JP2010154008A - Image processing unit and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately correct HSV target image data by HSV reference image data. <P>SOLUTION: A multifunctional machine executes processing below. More specifically, the multifunction machine calculates target representative values (OH, OS, OV) representing a group of pixels including hue in a specific range in HSV target image data (1) (S44). The multifunctional machine converts the target representative values (OH, OS, OV) to target coordinates values (Ov, Oa, Ob), namely values in a vab color space (2) (S48). The multifunctional machine calculates reference representative values (SH, SS, SV) representing a group of pixels having hue in a specific range in HSV reference image data (3) (S46). The multifunctional machine converts the reference representative values (SH, SS, SV) to reference coordinates values (Sv, Sa, Sb), namely values in the vab color space (4) (S50). The multifunctional machine corrects coordinates values (Pv, Pa, Pb) of each pixel included in a peripheral region of the target coordinates values (Ov, Oa, Ob) so that the coordinates values (Pv, Pa, Pb) approach the reference coordinates values (Sv, Sa, Sb) (5). <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing technique for correcting target image data using reference image data.

  Patent Document 1 below discloses a technique for correcting input image data using teacher image data. In this technique, image processing parameters are determined in accordance with the tendency of image reproduction indicated by the teacher image data (for example, hues of skin color, blue, green, etc.), and the input image data is corrected based on the image processing parameters.

JP 2006-80746 A JP-A-8-321957 JP 2002-223366 A JP 2005-197996 A Japanese Patent Laid-Open No. 10-173947

  The present inventor is engaged in the development of a technique for correcting the color in the hue in the specific range of the target image data so as to approach the color in the hue in the specific range of the reference image data (for example, blue). . The present inventor has found that the target image data may not be corrected properly when correcting the color within the hue in the specific range of the target image data. For example, there may be a plurality of pixels having a near-achromatic color (very small saturation) in the target image data. A pixel group having a color near an achromatic color has almost no difference in apparent color. In these pixel groups, there may be pixels having a hue to be corrected (the hue in the specific range) and pixels having no hue to be corrected. In this case, only the value of the former pixel is corrected, and the value of the latter pixel is not corrected. As a result, both pixels having almost no apparent color difference before correction become greatly different colors after correction.

  The present specification provides a technique capable of appropriately correcting the hue of a specific range of target image data using reference image data.

  The present specification discloses an image processing apparatus that corrects target image data using reference image data to generate corrected image data. The image processing apparatus includes a first determination unit, a first calculation unit, a second determination unit, a second calculation unit, and a correction unit. The first determining unit determines a first polar coordinate value that is a value representing the first pixel group having a hue in a specific range in the target image data and is a value in the color space of the polar coordinate system. The first calculation unit calculates a first orthogonal coordinate value that is a value in the color space of the first orthogonal coordinate system by orthogonally transforming the first polar coordinate value. The second determining unit determines a second polar coordinate value that is a value representative of the second pixel group having the hue in the specific range in the reference image data and is a value in the color space of the polar coordinate system. The second calculation unit calculates a second orthogonal coordinate value that is a value in the color space of the first orthogonal coordinate system by orthogonally transforming the second polar coordinate value. The correction unit converts the coordinate value of each pixel included in the area around the first orthogonal coordinate value in the color space of the first orthogonal coordinate system among all the pixels in the target image data to the second orthogonal coordinate value. By correcting so as to approach, corrected image data is generated from the target image data.

  In the image processing device, in the color space of the polar coordinate system, for each of the target image data and the reference image data, representative values (first polar coordinate value and second polar coordinate value) representing the pixel group having the hue in the specific range described above. ) Is determined. Since the representative value is determined in the color space of the polar coordinate system, the above-described identification in the target image data is performed using a correction vector that causes the first polar coordinate value to approach the second polar coordinate value in the color space of the polar coordinate system. It is also conceivable to employ a method of correcting the value of each pixel having a hue in a range. However, in this method, when there is a pixel group having an achromatic color with almost no apparent color difference, only the pixel having the hue in the specific range is corrected from the pixel group. An event occurs. In order to suppress the occurrence of this phenomenon, in the image processing apparatus described above, the value of each pixel of the target image data is corrected in the color space of the orthogonal coordinate system (first orthogonal coordinate system). That is, the first polar coordinate value is converted into the first orthogonal coordinate value, and the values of the pixels included in the area around the first orthogonal coordinate value are corrected. A pixel group having an achromatic color and having almost no apparent color difference is likely to have coordinate values close to each other in the color space of the orthogonal coordinate system (first orthogonal coordinate system). For this reason, even when there is a pixel having no hue in the specific range in the pixel group having a color near the achromatic color, the value of the pixel can be corrected. The occurrence of the above event can be suppressed.

  The target image data may be one in which the value of each pixel is represented by a value in the color space of an orthogonal coordinate system (for example, the first orthogonal coordinate system or the second orthogonal coordinate system described later). However, the value of each pixel may be represented by a value in the color space of the polar coordinate system. In the former case, the first determination unit and the second determination unit convert the value of each pixel in the target image data into a value in the color space of the polar coordinate system, and then convert the first polar coordinate value and the second polar coordinate value. You may decide. In the latter case, the correction unit may perform correction after converting each pixel in the target image data into a value in the color space of the first orthogonal coordinate system.

  The shape of the surrounding area is not particularly limited. The surrounding area may be a specific area including both the first orthogonal coordinate value and the second orthogonal coordinate value, for example. In this case, the specific area may be an area within an ellipsoid having the first orthogonal coordinate value and the second orthogonal coordinate value as two focal points. On the other hand, the surrounding area may be an area including only the first orthogonal coordinate value, for example.

  In human visual characteristics, for example, the color change of green is difficult to recognize. Therefore, for example, when correcting the hue of green, the range of the value of the pixel to be corrected (that is, the surrounding area) is relatively large so that it can be recognized that the green color has changed after the correction. It is preferable. On the other hand, for example, the skin color is easy to recognize the color change. Therefore, for example, when correcting the hue of the flesh color, the range of pixel values to be corrected (that is, the above surrounding area) is compared so that the flesh color does not change greatly after the correction and does not give a sense of incongruity. It is preferable to make it small. This technical idea can be expressed as follows. That is, the size of the surrounding area set when the specific range is the first range is the size of the surrounding area set when the specific range is a second range different from the first range. It may be different from the size. According to this configuration, the size of the surrounding area can be changed according to the hue to be corrected. For example, in the case where the specific range is a first range showing a green hue, the surrounding area can be made larger than in the case where the specific range is a second range showing a flesh-colored hue. .

  Even when two pixels having similar appearance colors in the target image data are adjacent to each other, only the value of one pixel may be corrected and the value of the other pixel may not be corrected. In this case, the color may be greatly different between the corrected pixel and the uncorrected pixel. In order to suppress the occurrence of this phenomenon, the following configuration may be employed. That is, the correction unit includes a first distance between the first orthogonal coordinate value and the coordinate value of the correction target pixel in the color space of the first orthogonal coordinate system, and the second orthogonal coordinate value and the correction target pixel. Coordinates in the color space of the first orthogonal coordinate system of the pixel to be corrected so that the correction amount increases as the sum of the second distance between the coordinate values in the color space of the one orthogonal coordinate system decreases. The value may be corrected.

