the color of an object or background according to the real world, he does not evaluate it as a beautiful image. This is because the reproduced color is different from the color that the human imaged (memory color), and it is recognized that the color is light.

For this reason, many image processing apparatuses perform a correction process for reproducing a color imaged by a human for each object or background. This correction process is called memory color correction. There are generally the following two methods for performing the memory color correction.
Dedicated circuit (spot color correction function)
3D lookup table

  First, (1) memory color correction using a dedicated circuit will be described. Here, the dedicated circuit performs processing specialized for memory color correction. Specifically, a specific color to be corrected is set for the dedicated circuit in the coordinate format of the movement source and the movement destination on the color space such as Cb / Cr or a * / b *. The dedicated circuit corrects the color area including the peripheral color area in accordance with this setting. However, since the dedicated circuit is a circuit specialized for memory color correction, other processing cannot be performed. Furthermore, when performing spot color correction in two or more areas, a plurality of dedicated circuits are required.

  Next, (2) memory color correction using a three-dimensional lookup table will be described. Information of the three-dimensional lookup table is stored in an arbitrary memory. In the three-dimensional lookup table, table data corresponding to coordinate points such as 9 × 9 × 9 or 17 × 17 × 17 representing mainly a three-dimensional RGB color space is stored. The three-dimensional lookup table stores the correspondence between the coordinates of each point on the color space and the coordinates to which the coordinates are converted. For example, when the three-dimensional lookup table has table data corresponding to coordinate points of 17 × 17 × 17, the RGB coordinates (0, 0, 0), the RGB coordinates of the conversion destination, and the RGB coordinates (16, 16, 16 ) And the RGB coordinates of the conversion destination, etc. are stored respectively. Then, an arbitrary processing unit refers to the three-dimensional lookup table, and moves each coordinate existing in the color space to a conversion destination coordinate by various interpolation processes. For example, the processing unit converts the conversion destination coordinates of the RGB coordinates (8, 8, 8), the conversion destination coordinates of the RGB coordinates (0, 0, 0), and the conversion destination coordinates of the RGB coordinates (16, 16, 16). To calculate.

  The three-dimensional lookup table merely holds conversion rules and does not have any function in itself. However, by setting appropriate table data for the three-dimensional lookup table and performing processing according to the conversion rules in the table, various types of color gamut mapping, six-axis correction, memory color correction, gamma correction, etc. A color management function can be realized. Here, the table data set in the three-dimensional lookup table can be changed as appropriate. Therefore, only by providing a single processing circuit (or processing function by software), necessary functions (including simultaneous realization of multiple area processing and multiple other functions) can be realized according to the situation.

  As described above, the correction using the three-dimensional lookup table has a much higher degree of freedom than the correction by the dedicated circuit function (spot color correction function). Therefore, the configuration using the three-dimensional lookup table can improve many cost advantages such as reduction of development resources and reduction of circuit scale, and improvement of usability. Furthermore, when a problem occurs or when improving the performance, the problem can be solved without changing the hardware by replacing the table data of the three-dimensional lookup table.

  Hereinafter, documents disclosing image processing techniques related to the above-described matters will be described. Patent Document 1 discloses a color signal processing system that performs color management processing on an input moving image signal using a three-dimensional lookup table. The system includes a color conversion unit that converts color information of an input image signal using a multidimensional lookup table, and a composite image obtained by combining the image signal converted by the color conversion unit and the input image signal at an arbitrary ratio. An interpolation unit that outputs a signal; and a lookup table rewriting unit that changes data of a multidimensional lookup table in the color conversion unit. With this configuration, this system can realize highly accurate color management processing even when the data in the lookup table is rewritten.

  Patent Document 2 discloses an image processing apparatus that performs automatic color adjustment with few side effects on memory color correction. The image processing apparatus includes: an intensity determining unit that generates a small correction intensity around a specific color range set based on two chromaticity components (a * / b *); Means for performing correction according to the correction intensity.

JP 2010-118881 A WO2004 / 032524

  However, correction processing on a color space using a conversion rule such as a three-dimensional lookup table requires a high level of knowledge and know-how to set conversion rules including memory color correction. there were. The subject is described below.

  A correction processing system using a three-dimensional lookup table generally has only a memory for storing table data and a circuit for calculating a conversion value from the table data. Therefore, the degree of freedom for setting the table data of the three-dimensional lookup table is very high. The setting of the table data of the three-dimensional lookup table includes not only setting a color (target color) to be subjected to memory color correction and a conversion destination of the color, but also colors existing around the target color. It is necessary to set how far the correction range is set and the conversion destination coordinates of each color within the correction range. Therefore, for example, when setting the table data relating to the memory color correction described above, the setter cannot perform a correct setting unless he / she has advanced knowledge and know-how regarding the memory color correction.

If the setter does not have sufficient knowledge and know-how to set table data and perform memory color correction, continuity in the three-dimensional color space is not guaranteed, and spatial distortion occurs in the corrected image. May occur. When an image having a gentle gradation of color and brightness is input to an image processing apparatus having such table data, the image processing apparatus can display an image having a contour that should not exist or gradation characteristics. There is a risk of outputting an image in which the image is damaged.

  The image correction methods described in Patent Literature 1 and Patent Literature 2 are both methods based on the premise that a setter sequentially sets table data. Therefore, if the setter does not have sufficient know-how and knowledge regarding color correction, there is a possibility that the intended correction cannot be performed.

  In the above description, when memory color correction is performed with reference to a three-dimensional lookup table, a problem occurs when a lookup table setter does not have sufficient know-how and knowledge. . However, this problem is not limited to correction using a three-dimensional lookup table, but is a problem that occurs when setting an arbitrary conversion rule that is necessary when performing color correction in a color space. Further, the problem is not limited to correction for the purpose of memory color correction, and is a problem common to correction of pixel data included in a video (coordinate movement in a color space).

  That is, when correcting pixel data (pixel value) included in an image (coordinate movement in a color space), calculation and movement of a correction range (range of coordinate points where movement occurs) that matches a desired correction purpose is performed. Advanced knowledge is required to set the amount and direction of movement of each coordinate point. If the user does not have advanced knowledge, there is a problem that appropriate settings cannot be made.

One aspect of the image processing apparatus according to the present invention is as follows.
A conversion unit that converts color information of the input image signal based on a conversion rule;
A correction range calculation unit that calculates a correction range in the color space based on the positional relationship between the movement source coordinates and the movement destination coordinates in the preset color space;
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion Each point movement amount calculation unit to be reflected in the rule.

One aspect of the image processing method according to the present invention is as follows.
An image processing method for converting color information of an input image signal based on a conversion rule,
Based on the positional relationship between the movement source coordinates and the movement destination coordinates in the preset color space, a correction range in the color space is calculated,
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion It is what is reflected in the rules.

One aspect of the image processing program according to the present invention is as follows:
On the computer,
A process of converting the color information of the input image signal based on a conversion rule;
A process of calculating a correction range in the color space based on the positional relationship between the movement source coordinates and the movement destination coordinates in a preset color space;
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion This is a program for executing processing to be reflected in the rules.

  In the present invention, the correction range calculation unit automatically calculates the correction range from the positional relationship between the movement source coordinates and the movement destination coordinates. Furthermore, each point movement amount calculation unit calculates the position of each lattice point in the correction range from the positional relationship between the movement source coordinate and the movement destination coordinate and the positional relationship between the coordinate of each point in the correction range and the movement source coordinate. The conversion destination coordinates are calculated, and the conversion destination coordinates are reflected in the conversion rule (a three-dimensional lookup table in the present embodiment). Here, the user only needs to provide information on the movement source coordinates and the movement destination coordinates. Thereby, the user can realize the correction process in the color space without having knowledge about the setting of the correction range, knowledge about the movement amount of each point, and the like.

  According to the present invention, when correction of pixel data (pixel value) included in an image (coordinate movement in a color space) is performed, a correction range that matches a desired correction purpose without requiring advanced knowledge ( It is possible to provide an image processing apparatus, an image processing method, and an image processing program capable of calculating the range of coordinate points where movement occurs and setting the amount and direction of movement of each coordinate point to be moved.

1 is a block diagram showing a configuration of an image processing apparatus 1 according to a first embodiment of the present invention. It is a block diagram which shows the structure of the table data generation part 100 concerning Embodiment 1 of this invention. It is a block diagram which shows the structure of the correction range calculation part 110 concerning Embodiment 1 of this invention. It is a figure which shows the concept of distance Lto of the movement origin coordinate in a color space (GB cross section) and a movement destination coordinate. It is a figure which shows the concept of rotation of the correction range by the correction range calculation part 110 concerning Embodiment 1 of this invention. It is the figure which showed the relationship of the ellipsoid which shows a movement origin coordinate, a movement destination coordinate, and a correction range three-dimensionally. It is a block diagram which shows the structure of the lattice point movement amount calculation part 120 concerning Embodiment 1 of this invention. It is a figure which shows the relationship between the influence Ef and the movement amount characteristic ME. It is a figure which shows the concept of the correction direction in color space. It is a conceptual diagram which shows the effect of rotation of a correction range. It is a conceptual diagram which shows the memory color correction | amendment of skin color. It is a block diagram which shows the structure of the image processing apparatus 1 concerning Embodiment 2 of this invention. It is a figure which shows the analysis of the histogram of the whole image by the histogram analysis part 304 concerning Embodiment 2 of this invention. It is a conceptual diagram which shows the interference of a correction range. It is a conceptual diagram which shows memory color correction by the image processing apparatus 1 concerning Embodiment 1 and Embodiment 2 of this invention. It is a figure which shows the concept of rotation of the correction range by the correction range calculation part 110 concerning Embodiment 3 of this invention. It is a figure which shows the concept of the reset which calculates the number of grid points contained in a correction range, and reduces an ellipsoid. It is a block diagram which shows the structure of the correction range calculation part 110 concerning Embodiment 4 of this invention. It is a conceptual diagram which shows the operation | movement image of the image processing apparatus 1 concerning Embodiment 4 of this invention. It is a conceptual diagram which shows the operation | movement image of the image processing apparatus 1 concerning Embodiment 4 of this invention. It is a block diagram which shows the structure of the image processing apparatus 1 concerning Embodiment 5 of this invention. It is a figure which shows the example of an image made into the process target of the image processing apparatus 1 concerning Embodiment 5 of this invention. It is a figure which shows the correction range of the blue sky in a GB cross section, and the correction range of a blue flower. It is a block diagram which shows the structure of the color space conversion part 500 concerning Embodiment 6 of this invention. It is a figure which shows an example of the hardware constitutions of the computer which performs the process of each process part of the image processing apparatus 1 as a program.

<Embodiment 1>
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the image processing apparatus according to this embodiment.

The image processing apparatus 1 includes a table data generation unit 100 and a three-dimensional lookup table unit 200. The three-dimensional lookup table unit 200 includes an address generation unit 210, a table data memory unit 220, and an interpolation unit 230.

