JP2007324665A - Image correction apparatus and video display apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image correction apparatus and a video display apparatus capable of reducing unnaturalness of display on a vision on the basis of a difference of a color reproduction range between an imaging system and a display system while effectively utilizing the wide reproduction color range of the display system. <P>SOLUTION: The image correction apparatus does not uniformly compress the wide color reproduction rang of the display system independently of an input image but calculates an unnaturalness coefficient being an index of unnaturalness on the vision as a correction coefficient dynamically in response to the input image. The image correction apparatus determines a hue range likely to be seen unnatural as a particular hue range and calculates the unnaturalness coefficient on the basis of at least one of the saturation of image parts with hues within the particular hue range, a rate of the image parts shared in the whole image, and the luminance of the image parts. Further, the image correction apparatus concentratingly corrects the image parts likely to be seen unnatural while suppressing lower a color correction amount of the image parts unlikely to be seen unnatural. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image correction apparatus and an image correction method for performing correction such as color correction on an input image. The present invention also relates to a video display apparatus adopting the image correction apparatus and the image correction method.

  The color reproduction range represented by an output signal from an imaging system (video input system) such as an imaging device may differ from the color reproduction range that can be expressed by a display system (video output system) such as a video display device. Many. In particular, in recent years, the development of expanding the color reproduction range in video display devices has been extensive, and it has become possible to display delicate colors that were difficult to express in the past.

  However, when the color reproduction range of the imaging system and the color reproduction range of the display system are different, if an image is displayed using the output signal from the imaging system as it is in the display system, an accurate color cannot be reproduced.

  This will be supplementarily described with reference to FIG. 13 showing the xy coordinate space. In FIG. 13, reference numeral 90 denotes a chromaticity diagram boundary representing a pure color (monochromatic light). Also, a broken-line triangle attached with reference numeral 91 represents the color reproduction range of the imaging system, and a solid-line triangle attached with reference numeral 92 represents the color reproduction range of the display system. In addition, a portion with a point 94 located substantially in the center of the chromaticity diagram corresponds to white (achromatic color). The color vividness (ie, saturation) increases from white to pure color.

  As shown in FIG. 13, in a state where the color reproduction range 91 of the imaging system and the color reproduction range 92 of the display system are different, for example, a video signal representing a triangular vertex 95 of the color reproduction range 91 is given to the display system as it is. To do. Regarding the direction from the point 94 to the vertex 95, the video signal representing the vertex 95 represents the maximum saturation for the imaging system. When the video signal representing the maximum saturation is supplied to the display system as it is, the display system tries to express the maximum saturation for itself. That is, in the display system, a color corresponding to the triangular vertex 96 of the color reproduction range 92 is generally displayed. The vertex 95 corresponds to a color that is a mixture of yellow and green, and the vertex 96 corresponds to a slightly deep green color. As a result, the color is not accurately reproduced, and the user who views the output of the display system Feels uncomfortable.

  Conventionally, the display system color reproduction range 92 is uniformly compressed so as to match the color reproduction range 91 as shown in FIG. That is, for example, when a video signal corresponding to the vertex 95 is received from the imaging system, a color correction process is performed on the video signal to generate a corrected video signal representing the color corresponding to the vertex 95, and the corrected video signal is displayed. By giving to the system, the color corresponding to the vertex 95 is displayed on the display system. This color correction is uniformly performed on the video signal corresponding to each color within the color reproduction range.

  A technique for performing this type of color correction is described in, for example, Patent Document 1 below. In the method described in Patent Document 1 below, an input color in the first color space is converted into a desired color in the second color space using a certain color conversion matrix. The color conversion matrix is fixedly set at the product manufacturing stage.

JP 2005-79834 A

  However, as described above, if the color reproduction range is uniformly compressed, uniform color correction is performed even on a video signal (input image) that does not require much color correction, and the wide color of the display system. The reproduction range cannot be fully utilized.

  Therefore, an object of the present invention is to provide an image correction apparatus, an image correction method, and a video display apparatus that can reduce visual unnaturalness while effectively utilizing a color reproduction range.

  In order to achieve the above object, a first image correction apparatus according to the present invention is an image correction apparatus that corrects an input image according to a correction coefficient, based on the hue of each pixel forming the input image. Correction coefficient calculation means for calculating the correction coefficient for the image, and the correction coefficient calculation means is included in the first, second,..., Nth specific hue range among the pixels forming the input image. Are specified as the first, second,..., Nth specific pixels (n is an integer equal to or greater than 1), the saturation of each mth specific pixel, and the mth with respect to the total number of pixels of the input image. The correction coefficient is calculated based on at least one of the ratio of the number of specific pixels and the luminance of each mth specific pixel (where m is an integer between 1, 2,... N). ).

  In order to reduce visual unnaturalness due to differences in the color reproduction range, it is necessary to correct the input image. However, the degree of unnaturalness is the degree of saturation in relation to the hue of each pixel of the input image. Depending on the distribution of hues in the input image and the level of brightness in relation to the hue of each pixel in the input image. Focusing on this, referring to the hue of each pixel of the input image, a correction coefficient for the input image is dynamically calculated, and correction is performed using the correction coefficient. For example, with reference to the hue of each pixel of the input image, a correction coefficient is calculated so that correction of an image that is difficult to look unnatural is suppressed to a low level while an image that looks unnatural is emphasized. Accordingly, it is possible to reduce visual unnaturalness while making use of the color reproduction range by suppressing correction more than necessary.

  For example, a hue range that contributes relatively greatly to the occurrence of visual unnaturalness is set as the specific hue range. Then, the saturation of an image part having a hue within a specific hue range, the ratio of the image part to the entire screen, and the luminance of the image part contribute to the occurrence of visual unnaturalness, respectively. To do. Therefore, an appropriate correction coefficient is calculated based on at least one of these. As a result, appropriate correction according to the input image can be performed.

  Also, for example, when the pixels included outside the first to nth specific hue ranges are non-specific pixels, the correction coefficient calculation means has a degree of correction performed on the first to nth specific pixels. The correction coefficient is calculated so as to be larger than that for the non-specific pixel.