  According to the above configuration, the closer the coordinate value of the correction target pixel with respect to both the first orthogonal coordinate value and the second orthogonal coordinate value is, the larger the correction amount is. Conversely, the correction amount decreases as the coordinate value of the pixel to be corrected with respect to both the first orthogonal coordinate value and the second orthogonal coordinate value increases. This means that the correction amount of the coordinate value of the pixel existing near the boundary of the surrounding area is small. The fact that the apparent colors of the two pixels in the target image data are similar means that the two pixels may have coordinate values that are close to each other in the color space of the orthogonal coordinate system (first orthogonal coordinate system). Is expensive. Nevertheless, only one pixel is corrected when one of the pixels has a coordinate value that exists inside near the boundary of the surrounding area, and the other pixel has the surrounding value. There is a high possibility of having a coordinate value existing outside in the vicinity of the boundary of the region. According to the above configuration, the correction amount of the pixel having the coordinate value existing inside near the boundary of the surrounding region is small. For this reason, even if only one of the two adjacent pixels that are similar in appearance color is corrected, the correction amount is small, so that a large color is generated between the two pixels. Different things can be suppressed.

  When the coordinate value in the color space of the first orthogonal coordinate system of the pixel to be corrected is the same as the first orthogonal coordinate value, the correction unit coordinates values in the color space of the first orthogonal coordinate system of the pixel to be corrected May be corrected to a third orthogonal coordinate value between the first orthogonal coordinate value and the second orthogonal coordinate value. In addition, when the coordinate value in the color space of the first orthogonal coordinate system of the correction target pixel is a fourth orthogonal coordinate value between the first orthogonal coordinate value and the second orthogonal coordinate value, The coordinate value in the color space of the first orthogonal coordinate system of the pixel to be corrected may be corrected to a fifth orthogonal coordinate value between the third orthogonal coordinate value and the second orthogonal coordinate value.

  In addition, the method for the 1st determination part and the 2nd determination part to determine a 1st polar coordinate value and a 2nd polar coordinate value is not specifically limited. For example, the first determination unit may determine the first polar coordinate value by calculating an average value of coordinate values in the color space of the polar coordinate system of each pixel included in the first pixel group. Further, the second determination unit may determine the second polar coordinate value by calculating an average value of coordinate values in the color space of the polar coordinate system of each pixel included in the second pixel group. The first polar coordinate value and the second polar coordinate value may be values obtained by other methods (for example, median value).

The above image processing apparatus is configured to convert specific image data in which the value of each pixel is represented by coordinate values in a color space of a second orthogonal coordinate system different from the first orthogonal coordinate system, using coordinate values in the color space of a polar coordinate system. You may further provide the conversion part which converts into the target image data in which the value of a pixel is represented. In this case, the correction unit may generate corrected image data in which the value of each pixel is represented by the coordinate value in the color space of the second orthogonal coordinate system by executing the following processes.
(1) The coordinate value of each pixel in the target image data is converted into a specific orthogonal coordinate value that is a coordinate value in the color space of the first orthogonal coordinate system.
(2) Each specific orthogonal coordinate value included in the surrounding area is corrected so as to approach the second orthogonal coordinate value.
(3) Each corrected specific orthogonal coordinate value is converted into a specific polar coordinate value that is a coordinate value in the color space of the polar coordinate system.
(4) Each specific polar coordinate value is converted into an orthogonal coordinate value that is a coordinate value in the color space of the second orthogonal coordinate system.

  A computer program for realizing the image processing apparatus is also useful. By using this computer program, it is possible to realize an image processing apparatus that can appropriately correct the hue of a specific range of target image data using reference image data.

Here, some of the techniques described in the following examples are listed.
(Mode 1) The polar coordinate system may be a cylindrical coordinate system. The cylindrical coordinate system may have an axis indicating luminance. In the above cylindrical coordinate system, the saturation may be expressed by the radius. In the above cylindrical coordinate system, the hue may be expressed by a declination angle.

(Mode 2) The correction unit determines an upper limit value (for example, Uv in the embodiment) and a lower limit value (for example, Lv in the embodiment) based on the first orthogonal coordinate value and the second orthogonal coordinate value. Also good. In addition, when the value of the pixel to be corrected is between the upper limit value and the lower limit value, the correction unit corrects the correction reference value (for example, Curvv in the embodiment) based on the first orthogonal coordinate value and the second orthogonal coordinate value. May be determined.

(Mode 3) The correction unit may calculate a distance between the first orthogonal coordinate value and the second orthogonal coordinate value (for example, the range of the embodiment). The correction unit may determine the correction amount based on the distance between the first orthogonal coordinate value and the second orthogonal coordinate value and the sum of the first distance and the second distance.

(First embodiment)
Embodiments will be described with reference to the drawings. FIG. 1 shows a schematic diagram of a multi-function device 10 of this embodiment. The multi-function device 10 includes an operation unit 12, a display unit 14, a slot unit 16, a scanner unit 20, a printing unit 22, a network interface 24, a control unit 28, and the like. The operation unit 12 includes a plurality of keys. The user can input various instructions and commands to the multi-function device 10 by operating the operation unit 12. The display unit 14 can display various information. The slot portion 16 has a space for accommodating the memory card 18. The user can insert the memory card 18 storing the image data into the slot unit 16. In this embodiment, the image data stored in the memory card 18 is RGB format image data.

  The scanner unit 20 has a reading mechanism such as a CCD or CIS. The scanner unit 20 can generate image data by color scanning a document, a photograph, or the like. In this embodiment, the image data generated by the scanner unit 20 is RGB format image data. The printing unit 22 has a printing mechanism such as an ink jet type or a laser type. The printing unit 22 can print based on image data stored in the memory card 18 or image data generated by the scanner unit 20. The printing unit 22 can print based on data input to the network interface 24. The network interface 24 is connected to the LAN 26.

  The control unit 28 includes a CPU 30, a ROM 32, a RAM 40, and the like. The CPU 30 executes various processes according to the programs 34 to 38 stored in the ROM 32. The ROM 32 stores various programs 34 to 38. The basic program 34 is a program for controlling basic operations of the multi-function device 10. The basic program 34 includes, for example, a program for generating data displayed on the display unit 14. The basic program 34 includes, for example, a program for controlling the scanner unit 20, the printing unit 22, and the like. The correction program 36 is a program for correcting image data. The contents of the processing executed by the CPU 30 according to the correction program 36 will be described in detail later. The program 38 is a program other than the programs 34 and 36 described above. The RAM 40 can temporarily store data generated while the CPU 30 executes processing.

  The multi-function device 10 of the present embodiment can correct the image data (image data selected by the user) stored in the memory card 18 using the image data generated by the scanner unit 20. More specifically, the multi-function device 10 stores the image data stored in the memory card 18 so as to approach the colors in the hue (the hue selected by the user) of the image data generated by the scanner unit 20. Colors within the hue of can be corrected.

  Hereinafter, the image data to be corrected (image data stored in the memory card 18) is referred to as RGB target image data. Further, image data serving as a correction reference (image data generated by the scanner unit 20) is referred to as RGB reference image data. FIG. 2 simply shows how the RGB target image data 50 is corrected using the RGB reference image data 60. Details of the processing will be described later in detail, but here, a procedure for correcting the RGB target image data 50 will be briefly described.