  The three-dimensional lookup table unit 200 holds a three-dimensional lookup table and performs a conversion process on the input RGB signal. The three-dimensional lookup table unit 200 performs general memory color correction using a three-dimensional lookup table. Here, the three-dimensional lookup table held by the three-dimensional lookup table unit 200 is a general three-dimensional RGB space having a lattice point of 9 × 9 × 9 or 17 × 17 × 17 having an 8-bit or 10-bit accuracy. Lookup table. The three-dimensional lookup table unit 200 holds a correspondence relationship such as moving the grid point coordinates (R1, G1, B1) in the RGB space to (R2, G2, B2) by the number of grid points. A method of interpolating lattice points in the three-dimensional lookup table unit 200 (a method of calculating a movement destination of coordinates on a color space not included in the lattice points) is a linear interpolation method or cubic in a general three-dimensional space. An interpolation method may be used.

  In the memory color correction process, correction is performed by designating a specific color region in the three-dimensional color space. Therefore, correction with higher accuracy can be performed when the number of grid points is larger. Therefore, the number of grid points in the three-dimensional lookup table should be 9 × 9 × 9 or more. As a result, it is possible to perform memory color correction that can withstand actual use.

  Next, the configuration of the table data generation unit 100 will be described with reference to FIG. The table data generation unit 100 includes a correction range calculation unit 110 and a lattice point movement amount calculation unit 120. The user sets three set values of the movement source coordinates (coordinate points on the RGB coordinates), the movement destination coordinates (coordinate points on the RGB coordinates), and the correction direction (compression direction / parallel direction) of the color to be subjected to memory color correction. Is input to the table data generation unit 100. Note that the user may further set the influence range gain / offset and movement characteristics. If not set, the default value may be used. Similarly, when the correction direction is not set, a default value (for example, the compression direction) may be used.

  The correction range calculation unit 110 is a processing unit that calculates a range (correction range) on a color space for performing memory color correction based on the above-described three setting values (movement source coordinates, movement destination coordinates, and correction direction). . If the user feels that the correction range calculated by the correction range calculation unit 110 is not appropriate (too wide or too narrow), the user can adjust the correction range by adjusting the influence range gain / offset. it can. Details will be described later.

  The correction range calculation unit 110 supplies the calculated correction range (a range in which coordinates in the color space change with memory color correction) to the lattice point movement amount calculation unit 120. The lattice point movement amount calculation unit 120 automatically calculates the movement amount of each lattice point within the correction range, and generates table data of a three-dimensional lookup table based on the movement amount.

Specifically, the grid point movement amount calculation unit 120 is an influence (an index value of the movement amount of each grid point) based on the distance between the movement source coordinates serving as the center of movement and the coordinates of each grid point. Details will be described later). The degree of influence calculated from the grid point close to the movement source coordinate becomes large. The degree of influence calculated from a grid point having a large distance from the movement source coordinate becomes small. The degree of influence calculated from the grid points outside the influence range (outside the correction range) is zero.

  The lattice point movement amount calculation unit 120 calculates the influence degree for each lattice point, and calculates the movement amount of each lattice point according to the influence degree. The lattice point movement amount calculation unit 120 generates table data by adding the movement amount calculated for each lattice point to the coordinate value of each lattice point.

  When the user wants to adjust the movement amount calculated from the degree of influence, the user may adjust the input value of the movement characteristic. Details of the movement characteristics will be described later.

  Subsequently, prior to the detailed configuration and operation of the correction range calculation unit 110 and the grid point movement amount calculation unit 120, the correction range and each grid point are determined based on the input (movement source coordinates, movement destination coordinates) described above. The significance of calculating the movement amount will be described.

  In the image processing apparatus 1 according to the present embodiment, when performing memory color correction, the user sets a movement source color coordinate that is the center of movement, and indicates a movement destination that indicates which color the color is to be used for. Set the color coordinates. In response to the setting, the correction range calculation unit 110 calculates the correction range. That is, the correction range calculation unit 110 also selects a color (coordinate) existing around two set color coordinates as a movement target. There are two reasons for this selection.

  The first reason is that by ensuring continuity in the three-dimensional color space, spatial distortion is prevented and gradation characteristics are not deteriorated.

The second reason is that the color to be corrected is not concentrated on a certain point in the three-dimensional space but needs to be corrected for a certain color area. For example, when a blue memory color correction is performed, the blue color to be corrected does not exist in a concentrated manner at a certain point in the three-dimensional space, and a certain color region around the certain point is a correction target. Because.

  In general, for example, when setting a skin color area to be subjected to memory color correction, skin color sample data existing in the world is acquired. In addition to the acquired data, statistical data of preferable skin color is collected. In addition, by analyzing these data using advanced know-how, it is defined which skin color is the center (movement source), up to a certain range (correction range) as the skin color, and which color is the preferred skin color (movement) I decided to do it first).

  However, there are various races even if it is called skin color. For this reason, when all skin color regions are defined, the region becomes considerably wide. Furthermore, the skin color changes due to the light source in the shooting environment, and the skin color appearance changes depending on the background color. Therefore, it is impossible to make an absolutely preferable setting for any case. Furthermore, memory colors are often vague definitions in the first place. Therefore, the memory color may vary greatly depending on the user's preference and regional characteristics. In short, setting the movement source coordinates and the movement destination coordinates for correcting the memory color is often very difficult. However, it is also true that there is a certain correction effect when the color of the image is corrected to the preferred color and brightness trends.

  Therefore, in the image processing apparatus according to the present embodiment, each user sets that “I want to correct this color so that it approaches this color” (that is, set the above-described movement source and movement destination coordinates). Then, the correction range calculation unit 110 automatically calculates the correction range. Further, the lattice point movement amount calculation unit 120 automatically calculates the movement amount of each lattice point within the correction range. As a result, the image processing apparatus 1 according to the present embodiment can easily provide a means for correcting the memory color that is not problematic in actual operation even when the user does not hold the above statistical data or advanced adjustment know-how. To.

  Next, a detailed configuration of the correction range calculation unit 110 will be described with reference to FIG. FIG. 3 is a block diagram illustrating a configuration of the correction range calculation unit 110. The correction range calculation unit 110 includes a movement destination-movement source distance calculation unit 111, a correction range gain / offset adjustment unit 112, a movement destination-movement source linear inclination calculation unit 113, a rotation angle calculation unit 114, and a correction range ellipse. A spherical calculation unit 115. 4 to 6 are diagrams illustrating the relationship between the movement source coordinates, the movement destination coordinates, and the correction range in the color space. The following description will be given with reference to these drawings as appropriate.

To the movement destination-movement source distance calculation unit 111, the movement source coordinates and the movement destination coordinates set by the user are input. The destination-source distance calculation unit 111 calculates the distance in the color space from the destination coordinate to the source coordinate by subtracting the source coordinate from the destination coordinate. The movement source coordinates, the movement destination coordinates, and the distance from the movement destination coordinates to the movement source coordinates are defined as follows, and a calculation formula is shown.
Source coordinates: O (Ro, Go, Bo)
Destination coordinates: T (Rt, Gt, Bt)
Source / Source Distance: Lto (Rlto, Glto, Blto)
Lto = (Rt-Ro, Gt-Go, Bt-Bo)

  FIG. 4 shows the concept of the distance Lto between the movement source coordinates and the movement destination coordinates in the color space (GB cross section). In FIG. 4, a two-dimensional cross section is shown due to the restriction of the drawing, but the movement destination-movement source distance calculation unit 111 calculates a distance in a three-dimensional space.

  The movement destination-movement source distance calculation unit 111 supplies the calculated movement source / movement source distance (Lto) to the correction range gain / offset adjustment unit 112.

The correction range gain / offset adjustment unit 112 determines a correction range for each axis of RGB based on the supplied movement source / movement source distance (Lto). The correction range gain and the correction range offset are appropriately set by the user. If no setting is made, the default value is used. Hereinafter, the correction range E, the correction range gain Egain, and the correction range offset Eofs are defined, and calculation formulas (Expression 1) to (Expression 3) of the correction range E used by the correction range gain / offset adjustment unit 112 are described.
Correction range: E (Re, Ge, Be)
Correction range gain: Egain
Correction range offset: Eofs

  The correction range gain Egain can be adjusted by the user between “1” and “5”, but it is assumed that the correction range gain Egain is set to about “3”. The correction range offset Eofs can be adjusted by the user between 8 and 64 in the case of 8-bit (256) gradation, but it is assumed that the correction range offset Eofs is set to about 16.

  The user can automatically set the size of the correction range E according to the distance Lto between the movement source coordinates and the movement destination coordinates by adjusting the correction range gain Egain and the correction range offset Eofs. The larger the correction range gain Egain and the correction range offset Eofs, the wider the range of the correction range E (in FIG. 4, the elliptical sphere range becomes larger). Since the correction range gain Egain does not take a value of 1 or less, the movement source coordinates and the movement destination coordinates are included in the correction range E in principle. The correction range gain / offset adjustment unit 112 supplies the calculated correction range E to the correction range elliptic sphere type calculation unit 115.

The movement destination-movement source linear inclination calculation unit 113 calculates the inclination (Δrg) of the RG cross section and the inclination (Δgb) of the GB cross section from the movement destination coordinates and the movement destination coordinates. Specifically, the movement destination-movement source linear inclination calculation unit 113 includes an R coordinate (Ro) and a G coordinate (Go) of the movement source coordinate, an R coordinate (Rt) and a G coordinate (Gt) of the movement destination coordinate, The slope (Δrg) of the RG cross section is calculated from the positional relationship. Similarly, the movement destination-movement source linear inclination calculation unit 113 calculates the positions of the movement coordinates G coordinates (Go) and B coordinates (Bo) and the movement destination G coordinates (Gt) and B coordinates (Bt). The inclination (Δgb) of the GB cross section is calculated from the relationship. Specifically, the destination-source linear slope calculation unit 113 calculates the slope of the RG cross section (Δrg) and the slope of the GB cross section (Δgb) using the following formulas (Equation 4) and (Equation 5). To do.

The destination-movement source straight line inclination calculation unit 113 supplies the calculated RG cross section inclination (Δrg) and GB cross section inclination (Δgb) to the rotation angle calculation unit 114. In FIG. 4, the inclination (Δgb) of the GB cross section is shown.

The rotation angle calculation unit 114 calculates the angles (θrg, θgb) for rotating the elliptic sphere formed by the outer periphery of the correction range E from the supplied RG cross section inclination (Δrg) and GB cross section inclination (Δgb). Specifically, the rotation angle calculation unit 114 calculates the rotation angle (θrg, θgb) using the following calculation formulas (Equation 6) and (Equation 7).

The correction range elliptic sphere type calculation unit 115 is a processing unit that calculates an elliptic sphere type of the correction range. The correction range elliptic sphere type calculation unit 115 applies the elliptic sphere indicated by the outer periphery of the correction range E calculated by the correction range gain / offset adjustment unit 112 to each axis by the angles (θrg, θgb) calculated by the rotation angle calculation unit 114. The final ellipsoidal correction range E is calculated by rotating the shaft. Specifically, the correction range ellipsoidal sphere calculation unit 115 calculates coordinates (R1, G1, *) obtained by rotating the coordinate variable (R0, G0, B0) using the following formula (Equation 8). To do. Then, the correction range ellipsoidal sphere formula calculation unit 115 calculates the coordinates (R1, G2, B1) by rotating the coordinates (R1, G1, *) by the RG axis using the following formula (Equation 9). The correction range elliptic sphere formula calculation unit 115 calculates the elliptic sphere formula (Equation 10) after the axis rotation from the coordinate points (R1, G2, B1).