  This makes it possible to reduce the degree of correction of images that are difficult to look unnatural and effectively use the color reproduction range, while reducing the unnaturalness by focusing on the images that are likely to look unnatural. It becomes.

  Further, specifically, for example, the correction coefficient calculating means corrects the correction so that a greater degree of correction is applied to the entire mth specific pixel as the saturation of each mth specific pixel increases. A coefficient is calculated (where m is an integer between 1, 2,... N).

  If the saturation of an image portion having a hue within a specific hue range is relatively high, it is likely to look unnatural, and conversely if it is relatively low, it is difficult to look unnatural. Therefore, the configuration is as described above. As a result, it is possible to expect both effective use of the color reproduction range and reduction of unnaturalness.

  More specifically, for example, the correction coefficient calculation means may perform a greater degree of correction on the entire mth specific pixel as the ratio of the number of the mth specific pixel increases. The correction coefficient is calculated (where m is an integer between 1, 2,... N).

  If the ratio of the image portion having the hue within the specific hue range to the entire screen is relatively high, it is likely to look unnatural, and conversely if it is relatively low, it is difficult to look unnatural. Therefore, the configuration is as described above. As a result, it is possible to expect both effective use of the color reproduction range and reduction of unnaturalness.

  Also, specifically, for example, the correction coefficient calculation means may correct the correction coefficient so that a greater degree of correction is applied to the entire mth specific pixel as the luminance of each mth specific pixel increases. (Where m is an integer between 1, 2,... N).

  If the luminance of an image portion having a hue within a specific hue range is relatively high, it tends to look unnatural, and conversely if it is relatively low, it is difficult to look unnatural. Therefore, the configuration is as described above. As a result, it is possible to expect both effective use of the color reproduction range and reduction of unnaturalness.

  In order to achieve the above object, the second image correction apparatus according to the present invention is an image correction apparatus that corrects an input image, and the input image is displayed on the display unit without correction. The degree of correction for a pixel having a relatively large visual unnaturalness is larger than that for a pixel having a relatively small unnaturalness.

  By performing correction in consideration of the degree of unnaturalness that may occur instead of performing uniform correction, an effect of reducing visual unnaturalness while effectively utilizing the color reproduction range can be expected.

  And it is good to form the video display apparatus provided with said 1st or 2nd image correction apparatus and the display part which displays the image after correction | amendment by an image correction apparatus.

  In order to achieve the above object, an image correction method according to the present invention is an image correction method for correcting an input image according to a correction coefficient, wherein the input image is referred to by referring to the hue of each pixel forming the input image. Among the pixels forming the image, the pixels included in the first, second,... Nth specific hue range are specified as the first, second,... Nth specific pixels, respectively (n Is an integer greater than or equal to 1), the saturation of each mth specific pixel, the ratio of the number of the mth specific pixel to the total number of pixels of the input image, and the luminance of each mth specific pixel The correction coefficient is calculated (where m is an integer between 1, 2,... N).

  According to the present invention, it is possible to reduce visual unnaturalness while effectively utilizing the color reproduction range.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings. In the drawings to be referred to, the same parts are denoted by the same reference numerals. The first and second embodiments will be described later, but matters common to each embodiment will be described first.

  FIG. 1 is a configuration block diagram of a video display apparatus 1 according to an embodiment of the present invention. The video display device 1 includes a video processing unit 2, a driver 3, and a display unit 4.

  The video display device 1 is provided in, for example, a digital still camera or a digital video camera. For example, the video display device 1 is a television receiver.

  A video signal is given to the video display device 1 from the outside. This video signal is, for example, a signal representing a still image or a moving image shot by a digital still camera or a digital video camera, or a video signal representing a video (image) shot and broadcast by a broadcasting station. A video signal representing a video (image) from a broadcasting station is transmitted to the video display device 1 via an antenna or a cable.

  The video signal given to the video display device 1 is transmitted to the video processing unit 2, and the video processing unit 2 performs appropriate video processing on the video signal. The video processing unit 2 includes a color correction processing unit, and the video processing includes color correction processing by the color correction processing unit. The driver 3 receives the video signal after the color correction processing from the video processing unit 2 and causes the display unit 4 to display a video (image) represented by the video signal after the color correction processing.

  The display unit 4 is an arbitrary display panel such as a liquid crystal display panel, a plasma display panel, an organic EL (electro-luminescence) display panel, an inorganic EL display panel, a rear projection display, a front projector, or a cathode ray tube.

  When the color reproduction range of the video signal supplied to the video display device 1 is relatively narrow as shown in the color reproduction range 91 of FIG. 13 and the color reproduction range of the display unit 4 is relatively wide as shown in the color reproduction range 92 of FIG. Is assumed.

  In each embodiment described later, each image to be subjected to color correction processing by the color correction processing unit is referred to as an input image (or simply input image) to the color correction processing unit, and after the color correction processing by the color correction processing unit. Each image is called an output image (or simply an output image) of the color correction processing unit. FIG. 2 shows the arrangement of each pixel forming the input image (or output image). The arrangement of each pixel is regarded as a matrix of I rows and J columns with reference to the origin X of the image, and each pixel is represented by P [i, j]. Here, I and J are each an integer of 2 or more, i is an integer between 1 and I, and j is an integer between 1 and J.

  The input image and the output image are represented by a luminance signal Y representing luminance and color difference signals RY and BY representing colors, respectively. The color difference signal RY represents the difference between the luminance and the red component, and the color difference signal BY represents the difference between the luminance and the blue component. Regarding a certain pixel, as the color corresponding to the pixel goes from white to pure color, the values of the color difference signals RY and BY corresponding to the pixel increase.

  For the pixel P [i, j] of the input image, the value of the luminance signal, the value of the color difference signal RY, and the value of the color difference signal BY are respectively expressed as Y [i, j], RY [i, j] and BY [i , J]. Further, the value of the color difference signal RY and the value of the color difference signal BY for the pixel P [i, j] of the output image are represented by RYc [i, j] and BYc [i, j], respectively. The luminance signal of the input image and the output image is the same.