  The RGB target image data 50 has a plurality of pixels. The value of each pixel is represented by a value in a color space of a three-axis orthogonal coordinate system (hereinafter sometimes referred to as an RGB color space). Specifically, the value of each pixel includes a value (0 to 255) indicating R (red), a value (0 to 255) indicating G (green), and a value (0 to 0) indicating B (blue). 255). The multi-function device 10 converts the RGB target image data 50 into image data 52 in the HSV format. Hereinafter, this image data 52 is referred to as HSV target image data 52. The multi-function device 10 generates the HSV target image data 52 by converting the value (orthogonal coordinate value) of each pixel constituting the RGB target image data 50 into a polar coordinate value. Therefore, the number of pixels included in the HSV target image data 52 is equal to the number of pixels included in the RGB target image data 50. The value of each pixel included in the HSV target image data 52 is represented by a value in a color space of a cylindrical polar coordinate system (hereinafter sometimes referred to as an HSV color space). Specifically, the value of each pixel indicates a value (0 to 360 °) indicating hue (H), a value (0 to 1.0) indicating saturation (S), and luminance (V). It is represented by a value (0 to 1.0).

  The multi-function device 10 converts the RGB reference image data 60 into the image data 62 in the HSV format in the same manner as described above for converting the RGB target image data 50 into the HSV target image data 52. Hereinafter, the image data 62 is referred to as HSV reference image data 62. The value of each pixel included in the RGB reference image data 60 is represented by a value in the RGB color space. Further, the value of each pixel included in the HSV reference image data 62 is represented by a value in the HSV color space.

  The multi-function device 10 corrects the HSV target image data 52 using the HSV reference image data 62 by executing various processes described later. Thereby, the corrected image data 56 is generated. Hereinafter, this image data 56 is referred to as corrected RGB image data 56. The value of each pixel included in the corrected RGB image data 56 is represented by a value in the RGB color space.

  Next, the content of the correction printing process executed by the multi-function device 10 will be described in detail. The user can select the corrected printing mode by operating the operation unit 12. When this operation is executed, the correction printing process is executed. FIG. 3 shows a flowchart of the correction printing process. The correction printing process is executed by the CPU 30 according to the basic program 34 and the correction program 36. Of the processes shown in the flowchart of FIG. 3, the process of S <b> 18 is executed according to the correction program 36, and the other processes are executed according to the basic program 34.

  The user can select one image data from among a plurality of image data stored in the memory card 18 by operating the operation unit 12. The image data selected here becomes image data to be corrected (that is, RGB target image data 50). The CPU 30 waits until the RGB target image data 50 is selected (S10).

  The user can cause the scanner unit 20 to scan a reference image (for example, a photograph) used to correct the RGB target image data 50 by operating the operation unit 12. The CPU 30 waits until this operation is executed (S12). When an operation is executed by the user, the CPU 30 instructs the scanner unit 20 to scan (S14). Accordingly, the scanner unit 20 generates image data (that is, RGB reference image data 60) by performing color scanning on the reference image selected by the user.

  Next, the CPU 30 displays a character string indicating a plurality of correction modes on the display unit 14. Specifically, the CPU 30 causes the display unit 14 to display character strings “empty mode”, “green mode”, and “skin mode”. Each correction mode will be briefly described. The sky mode is a mode for correcting the color in the blue hue of the RGB target image data 50 so as to approach the color in the blue hue of the RGB reference image data 60. The green mode is a mode for correcting the color in the green hue of the RGB target image data 50 so as to approach the color in the green hue of the RGB reference image data 60. The skin mode is a mode for correcting the color in the skin color hue of the RGB target image data 50 so as to approach the color in the skin color hue of the RGB reference image data 60. The user can select one correction mode from the three correction modes displayed on the display unit 14 by operating the operation unit 12. The CPU 30 waits until this operation is executed (S16). Information indicating the correction mode selected by the user is stored in the RAM 40.

  Next, the CPU 30 executes a correction process (S18). The contents of the correction process will be described in detail later. The corrected RGB image data 56 is obtained by the correction process. The CPU 30 converts the corrected RGB image data 56 into print data (CMYK format bitmap data) (S20). Subsequently, the CPU 30 instructs the printing unit 22 to perform printing based on the print data generated in S20 (S22). As a result, the printing unit 22 prints on the print medium based on the print data. The user can obtain a print medium on which the corrected RGB image data 56 is printed.

  Next, the details of the correction process executed in S18 will be described in detail. 4 and 5 show a flowchart of the correction process. The CPU 30 converts the RGB target image data 50 into HSV target image data 52 (S40). FIG. 6 shows mathematical formulas for converting coordinate values in the RGB color space into coordinate values in the HSV color space. The CPU 30 selects one pixel (conversion target pixel) constituting the RGB target image data 50. As shown in FIG. 6A, the coordinate value of the pixel to be converted is expressed as (R, G, B), and the coordinate value in the HSV color space converted from the coordinate value is expressed as (H, S, V). As shown in FIG. 6B, the CPU 30 calculates R / 255, G / 255, and B / 255, respectively, and specifies the largest value among them as V. Further, the CPU 30 calculates S according to the mathematical formula shown in FIG. Next, the CPU 30 calculates parameters r, g, and b according to the mathematical formula shown in FIG. The CPU 30 calculates H according to the mathematical formula shown in FIG. When the calculated H is a negative value, the CPU 30 calculates new H by adding 360 to H as shown in FIG. Thereby, the coordinate value (H, S, V) in the HSV color space can be obtained from the coordinate value (R, G, B) of the pixel to be converted.

  FIG. 7 schematically shows the HSV color space. It can be seen from FIG. 7 that the HSV color space is a cylindrical coordinate system color space. H indicating the hue is any value from 0 to 360 °. S indicating saturation is any value from 0 to 1.0. V which shows a brightness | luminance is any value of 0-1.0.

  The CPU 30 calculates the coordinate values (H, S, V) in the HSV color space for each of all other pixels of the RGB target image data 50 according to the mathematical expressions of FIGS. 6 (a) to 6 (f). . Thereby, the HSV target image data 52 is obtained. The CPU 30 stores the HSV target image data 52 in the RAM 40.

  Subsequently, the CPU 30 converts the RGB reference image data 60 into HSV reference image data 62 (S42). The process of S42 is performed by the same method as the process of S40 described above. That is, the CPU 30 calculates the coordinate values (H, S, V) in the HSV color space for each of all the pixels of the RGB reference image data 60 according to FIGS. 6 (a) to 6 (f). Thereby, the HSV reference image data 62 is obtained. The CPU 30 stores the HSV reference image data 62 in the RAM 40.

  Next, the CPU 30 calculates a representative value from the HSV target image data 52 (S44). Hereinafter, this representative value is referred to as a target representative value (OH, OS, OV). The process of S44 is executed according to the correction mode selected in S16 of FIG. FIG. 8 shows an example of the correspondence relationship between the correction mode and the ranges of H, S, and V. For example, when the correction mode is the empty mode, H is 180 to 240 °, S is 0 to 1.0, and V is 0 to 1.0. This is based on the pixel group in which H is 180 to 240 °, S is 0 to 1.0, and V is 0 to 1.0 when the correction mode is the empty mode. This means that a representative value is calculated.