The correction range elliptic sphere calculation unit 115 supplies the calculated elliptic sphere type of the correction range E to the lattice point movement amount calculation unit 120 (intersection point calculation unit 122).

FIG. 5 is a diagram showing the concept of rotation of the correction range E. FIG. 5 shows the concept of rotation of the RG cross section. As illustrated, the correction range E is rotated according to the positional relationship between the movement source coordinates and the movement destination coordinates. That is, the correction range elliptic sphere type calculation unit 115 rotates the correction range E so that the straight line connecting the movement source coordinates and the movement destination coordinates is equal to the long side direction of the elliptic sphere.

FIG. 6 is a diagram three-dimensionally showing the relationship between the movement source coordinates, the movement destination coordinates, and the elliptical sphere indicating the correction range. As shown in the figure, the elliptical sphere has movement source coordinates and movement destination coordinates inside.

As described above, the user sets the movement source coordinates and the movement destination coordinates one by one. The correction range calculation unit 110 automatically calculates the correction range E through the series of processes described above. In the above-described method for calculating the correction range E, the center of the ellipsoid indicating the correction range is the correction source coordinate. However, the above calculation method is merely an example of a method for calculating an elliptical sphere of the correction range E, and the center of the elliptical sphere may be an intermediate coordinate between the movement source coordinate and the movement destination coordinate. The coordinates in the vicinity of the (destination coordinates) may be used.

Furthermore, in the above-described method for calculating the correction range E, the correction range is indicated by an elliptical sphere. However, the formula indicating the correction range E is not necessarily a complicated formula such as an elliptic sphere formula, and shows a range corresponding to the above elliptic sphere formula (that is, the positional relationship between the movement source coordinates and the movement destination coordinates). As long as it indicates a range calculated based on the above, a simple mathematical expression such as a linear expression may be used. The correction range E indicated by the linear formula includes, for example, a regular cube (octagon), a long cube (octagon, 10-gon, 12-gon).

Next, the configuration and operation of the lattice point movement amount calculation unit 120 will be described with reference to FIG. FIG. 7 is a block diagram illustrating a configuration of the lattice point movement amount calculation unit 120. First, an outline of the processing of the lattice point movement amount calculation unit 120 will be described.

The grid point movement amount calculation unit 120 calculates the movement amount of each grid point in the elliptic sphere indicated by the elliptic sphere formula of the correction range E input from the correction range elliptic sphere type calculation unit 115. For example, the grid point movement fee calculation unit 20 calculates the movement amount of each grid point within the correction range of FIG.

Specifically, first, the lattice point movement amount calculation unit 120 calculates an intersection point between each lattice point and a straight line passing through the movement source coordinates and the elliptic sphere indicated by the input elliptic sphere formula. Then, the lattice point movement amount calculation unit 120 calculates the distance (normalized distance) between the intersection point and the movement source coordinates. The grid point movement amount calculation unit 120 also calculates the distance between the movement source coordinates and the grid point. The grid point movement amount calculation unit 120 calculates the movement influence degree from the ratio between the distance (the distance between the movement source coordinate and the grid point) and the normalized distance. Then, the lattice point movement amount calculation unit 120 calculates the movement amount for each RGB coordinate of the lattice point from the relationship between the degree of influence and the movement source coordinate, the movement destination coordinate, and the lattice point coordinate. The lattice point movement amount calculation unit 120 adds the movement amount for each RGB coordinate to the lattice point coordinates to calculate the coordinates after the movement of the lattice points. The coordinates after the movement become table data for memory color correction.

In the following description, it is assumed that grid points (coordinate arrangements where the distances between the top, bottom, left, and right are all uniform) are handled, but the present invention is not necessarily limited thereto. The lattice point movement amount calculation unit 120 (each point movement amount calculation unit) may be configured to calculate the movement amount of each coordinate point having an arrangement other than the lattice points. In this case, the three-dimensional lookup table unit 200 may perform interpolation processing with reference to the distance relationship between the coordinate points.

Next, the detailed configuration of the grid point movement amount calculation unit 120 and the operation of each configuration processing unit will be described. The lattice point movement amount calculation unit 120 includes a movement source-grid point linear equation calculation unit 121, an intersection point calculation unit 122, a movement source-intersection point distance calculation unit 123, a movement source-grid point distance calculation unit 124, and an influence degree. A calculation unit 125, a movement characteristic setting unit 126, and a coordinate calculation unit 127 are provided.

The movement source-grid point linear equation calculation unit 121 calculates a linear equation of a straight line connecting the movement source coordinates and each coordinate included in the correction range E. Here, when an arbitrary grid point coordinate D and coordinate variable INPUT included in the correction range E are defined as follows, the linear equation is calculated as the following equation (Equation 11) by a general three-dimensional linear equation. The
Lattice point coordinates within the correction range ellipsoid: D (Rd, Gd, Bd)
Coordinate variable: (R0, G0, B0)

The movement source-grid point linear equation calculation unit 121 supplies the calculated linear equation (Equation 11) for each lattice point to the intersection calculation unit 122.

The intersection calculation unit 122 includes an elliptical sphere type (ellipsoidal type after axis rotation, (Equation 10)) of the correction range E input from the correction range elliptical sphere type calculation unit 115, and a linear formula calculated from each lattice point. (Expression 11) is input. The intersection calculation unit 122 calculates the intersection of the elliptic sphere (the elliptic sphere in the correction range E) and the straight line (the straight line connecting the movement source coordinates and the arbitrary lattice point) by substituting the linear expression into the elliptic sphere. The substitution formula is shown below. First, an assignment formula (Expression 12) relating to R1 is shown.

  In (Equation 12), some terms are replaced with variable A. Similarly, the substitution formula (Formula 13) concerning G2 is shown. Since the substitution method is the same as in (Equation 12), details of the substitution process will be described in a simplified manner.

  In (Equation 13), some terms are replaced with variable B. Similarly, the substitution formula (Formula 14) concerning B1 is shown. In (Equation 14), some terms are replaced with variable C.

From the above equation, the intersection point Cr can be obtained by the following equation (Equation 15).
Intersection Cr (Rcr, Gcr, Bcr)

  Further, when Rd = Ro, division by zero occurs in the linear equation (Equation 11). Therefore, Rcr and B0 described above are expressed by the following (Equation 16).

  Substituting (Equation 16) into the elliptic sphere equation (Equation 10), the following equations (Equation 17) to (Equation 19) are obtained.

  From the above (Equation 17) to (Equation 19), the intersection Cr can be obtained by the following equation (Equation 20).

Further, when Gd = Go, division by zero occurs in the linear equation (Equation 11). Therefore, Rcr and Gcr described above are expressed by the following (Equation 21).

Substituting (Equation 21) into the elliptic sphere equation (Equation 10), the following equations (Equation 22) to (Equation 24) can be obtained.

  From the above (Equation 22) to (Equation 24), the intersection Cr can be obtained by the following equation (Equation 25).

  As described above, the intersection calculation unit 122 calculates the intersection Cr corresponding to each lattice point. The intersection calculation unit 122 supplies the coordinates of the intersection Cr (the intersection of the ellipsoid and the straight line connecting the lattice point and the movement source coordinate) corresponding to each calculated lattice point to the movement source-intersection distance calculation unit 123.

  The movement source-intersection distance calculation unit 123 calculates a distance Loc between the movement source coordinates and the coordinates of the intersection corresponding to each grid point (the intersection of the ellipsoid and the straight line connecting the grid point and the movement source coordinate). To do. The source-intersection distance calculation unit 123 calculates the distance Loc from the following (Equation 26) using the three-square theorem.

  The distance Loc indicates a range in which movement around the movement source coordinate occurs. Therefore, the distance Loc is set as a normalized distance of the movement range. The movement source-intersection distance calculation unit 123 supplies the normalized distance Loc calculated for each lattice point to the influence calculation unit 125.

  The movement source-grid point distance calculation unit 124 calculates the distance between the movement source coordinate O and the arbitrary grid point D. The movement source-grid point distance calculation unit 124 calculates the distance Lod from the following (Equation 27) using the three-square theorem.

  The movement source-grid point distance calculation unit 124 supplies the distance Lod calculated for each lattice point to the influence degree calculation unit 125. FIG. 4B is a diagram illustrating the relationship between the normalized distance Loc and the distance Lod.

  The influence calculation unit 125 calculates an influence Ef used for calculating the movement amount of each grid point. The influence degree calculation unit 125 receives the normalized distance Loc and the distance Lod of each lattice point described above. As the influence Ef increases, the amount of movement of each lattice point increases. The degree of influence Ef is maximum when calculated from the vicinity of the movement source coordinates, and the intersection point Cr is 0 (a value at which movement does not occur). This is because the movement source coordinates are the center of movement. The influence degree calculation unit 125 calculates the influence degree Ef by the following (Equation 28).

  The influence degree calculation unit 125 performs the above-described influence degree Ef calculation process for each lattice point in the correction range E.

  As described above, the movement amount of the lattice point close to the movement source coordinate is increased, and the movement amount is decreased as the distance from the movement source coordinate is increased. The movement amount (influence Ef) of the lattice points on the correction range ellipsoid is zero. This ensures continuity in the three-dimensional color space. By guaranteeing continuity, it is possible to perform memory color correction without color distortion.

  The influence degree calculation unit 125 supplies the calculated influence degree Ef of each lattice point to the movement characteristic setting unit 126.

  The movement characteristic setting unit 126 calculates a movement amount characteristic ME from the degree of influence Ef of each grid point. The movement amount characteristic ME is a coefficient of the movement amount of each lattice point. When the influence degree Ef of each lattice point described above is used as the movement amount, the movement amount (movement amount characteristic ME) decreases as the distance from the movement source coordinate increases. The movement characteristic setting unit 126 calculates the movement amount characteristic ME from the influence degree Ef of each grid point, for example, by the following equation (Equation 29).

  FIG. 8 is a diagram showing the relationship between the degree of influence Ef and the movement amount characteristic ME. By the above (Equation 29), the movement amount (movement amount characteristic ME) is controlled according to the influence degree Ef. As shown in FIG. 8, the movement amount (movement amount characteristic ME) is larger than the linear change in the degree of influence Ef. Thereby, a more appropriate amount of movement can be set.

  When the user is not satisfied with the movement amount characteristic ME calculated by the movement characteristic setting unit 126, the user may be able to adjust the movement amount characteristic ME as appropriate. That is, the calculation formula used by the movement characteristic setting unit 126 may be appropriately set by the user.

  The movement characteristic setting unit 126 supplies the calculated movement amount characteristic ME of each grid point to the coordinate calculation unit 127.