  Each embodiment will be described by taking the case where the signal representing the color information of the input image and the output image is a color difference signal as an example, but it is naturally possible to use an RGB signal or the like instead of the color difference signal. . Further, in the following description, when simply referring to a pixel, it means a pixel in the input image. In addition, unless otherwise specified, the operation and the like will be described focusing on one input image.

<< First Example >>
First, the first embodiment will be described. FIG. 3 is an internal block diagram of the color correction processing unit 10 according to the first embodiment. The color correction processing unit 10 in FIG. 3 can be applied as a color correction processing unit in the video processing unit 2 in FIG.

  The color correction processing unit 10 includes a hue / saturation calculation unit 11, an unnaturalness coefficient calculation unit 12, and a color correction unit 13.

  The hue / saturation calculation unit 11 calculates the hue and saturation of each pixel of the input image based on the color difference signal of the input image. The calculated hue and saturation of each pixel are sent to the unnaturalness coefficient calculation unit 12 and the color correction unit 13.

  As is generally known, hue is one of the three attributes of perceptual color that is characterized by being expressed in terms such as red, yellow, green, cyan, blue, and magenta. As shown in FIG. 4, the hue is represented by an angle of 0 to 360 ° with respect to the color difference signal BY on the hue circle. Note that the color wheel is a color chart in which color charts are arranged in a ring shape, as is generally known.

  The hue for the pixel P [i, j] is represented by H [i, j]. The hue / saturation calculation unit 11 calculates the hue H [i, j] for each pixel according to the following equation (1). Although derived from the definition of hue, the hue is represented by an angle formed by the color difference signals RY and BY. This angle is represented by θ. For example, when the angle θ corresponding to a certain pixel is 0 °, the pixel has a color between blue and magenta (see FIG. 4).

  The saturation indicates the vividness of the color (color purity), as is generally known. When the saturation is high, the color approaches a pure color, and when the saturation is low, the color approaches a cloudy color (gray, etc.). In the present specification, the saturation value takes a value in the range of 0 to 100%. The saturation of the achromatic color is 0%, and the saturation of the pure color is 100%.

  The saturation for the pixel P [i, j] is represented by S [i, j]. The hue / saturation calculation unit 11 calculates the saturation S [i, j] for each pixel according to the following equation (2). As derived from the definition of saturation, saturation is represented by the square root of the sum of squares of the color difference signals RY and BY.

  The color correction unit 13 in FIG. 3 performs color correction processing on the color difference signal for each pixel of the input image, and outputs the color difference signal after the color correction processing. The output image of the color correction processing unit 10 is represented by the color difference signal after the color correction processing and the luminance signal. The color correction (color correction processing) in the color correction unit 13 is performed for the hue and saturation for each pixel calculated by the hue / saturation calculation unit 11 and the arbitrary hue calculated by the unnaturalness coefficient calculation unit 12. And the unnaturalness coefficient γ [θ].

  Although the calculation method will be described in detail later, the unnaturalness coefficient γ [θ] is expressed as a function of θ representing the angle formed by the color difference signals RY and BY. The unnaturalness coefficient γ [θ] is an index representing unnaturalness of display by the display unit 4 when color correction is not performed on the input image that is the basis of the unnaturalness coefficient γ [θ]. In the subjective evaluation, it is an index regarding the degree that the user feels that the color is different from the color that is stored or recognized, and the degree of unnaturalness varies depending on the hue.

  Further, the color correction unit 13 performs color correction with a greater degree (correction amount) as the unnaturalness coefficient γ [θ] increases. Therefore, the unnatural coefficient calculation unit 12 calculates an unnatural coefficient γ [θ] having a relatively large value for an input image that requires a relatively large degree of color correction in order to eliminate the unnaturalness. I am trying to calculate. In order to realize this, the unnatural coefficient calculation unit 12 performs a calculation process in consideration of what input image needs a lot of color correction.

[FIG. 5: Unnatural coefficient calculation unit 12]
FIG. 5 shows an internal block diagram of the unnatural coefficient calculation unit 12. The unnaturalness coefficient calculation unit 12 includes a high saturation degree calculation unit 14 within a specific hue range (hereinafter abbreviated as a calculation unit 14) and an unnaturalness coefficient calculation unit 15 within a specific hue range (hereinafter referred to as a calculation unit 15). Abbreviated) and an unnatural coefficient calculation unit 16 for each hue (hereinafter abbreviated as calculation unit 16).

  The specific hue range means a hue range that requires a relatively large degree of color correction. Therefore, when color correction is not performed on a color difference signal of a pixel having a hue included in a specific hue range with respect to a certain input image, the visual unnaturalness of the image portion including the pixel becomes relatively large. . Conversely, when color correction is not performed on the color difference signal of a pixel having a hue outside the specific hue range, the visual unnaturalness of the image portion including the pixel is relatively small.

  The specific hue range is considered in consideration of the color reproduction range of the input image, the color reproduction range of the display unit 4 in FIG. 1, and the like so that visual unnaturalness is reduced as much as possible by color correction by the color correction unit 13. Are preset. In this embodiment, a hue range from red to yellow corresponding to 103.5 ° ≦ θ <167.6 ° (hereinafter referred to as RY hue range) and green corresponding to 240.6 ° ≦ θ <283.5 °. To cyan hue range (hereinafter referred to as GC hue range) and blue to magenta hue range (hereinafter referred to as BM hue range) corresponding to 347.6 ° ≦ θ <0 ° and 0 ° ≦ θ <60.6 °. ) Are selected as specific hue ranges (a total of three specific hue ranges).

  Therefore, in this case, the hue range from yellow to green (hereinafter referred to as YG hue range), the hue range from cyan to blue (hereinafter referred to as CB hue range), and the hue range from magenta to red (hereinafter referred to as MR hue range). ) Is outside the specified hue range. In this embodiment, as described above, it is assumed that the entire hue range is divided into six hue ranges.

―――― Calculation of saturation parameters β _RY , β _GC and β _BM ――――
5 calculates a saturation parameter representing the degree of high saturation within the specific hue range for each specific hue range based on the hue and saturation of each pixel calculated by the hue / saturation calculation unit 11. calculate. More specifically, the saturation parameters β _RY , β _GC, and β _BM are calculated as the saturation parameters corresponding to the RY hue range, the GC hue range, and the BM hue range, respectively.