  More specifically, when the correction mode is the empty mode, the CPU 30 is a pixel group in which H is 180 to 240 ° (see FIG. 8A) among all the pixels constituting the HSV target image data 52. Is identified. Here, S and V are not considered. Each range of S and V in the sky mode is 0 to 1.0 (see FIG. 8A), and S and V are included in the range of 0 to 1.0 in any pixel. Because. Next, the CPU 30 specifies OH by calculating an average value of H of the specified pixel group, specifies an OS by calculating an average value of S of the specified pixel group, and specifies the specified pixel group OV is specified by calculating the average value of V. Thereby, the target representative values (OH, OS, OV) are obtained. When the correction mode selected in S16 in FIG. 3 is the green mode, the CPU 30 determines the target representative value (OH, OS, OV) from the pixel group in which H is 50 to 170 ° (see FIG. 8B). Is identified. Further, when the correction mode selected in S16 of FIG. 3 is the skin mode, the CPU 30 determines that H is 10 to 40 °, S is 0.1 to 0.6, and V is 0.2 to The target representative value (OH, OS, OV) is specified from the pixel group that is 1.0 (see FIG. 8C).

  Subsequently, the CPU 30 calculates a representative value from the HSV reference image data 62 (S46). Hereinafter, this representative value is referred to as a reference representative value (SH, SS, SV). The process of S46 is executed similarly to the process of S44. For example, when the correction mode selected in S16 of FIG. 3 is the empty mode, the CPU 30 has H of 180 to 240 ° (see FIG. 8A) among all the pixels constituting the HSV reference image data 62. A certain pixel group is specified. Next, the CPU 30 specifies SH by calculating an average value of H of the specified pixel group, specifies SS by calculating an average value of S of the specified pixel group, and specifies the specified pixel group SV is specified by calculating the average value of V. Thereby, the reference representative values (SH, SS, SV) are obtained.

  Next, the CPU 30 converts the target representative value (OH, OS, OV) specified in S44 into a coordinate value in the vab color space (S48). Hereinafter, these coordinate values are referred to as target coordinate values (Ov, Oa, Ob). The vab color space is a three-axis orthogonal coordinate color space different from the RGB color space. The vab color space is a color space unique to the present embodiment, and is not a well-known color space. The contents of the processing of S48 will be described with reference to FIG. FIG. 9 shows mathematical formulas for converting coordinate values in the HSV color space into coordinate values in the vab color space. In FIG. 9, the coordinate value to be converted is expressed as (H, S, V), and the coordinate value in the vab color space converted from the coordinate value is expressed as (v, a, b). In the equation shown in FIG. 9, when (H, S, V) is replaced with (OH, OS, OV) and (v, a, b) is replaced with (Ov, Oa, Ob), the target coordinate value ( Equations for obtaining Ov, Oa, Ob) are obtained. The CPU 30 calculates target coordinate values (Ov, Oa, Ob) according to this mathematical formula.

  Next, the CPU 30 converts the reference representative values (SH, SS, SV) specified in S46 into coordinate values in the vab color space (S50). Hereinafter, this coordinate value is referred to as a reference coordinate value (Sv, Sa, Sb). The process of S50 is executed similarly to the process of S48. In the equation shown in FIG. 9, when (H, S, V) is replaced with (SH, SS, SV) and (v, a, b) is replaced with (Sv, Sa, Sb), the reference coordinate value ( Equations for obtaining Sv, Sa, Sb) are obtained. The CPU 30 calculates reference coordinate values (Sv, Sa, Sb) according to this mathematical formula.

  Subsequently, the CPU 30 selects one pixel constituting the HSV target image data 52 (S52). The coordinate value of the pixel selected in S52 (hereinafter sometimes referred to as a pixel value) is expressed as (PH, PS, PV). The CPU 30 converts the coordinate values (PH, PS, PV) of the pixel selected in S52 into coordinate values (Pv, Pa, Pb) in the vab color space (S54). The process of S54 is performed similarly to the process of S48. In the equation shown in FIG. 9, when (H, S, V) is replaced with (PH, PS, PV) and (v, a, b) is replaced with (Pv, Pa, Pb), the coordinate value (Pv , Pa, Pb) is obtained. The CPU 30 calculates coordinate values (Pv, Pa, Pb) according to this mathematical formula.

  Next, the CPU 30 calculates the correction amount (Cv, Ca, Cb) of the coordinate values (Pv, Pa, Pb) (S56). FIG. 10 shows mathematical formulas for calculating the correction amount. The CPU 30 calculates Cv based on Pv calculated in S54, Ov calculated in S48, and Sv calculated in S50. First, the CPU 30 calculates the lower limit value Lv and the upper limit value Uv according to the mathematical formula shown in FIG. Next, the CPU 30 calculates Curvv according to the mathematical formula shown in FIG. It should be noted that AdjS and AdjO included in the formula shown in FIG. 10B are not constant values depending on Pv, Ov, and Sv, but are preset constants. Subsequently, the CPU 30 calculates Cv according to the mathematical formula shown in FIG.

  FIG. 11 shows a graph in which Pv is the horizontal axis and the sum of Pv and Cv (referred to herein as Mv) is the vertical axis. The graph of FIG. 11 exemplifies a case where Ov is 0.25 and Sv is 0.38. Mv is a value after correction when Pv is corrected by Cv. As will be described in detail later, in this embodiment, Mv is not obtained only by adding Cv to Pv. However, in order to facilitate understanding of Cv, in FIG. 11, the sum of Pv and Cv is Mv.

  As is apparent from the graph of FIG. 11, when Pv is a value between the lower limit value Lv and the upper limit value Uv, Pv is a correction curve that approaches Sv. This correction curve is divided into a region convex upward and a region convex downward with Sv as a boundary. For example, if Pv is equal to Ov, Cv is C1. In this case, Mv is V1, which is the sum of Ov and C1. V1 is a value between Ov and Sv. For example, when Pv is a value V2 between Ov and Sv, Cv is C2. In this case, Mv is V3 which is the sum of V2 and C2. V3 is a value between V2 and Sv. More specifically, V3 is a value between V1 and Sv. That is, the closer the value Pv before correction is to Sv, the closer the value Mv after correction is to Sv. It should be noted that FIG. 12A also shows how the correction is made to V1 when Pv is Ov, and how the correction is made to V3 when Pv is V2.

  The CPU 30 calculates the correction amount Ca using the same method as that for calculating the correction amount Cv. The CPU 30 calculates Ca based on Pa, Oa, and Sa. Ca is calculated according to a mathematical formula similar to the mathematical formula shown in FIG. Specifically, in FIGS. 10A and 10B, Lv is changed to La, Uv is changed to Ua, Ov is changed to Oa, Sv is changed to Sa, and Curvv is changed to Curva. By changing to, Curva can be calculated. The CPU 30 calculates Ca according to the mathematical formula shown in FIG.

  When a graph is drawn in which Pa is the horizontal axis and the sum of Pa and Ca (herein referred to as Ma) is the vertical axis, a graph similar to the graph of FIG. 11 is obtained. That is, when Pa is a value between the lower limit value La and the upper limit value Ua, Pa is corrected so as to approach Sa. FIG. 12B shows an example of how Pa is corrected. When Pa is equal to Oa, Pa is corrected to a value A1 between Oa and Sa. When Pa is a value A2 between Oa and Sa, Pa is corrected to a value A3 between A1 and Sa. That is, the closer the value Pa before correction is to Sa, the closer the value after correction is to Sa.