  The coordinate calculation unit 127 calculates the movement amount M (Mr, Mg, Mb) of each grid point based on the movement amount characteristic ME of each grid point, and adds the movement amount M to each grid point coordinate D. The corrected coordinates DC (DCr, DCg, DCb) of each grid point are calculated. Here, in addition to the movement amount characteristic ME of each grid point and each grid point coordinate D, the correction direction is input to the coordinate calculation unit 127. The correction direction is designated by the user, and one of the compression direction and the parallel direction is designated. Hereinafter, the correction direction will be described with reference to FIG.

  As shown in FIG. 9A, the correction in the compression direction is a correction for changing the coordinate position so as to concentrate on the coordinates of one point designated by the user. Thereby, no matter what color is input, the memory color correction is performed so as to approach the preferable color (preferable coordinates). In other words, this is a correction that narrows the color reproduction range.

  As shown in FIG. 9B, the correction in the parallel direction is a correction for moving all the lattice points in the correction range in the same direction. The movement direction is a direction connecting the movement source coordinates designated by the user and the movement destination coordinates. In other words, since the correction in the parallel direction moves all the lattice points in the correction range in the same direction, the color reproduction range does not become narrower than the correction in the compression direction.

  The user designates a correction direction according to the purpose of memory color correction. The coordinate calculation unit 127 calculates the movement amount M (Mr, Mg, Mb) of each grid point and the corrected coordinates DC (DCr, DCg, DCb) of each grid point from the following calculation formulas according to the specified correction direction. ) Is calculated.

  The calculation formula of the compression direction is shown below (Equation 30).

  The calculation formula in the parallel direction is shown below (Equation 31).

The coordinate calculation unit 127 supplies the post-correction coordinates DC of each grid point within the correction range E to the table data memory unit 220. Accordingly, the table data memory unit 220 holds a lookup table for converting the arbitrary grid point D into the corresponding corrected coordinate DC.

Next, effects of the image processing apparatus 1 according to the present embodiment will be described. As described above, the user only specifies the movement source coordinates and the movement destination coordinates in the color space. In response to this designation, the correction range calculation unit 110 automatically calculates the correction range E, the grid point movement amount calculation unit 120 calculates the conversion destination coordinates of each grid point in the correction range E, and the conversion destination coordinates Are reflected in the conversion rule (a three-dimensional lookup table in this embodiment). That is, the user can easily realize the color conversion process on the color space even if he / she has no knowledge and know-how about the color conversion (correction) process on the color space (in the above example, the memory color correction is performed). Can be realized).

Further, the user does not need to set conversion rules sequentially. That is, the user does not need to sequentially set each table data of the lookup table for the memory color correction. The user sets only the movement source coordinates and the movement destination coordinates. Thereby, the man-hour of a user's setting can be reduced.

Furthermore, the correction range calculation unit 110 according to the present embodiment rotates the correction range E (rotation on the RG axis plane, rotation on the GB axis plane) according to the positional relationship between the movement source coordinates and the movement destination coordinates. (FIG. 5). Specifically, the correction range calculation unit 110 rotates the correction range E so that a straight line connecting the movement source coordinates and the movement destination coordinates is in the major axis direction. Thereby, memory color correction without color space distortion can be performed. Details will be described with reference to FIG.

FIG. 10A is a diagram illustrating a change in coordinates on the color space when the correction range is not rotated. As shown in FIG. 10A, the movement amount of each coordinate located in the vicinity of the movement destination coordinates is changing rapidly.

FIG. 10B is a diagram illustrating coordinate changes on the color space when the correction range is rotated (that is, the configuration of the image processing apparatus 1 according to the present embodiment). As shown in the figure, as the movement source coordinates and the grid point coordinates move away, the movement amount (the magnitude of the movement vector) of the grid point coordinates changes gently (for example, the correction shown in FIGS. 10B and 10B). The amount is larger than the correction amount shown in FIGS. This is because the movement amount of each coordinate within the correction range is controlled so as to gradually decrease as the distance from the movement source coordinate increases. Since the color change does not occur abruptly, color space distortion can be avoided.

Furthermore, according to the image processing apparatus 1 according to the present embodiment, it is possible to correct only the color to be corrected. This effect will be described with reference to FIG.

FIG. 11 is a conceptual diagram showing a skin color memory color correction. FIG. 11A is a diagram showing the correction range E when the correction range E is not rotated. When the correction range E is not rotated, coordinates indicating colors other than the color to be corrected (skin color) are also subject to correction. Therefore, the correction (coordinate movement in the color space) for the color that the user does not want to correct is performed.

On the other hand, FIG. 11B is a diagram showing the correction range E when the correction range E is rotated (that is, the configuration of the image processing apparatus 1 according to the present embodiment). When the correction range E is rotated, only the coordinates indicating the correction target color (skin color) are set as correction targets. Furthermore, the coordinates of the conversion destination of each grid point are also within the skin color existing range. Therefore, the image processing apparatus 1 according to the present embodiment can appropriately correct only the color to be corrected without correcting the color that is not desired to be corrected.

  As described with reference to FIGS. 10 and 11, the image processing apparatus 1 according to the present embodiment can improve the correction quality in addition to the effect of realizing the correction process without know-how and reducing the man-hours. There is an effect. When the user manually sets, the focus is on optimization of each setting item, which may cause final color space distortion. On the other hand, the image processing apparatus according to the present embodiment determines the correction range E from the relationship between the movement source coordinates and the movement destination coordinates. The movement amount of the coordinates within the correction range E is controlled so as to decrease as the distance from the movement source coordinates increases. Therefore, the image processing apparatus can generate an optimal image without color space distortion as described above.

Note that the lattice point movement amount calculation unit 120 in the image processing apparatus 1 of the present embodiment is configured to include the movement characteristic setting unit 126, but the present invention is not limited to this, and the conversion destination coordinates are calculated directly from the degree of influence Ef. May be.

<Embodiment 2>
The image processing apparatus according to the present embodiment is characterized in that memory color correction can be realized in consideration of a picture for each image (for example, for each frame for moving images). The image processing apparatus according to this embodiment will be described below.

FIG. 12 is a block diagram showing the configuration of the image processing apparatus according to this embodiment. The image processing apparatus 1 includes a pattern adaptation unit 300 in addition to the table data generation unit 100 and the three-dimensional lookup table unit 200. Since the configuration and processing of the table data generation unit 100 and the three-dimensional lookup table unit 200 are the same as those in the first embodiment, the configuration and operation of the picture adaptation unit 300 will be described below.

  The pattern adaptation unit 300 includes a memory color correction target detection unit 301, a memory color correction determination unit 302, a three-dimensional histogram generation unit 303, a histogram analysis unit 304, and a movement source / destination calculation unit 305.

The memory color correction target detection unit 301 detects objects, backgrounds, and the like (in the following description, the objects, backgrounds, etc. are collectively referred to as memory color correction targets) in the image. Examples of the memory color correction target include face, sky, sea, maple, cherry blossom, and green trees. The memory color correction target detection unit 301 detects and analyzes the shape, color, brightness, position in the image (for example, screen upper part: sky, screen lower part: sea), and the like. For the detection process, a generally used image recognition technique may be used. In recent years, image recognition techniques for detecting a unique memory color correction target have been widely developed in various fields. In particular, many products such as digital still cameras have functions such as face detection.

The memory color correction target detection unit 301 uses these general memory color correction target detection techniques, and detects the detected information (whether the memory color correction target exists, shape, color, luminance, position in the image). The data is supplied to the memory color correction determination unit 302 and the three-dimensional histogram acquisition unit 303.

The memory color correction determination unit 302 is a processing unit that determines whether or not to perform memory color correction. The memory color correction determination unit 302 receives information on a memory color correction target that the user wants to correct. In the present embodiment, the user designates a memory color correction target to be corrected. Here, the user may designate a plurality of memory color correction targets. When a plurality of settings are made, the user sets a preset value for the movement source coordinates and a preset value for the movement destination coordinates for each memory color correction target. For example, when performing memory color correction for blue sky and cherry blossoms at the same time, the user can preset the source coordinate preset value and destination coordinate preset value for the blue sky, and the source coordinate preset value and destination coordinate preset for the cherry blossom. Set the value.

The memory color correction determination unit 302 includes information detected by the memory color correction target detection unit 301 (whether the memory color correction target exists in the image, shape, color, luminance, position in the image), and memory specified by the user. The color correction target is compared, and it is determined whether or not the memory color correction is performed.

If there is no memory color correction target designated by the user in the image, there is a possibility that the effect of the correction will not occur even if the memory color correction is performed. Furthermore, there is a possibility that memory color correction is performed on a memory color correction target that is not the target. For these reasons, the memory color correction determination unit 302 determines whether or not to perform memory color correction.

The determination by the memory color correction determination unit 302 is determined for each image (frame). However, in a system that holds a plurality of three-dimensional lookup tables (described later) or a system that can specify a correction area to be corrected by a three-dimensional lookup table, the memory color correction determination unit 302 is described above for each screen area. Determination (determination of whether or not to perform memory color correction) may be performed. Thereby, a more accurate memory color correction can be realized.

The memory color correction determination unit 302 supplies the determination result to the movement source / destination calculation unit 305.

The three-dimensional histogram generation unit 303 receives the RGB values of the entire input image and the information detected by the memory color correction target detection unit 301. The three-dimensional histogram acquisition unit 303 generates a three-dimensional histogram of the entire screen and the image area where the memory color correction target is detected. The three-dimensional histogram generation unit 303 supplies the two types of generated three-dimensional histograms to the histogram analysis unit 304.

The histogram analysis unit 304 analyzes two types of input three-dimensional histograms (a three-dimensional histogram of an image area in which a memory color correction target is detected and a three-dimensional histogram of the entire image). First, the analysis of the three-dimensional histogram of the image area where the memory color correction target is detected will be described.

The same object or the same background is displayed in the image area where the memory color correction target is detected. For example, in the area where the blue sky is detected, there are pixels constituting the blue sky. Therefore, in the histogram on the color space, it is assumed that the values are concentrated in a certain area in the three-dimensional space. Here, color coordinates having a certain frequency or less are ignored in the histogram. Then, an elliptical spherical histogram centered on a certain color coordinate in the three-dimensional space can be obtained. The histogram analysis unit 304 detects the coordinates of the center of the elliptic sphere shape, that is, the coordinates of the central color of the memory color correction target as an analysis result. There are mainly two types of detection methods, ellipsoidal position center detection and ellipsoidal centroid detection. First, ellipsoidal position center detection will be described.