  The degree of high saturation within the specific hue range represents the overall height (average height of the entire input image) of the saturation of the pixels included in the specific hue range. When the degree (saturation parameter) is low, it does not look unnatural even if color correction is not performed so much. In other words, if the saturation of an image portion having a hue within a specific hue range is overall low (if it is generally whitish), it does not look unnatural even if color correction is not performed so much. Conversely, if the degree (saturation parameter) is high, a relatively large degree of color correction is required.

Considering this, the saturation parameters β _RY , β _GC and β _BM are calculated. As will be apparent from the following description, as the saturation parameter increases, a greater degree of color correction is applied to the entire input image.

The “calculation method of saturation parameter β _RY ” by the calculation unit 14 will be described.
(A) First, based on each calculated hue H [i, j], it is detected whether the hue of the pixel is included in the RY hue range for each pixel forming the input image.

(B) Next, a saturation specific gravity value is calculated for a pixel detected to be included in the RY hue range. The saturation specific gravity value for the pixel P [i, j] is denoted as W ₁ [i, j]. The specific gravity value W ₁ [i, j] is determined according to a function represented by a straight line 71 in FIG. That is, for example, the specific gravity value W ₁ [i, j] is calculated according to the following equation (3). Here, 0 <k ₁ ≦ 1 holds.

(C) Then, the total _BRY of the specific gravity values W ₁ [i, j] for all the pixels detected to be included in the RY hue range is divided by the total number A _RY of the pixels included in the RY hue range. To calculate the saturation parameter β _RY . The saturation parameter β _RY is expressed by the following equation (4). As is clear from the method of deriving the saturation parameter β _RY , 0 ≦ β _RY ≦ 1 holds. In addition, when using Expression (4), the specific gravity value W ₁ [i, j] for pixels not included in the RY hue range is set to zero.

Similarly, saturation parameters β _GC and β _BM are calculated. When calculating the saturation parameters β _GC and β _BM , “RY” in the description of the above-mentioned “Calculation method of the saturation parameter β _RY ” including the above equation (4) is replaced with “GC” and “ It may be read as “BM”. However, the coefficient k ₁ for the RY hue range, the coefficient k ₁ for the GC hue range, (see the equation (3)) coefficients k ₁ for the BM hue range, which is the same or different values.

―――― Calculation of unnaturalness coefficient γ _RY , γ _GC and γ _BM for each specific hue range ――――
The calculating unit 15 in FIG. 5 calculates an unnaturalness coefficient for each specific hue range based on the saturation parameters β _RY , β _GC, and β _BM calculated by the calculating unit 14. In other words, it calculates the unnaturalness coefficient gamma _RY corresponding to RY color gamut, and unnaturalness coefficient gamma _GC corresponding to GC hue range, the unnaturalness coefficient gamma _BM corresponding to BM hue range. The unnaturalness coefficients γ _RY , γ _GC, and γ _BM are values of 0 or more and 1 or less, respectively. Specifically, for example, the unnaturalness coefficients γ _RY , γ _GC, and γ _BM for each specific hue range are calculated according to the following formulas (5a), (5b), and (5c).

  Further, the unnaturalness coefficient corresponding to outside the specific hue range is also calculated by the calculation unit 15. For example, the unnaturalness coefficient corresponding to outside the specific hue range is set to zero. FIG. 7 shows the unnaturalness coefficient corresponding to each hue range calculated by the calculation unit 15. In FIG. 7, the horizontal axis represents the hue, and the vertical axis represents the unnaturalness coefficient.

―――― Calculation of unnaturalness coefficient γ [θ] for arbitrary hue ――――
The calculation unit 16 in FIG. 5 calculates the unnaturalness coefficient γ [θ] (in other words, an arbitrary hue) based on the unnaturalness coefficient for each hue range (total of 6) calculated by the calculation unit 15. , The unnaturalness coefficient γ [θ]) for each hue is calculated.

  Specifically, for example, first, a representative hue is determined for each of the MR hue range, the RY hue range, the YG hue range, the GC hue range, the CB hue range, and the BM hue range. For example, a representative hue is determined as an intermediate hue in each hue range. That is, for example, the angle θ representing the representative hue for the RY hue range is (103.5 ° + 167.6 °) / 2 (see FIG. 4 and the like).

Further, the unnaturalness coefficient corresponding to each representative hue is set as the unnaturalness coefficient calculated by the calculation unit 15. That is, the unnaturalness coefficients corresponding to the representative hues of the MR hue range, the RY hue range, the YG hue range, the GC hue range, the CB hue range, and the BM hue range are set to 0, γ _RY , 0, γ _GC , 0, respectively. And γ _BM .

Then, as shown in FIG. 8, in the coordinate system in which the horizontal axis is the hue and the vertical axis is the unnaturalness coefficient, the adjacent hue ranges so that the unnaturalness coefficient continuously changes between adjacent hue ranges. The unnaturalness coefficient γ [θ] for an arbitrary hue is determined by linearly interpolating the unnaturalness coefficient between the representative hues. Thus, for example, the unnaturalness coefficient (γ [θ] = γ [103.5 °]) corresponding to the red hue is (0 + γ _RY ) / 2. Note that 0 ≦ γ [θ] ≦ 1 always holds for any hue.

  The calculated unnaturalness coefficient γ [θ] for an arbitrary hue is transmitted to the color correction unit 13 in FIG. At this time, for example, the entire hue range is divided into minute hue blocks (for example, every 1 ° of the angle θ), and the unnaturalness coefficient for each hue block is transmitted to the color correction unit 13. In this case, the unnaturalness coefficient γ [θ] transmitted to the color correction unit 13 is composed of discrete values corresponding to the number of hue blocks.

  In addition, a function equation or the like for calculating the value of the unnaturalness coefficient γ [θ] according to the hue may be determined by the calculation unit 16, and the function equation or the like may be transmitted to the color correction unit 13. . In this case, the color correction unit 13 calculates the actual value of the unnaturalness coefficient γ [θ] using this functional equation, and uses this value for color correction.