  Further, the CPU 30 calculates the correction amount Cb by using the same method as that for calculating the correction amount Cv. The CPU 30 calculates Cb based on Pb, Ob, and Sb. 10 (a) and 10 (b), when Lv is changed to Lb, Uv is changed to Ub, Ov is changed to Ob, Sv is changed to Sb, and Curvv is changed to Curvb, Curvb is changed to Curvb. Can be calculated. The CPU 30 calculates Cb according to the mathematical formula shown in FIG.

  When a graph is drawn in which Pb is the horizontal axis and the sum of Pb and Cb (referred to herein as Mb) is the vertical axis, a graph similar to the graph of FIG. 11 is obtained. That is, when Pb is a value between the lower limit value Lb and the upper limit value Ub, Pb is corrected so as to approach Sb. FIG. 12C shows an example of how Pb is corrected. When Pb is equal to Ob, Pb is corrected to a value B1 between Ob and Sb. When Pb is a value B2 between Ob and Sb, Pb is corrected to a value B3 between B1 and Sb. That is, the closer the value Pb before correction is to Sb, the closer the value after correction is to Sb.

  As is clear from FIG. 12, when the coordinate values (Pv, Pa, Pb) of the correction target pixel (the pixel selected in S52 of FIG. 4) are equal to the target coordinate value (Ov, Oa, Ob), the coordinate value (Pv, Pa, Pb) is corrected to coordinate values (V1, A1, B1) between the target coordinate values (Ov, Oa, Ob) and the reference coordinate values (Sv, Sa, Sb). Further, the coordinate values (Pv, Pa, Pb) of the correction target pixel are coordinate values (V2, A2, B2) between the target coordinate values (Ov, Oa, Ob) and the reference coordinate values (Sv, Sa, Sb). The coordinate values (V2, A2, B2) are corrected to coordinate values (V3, A3, B3) between the coordinate values (V1, A1, B1) and the reference coordinate values (Sv, Sa, Sb). The

  After calculating the correction amount (Cv, Ca, Cb), the CPU 30 proceeds to S58 in FIG. In S58, the CPU 30 calculates a parameter d and a parameter range. FIG. 13A shows mathematical formulas for calculating the parameter d and the parameter range. The CPU 30 calculates the parameter d and the parameter range according to the mathematical formula shown in FIG. The parameter d includes the distance between the coordinate value (Pv, Pa, Pb) of the correction target pixel and the target coordinate value (Ov, Oa, Ob), and the coordinate value (Pv, Pa, Pb) of the correction target pixel. And the distance between the reference coordinate value (Sv, Sa, Sb). The parameter range is a distance between the target coordinate value (Ov, Oa, Ob) and the reference coordinate value (Sv, Sa, Sb).

  Next, the CPU 30 calculates the parameter w based on the parameter d and the parameter range (S60). FIG. 13B shows mathematical formulas for calculating w. In the equation of FIG. 13B, the value of sui is set according to the correction mode selected in S16 of FIG. FIG. 13C shows the correspondence between the correction mode and the value of sui. The value of sui in the green mode is the largest, and the value of sui in the skin mode is the smallest. The value of sui in the case of the sky mode is a value between the value of sui in the case of the green mode and the value of sui in the case of the skin mode. As can be seen from FIG. 13B, when d is equal to range, w is the largest value (1.0). Further, when d is larger than range, w decreases as d increases. If d is greater than the value obtained by multiplying range and sui, then w is the smallest value (0). According to the equation in FIG. 13B, it can be said that w is a value that decreases as d increases. Conversely, it can be said that w is a value that increases as d decreases.

  FIG. 14 shows a graph where Pv is the horizontal axis and w is the vertical axis. The graph of FIG. 14 exemplifies a case where Ov is 0.2 and Sv is 0.31. As is clear from the graph of FIG. 14, when Pv is a value between Ov and Sv (that is, when d = range), w is 1.0. As Pv becomes smaller than Ov, w decreases smoothly. Also, as Pv becomes larger than Sv, w becomes smaller smoothly. As Pv becomes farther from both Ov and Sv, w becomes smaller. In other words, w is larger as Pv is closer to both Ov and Sv. A graph in which Pa is the horizontal axis and w is the vertical axis, and a graph in which Pb is the horizontal axis and w is the vertical axis also show the same shape as the graph of FIG.

  The CPU 30 corrects the coordinate value (Pv, Pa, Pb) of the pixel to be corrected based on the parameter w and the correction amount (Cv, Ca, Cb) (S62). As a result, corrected coordinate values (Mv, Ma, Mb) are obtained. FIG. 15 shows mathematical formulas for calculating the corrected coordinate values (Mv, Ma, Mb). The CPU 30 adds the coordinate value obtained by multiplying the correction amount (Cv, Ca, Cb) by the parameter w to the coordinate value (Pv, Pa, Pb) of the correction target pixel, thereby correcting the corrected coordinate. Values (Mv, Ma, Mb) are calculated.

  As described above, the actual correction amount is calculated by multiplying w by the correction amount (Cv, Ca, Cb). w is calculated according to the mathematical formula shown in FIG. 13B, and is a value that increases as d decreases. As d decreases, the actual correction amount (w × (Cv, Ca, Cb)) increases. That is, the smaller the d is, the larger the coordinate values (Pv, Pa, Pb) of the pixel to be corrected are corrected. FIG. 16 shows the target coordinate value O and the reference coordinate value S on the va plane. FIG. 16 means that the closer the color is to white, the greater the correction is made. Conversely, FIG. 16 means that the color is corrected to be smaller as it approaches black. A completely black area means no correction. FIG. 16 also shows that the smaller the parameter d is, the larger the coordinate values (Pv, Pa, Pb) of the pixel to be corrected are corrected.

  In FIG. 16, a completely black region is a region outside the ellipse having the target coordinate value O and the reference coordinate value S as two focal points. That is, when the coordinate value (Pv, Pa, Pb) of the pixel to be corrected exists in a region outside the ellipse having the target coordinate value O and the reference coordinate value S as two focal points, the coordinate value is not corrected. In other words, if the coordinate value (Pv, Pa, Pb) of the pixel to be corrected exists in the area inside the ellipse having the target coordinate value O and the reference coordinate value S as two focal points, the coordinate value is It is corrected. Although FIG. 16 is represented by the va plane, the same applies to the vb plane and the ab plane. That is, in the vab space, when the coordinate value (Pv, Pa, Pb) of the correction target pixel exists in an area outside the ellipsoid having the target coordinate value O and the reference coordinate value S as two focal points, The value is not corrected. In other words, in the vab space, the coordinate value (Pv, Pa, Pb) of the pixel to be corrected exists in the area inside the ellipsoid having the target coordinate value O and the reference coordinate value S as two focal points. The coordinate value is corrected.