The histogram analysis unit 304 defines the maximum value, the minimum value, and the center value of each of the RGB axes in the elliptic sphere relating to the above-described histogram as follows, and calculates the center value of the elliptic sphere using the following formula.
Maximum value in elliptic sphere Rmax, Gmax, Bmax
Minimum value in ellipsoidal sphere Rmin, Gmin, Bmin
Median value in elliptic sphere Rcen, Gcen, Bcen

Next, the ellipsoidal centroid detection will be described. The histogram analysis unit 304 defines the cumulative value and the elliptical centroid value of each of the RGB axes in the elliptical sphere as follows, and calculates the elliptical centroid value by the following equation (Equation 33).
Cumulative value in elliptic sphere Rsgm, Gsgm, Bsgm
Total pixels Dot
The center of gravity of the ellipsoid sphere Rgra, Ggra, Bgra

Note that the histogram analysis unit 304 may calculate the center / centroid of the elliptic sphere in the histogram (that is, the coordinates of the central color of the object or background) by a method other than the above. The histogram analysis unit 304 supplies the coordinates obtained by the analysis to the movement source / destination calculation unit 305.

Next, analysis of the histogram of the entire image by the histogram analysis unit 304 will be described. The histogram analysis unit 304 determines the shape of the histogram between the preset value of the movement source coordinates and the preset value of the movement destination coordinates set by the user for each memory color correction target and in the surrounding color coordinates. To analyze. Specifically, the histogram analysis unit 304 determines whether a color already exists in the vicinity of the preset value of the destination coordinate in the state before the memory color correction (a plurality of coordinates having colors similar to the destination color exist). Is analyzed). Further, the histogram analysis unit 304 analyzes the color distribution ratio between the preset value of the movement source coordinates and the preset value of the movement destination coordinates.

Details of the histogram analysis of the entire image by the histogram analysis unit 304 will be described in detail with reference to FIG. In FIG. 13, seven areas are set between the preset value OP of the movement source coordinates and the preset value TP of the movement destination coordinates. In FIG. 13, for convenience of explanation, a two-dimensional coordinate space is shown, but the histogram analysis unit 304 actually handles a three-dimensional coordinate space.

The histogram analysis unit 304 analyzes the histogram for each of the seven areas. The histogram analysis unit 304 determines whether or not the number of pixels located in each area exceeds a predetermined threshold value. If it exceeds, the histogram analysis unit 304 determines that a color exists in the area before correction (a predetermined number or more pixels exist in the area before correction). This determination is performed in order from area 1 to area 7 close to the preset value of the movement source coordinates. The histogram analysis unit 304 ends the determination process when detecting an area where a predetermined number of pixels or more can be confirmed. The histogram analysis unit 304 detects the representative coordinates of an area where a predetermined number of pixels or more can be confirmed as existence coordinates TE (TEr, TEg, TEb). The histogram analysis unit 304 supplies the existing coordinates TE (TEr, TEg, TEb) to the movement source / destination calculation unit 305.

For example, when there are no more than a predetermined number of pixels in area 1 and there are more than a predetermined number of pixels in area 2, the histogram analysis unit 304 does not make a determination after area 3. Then, the histogram analysis unit 304 detects the representative coordinates of the area 2 as the existing coordinates TE (TEr, TEg, TEb). Note that the determination up to the area 7 is performed, and if there is no area where a predetermined number of pixels or more exist, the histogram analysis unit 304 considers that the existence coordinates TE do not exist.

In the description of FIG. 13, the number of areas in the histogram is seven, but the number is not necessarily limited to this. When the number of areas is increased, the histogram analysis unit 304 can perform detailed presence coordinate detection, that is, improve the accuracy of presence determination. As a result, it is possible to improve the setting accuracy of the destination coordinates for memory color correction described later.

Next, the movement source / destination calculation unit 305 will be described. In the present embodiment, the user sets a preset value OP of the movement source coordinate O and a preset value TP of the movement destination coordinate T for each object and background to be corrected. The movement source / movement destination calculation unit 305 supplies the movement source coordinates O supplied to the table data generation unit 100 based on the preset value, the determination result of the memory color correction determination unit 302, and the analysis result of the histogram analysis unit 304. And the destination coordinate T are calculated. That is, in this embodiment, the movement source coordinates and the movement destination coordinates input by the user are not used as they are, but the coordinate positions are adjusted and supplied to the table data generation unit 100.

First, the calculation process of the movement source coordinate O by the movement source / destination calculation unit 305 will be described. The user sets a preset value OP (OPr, OPg, ORb) of the movement source coordinate O for each memory color correction target. The movement source / movement destination calculation unit 305 receives the analysis result of the histogram analysis unit 304. The analysis result is the center color coordinate OC (OCr, OCg, OCb) of the memory color correction target specified by the user. The movement source / destination calculation unit 305 sets the movement source coordinates O (Ro, Go, Bo) between the preset value OP (OPr, OPg, ORb) and the center color coordinates OC (OCr, OCg, OCb). This is calculated by the following formula (Equation 34). The user can set a follow-up degree F1 (0 to 100%) that is an index of how close the preset value of the movement source coordinate OP is to the center color coordinate OC (OCr, OCg, OCb). When the follow-up degree F1 is set to 0%, the movement source coordinate O becomes equal to the preset value OP (OPr, OPg, ORb) of the movement source coordinate O. When the follow-up degree F1 is 100%, the movement source coordinate O is equal to the center color coordinate OC (OCr, OCg, OCb). When the tracking degree F1 is 50%, the movement source coordinate O is an intermediate value between the preset value OP (OPr, OPg, ORb) of the movement source coordinate O and the center color coordinate OC (OCr, OCg, OCb). Become. Hereinafter, a calculation formula (Formula 34) of the movement source coordinates O (Ro, Go, Bo) is shown.

The follow-up degree F1 is set on the assumption of about 50%, and the default value when the user does not specify explicitly is also about 50%.

Next, the processing for calculating the movement destination coordinates by the movement source / movement destination calculation unit 305 will be described. The user sets preset values TP (TPr, TPg, TRb) of the movement source coordinates for each memory color correction target. The movement source / destination calculation unit 305 receives the analysis result of the histogram analysis unit 304 and determines whether or not there are pixels of a certain number or more in the direction of the movement destination in the state before correction. If it exists, the movement source / movement destination calculation unit 305 is based on the input existence coordinates TE (TEr, TEg, TEb) and the preset values TP (TPr, TPg, TRb) of the movement destination coordinates T. The final movement destination coordinate T is calculated by the following equation (Equation 35). The user can set a follow-up degree F2 (0 to 100%) that is an index of how close the preset value of the destination coordinate T is to the existing coordinate TE. When the follow-up degree F2 is set to 0%, the movement destination coordinate T becomes equal to the preset value TP (TPr, TPg, TRb) of the movement destination coordinate T. When the follow-up degree F2 is 100%, the movement destination coordinate T becomes equal to the existence coordinate TE (TEr, TEg, TEb). When the follow-up degree F2 is 50%, the destination coordinate T is an intermediate value between the preset value TP (TPr, TPg, TRb) of the destination coordinate T and the existing coordinate TE (TEr, TEg, TEb). . Hereinafter, a calculation formula (Expression 35) of the movement destination coordinates T (Rt, Gt, Bt) is shown.

The follow-up degree F2 is set on the assumption of about 50%, and the default value when the user does not specify explicitly is also about 50%.

The movement source / movement destination calculation unit 305 supplies the calculated movement source coordinates O (Ro, Go, Bo) and movement destination coordinates T (Rt, Gt, Bt) to the table data generation unit 100.

Next, the problem of memory color correction that does not take into account the pattern and the effect of the image processing apparatus 1 according to the present embodiment will be described. First, the problem of memory color correction without considering a picture will be described.

In memory color correction that does not consider a pattern, correction processing is performed using a look-up table to which one stationary pattern is input regardless of whether or not there is a memory color correction target. For this reason, there is a risk of correcting a pixel relating to a memory color correction target that is not intended to be corrected.

Furthermore, even if there is a memory color correction target that the user wants to correct in the image, the memory color correction target (for example, a human face) is different from the color region that is assumed as the memory color correction target. Because of the color, there is a possibility that the pixel to be corrected cannot be sufficiently included in the correction range.

In memory color correction using a lookup table, correction is often performed simultaneously on a plurality of memory color correction targets. In this case, the correction range may interfere. The situation will be described with reference to FIG. As shown in the figure, interference may occur between a correction range related to a certain memory color correction target and a correction range related to another memory color correction target. In this case, it is unclear how to perform correction in the interfering area. For this reason, it is generally necessary to perform processing while suppressing the correction intensity from an ideal correction amount. That is, correction processing is performed after the correction range is reduced as shown in FIG. For this reason, a sufficient correction effect cannot be obtained.

Next, effects of the image processing apparatus 1 according to the present embodiment will be described. The image processing apparatus 1 according to the present embodiment detects a memory color correction target, and automatically calculates a movement source coordinate and a movement destination coordinate based on the memory color correction target and a histogram of the entire image. Thereby, the image processing apparatus 1 according to the first embodiment can obtain the following two effects.
Correction processing can be performed only for the memory color correction target that the user wants to correct.
A sufficient intensity correction can be performed on the memory color correction target that the user wants to correct.

First, the reason why the image processing apparatus 1 according to the present embodiment has an effect that “the correction process can be performed only on the memory color correction target that the user wants to correct” will be described.

The memory color correction determination unit 302 calculates whether there is a memory color correction target designated by the user. The movement source movement destination calculation unit 305 calculates the movement source coordinates O and the movement destination coordinates T only when there is a memory color correction target designated by the user. Thereby, the image processing apparatus 1 (three-dimensional lookup table unit 200) can perform only the memory color correction related to the memory color correction target, that is, the effect (1) described above can be achieved.

When there are not a plurality of memory color correction targets in one screen, that is, when there is only one, the image processing apparatus 1 performs the memory color correction for only one memory color correction target. Therefore, the image processing apparatus 1 does not need to cope with correction range interference.

When the memory color correction determination unit 302 determines that there are a plurality of memory color correction targets specified by the user and the correction ranges of the two interfere with each other, the image processing apparatus 1 intends for the color coordinates belonging to the interference area. It is possible that no conversion is performed. However, this problem is solved by the configuration of the fourth or fifth embodiment described later.

Next, in the image processing apparatus 1 according to the present embodiment, the reason is that “(2) a sufficient intensity correction can be performed on a memory color correction target that the user wants to correct”. explain. The reason is the following two points.

The first reason is as follows. As described above, the memory color correction determination unit 302 calculates whether there is a memory color correction target designated by the user. Then, the movement source movement destination calculation unit 305 sets a correction range only for the memory color correction target designated and detected by the user. That is, for example, in FIG. 14, the correction range can be limited to only those related to the memory color correction target designated by the user. Therefore, it is possible to reduce the number of elliptic spheres indicating the correction range in the color space. As a result, it is possible to reduce the need to consider interference in the correction range of the memory color correction, and to reduce the need to take measures such as reducing the correction range. Since the need to reduce the correction range is reduced, it is possible to perform sufficient intensity correction.

Next, a second reason for achieving the effect (2) described above will be described with reference to FIG. In the description of FIG. 15, it is assumed that the image to be processed includes an area displaying a blue sky and an area displaying a blue flower. It is assumed that the user wants to perform only blue sky memory color correction.