[FIG. 9; Color Correction Unit 13]
Next, the color correction unit 13 in FIG. 3 will be described. FIG. 9 is an internal block diagram of the color correction unit 13. The color correction unit 13 includes an unnatural coefficient calculation unit 17 for each pixel and a color correction calculation unit 18.

―――― Calculation of pixel unnaturalness coefficient δ [i, j] ――――
The unnaturalness coefficient calculation unit 17 for each pixel includes the hue and saturation of each pixel calculated by the hue / saturation calculation unit 11 in FIG. 3 and the unnaturalness calculated by the calculation unit 16 in FIG. Based on the coefficient γ [θ], an unnatural coefficient is calculated for each pixel. The unnaturalness coefficient for the pixel P [i, j] calculated here is represented by δ [i, j], and is hereinafter referred to as a pixel unnaturalness coefficient δ [i, j]. The greater the pixel unnaturalness coefficient δ [i, j], the more color correction is performed.

A method for calculating the pixel unnaturalness coefficient δ [i, j] will be described.
(A) first calculates each pixel P [i, j] the hue H [i, j] of unnaturalness coefficients corresponding to the gamma to [theta], as an interim unnaturalness coefficient γ _H [i, j]. For example, if the angle θ representing the hue of a certain pixel is 103.5 ° corresponding to the red hue, the provisional unnatural coefficient for that pixel is (0 + γ _RY ) / 2.

(B) Next, the specific gravity value W ₁ [i, j] of the saturation of each pixel is calculated. The specific gravity value W ₁ [i, j] is the same as the above specific gravity value described with reference to FIG. The specific gravity value W ₁ [i, j] is calculated, for example, according to the above equation (3).

(C) Finally, the pixel unnaturalness coefficient δ [i, j] is calculated for each pixel according to the following formula (6). The pixel unnaturalness coefficient δ [i, j] represents the degree of color correction (that is, the correction amount) for each pixel.

―――― Color correction calculation ――――
The color correction calculation unit 18 corrects the color difference signals RY [i, j] and BY [i, j] using the pixel unnaturalness coefficient δ [i, j] calculated for each pixel, and performs the corrected The color difference signals RYc [i, j] and BYc [i, j] are output as color difference signals representing the color information of the output image of the color correction processing unit. A greater degree of color correction is performed on a pixel having a large pixel unnatural coefficient δ [i, j]. Specifically, for example, the color difference signals RYc [i, j] and BYc [i, j] are calculated according to the following equations (7a) and (7b).

  The unnaturalness coefficient γ [θ] represents the degree of color correction to be performed on a pixel having a hue corresponding to the angle θ, and as understood from the above description, if there was no color correction. For example, the unnaturalness coefficient γ [θ] for a hue that tends to appear unnatural is a relatively large value (see FIGS. 7 and 8). However, even if the hue tends to look unnatural, if the saturation is low, it does not look unnatural even if the color correction is not performed so much.

  Therefore, as described above, the pixel unnaturalness coefficient δ [i, j] for each pixel is calculated according to the saturation while using the unnaturalness coefficient γ [θ] according to the hue as a basis (so to be an upper limit). To do. At this time, the pixel unnaturalness coefficient δ [i, j] is continuously changed in accordance with the saturation, and the lower the saturation, the lower the pixel unnaturalness coefficient δ [ i, j] is reduced.

  Then, the color correction calculation unit 18 corrects the saturation of the input image. At this time, a large degree of correction is applied to a pixel having a large pixel unnatural coefficient δ [i, j]. Since the pixel unnaturalness coefficient δ [i, j] for a pixel that tends to look unnatural unless color correction is performed is larger than that for a pixel that does not, the degree of color correction for the former is higher than that for the latter. Also grows.

  Further, the unnaturalness coefficient γ [θ] that is the basis of the pixel unnaturalness coefficient δ [i, j] is determined according to the saturation tendency of the entire input image. When an input image is viewed as a whole, it is not suitable for an input image in which the saturation of pixels corresponding to a specific hue range is relatively high, that is, an input image that looks unnatural unless color correction is performed. The naturalness coefficient γ [θ] is made relatively large, and the unnaturalness coefficient γ [θ] is made relatively small for input images that are not.

  In this way, instead of compressing the color reproduction range uniformly regardless of the input image as in the prior art, the degree of color correction and the color range to be color-corrected are dynamically determined according to the tendency of the input image. To do. As a result, a relatively small degree of color correction (or no color correction is performed) is applied to an input image or an image portion (color range) that is difficult to look unnatural, and it tends to look unnatural. A relatively large degree of color correction is applied to the input image or image portion (color range). As a result, it is possible to reduce the unnaturalness of display due to the difference in the color reproduction range between the imaging system and the display system while making use of the wide color reproduction range of the display unit 4 in FIG.

<< Second Example >>
Next, a second embodiment will be described. FIG. 10 is an internal block diagram of the color correction processing unit 10a according to the second embodiment. The color correction processing unit 10a in FIG. 10 can be applied as a color correction processing unit in the video processing unit 2 in FIG. The matters described in the first embodiment are also applied to the second embodiment as long as there is no contradiction. The overlapping description of the matters described in the first embodiment is omitted in the second embodiment.

  The color correction processing unit 10 a includes a hue / saturation calculation unit 11, an unnaturalness coefficient calculation unit 12 a, and a color correction unit 13. The hue / saturation calculation unit 11 and the color correction unit 13 in the color correction processing unit 10a are the same as those in the first embodiment (FIG. 3).

[FIG. 11: Unnatural coefficient calculation unit 12a]
FIG. 11 shows an internal block diagram of the unnatural coefficient calculation unit 12a. The unnaturalness coefficient calculation unit 12a includes a high saturation degree calculation unit 14 within a specific hue range (hereinafter abbreviated as a calculation unit 14 as in the first embodiment), and an unnaturalness coefficient calculation unit 15a within a specific hue range. (Hereinafter abbreviated as calculation unit 15a) and an unnaturalness coefficient calculation unit 16 for each hue (hereinafter abbreviated as calculation unit 16 as in the first embodiment). The calculation unit 14 and the calculation unit 16 are the same as those in the first embodiment (FIG. 5).