  After calculating the corrected coordinate values (Mv, Ma, Mb) in S62, the CPU 30 proceeds to S64. In S64, the CPU 30 converts the corrected coordinate values (Mv, Ma, Mb) into coordinate values (MH, MS, MV) in the HSV color space. FIG. 17 shows mathematical formulas for converting coordinate values in the vab color space into coordinate values in the HSV color space. In FIG. 17, the coordinate value to be converted is expressed as (v, a, b), and the coordinate value in the HSV color space converted from the coordinate value is expressed as (H, S, V). In the equation shown in FIG. 17, when (v, a, b) is replaced with (Mv, Ma, Mb) and (H, S, V) is replaced with (MH, MS, MV), the coordinate value (MH , MS, MV). The CPU 30 calculates coordinate values (MH, MS, MV) according to this mathematical formula.

  Next, the CPU 30 converts the coordinate values (MH, MS, MV) specified in S64 into coordinate values (MR, MG, MB) in the RGB color space (S66). FIG. 18 shows mathematical formulas for converting coordinate values in the HSV color space into coordinate values in the RGB color space. In FIG. 18, the coordinate value to be converted is expressed as (H, S, V), and the coordinate value in the RGB color space converted from the coordinate value is expressed as (R, G, B). In the mathematical formula shown in FIG. 18, when (H, S, V) is replaced with (MH, MS, MV) and (R, G, B) is replaced with (MR, MG, MB), the coordinate value (MR , MG, MB) is obtained. As shown in FIG. 18A, the CPU 30 calculates parameter variables in and fl from MH. As shown in FIG. 18B, when the parameter variable in is an even number, the CPU 30 subtracts fl from 1 to calculate a new fl. Next, the CPU 30 calculates parameter variables m and n from the MS and the MV according to the mathematical formula shown in FIG. CPU30 calculates MR, MG, and MB from MV, m, and n according to the numerical formula shown by FIG.18 (d). Thereby, coordinate values (MR, MG, MB) in the RGB color space are obtained.

  By executing the processing of S54 to S66, the coordinate value (PH, PS, PV) of one pixel in the HSV target image data 52 is corrected, and the coordinate value in the RGB color space is corrected as the coordinate value. Values (MR, MG, MB) are obtained. The CPU 30 determines whether or not the processing of S54 to S66 has been executed for all the pixels in the HSV target image data 52 (S68). In the case of NO here, the CPU 30 selects the next pixel in the HSV target image data 52 (S70). CPU30 performs the process of S54-S66 about the pixel selected by S70. On the other hand, if YES in S68, the correction process ends. In this case, corrected RGB image data 56 (see FIG. 2) in which the value of each pixel is expressed by the coordinate value in the RGB color space obtained by the processing of S54 to S66 is obtained.

  The multi-function device 10 of the present embodiment has been described in detail. The multi-function device 10 executes processing in the HSV color space (see FIG. 7) in S44 and S46 of FIG. For example, in S44 of FIG. 4, the multi-function device 10 identifies a pixel group having a hue H (for example, 180 to 240 ° in the sky mode) according to the correction mode desired by the user from the HSV target image data 52, A target representative value is calculated from the pixel group.

  Since the target representative value and the reference representative value are obtained in the HSV color space, it may be possible to adopt a method of correcting each pixel in the HSV target image data 52 in the HSV color space. However, in this case, the following problem occurs. FIG. 19 shows the HS plane in the HSV color space. Α shown in FIG. 19 indicates a hue H (for example, 180 to 240 ° in the sky mode) according to the correction mode. When the correction method in the HSV color space is employed, when two pixels 120 and 124 having a low saturation S are present at the positions shown in FIG. 19, the pixel 120 is corrected and the pixel 124 is not corrected. . As a result, the two pixels 120 and 124 that have almost no difference in the apparent color before correction become greatly different colors after correction. In order to cope with the above problem by the method of correcting in the HSV color space, it is necessary to correct the pixel group included in the area indicated by hatching in FIG. However, in order to define such a region including the V axis in the HSV color space, it is necessary to define S and V for each of various values of H, which requires a huge amount of calculation.

  On the other hand, the multi-function device 10 of the present embodiment executes correction in the vab color space of the orthogonal coordinate system. That is, the multi-function device 10 converts the target representative value (OH, OS, OV) into the target coordinate value (Ov, Oa, Ob) in the vab color space (S48 in FIG. 4), and the target coordinate value (Ov , Oa, Ob) are the values of the pixels included in the area around the object (the area within the ellipsoid having the target coordinate values (Ov, Oa, Ob) and the reference coordinate values (Sv, Sa, Sb) as two focal points). to correct. For this reason, even when there is a pixel 124 that does not have a hue corresponding to the correction mode, the value of the pixel 124 can be corrected. It is possible to prevent the two pixels 120 and 124 that have almost no difference in apparent color before correction from becoming significantly different colors after correction. The multi-function device 10 according to this embodiment can appropriately correct the HSV target image data 52 using the HSV reference image data 62.

  Further, as shown in the mathematical formula of FIG. 13, in the multi-function device 10, the range of the pixel value to be corrected (range × sui) increases as the value of sui set according to the correction mode increases. It is set to be. sui is the largest in the green mode and the smallest in the skin mode. Therefore, when correcting green, which is difficult to recognize color changes, the range of pixel values to be corrected becomes large, and when correcting skin colors that are easy to recognize color changes, pixels to be corrected The range of values becomes smaller. The range of pixel values to be corrected is appropriately set in view of human visual characteristics.

  Note that the correction curve shown in FIG. 11 is an example when w is 1.0. In the example of FIG. 11, V4 is not corrected, and V5 is corrected to V6. Since the values of V4 and V5 are close, the appearance colors are similar. When the pixel corresponding to V4 and the pixel corresponding to V5 are adjacent to each other, the color is greatly different between the former pixel that is not corrected (ie, V = V4) and the latter pixel that is corrected (ie, V = V6). there is a possibility. In order to suppress the occurrence of such an event, the following configuration is adopted in this embodiment. That is, the distance between the target coordinate value (Ov, Oa, Ob) and the coordinate value (Pv, Pa, Pb) of the correction target pixel, the reference coordinate value (Sv, Sa, Sb) and the correction target pixel. The parameter w is employed so that the correction amount increases as the sum d with the distance between the coordinate values (Pv, Pa, Pb) decreases. The parameter w is a value that decreases as d increases (see FIG. 13B). Since the pixel corresponding to V5 shown in FIG. 11 is located far from both the target coordinate value (Ov in FIG. 11) and the reference coordinate value (Sv in FIG. 11), the correction amount is small. As a result, even when the pixel corresponding to V4 and the pixel corresponding to V5 are adjacent to each other, it is possible to suppress a large difference in color between the pixels.