  FIG. 15A is a conceptual diagram illustrating memory color correction by the image processing apparatus 1 according to the first embodiment. In this case, the image processing apparatus 1 determines the correction range using the movement source coordinates and the movement destination coordinates input by the user as they are. For this reason, when there are objects having similar colors such as blue sky and blue flowers in the image, the image processing apparatus 1 changes the color of the blue flowers that are not intended to be corrected.

FIG. 15B is a conceptual diagram showing memory color correction by the image processing apparatus 1 according to the present embodiment. The histogram analysis unit 304 determines whether or not a predetermined number or more pixels exist between the preset value of the movement source coordinate and the preset value of the movement destination coordinate in the color space. If there are no more than a predetermined number of pixels, the movement source / destination calculation unit 305 uses the movement source coordinate preset value as it is as the movement source coordinate, and uses the movement destination coordinate preset value as it is as the movement destination coordinate. Thereby, the image processing apparatus 1 can perform memory color correction with sufficient intensity.

On the other hand, when there are a predetermined number of pixels or more, the movement source / destination calculation unit 305 moves the preset value (coordinates) of the movement destination coordinates in a direction closer to the preset value of the movement source coordinates by the above formula, The movement point is determined as a movement destination coordinate. For example, when a blue sky and a blue flower are displayed in the image, the image processing apparatus 1 performs memory color correction after making the preset value of the movement destination coordinate close to the preset value of the movement source coordinate (FIG. 15B )reference). Here, the movement source / destination calculation unit 305 appropriately adjusts the movement amount of the preset value (coordinates) according to an area where a predetermined number of pixels or more exist. As a result, as shown in FIG. 15B, the image processing apparatus 1 automatically calculates a range that does not affect the object (background) that is not the correction target, and corrects the memory color of the correction target object (background). It can be performed.

As described above, a general image processing apparatus uniformly reduces the correction range of the memory color correction in consideration of adverse effects on other objects (backgrounds). However, the image processing apparatus 1 according to the present embodiment may have another object that is not intended to be corrected by analyzing the position of the color in the color space (by performing the above-described histogram analysis). The optimal correction range can be determined for each frame (image). Thereby, the image processing apparatus 1 according to the present embodiment can realize memory color correction having a sufficient correction amount according to the pattern of the image (frame).

The design of the moving image changes with time. The image processing apparatus 1 according to the present embodiment appropriately adjusts the movement source coordinates and the movement destination coordinates used for memory color correction following the change in the pattern. Therefore, the image processing apparatus 1 according to the present embodiment can realize accurate memory color correction for moving images.

  In the above description, the pattern adaptation unit 300 adjusts the preset value of the movement source coordinates and the preset value of the movement destination coordinates, but is not necessarily limited to this, and only one of the adjustments may be performed. For example, the picture adaptation unit 300 only adjusts the preset value of the movement source coordinates, and adopts the user input as it is for the movement destination coordinates. Even with such a configuration, the image processing apparatus 1 can achieve certain effects.

<Embodiment 3>
The image processing apparatus according to the present embodiment is characterized in that the number of table data held in the three-dimensional lookup table can be limited. The image processing apparatus according to the present embodiment is premised mainly on dynamic processing (processing for analyzing a picture and performing memory color correction according to the picture) for a moving image. Since the image processing apparatus according to the present embodiment is a modification of the second embodiment, a description will be given focusing on differences from the second embodiment (configuration and operation of the correction range calculation unit 110).

  FIG. 16 is a block diagram illustrating a configuration of the correction range calculation unit 110 according to the present embodiment. The correction range calculation unit 110 includes a conversion coordinate number limiting unit 116 in addition to the configuration of FIG.

  The user sets the table data limit number set in the three-dimensional lookup table in the conversion coordinate number limit unit 116. Here, the table data limit number means the upper limit number of grid points set as a conversion target in the three-dimensional lookup table. For example, the user sets a table data limit number to be set in consideration of the type of moving image data subject to memory color correction, the image data size (number of vertical and horizontal pixels) of the moving image data, the required operation speed, etc. do it.

  The conversion coordinate number limiting unit 116 is input with the elliptical spherical type to be corrected for each memory color calculated by the correction range elliptical spherical type calculating unit 115. For example, when the blue sky memory color correction and the cherry blossom memory color correction are performed, two elliptic sphere types are input. The conversion coordinate number limiting unit 116 calculates the total number of lattice points to be corrected based on these elliptic sphere formulas. From the ellipsoidal formula, the number of grid points included in the ellipsoidal sphere can be calculated based on the grid point distance (FIG. 17A). Here, the conversion coordinate number limiting unit 116 calculates the total number of lattice points to be corrected based on all elliptic spheres to be corrected. The conversion coordinate number restriction unit 116 compares the table data restriction number input from the user with the total number of lattice points to be corrected, which are calculated based on the ellipsoidal formula.

  When the table data limit number is equal to or greater than the total number of grid points to be corrected, the conversion coordinate number limit unit 116 instructs the correction range elliptic sphere type calculation unit 115 to output an elliptic sphere type.

  On the other hand, when the table data limit number is smaller than the total number of grid points to be corrected, the conversion coordinate number limit unit 116 instructs the correction range gain / offset adjustment unit 112 to readjust the gain and / or offset. Here, the conversion coordinate number restriction unit 116 notifies the correction range gain / offset adjustment unit 112 together with the ratio of the table data restriction number and the total number of grid points to be corrected. Then, the conversion coordinate number limiting unit 116 instructs the correction range elliptic sphere type calculation unit 115 to prohibit the output of the elliptic sphere type.

  The correction range gain / offset adjustment unit 112 refers to the notified ratio and resets the gain and / or offset so that the elliptical sphere becomes smaller (FIG. 17B). Since the correction is made with reference to the ratio, the correction range gain / offset adjustment unit 112 can avoid making the ellipsoid too small more than necessary.

  After resetting the gain and / or offset, the correction range elliptic sphere formula calculation unit 115 calculates the elliptic sphere formula again.

  In the above description, the correction range is reduced by adjusting the gain and / or offset, but the present invention is not necessarily limited thereto. For example, the correction range may be reduced by moving the movement destination coordinates and / or movement destination coordinates related to the memory color correction target. For example, in the example of FIG. 17B, the correction range of the memory color correction target is reduced by moving the movement destination coordinates to the position indicated by the dotted line. For example, the conversion coordinate number restriction unit 116 may instruct the design adaptation unit 300 to change the position of the movement source coordinate and / or the movement destination coordinate of the target storage path object.

  Next, effects of the image processing apparatus according to this embodiment will be described. As a general problem of dynamic processing using a three-dimensional lookup table, there is a large number of table data to be set. Since the number of table data is large, it is difficult to use a three-dimensional lookup table for processing moving image data in which data is sequentially input.

  The image processing apparatus according to the present embodiment limits the number of table data so as to be less than or equal to the table data limit number input by the user. That is, the image processing apparatus 1 according to the present embodiment stores the number of data sizes stored in a lookup table having a table size such as 729 (9 × 9 × 9) and 4913 (17 × 17 × 17), for example. Less than the table data limit. For example, the number of table data stored for a lookup table having a capacity of 729 (9 × 9 × 9) is limited to 200. As a result, an increase in the number of data transferred within a unit time can be prevented, and the processing speed of the entire system can be increased.

  In the above description, the number of data stored in the three-dimensional lookup table is reduced by reducing the size of the elliptic sphere. However, the present invention is not limited to this. For example, the range ellipsoidal calculation unit 115 may reduce the number of data by reducing the memory color correction target to be processed.

<Embodiment 4>
The image processing apparatus according to the present embodiment performs memory color correction in consideration of priority when there are a plurality of memory color correction targets. As described above, in memory color correction using one three-dimensional lookup table, there are cases where interference of color regions occurs. As a result, the memory color correction for a plurality of objects may not be performed simultaneously, or sufficient correction may not be performed. In this embodiment, the user sets the priority order of memory color correction for each object or background in the image processing apparatus. The image processing apparatus according to the present embodiment performs optimal memory color correction according to this priority order even when a plurality of objects are subjected to memory color correction. Hereinafter, the configuration of the image processing apparatus according to the present embodiment will be described with reference to FIG. Further, an operation image according to the present embodiment will be described with reference to FIG. 19 as appropriate.

The correction range calculation unit 110 includes a correction range overlap detection unit 117 in addition to the configuration of FIG. The correction range overlap detection unit 117 is supplied with the priority order of the memory color correction target (the smaller the numerical value, the higher the priority order). This priority is set by the user. For example, the user inputs “priority 1: blue sky”, “priority 2: blue flowers”, and “priority 3: cherry blossoms”.

  The correction range overlap detection unit 117 is further input with the elliptical sphere type for each memory color correction calculated by the correction range elliptical sphere type calculation unit 115. The correction range overlap detection unit 117 determines whether or not there is an overlapping portion in the elliptic sphere when there are a plurality of elliptic sphere types. For example, in the example of FIG. 19, there is an overlap between an oval sphere for memory color correction applied to an object with priority 1 and an oval sphere for memory color correction applied to an object with priority 2. It should be noted that the correction range duplication location detection unit 117 may determine whether or not there is duplication by a process in a known three-dimensional space from the given elliptical sphere.

  When there is an overlap, the correction range overlap detection unit 117 notifies the correction range gain / offset adjustment unit 112 of the low-priority memory color correction target and the size of the overlap range. Then, the correction range overlap detection unit 117 instructs the correction range ellipsoidal calculation unit 115 to prohibit the elliptical output.

  When there is no overlap, the correction range overlap detection unit 117 instructs the correction range elliptic sphere type calculation unit 115 to output an elliptic sphere type.

  When the correction range gain / offset adjustment unit 112 receives the notification from the correction range duplication detection unit 117, the correction range gain / offset adjustment unit 112 adjusts the gain or / and the offset so that the elliptical sphere for the memory color correction target with a low priority is reduced. Here, the correction range gain / offset adjustment unit 112 adjusts the gain or / and the offset according to the size of the overlapping range. Then, the correction range gain / offset adjustment unit 112 supplies the correction range for each RGB calculated after the adjustment to the correction range elliptic sphere type calculation unit 115.

  In the example of FIG. 19, the correction range gain / offset adjustment unit 112 changes the correction range of the priority order 2 (the correction range of the lower priority order) to the correction range indicated by the dotted line by adjusting the gain and / or offset, This correction range is supplied to the correction range elliptic sphere type calculation unit 115.

  When there is an overlap, the memory color correction for the memory color correction target with a low priority may not be performed.

  Next, effects of the image processing apparatus according to this embodiment will be described. First, a problem of memory color correction that does not consider priority will be described. As shown in FIG. 19, when dealing with a plurality of memory color correction targets, there is a case where overlap of correction ranges occurs. In this case, it is necessary to take measures such as giving up the correction of one of the memory color correction targets. Therefore, the intended memory color correction effect may not be obtained.

  On the other hand, when there are a plurality of memory color correction targets, the image processing apparatus according to the present embodiment can set a priority order for the memory color correction targets. The image processing apparatus according to the present embodiment can ensure that memory color correction with a desired intensity is reliably performed on a memory color correction target with a high priority. Furthermore, the image processing apparatus according to the present embodiment can perform memory color correction with sufficient correction strength even for a memory color correction target with a low priority when there is no overlap in the correction range.