  In addition, the unnatural coefficient calculating unit 12a includes a pixel number ratio calculating unit 21 (hereinafter abbreviated as a calculating unit 21) within a specific hue range and a high luminance degree calculating unit 22 (hereinafter referred to as a specific hue range). And abbreviated as calculation unit 22).

  Also in the second embodiment, as in the first embodiment, the specific hue range is assumed to be the RY hue range, the GC hue range, and the BM hue range.

―――― Calculation of saturation parameters β _RY , β _GC and β _BM ――――
Similar to the first embodiment, the calculation unit 14 in FIG. 11 is a saturation parameter that represents the degree of high saturation within a specific hue range based on the hue and saturation of each pixel calculated by the hue / saturation calculation unit 11. β _RY , β _GC and β _BM are calculated.

―――― Calculation of ratio parameters α _RY , α _GC and α _BM ――――
The calculation unit 21 in FIG. 11 refers to the hue of each pixel calculated by the hue / saturation calculation unit 11, and the ratio of the number of pixels in the specific hue range to the total number of pixels of the input image for each specific hue range. Is calculated. Assuming that the ratios calculated for the RY hue range, the GC hue range, and the BM hue range are the pixel ratio parameters α _RY , α _GC, and α _BM , respectively, the pixel ratio parameters α _RY , α _GC, and α _BM are respectively It calculates according to following formula (8a), (8b), and (8c). When the pixel ratio parameter is low, it does not look unnatural even if color correction is not performed so much.

Here, A _RY , A _GC , and A _BM are included in the total number of pixels having a hue included in the RY hue range, the total number of pixels having a hue included in the GC hue range, and the BM hue range, respectively. This is the total number of pixels having a hue. A _TOTAL is the total number of pixels of the input image (ie, I × J).

―――― Calculation of luminance parameters φ _RY , φ _GC and φ _BM ――――
11 calculates a specific hue range based on the hue of each pixel calculated by the hue / saturation calculation unit 11 and the luminance signal Y [i, j] representing the luminance of each pixel of the input image. Each time, a luminance parameter indicating the degree of high luminance within the specific hue range is calculated. More specifically, the luminance parameters φ _RY , φ _GC, and φ _BM are calculated as the luminance parameters corresponding to the RY hue range, the GC hue range, and the BM hue range, respectively.

  The degree of high luminance within the specific hue range represents the overall height (average height of the entire input image) of the luminance of the pixels included in the specific hue range. When the degree (brightness parameter) is low, it does not look unnatural even if color correction is not performed so much.

The “calculation method of luminance parameter φ _RY ” by the calculation unit 22 will be described.
(A) First, based on each calculated hue H [i, j], it is detected whether the hue of the pixel is included in the RY hue range for each pixel forming the input image.

(B) Next, the specific gravity value of the luminance is calculated for the pixel detected to be included in the RY hue range. The specific gravity value of the luminance for the pixel P [i, j] is denoted as W ₂ [i, j]. The specific gravity value W ₂ [i, j] is determined according to a function represented by a straight line 72 in FIG. That is, for example, the specific gravity value W ₂ [i, j] is calculated according to the following equation (9). For each pixel, Y [i, j] representing the value of the luminance signal increases as the luminance increases. Further, for example, the coefficient k ₂ is set so that 0 ≦ W ₂ [i, j] ≦ 1 is always established for all luminance signal values.

(C) Then, the sum V _RY of the specific gravity values W ₂ [i, j] for all the pixels detected to be included in the RY hue range is divided by the total number A _RY of pixels included in the RY hue range. To calculate the luminance parameter φ _RY . The luminance parameter φ _RY is expressed by the following formula (10). In using the equation (10), the specific gravity value W ₂ [i, j] for the pixels not included in the RY hue range is set to zero.

It is also possible to calculate the average luminance of the pixels included in the RY hue range as the luminance parameter _φRY . This is equivalent to calculating the brightness parameter φ _RY after replacing W ₂ [i, j] in equation (10) with Y [i, j]. In this case, however, the numerical level of Y [i, j] is normalized so that the obtained luminance parameter φ _RY always takes a value between 0 and 1.

Similarly, luminance parameters φ _GC and φ _BM are calculated. When calculating the luminance parameters φ _GC and φ _BM , “RY” in the description of the “method of calculating the luminance parameter φ _RY ” including the above equation (10) is replaced with “GC” and “BM”, respectively. Should be read as However, the coefficient k ₂ for the RY hue range, the coefficient k ₂ for the GC hue range, (see above equation (9)) coefficient k ₂ for the BM hue range, which is the same or different values.

―――― Calculation of unnaturalness coefficient γ _RY , γ _GC and γ _BM for each specific hue range ――――
The calculation unit 15a in FIG. 11 includes the saturation parameters β _RY , β _GC and β _BM calculated by the calculation unit 14, the pixel ratio parameters α _RY , α _GC and α _BM calculated by the calculation unit 21, Based on the luminance parameters φ _RY , φ _GC and φ _BM calculated by the calculation unit 22, an unnatural coefficient γ _RY corresponding to the RY hue range, and an unnatural coefficient γ _GC corresponding to the GC hue range, And an unnaturalness coefficient γ _BM corresponding to the BM hue range. The unnaturalness coefficients γ _RY , γ _GC, and γ _BM are values of 0 or more and 1 or less, respectively. Specifically, for example, the unnaturalness coefficients γ _RY , γ _GC and γ _BM are calculated according to the following formulas (11a), (11b) and (11c).

  Further, the unnaturalness coefficient corresponding to outside the specific hue range is also calculated by the calculation unit 15a. For example, the unnaturalness coefficient corresponding to outside the specific hue range is set to zero. As a result, as in the first embodiment, the unnaturalness coefficient corresponding to each hue range as shown in FIG. 7 is calculated. Each calculated unnaturalness coefficient is sent to the calculation unit 16.

  Similar to the first embodiment (see FIG. 5), the calculation unit 16 in FIG. 11 uses the unnaturalness coefficient γ [for any hue based on the unnaturalness coefficient for each hue range calculated by the calculation unit 15a. θ] is calculated and given to the color correction unit 13 in FIG. The color correction unit 13 in FIG. 10 corrects the color difference signals RY and BY in the same manner as in the first embodiment (see FIG. 9).