  The relationship between each element described in the claims and the first embodiment will be described. The image processing apparatus according to the claims corresponds to the multi-function device 10. The target image data in the claims corresponds to the HSV target image data 52. The reference image data in the claims corresponds to the HSV reference image data 62. The corrected image data in the claims corresponds to the corrected RGB image data 56. The specific range of the claims corresponds to the range of hue H (see FIG. 8) determined according to each correction mode. The polar coordinate system color space in the claims corresponds to the HSV color space. The first polar coordinate value in the claims corresponds to the target representative value (OH, OS, OV). The first determination unit in the claims corresponds to the CPU 30 that executes the process of S44 of FIG. The first orthogonal coordinate value in the claims corresponds to the target coordinate value (Ov, Oa, Ob). The color space of the first orthogonal coordinate system in the claims corresponds to the vab color space. The first calculation unit in the claims corresponds to the CPU 30 that executes the process of S48 of FIG. The second polar coordinate value in the claims corresponds to the reference representative value (SH, SS, SV). The second determination unit in the claims corresponds to the CPU 30 that executes the process of S46 of FIG. The second orthogonal coordinate value in the claims corresponds to the reference coordinate value (Sv, Sa, Sb). The second calculation unit in the claims corresponds to the CPU 30 that executes the process of S50 of FIG. The area around the first orthogonal coordinate value (the specific area and the area within the ellipsoid) in the claims corresponds to an area of w> 0 (an area other than black in FIG. 16) obtained according to the mathematical formula of FIG. The correction unit in the claims corresponds to the CPU 30 that executes the processes of S52 to S56 in FIG. 4 and S58 to S70 in FIG.

  The sum of the first distance and the second distance in the claims corresponds to the parameter d. The third orthogonal coordinate value in the claims corresponds to (V1, A1, B1) in FIG. The fourth orthogonal coordinate value in the claims corresponds to (V2, A2, B2) in FIG. The fifth orthogonal coordinate value in the claims corresponds to (V3, A3, B3) in FIG. The color space of the second orthogonal coordinate system in the claims corresponds to the RGB color space. The specific image data in the claims corresponds to the RGB target image data 50. The conversion unit in the claims corresponds to the CPU 30 that executes the process of S40 of FIG. The specific orthogonal coordinate value in the claims corresponds to the coordinate value (Pv, Pa, Pb). The specific polar coordinate value in the claims corresponds to the coordinate value (MH, MS, MV).

(Second embodiment)
Subsequently, a second embodiment will be described. In the present embodiment, the content of the correction process in S18 of FIG. 3 is different from that of the first embodiment. In the first embodiment, the coordinate value of the correction target pixel is corrected in the vab color space. In contrast, in this embodiment, the coordinate value of the correction target pixel is corrected in the RGB color space.

  FIG. 21 shows a flowchart of the correction process of this embodiment. The CPU 30 calculates the target representative value (OH, OS, OV) and the reference representative value (SH, SS, SV) by executing the processes of S40 to S46 in FIG. Next, the CPU 30 converts the target representative values (OH, OS, OV) into target coordinate values (OR, OG, OB) in the RGB color space (S148). Further, the CPU 30 converts the reference representative values (SH, SS, SV) into target coordinate values (SR, SG, SB) in the RGB color space (S150). The processing of S148 and S150 is executed according to the mathematical formula shown in FIG.

  Next, the CPU 30 selects one pixel constituting the RGB target image data 50 (S152). Hereinafter, the coordinate value of the pixel selected in S152 is expressed as (PR, PG, PB). The CPU 30 calculates the correction amount (CR, CG, CB) of the coordinate values (PR, PG, PB) of the pixel selected in S152. This correction amount is calculated by the mathematical formula shown in FIG. Specifically, in FIG. 10, Lv is changed to LR, Uv is changed to UR, Ov is changed to OR, Sv is changed to SR, Pv is changed to PR, and Curvv is changed to CurvR. If Cv is changed to CR, CR can be calculated. In this embodiment, the description “Lv = 1.0 when Lv> 1.0” in FIG. 10A is changed to “LR = 255 when LR> 255”. Further, the description “Uv = 1.0 when Uv> 1.0” in FIG. 10A is changed to “UR = 255 when UR> 255”. The method for calculating CG and CB is the same as the method for calculating CR.

  Subsequently, the CPU 30 calculates a parameter d and a parameter range (S158). The parameter d and the parameter range are calculated according to the mathematical formula shown in FIG. The parameter d includes the distance between the coordinate value (PR, PG, PB) of the pixel to be corrected and the target coordinate value (OR, OG, OG), and the coordinate value (PR, PG, PB) of the pixel to be corrected. And the distance between the reference coordinate values (SR, SG, SB). The parameter range is a distance between the target coordinate value (OR, OG, OB) and the reference coordinate value (SR, SG, SB).

  Next, the CPU 30 calculates the parameter w based on the parameter d and the parameter range (S160). The parameter w is calculated according to the mathematical formula shown in FIG. Subsequently, the CPU 30 corrects the coordinate values (PR, PG, PB) of the pixel to be corrected based on the parameter w and the correction amount (CR, CG, CB) (S162). As a result, corrected coordinate values (MR, MG, MB) are obtained. The process of S162 is calculated according to the mathematical formula shown in FIG. That is, the CPU 30 adds the coordinate value obtained by multiplying the parameter w by the correction amount (CR, CG, CB) to the coordinate value (PR, PG, PB) of the pixel to be corrected, thereby correcting the corrected value. Are calculated (MR, MG, MB).

  After calculating the corrected coordinate values (MR, MG, MB) in S162, the CPU 30 proceeds to S168. In S168, the CPU 30 determines whether or not the processing in S156 to S162 has been executed for all the pixels in the RGB target image data 50. In the case of NO here, the CPU 30 selects the next pixel in the RGB target image data 50 (S170), and executes the processing of S156 to S162. On the other hand, if YES in S168, the correction process ends. In this case, corrected RGB image data 56 (see FIG. 2) is obtained.

  In the first embodiment, after the corrected coordinate values (Mv, Ma, Mb) are calculated in S62 of FIG. 5, the coordinate values (Mv, Ma, Mb) are converted into the coordinate values (MH, MS) in the HSV color space. , MV), and a process (S66) for converting the coordinate values (MH, MS, MV) to the coordinate values (MR, MG, MB) in the RGB color space. On the other hand, in this embodiment, since the coordinate values (MR, MG, MB) obtained by the process of S162 in FIG. 21 are coordinate values in the RGB color space, no conversion process is required.

  Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above. The modifications of the above embodiment are listed below.

(1) In the first embodiment, when the parameter w calculated in S60 of FIG. 5 is 0, no correction is performed on the pixel selected in S52 of FIG. Therefore, when the parameter w is 0, S62 and S64 in FIG. 5 may be skipped. In this case, the CPU 30 may execute the process of S66 in FIG. 5 with the coordinate values (PH, PS, PV) of the pixel selected in S52 of FIG. 4 as the coordinate values (MH, MS, MV). Note that if the coordinate values in the RGB target image data 50 corresponding to the coordinate values (PH, PS, PV) are known, S66 can be skipped.

(2) In each of the above embodiments, the image data stored in the memory card 18 is the RGB target image data 50 to be corrected. However, other image data may be used as the RGB target image data 50. For example, the image data generated by the scanner unit 20 may be RGB target image data 50. In each of the above embodiments, the image data generated by the scanner unit 20 is the RGB reference image data 60. However, other image data may be used as the RGB reference image data 60. For example, the image data stored in the memory card 18 may be the RGB reference image data 60.

(3) In each of the above-described embodiments, a cylindrical coordinate system color space (HSV color space) is adopted as the polar coordinate system color space. However, a spherical coordinate system color space may be adopted as the polar coordinate system color space.