Furthermore, when there is an overlap in the correction range, the correction range gain / offset adjustment unit 112 adjusts the gain and / or offset to the extent that there is no overlap. Therefore, the image processing apparatus according to the present embodiment can perform a fixed memory color correction on a memory color correction target with a low priority even when there is an overlap in the correction range.

That is, according to the image processing apparatus according to the present embodiment, it is possible to perform memory color correction as intended by the user on a plurality of memory color correction targets belonging to the same (approximate) color region.

Furthermore, the configuration according to the third embodiment and the configuration according to the fourth embodiment can be combined. That is, the image processing apparatus 1 may adjust the correction range to be narrowed according to the priority order so that the number of table data is equal to or less than the table data limit number. In this case, the correction range is reduced even when there is no overlap in the correction range. The concept will be described below.

First, the correction range calculation unit 110 calculates a correction range (elliptical sphere) for each memory color correction target without considering the table data limit number. In this case, it is expected that the number of table data to be set for each frame exceeds the limit number of table data (exceeds system capacity). However, when the user has determined the priority order of each memory color correction target, the image processing apparatus 1 increases the reduction ratio of the correction range as the priority order decreases so that the table data limit number or less. Adjust each correction range as follows.

The operation concept will be described with reference to FIG. In the example of FIG. 20, three memory color correction targets are set. In FIG. 20A, the memory color correction target of priority 1 does not exist in the frame. For this reason, the correction range calculation unit 110 adjusts the correction range without greatly reducing the correction range of the memory color correction target of priority 2 and the correction range of the memory color correction target of priority 3. .

On the other hand, in FIG. 20B, the memory color correction target with priority 1 exists in the frame. For this reason, the correction range calculation unit 110 significantly reduces the correction range of the memory color correction target of priority 2 and the correction range of the memory color correction target of priority 3. On the other hand, the correction range calculation unit 110 minimizes the reduction of the correction range of the priority 1 memory color correction target.

Note that a specific numerical value of the reduction rate according to the priority order can be appropriately specified by the user.

As described above, by considering both the priority order and the table data limit number, the image processing apparatus 1 sufficiently sets the correction intensity of the memory color correction target to be subjected to memory color correction while limiting the table data to a desired number. Can be secured.

<Embodiment 5>
The image processing apparatus according to the present embodiment is characterized in that it can hold a plurality of three-dimensional lookup tables. Since the image processing apparatus according to the present embodiment is a modification of the image processing apparatus according to the second embodiment, differences from the image processing apparatus according to the second embodiment will be mainly described.

  FIG. 21 is a block diagram showing a configuration of the image processing apparatus 1 according to the present embodiment. The image processing apparatus 1 includes a plurality of three-dimensional lookup table units 200 to 202 and a selector 400 in addition to the configuration of FIG.

  The memory color correction target detection unit 301 detects the memory color correction target and detects in which area in the image each memory color correction target exists, and the display area in each memory color correction target image. To do. With reference to FIG. 22, the process of the memory color correction target detection unit 301 will be described.

  In the following example, the memory color correction target detection unit 301 detects three “blue sky”, “blue flower”, and “human skin” as memory color correction targets. In addition, the user sets “blue sky” and “blue flower” as memory color correction targets. FIG. 22 is a diagram illustrating an example of an image on which correction processing is performed. As shown in the figure, there is a blue sky at the top of the image and a blue flower at the bottom left of the screen. The user sets a movement source coordinate (RGB value) and a movement destination coordinate (RGB value) in the color space for “blue sky” and “blue flower”. The memory color correction target detection unit 301 notifies the selector 400 which memory color correction target exists for each pixel in the image. For example, the memory color correction target detection unit 301 notifies that a blue sky exists in the pixel 1 (X1, Y1) in FIG. Similarly, the memory color correction target detection unit 301 notifies that a blue flower exists in the pixel 2 (X2, Y2) in FIG. Note that the memory color correction target detection unit 301 notifies the processing target pixel when there is no memory color correction target.

  The correction range calculation unit 110 calculates the correction range of each memory color correction target using the same method as in the first embodiment. FIG. 23 is a diagram illustrating a blue sky correction range and a blue flower correction range. Here, there is an overlapping range D1 between the blue sky correction range and the blue flower correction range.

  The grid point movement amount calculation unit 120 calculates the movement amount of each grid point for each input memory color correction target. In the example of FIG. 23, the lattice point movement amount calculation unit 120 calculates the conversion destination coordinates of each lattice point included in the blue sky correction range, and converts each lattice point included in the blue flower correction range. Calculate the destination coordinates. Therefore, the correction range movement amount calculation unit 120 calculates P2 as conversion coordinates when the coordinate P1 is handled as a grid point included in the blue sky correction range, and coordinates as a grid point included in the blue flower correction range. When P1 is handled, P3 is calculated as converted coordinates.

  The grid point movement amount calculation unit 120 calculates conversion coordinates of each grid point for each memory color correction target. Then, the lattice point movement amount calculation unit 120 sets a three-dimensional lookup table for each memory color correction target. In the following example, the grid point movement amount calculation unit 120 sets table data relating to blue sky memory color correction in the three-dimensional lookup table unit 200, and a table relating to blue flower memory color correction in the three-dimensional lookup table unit 201. Data is set, and other table data for color correction is set in the three-dimensional lookup table unit 202.

  Each look-up table unit (200 to 202) corrects each pixel in the color space according to the set look-up table. Each lookup table unit (200 to 202) outputs the RGB value of each pixel obtained by the correction to the selector 400.

  As described above, the selector 400 is notified from the memory color correction target detection unit 301 of the coordinates of each memory color correction target. The selector 400 selects and outputs one of the RGB values output from the look-up table unit (200 to 202) according to the relationship between the pixel position and the coordinates of each memory color correction target. For example, when the pixel 1 in FIG. 22 is being corrected, the selector 400 selects and outputs an output from the lookup table unit 200 (a processing unit in which blue sky memory color correction table data is set). . Similarly, when the correction of the pixel 2 in FIG. 22 is performed, the selector 400 selects the output from the lookup table unit 201 (the processing unit in which the blue flower memory color correction table data is set). Output.

  Next, effects of the image processing apparatus according to this embodiment will be described. As described above, the image processing apparatus 1 according to the present embodiment provides a lookup table for each memory color correction target. The image processing apparatus 1 selects an appropriate value from the conversion results of a plurality of lookup tables in accordance with the position where each memory color correction target exists in the image. As a result, as shown in FIG. 23, even when there is an overlap in the correction range in the color space, the image processing apparatus 1 can appropriately perform the memory color correction without being affected by the overlap in the correction range. it can.

  In the above description, the configuration in which the image processing apparatus 1 includes the lookup table for each memory color correction target has been described, but the configuration is not necessarily limited thereto. The image processing apparatus 1 may store the table data in another lookup table only when there is an overlap in the correction range. For example, consider a case where the blue sky correction range and the blue flower correction range overlap, but the blue sky correction range and the cherry blossom correction range do not overlap. In this case, in addition to the table data relating to the blue sky correction range, the table data relating to the cherry blossom correction range is stored in the first lookup table. And the table data concerning the correction range of the blue flower is stored in the second lookup table. As a result, it is possible to realize a memory color correction process that is not affected by overlapping correction ranges while minimizing an increase in the number of lookup tables.

<Embodiment 6>
The image processing apparatus according to the present embodiment is characterized in that it can handle image signals other than RGB. In the image processing apparatus according to the present embodiment, the user can set the movement source coordinates of the memory color correction target and the movement amount from the coordinates. The image processing apparatus according to the present embodiment will be described focusing on differences from the first embodiment.

  The image processing apparatus 1 according to the present embodiment includes a color space conversion unit 500 in front of the table data generation unit 100 (or the pattern adaptation unit 300). FIG. 24 is a block diagram illustrating a configuration of the color space conversion unit 500. The color space conversion unit 500 includes an RGB / HSV conversion unit 510, an adder 520, and an HSV / RGB conversion unit 530.

  The user designates the movement source coordinates O (Ro, Go, Bo) and the amount of change from the movement source coordinates. Here, the amount of change is specified for each item of hue (H), saturation (S), and luminance (V), for example.

  The RGB / HSV conversion unit 510 converts the movement source coordinates designated by the user from RGB values to HSV values. The conversion may be performed by a known method. The RGB / HSV conversion unit 510 outputs the converted HSV value (Ho, So, Vo) to the adder 520.

  The adder 520 adds the amount of change (ΔH, ΔS, ΔV) of each item of hue (H), saturation (S), and luminance (V) to the input HSV value (Ho, So, Vo). . The adder 520 outputs the added value (Ht, St, Vt) to the HSV / RGB conversion unit 530.

  The HSV / RGB conversion unit 530 converts an input HSV value (Ht, St, Vt), which is an added value, into an RGB value. This RGB value indicates the destination coordinate T (Rt, Gt, Bt). The HSV / RGB conversion unit 530 supplies the movement destination coordinate T to the table data generation unit 100. The table data generation unit 100 is also supplied with the source coordinates in RGB format specified by the user.

  Through the above processing, the movement source coordinate O and the movement destination coordinate T are set in the table data generation unit 100. The subsequent processes are the same as those in the first embodiment.

  Next, effects of the image processing apparatus 1 according to the present embodiment will be described. Even when it is desired to perform memory color correction, the target color coordinates may not be determined. In the present embodiment, the user can specify the amount of change from the source coordinate instead of the destination coordinate. For this reason, the user can make ambiguous designations such as “slightly yellow”, “more darker”, and “pretty bright”, and convert the designation into a change amount. This further reduces the user's setting burden.

  In the above-described example, the configuration is described in which HSV and RGB conversion is performed, but the configuration is not necessarily limited thereto. For example, the user inputs the movement source coordinates and the change amount in the RGB format. And the structure which an adder calculates a movement destination coordinate by an addition process may be sufficient.

  In the example of FIG. 24, the conversion between the HSV value and the RGB value has been described. However, the present invention is not necessarily limited to this. good.

  The image processing apparatus according to the present invention has been described above according to the embodiment. However, the present invention is not limited to the above-described embodiment, and can be changed as appropriate without departing from the spirit of the present invention. In the above-described embodiment, coordinate conversion processing is performed on the premise of the RGB space as the color space, but the present invention can also be applied to other color spaces. That is, the image processing apparatus may perform the above-described processes in a color space such as HSV, HSU, YUV, HLS, Lab, or YCM.

  The above-described image processing apparatus 1 is used by being incorporated in, for example, a printer device, a television receiver, a copier, a digital multifunction peripheral, a projector device, a mobile phone terminal, a digital still camera, a smartphone, a digital photo frame, a display device, or the like. . That is, the image processing apparatus 1 is used in a device that displays an image on a display unit, a device that can be connected to a device that includes a display unit, and a device that prints or displays an image.