  Consider the significance of color correction as described above. First, when the saturation parameter is low, it does not look unnatural even if the color correction is not performed so much. This is because if the saturation of an image portion having a hue within a specific hue range is totally low (if it is generally whitish), it does not look unnatural even if color correction is not performed so much.

  Second, even when the pixel ratio parameter is low, it does not look unnatural even if the color correction is not performed so much. If the ratio of pixels with a hue within a specific hue range to the entire screen is small, that is, if the number of pixels that greatly contribute to unnaturalness is small, it will not look unnatural even if color correction is not performed so much. is there.

  Third, even when the luminance parameter is low, it does not look unnatural even if the color correction is not performed so much. If the brightness of an image part having a hue within a specific hue range is low overall, that is, if pixels that contribute greatly to the unnaturalness are dark overall, the image will appear unnatural even if not so much color correction is applied. Because there is no.

  In the second embodiment, these factors are comprehensively considered, and the unnaturalness coefficient serving as an unnaturalness index is calculated using the above formula (11a) and the like, and color correction is performed in accordance with this. As a result, a relatively small degree of color correction (or no color correction is performed) is applied to an input image or an image portion (color range) that is difficult to look unnatural, and it tends to look unnatural. A relatively large degree of color correction is applied to the input image or image portion (color range). As a result, it is possible to reduce the unnaturalness of display due to the difference in the color reproduction range between the imaging system and the display system while making use of the wide color reproduction range of the display unit 4 in FIG.

  In the second embodiment, the unnaturalness coefficient is calculated in consideration of all of the saturation parameter, the pixel ratio parameter, and the luminance parameter. However, as is apparent from the above description, the saturation parameter and the pixel ratio are calculated. The parameter and the brightness parameter can individually be an index of unnaturalness. Therefore, the unnaturalness coefficient may be calculated by any one or any combination of the saturation parameter, the pixel ratio parameter, and the luminance parameter.

For example, when focusing on the RY hue range, the calculation unit 15a in FIG. 11 calculates the unnaturalness coefficient γ _RY corresponding to the RY hue range by any one of the following seven formulas (12a) to (12g). It is possible to calculate. The same applies to the unnaturalness coefficients γ _GC and γ _BM . In addition, the thing using Formula (12a) respond | corresponds to 1st Example, and the thing using Formula (12g) respond | corresponds to 2nd Example.

  The video display device described above is merely an example of an embodiment of the present invention, and the present invention includes various modifications. Below, the modification 1-the modification 4 are illustrated as a modification applicable to this invention. The contents described in each modification can be arbitrarily combined as long as there is no contradiction.

[Modification 1]
In each of the above-described embodiments, the case where the entire hue range is divided into six hue ranges is illustrated, but the number of divisions may be set to 7 or more as necessary, and conversely, may be set to 5 or less. Moreover, although the specific hue range is the RY hue range, the GC hue range, and the BM hue range, this is merely an example, and the present invention is not limited to such a division method.

[Modification 2]
Further, the color reproduction range of the video signal supplied to the video display device 1 is relatively narrow as shown in the color reproduction range 91 in FIG. 13, and the color reproduction range of the display unit 4 is comparatively as shown in the color reproduction range 92 in FIG. 13. Although a wide case is assumed, the present invention can be applied to the opposite case. In other words, the color reproduction range of the video signal supplied to the video display device 1 is relatively wide as the color reproduction range 92 in FIG. 13, and the color reproduction range of the display unit 4 is relatively as the color reproduction range 91 in FIG. Even in a narrow case, color correction can be performed by a method substantially equivalent to the method of each of the embodiments described above.

[Modification 3]
In each of the above-described embodiments, the saturation of each pixel is corrected by the color correction unit 13 in FIG. 3 or the like according to the unnaturalness coefficient γ [θ] as the correction coefficient (the above formula (7a) ) And (7b)). However, the image information to be corrected is not limited to saturation in order to reduce visual unnaturalness. That is, according to the unnaturalness coefficient γ [θ], the color correction unit 13 may correct at least one of the hue, saturation, and luminance of each pixel of the input image.

  When the hue of a certain pixel in the input image is corrected by the color correction unit 13, for example, as the unnaturalness coefficient γ [θ] corresponding to the hue of the pixel increases, the hue correction of the pixel is performed. The degree is increased (that is, the angle θ is changed more greatly by correction). When the luminance of a certain pixel in the input image is corrected by the color correction unit 13, for example, as the unnaturalness coefficient γ [θ] corresponding to the hue of the pixel increases, the luminance of the pixel is corrected. The degree is increased (that is, the luminance is changed more greatly by correction).

  When the color correction unit 13 corrects at least one of saturation and hue, the correction performed by the color correction unit 13 can be called color correction. However, when only the luminance is corrected by the color correction unit 13, the correction performed by the color correction unit 13 is not color correction but luminance correction. For this reason, when only luminance correction is performed, the color correction unit 13 is a part to be called a luminance correction unit (or correction unit). In consideration of these, the color correction processing unit denoted by reference numeral 10 or 10a in FIG. 3 or 10 can also be called an image correction processing unit.

[Modification 4]
Further, each of the above-described examples is merely an example of an embodiment for reducing visual unnaturalness, and various other techniques can be adopted as a technique contributing to the achievement of this object.

  For example, assuming a pixel that contains a lot of red components and a pixel that contains a lot of green components exist in the input image with approximately equal brightness (such as an input image in which bright flowers and green leaves are mixed) The former pixel is called a first pixel, and the latter pixel is called a second pixel. In this case, the phenomenon that the display image is glaring and looks unnatural often occurs. On the other hand, there is little unnaturalness in the case of a combination of red and yellow, or in the case where the luminance between contrasting pixels is relatively different.