(4) In each of the above embodiments, the hue range to be corrected is determined according to the correction mode selected by the user. However, in the correction printing process, only one correction mode may exist (that is, the hue range to be corrected may be fixed). In the correction printing process, the correction process may be sequentially executed for each of a plurality of correction modes (three correction modes in the above embodiment). In this case, the multi-function device 10 displays a plurality of corrected image data, allows the user to select one of the image data, and executes printing based on the image data selected by the user. May be.

(5) In each of the embodiments described above, it is not necessary to correct using w. That is, in the mathematical formula shown in FIG. In this case, only the value between the lower limit value Lv and the upper limit value Uv shown in FIG. 11 is corrected, and other values are not corrected (the same is corrected in the case of the a-axis and b-axis). In this modification, the area around the claims corresponds to the area between the lower limit value Lv and the upper limit value Uv.

  The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology illustrated in the present specification or the drawings achieves a plurality of objects at the same time, and has technical utility by achieving one of the objects.

The configuration of the multi-function device is shown. A brief description of how the image data is corrected is shown. 5 shows a flowchart of correction printing processing. The flowchart of a correction process is shown. FIG. 5 is a flowchart continued from FIG. 4. FIG. An equation for converting from RGB to HSV is shown. The HSV color space is shown. An example of the correspondence between the correction mode and the HSV range is shown. An expression for converting from HSV to vab is shown. An equation for calculating the correction amount is shown. An example of a graph in which Pv is the horizontal axis and the sum of Pv and Cv is the vertical axis is shown. The manner in which coordinate values are corrected is shown. An equation for calculating the parameter w is shown. An example of a graph in which Pv is the horizontal axis and the parameter w is the vertical axis is shown. Formulas for calculating the corrected coordinate values (Mv, Ma, Mb) are shown. In the va plane, the larger the correction amount, the closer to white. An expression for converting vab to HSV is shown. An equation for converting from HSV to RGB is shown. The SH plane is shown. The HSV color space is shown. The flowchart of the correction process of 2nd Example is shown.

Explanation of symbols

10: Multi-function device, 12: Operation unit, 14: Display unit, 16: Slot unit, 18: Memory card, 20: Scanner unit, 22: Printing unit, 24: Network interface, 28: Control unit, 30: CPU, 32: ROM, 34: Basic program, 36: Correction program, 50: RGB target image data, 52: HSV target image data, 56: Corrected RGB image data, 60: RGB reference image data, 62: HSV reference image data

Claims (9)

An image processing apparatus that corrects target image data using reference image data to generate corrected image data,
A first determination unit for determining a first polar coordinate value representing a first pixel group having a specific range of hues in the target image data, which is a value in a color space of a polar coordinate system;
A first calculation unit that calculates a first orthogonal coordinate value that is a value in a color space of a first orthogonal coordinate system by orthogonally transforming the first polar coordinate value;
A second determination unit that determines a second polar coordinate value that is a value representing the second pixel group having the hue in the specific range in the reference image data and that is a value in the color space of the polar coordinate system;
A second calculation unit that calculates a second orthogonal coordinate value that is a value in a color space of the first orthogonal coordinate system by orthogonally transforming the second polar coordinate value;
Among all the pixels in the target image data, the coordinate value of each pixel included in the area around the first orthogonal coordinate value in the color space of the first orthogonal coordinate system is set as the second orthogonal coordinate value. An image processing apparatus comprising: a correction unit that generates the corrected image data from the target image data by correcting the image data so as to approach each other. The image processing apparatus according to claim 1, wherein the surrounding area is a specific area including both the first orthogonal coordinate value and the second orthogonal coordinate value. The image processing apparatus according to claim 2, wherein the specific region is a region in an ellipsoid having the first orthogonal coordinate value and the second orthogonal coordinate value as two focal points. The size of the surrounding area set when the specific range is the first range is different from the size of the surrounding area set when the specific range is a second range different from the first range. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. The correction unit includes a first distance between the first orthogonal coordinate value and a coordinate value of the pixel to be corrected in the color space of the first orthogonal coordinate system, the second orthogonal coordinate value, and the correction target. The first orthogonal coordinate system of the pixel to be corrected is such that the smaller the sum of the second distance between the pixel and the coordinate value in the color space of the first orthogonal coordinate system of the pixel, the larger the correction amount. The image processing apparatus according to claim 1, wherein the coordinate value in the color space is corrected. The correction unit is
When the coordinate value in the color space of the first orthogonal coordinate system of the pixel to be corrected is the same as the first orthogonal coordinate value, the coordinate value in the color space of the first orthogonal coordinate system of the pixel to be corrected is , Correcting to a third orthogonal coordinate value between the first orthogonal coordinate value and the second orthogonal coordinate value;
When the coordinate value in the color space of the first orthogonal coordinate system of the pixel to be corrected is a fourth orthogonal coordinate value between the first orthogonal coordinate value and the second orthogonal coordinate value, the correction target pixel The coordinate value of the pixel in the color space of the first orthogonal coordinate system is corrected to a fifth orthogonal coordinate value between the third orthogonal coordinate value and the second orthogonal coordinate value. 6. The image processing apparatus according to any one of items 1 to 5. The first determination unit determines the first polar coordinate value by calculating an average value of coordinate values in the color space of the polar coordinate system of each pixel included in the first pixel group,
The second determination unit determines the second polar coordinate value by calculating an average value of coordinate values in a color space of the polar coordinate system of each pixel included in the second pixel group. The image processing apparatus according to claim 1. The specific image data in which the value of each pixel is represented by the coordinate value in the color space of the second orthogonal coordinate system different from the first orthogonal coordinate system, and the value of each pixel is represented by the coordinate value in the color space of the polar coordinate system. Further comprising a conversion unit for converting into the target image data.
The correction unit is
(1) The coordinate value of each pixel in the target image data is converted into a specific orthogonal coordinate value that is a coordinate value in the color space of the first orthogonal coordinate system,
(2) correcting each specific orthogonal coordinate value included in the surrounding area so as to approach the second orthogonal coordinate value;
(3) Convert each specific orthogonal coordinate value after correction into a specific polar coordinate value that is a coordinate value in the color space of the polar coordinate system,
(4) By converting each specific polar coordinate value into an orthogonal coordinate value that is a coordinate value in the color space of the second orthogonal coordinate system, the coordinate value in the color space of the second orthogonal coordinate system The image processing apparatus according to claim 1, wherein the corrected image data representing a value is generated. An image processing program for generating corrected image data by correcting target image data using reference image data,
A first determination process for determining a first polar coordinate value representing a first pixel group having a specific range of hues in the target image data, which is a value in a color space of a polar coordinate system;
A first calculation process for calculating a first orthogonal coordinate value that is a value in a color space of a first orthogonal coordinate system by orthogonally transforming the first polar coordinate value;
A second determination process for determining a second polar coordinate value representing a second pixel group having the hue in the specific range in the reference image data, which is a value in the color space of the polar coordinate system;
A second calculation process for calculating a second orthogonal coordinate value that is a value in a color space of the first orthogonal coordinate system by orthogonally transforming the second polar coordinate value;
Among all the pixels in the target image data, the coordinate value of each pixel included in the area around the first orthogonal coordinate value in the color space of the first orthogonal coordinate system is set as the second orthogonal coordinate value. An image processing program for causing a computer to execute a correction process for generating the corrected image data from the target image data by correcting the data so as to approach each other.