Part or all of the processing of each processing unit (table data generation unit 100, three-dimensional lookup table unit 200, picture adaptation unit 300, selector 400, color space conversion unit 500) in the image processing apparatus 1 is implemented by a hardware circuit. Hardware configuration. Furthermore, a part or all of the processing of each processing unit (table data generation unit 100, three-dimensional lookup table unit 200, picture adaptation unit 300, selector 400, color space conversion unit 500) in the image processing apparatus 1 is arbitrary. You may implement | achieve as a program which operate | moves in this computer. The program may be stored using various types of non-transitory computer readable media and supplied to a computer. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (for example, flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (for example, magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / W and semiconductor memory (for example, mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (random access memory)) are included. The program may also be supplied to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

FIG. 25 shows an example of the hardware configuration of a computer that executes the processing of each processing unit of the image processing apparatus 1 as a program. The computer includes a central processing unit (CPU) 600 and a memory 610. The CPU 600 and the memory 610 are connected to a hard disk device (HDD) 620 as an auxiliary storage device via a bus. A storage medium such as the hard disk device 620 can store a computer program for giving an instruction to the CPU 600 or the like in cooperation with the operating system and performing each process of the image processing apparatus 1 described above.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 100 Table data generation part 110 Correction range calculation part 111 Destination-movement origin distance calculation part 112 Correction range gain / offset adjustment part 113 Destination-movement source linear inclination calculation part 114 Rotation angle calculation part 115 Correction range ellipse Spherical calculation unit 116 Conversion coordinate number limiting unit 117 Correction range overlap detection unit 120 Lattice point movement amount calculation unit 121 Movement source-grid point linear equation calculation unit 122 Intersection calculation unit 123 Movement source-intersection distance calculation unit 124 Movement source-grid Point distance calculation unit 125 Influence degree calculation unit 126 Movement characteristic setting unit 127 Coordinate calculation unit 200 Three-dimensional lookup table unit 210 Address generation unit 220 Table data memory unit 230 Interpolation unit 300 Pattern adaptation unit 301 Object / background detection unit 302 Memory color Correction determining unit 303 Three-dimensional histogram generating unit 304 Decoder 305 move source-destination calculator 400 selector 500 color space conversion unit 510 RGB / HSV converter 520 adder 530 HSV / RGB conversion section 600 CPU
610 Memory 620 HDD

Claims (22)

A conversion unit that converts color information of the input image signal based on a conversion rule;
A correction range calculation unit that calculates a correction range in the color space based on the positional relationship between the movement source coordinates and the movement destination coordinates in the preset color space;
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion Each point movement amount calculation unit to be reflected in the rule,
The correction range calculation unit
Accepting an input of the rule limit number of the conversion rule from the user, comparing the number of each point coordinate in the correction range calculated from the size of the correction range, and the rule limit number,
An image processing apparatus that recalculates the correction range so that the region range becomes smaller when the rule restriction number is smaller than the number of each point coordinate in the correction range. The image processing apparatus according to claim 1, wherein the point movement amount calculation unit reduces the movement amount of each point as the distance from the movement source coordinate increases.   The image processing apparatus according to claim 1, wherein the correction range is an ellipsoidal sphere in which a first straight line connecting the movement source coordinates and the movement destination coordinates is a major axis direction. The correction range calculation unit
A first distance calculation unit for calculating a distance between the movement source coordinates and the movement destination coordinates;
A straight line inclination calculating unit for calculating the inclination of the first straight line;
A correction range adjustment unit that calculates a coordinate range based on the preset adjustment value and the distance calculated by the first distance calculation unit;
A rotation angle calculation unit for calculating a rotation angle for rotating the coordinate range based on the inclination of the first straight line calculated by the linear inclination calculation unit;
The correction range elliptic sphere type calculation part which calculates the mathematical expression of the ellipsoidal sphere by rotating the coordinate range calculated by the correction range adjustment part by the rotation angle. Image processing device. The straight line inclination calculating unit, when the movement source coordinate of the color space is O (Ro, Go, Bo) and the movement destination coordinate is T (Rt, Gt, Bt), the first straight line in the color space. 5. The image processing apparatus according to claim 4, wherein the inclination Δ1 of the first cross section and the inclination Δ2 of the second cross section are calculated by the following equations (Equation 36) and (Equation 37).

The rotation angle calculation unit calculates the rotation angle θ1 at the first cross section and the rotation angle θ2 at the second cross section in the color space by the following equations (Equation 38) and (Equation 39). The image processing apparatus according to claim 5.

Each point movement amount calculation unit
A linear equation calculation unit for calculating a linear equation of a second straight line connecting the first coordinate point included in the correction range and the movement source coordinate;
An intersection calculation unit that calculates an intersection of the elliptic sphere and the second straight line based on the mathematical expression of the elliptic sphere and the linear equation of the second straight line;
A second distance calculation unit for calculating a distance between the movement source coordinates and the intersection;
A third distance calculating unit for calculating a distance between the first coordinate point and the movement source coordinate;
An influence degree calculating section that calculates an influence degree that serves as a reference for the amount of movement of the first coordinate point according to a ratio between the distance calculated by the second distance calculating section and the distance calculated by the third distance calculating section. When,
The image processing apparatus according to claim 3, further comprising: a coordinate calculation unit that calculates a conversion destination coordinate of the first coordinate point based on the degree of influence.   The image processing apparatus according to claim 7, wherein the coordinate calculation unit calculates a conversion destination coordinate of the first coordinate point according to a correction direction (compression direction or parallel direction) designated by a user. Each point movement amount calculation unit
A movement characteristic that is adjusted so that the degree of influence is increased more than a linear change is calculated, and the movement characteristic is set as a parameter when the coordinate calculation unit calculates a conversion destination coordinate of the first coordinate point. The image processing apparatus according to claim 7, further comprising a characteristic adjustment unit.   The coordinate of each point within the correction range processed by each point movement amount calculation unit is a grid point coordinate whose distances from adjacent points are all equal. The image processing apparatus according to item.   The image processing apparatus according to claim 1, wherein the movement source coordinates and the movement destination coordinates are input values designated by a user. The type of the memory color correction target, the first preset value of the movement source coordinate and the second preset value of the movement destination coordinate for each memory color correction target to be corrected are received from the user,
Determining whether or not the memory color correction target specified by the user exists in the image constituting the image signal, and analyzing the color distribution of the image;
At least one of the first preset value and the second preset value for each memory color correction target existing in the image is adjusted based on an analysis result of a color distribution of the image, and an adjustment value is stored in the memory color. The image processing apparatus according to claim 1, further comprising a picture adaptation unit that sets the movement source coordinates and the movement destination coordinates for each correction target. The pattern adaptation unit is:
A correction target detection unit that detects the presence or absence of an object or background that can be a target of memory color correction in the image, and a detection region of the memory color correction target in the image;
A memory color correction determination unit that determines a memory color correction target to be processed by comparing the detection result of the correction target detection unit with the type of the memory color correction target specified by the user;
A histogram generator for generating a color space histogram of the image;
Based on the color space histogram, a first analysis result relating to color distribution analysis in the detection region, and a second relating to color distribution analysis between color space coordinates indicated by the first preset value and the second preset value. A histogram analysis unit for calculating an analysis result;
Based on the memory color correction target determined by the memory color correction determination unit, the first and second analysis results, and the first and second preset values, for each memory color correction target existing in the image. A movement source movement destination calculation unit that calculates the movement source coordinates and the movement destination coordinates;
The image processing apparatus according to claim 12, further comprising: The histogram analysis unit
The central color coordinate indicating the central color of the detection area is the first analysis result,
The color space coordinates indicated by the first preset value and the second preset value are divided into a plurality of areas, and it is determined whether or not there are pixels having a predetermined number of pixels or more in order from an area close to the first preset value, The image processing apparatus according to claim 13, wherein existence coordinates that are representative coordinates of an area where a predetermined number of pixels or more exist are used as the second analysis result. The movement source movement destination calculation unit sets the center color coordinates to OC (OCr, OCg, OCb), the existence coordinates to TE (TEr, TEg, TEb), and the first preset value to OP (OPr, OPg, ORb). The second preset value is TP (TPr, TPg, TRb), the movement source coordinates are O (Ro, Go, Bo), the movement destination coordinates are T (Rt, Gt, Bt), and user-adjustable tracking When the degrees are F1 (0 to 100%) and F2 (0 to 100%), the movement source coordinates and the movement destination coordinates are calculated by the following equations (Equation 40) and (Equation 41). The image processing apparatus according to claim 14.

The correction range calculation unit
Receiving the priority order of the memory color correction target from the user, determining whether or not there is an overlap in the correction range for each memory color correction target;
When there is an overlap, the correction range for each memory color correction target is recalculated so that the reduction ratio of the correction range for the memory color correction target increases in the order of low priority. Item 15. The image processing apparatus according to Item 15. The correction range calculation unit
The image processing apparatus according to claim 16, wherein the recalculation is performed after adjusting at least one of the coordinates so that a distance between the movement source coordinates and the movement destination coordinates becomes small. The image processing apparatus includes at least a first conversion unit and a second conversion unit as the conversion unit,
The pattern adaptation unit detects a display area in each of the first memory color correction target and the second memory color correction target that are the memory color correction targets specified by the user,
The correction range calculation unit calculates the correction range corresponding to each of the first and second memory color correction targets,
Each point movement amount calculation unit calculates a conversion destination coordinate of each point in the correction range related to the first memory color correction target, reflects it in the conversion rule of the first conversion unit, and the second 18. The conversion destination coordinates of each point within the correction range for a memory color correction target are calculated and reflected in the conversion rule of the second conversion unit. The image processing apparatus according to item 1.   The image processing apparatus according to claim 1, wherein the color space is any one of RGB, HSV, HSU, YUV, HLS, Lab, and YCM.   The image processing apparatus according to claim 1, wherein the conversion rule is an LUT (Look Up Table). An image processing method for converting color information of an input image signal based on a conversion rule,
Based on the positional relationship between the movement source coordinates and the movement destination coordinates in the preset color space, a correction range in the color space is calculated,
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion Reflected in the rules,
In the calculation process of the correction range,
Accepting an input of the rule limit number of the conversion rule from the user, comparing the number of each point coordinate in the correction range calculated from the size of the correction range, and the rule limit number,
An image processing method for recalculating the correction range so that the area range becomes smaller when the rule limit number is smaller than the number of each point coordinate in the correction range. On the computer,
A process of converting the color information of the input image signal based on a conversion rule;
A process of calculating a correction range in the color space based on the positional relationship between the movement source coordinates and the movement destination coordinates in a preset color space;
Based on the positional relationship between the movement source coordinates and the movement destination coordinates, and the positional relationship between the coordinates of each point in the correction range and the movement source coordinates, the conversion destination coordinates of each point are calculated, and the conversion Process to be reflected in the rules,
In the process of calculating the correction range,
Accepting an input of the rule limit number of the conversion rule from the user, comparing the number of each point coordinate in the correction range calculated from the size of the correction range, and the rule limit number,
A program for causing the correction range to be recalculated so that the area range becomes smaller when the rule limit number is smaller than the number of point coordinates in the correction range.

Images