  Therefore, at least referring to the hue of each pixel of the input image, the first pixel and the second pixel are specified, and the correction coefficient suitable for the input image is calculated so that the unnaturalness is reduced, and Correction processing using a correction coefficient is performed. For example, calculation of a correction coefficient for changing the hue of the first pixel or the second pixel and correction processing using the correction coefficient are performed (for example, the color of the first pixel is changed from red to vermilion). At this time, the hues of both the first pixel and the second pixel may be corrected. Further, for example, the correction coefficient suitable for the input image is calculated and the correction process using the correction coefficient is performed so that the luminance difference between the first pixel and the second pixel is increased by the correction process. At this time, in the correction process, at least one of the luminances of the first pixel and the second pixel is corrected.

  Although the case where the first pixel is a pixel containing a lot of red components and the second pixel is a pixel containing a lot of green components is exemplified, the same correction processing is performed for other specific color combinations. Is effective and contributes to the reduction of unnaturalness. For color combinations such as red and yellow that have almost no unnaturalness without correction, the degree of correction may be reduced (extremely no correction).

<< Other variations >>
Also, the color correction processing units 10 and 10a shown in FIGS. 3 and 10 can be realized by hardware, software, or a combination of hardware and software, respectively.

  When the functions of the color correction processing units 10 and 10a are realized using software, FIGS. 3 and 10 represent functional block diagrams thereof. In addition, all or part of the functions realized by the color correction processing units 10 and 10a are described as a program, and the program is executed on a computer to realize all or part of the function. Good.

  In FIG. 3 or FIG. 10, the whole (or part) of the color correction processing unit 10 or 10a functions as an image correction device (or color correction device). The whole (or part) of the unnatural coefficient calculating unit 12 or 12a functions as a correction coefficient calculating unit that calculates a correction coefficient. The correction coefficient represents, for example, the degree of correction (correction amount) such as color correction or luminance correction. In each of the embodiments described above, the correction coefficient corresponds to, for example, the unnaturalness coefficient γ [θ].

1 is a configuration block diagram of a video display device according to an embodiment of the present invention. It is a figure showing the pixel arrangement | sequence of the input image or output image of the color correction process part contained in the video display apparatus of FIG. FIG. 2 is an internal block diagram of a color correction processing unit according to the first embodiment of the present invention that can be applied to the video display apparatus of FIG. 1. It is a figure showing a hue ring. It is an internal block diagram of the unnatural coefficient calculation part of FIG. It is a figure for demonstrating the calculation method of the saturation parameter calculated by the high saturation degree calculation part in the specific hue range in FIG. It is a figure showing the unnaturalness coefficient for every hue range calculated by the unnaturalness coefficient calculation part in the specific hue range in FIG. It is a figure showing the unnaturalness coefficient with respect to arbitrary hues calculated by the unnaturalness coefficient calculation part for every hue in FIG. FIG. 4 is an internal block diagram of a color correction unit in FIG. 3. It is an internal block diagram of the color correction process part which can be applied to the video display apparatus of FIG. 1 and which concerns on 2nd Example of this invention. It is an internal block diagram of the unnatural coefficient calculation part of FIG. It is a figure for demonstrating the calculation method of the luminance parameter calculated by the high-luminance degree calculation part in the specific hue range in FIG. It is a chromaticity diagram for explaining the difference in the color reproduction range between the imaging system and the display system. It is a figure for demonstrating the conventional color correction method with respect to the difference in the color reproduction range of an imaging system and a display system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Video display apparatus 2 Video processing part 3 Driver 4 Display part 10, 10a Color correction process part 11 Hue / saturation calculation part 12, 12a Unnatural coefficient calculation part 13 Color correction part

Claims (8)

In an image correction apparatus that corrects an input image according to a correction coefficient,
Correction coefficient calculating means for calculating the correction coefficient for the input image based on the hue of each pixel forming the input image;
The correction coefficient calculation means includes
Among the pixels forming the input image, pixels included in the first, second,..., Nth specific hue range are specified as first, second,..., Nth specific pixels, respectively. (N is an integer of 1 or more),
The correction coefficient is calculated based on at least one of the saturation of each mth specific pixel, the ratio of the number of the mth specific pixel to the total number of pixels of the input image, and the luminance of each mth specific pixel. (However, m is an integer between 1, 2, ... n)
An image correction apparatus characterized by that. When the pixels included outside the first to nth specific hue ranges are non-specific pixels,
The correction coefficient calculation unit calculates the correction coefficient so that a degree of correction performed on the first to nth specific pixels is larger than that on the non-specific pixels. The image correction apparatus according to 1. The correction coefficient calculation means calculates the correction coefficient such that a greater degree of correction is applied to the entire mth specific pixel as the saturation of each mth specific pixel increases (however, m is an integer between 1, 2, ... n)
The image correction apparatus according to claim 1, wherein the image correction apparatus is an image correction apparatus. The correction coefficient calculation means calculates the correction coefficient so that a greater degree of correction is applied to the entire mth specific pixel as the ratio with respect to the number of the mth specific pixel increases. (However, m is an integer between 1, 2, ... n)
The image correction apparatus according to claim 1, wherein the image correction apparatus is an image correction apparatus. The correction coefficient calculation means calculates the correction coefficient so that a greater degree of correction is applied to the entire mth specific pixel as the luminance of each mth specific pixel increases (however, m Is an integer between 1, 2, ... n)
The image correction apparatus according to claim 1, wherein the image correction apparatus is an image correction apparatus. In an image correction apparatus that corrects an input image,
When the input image is displayed on the display unit without correction, the degree of correction for a pixel having a relatively large visual unnaturalness is set to be larger than that for a pixel having a relatively small unnaturalness. A featured image correction apparatus. An image correction apparatus according to any one of claims 1 to 6,
A video display device comprising: a display unit that displays an image corrected by the image correction device. In an image correction method for correcting an input image according to a correction coefficient,
With reference to the hue of each pixel forming the input image, the pixels included in the first, second,... Nth specific hue ranges among the pixels forming the input image 1st, 2nd, ... specified as the nth specific pixel (n is an integer of 1 or more),
The correction coefficient is calculated based on at least one of the saturation of each mth specific pixel, the ratio of the number of the mth specific pixel to the total number of pixels of the input image, and the luminance of each mth specific pixel. (However, m is an integer between 1, 2, ... n)
An image correction method characterized by that.