JP2010118881A - Color management module, color management device, integrated circuit, and display device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce an adverse effect and an uncomfortable feeling by a set profile of color management to optimize a display image scene into a hue condition, in a color signal processing system for subjecting an input moving image signal to color management processing using a three-dimensional LUT (Look Up Table). <P>SOLUTION: This color management module includes: a color conversion part for converting color information of an input image signal using a multidimensional look-up table; an interpolation part for combining an image signal converted by the color conversion part with the input image signal at an optional ratio to output the combined image signal; and a look-up table rewriting part for changing data of the multidimensional look-up table in the color conversion part. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention is a color management module, a color management device, an integrated circuit, and a display device that perform color management processing such as color conversion, color correction, and color gamut conversion of an input image signal. The present invention relates to a device that performs color management processing in real time according to the characteristics of an input image signal.

  Conventionally, in color printers, color copiers, display devices, etc., color signals are converted to different color spaces, converted to different color gamuts, or hue, saturation, brightness, etc. are individually adjusted for specific colors. When various color management processes with a high degree of freedom are performed accurately and in detail, color management processes using a three-dimensional lookup table (Look Up Table, hereinafter referred to as LUT) are performed.

  In such a multidimensional LUT, it is easy to provide a large number of LUTs to optimize each according to the media, signal source, color conversion processing mode, etc. Techniques have been developed that realize color conversion processing in accordance with these various modes by using a combination of typical LUT and matrix calculation. In particular, when color management processing is performed on a moving image, having a plurality of LUTs according to the color conversion processing mode and the signal source as described above rewrites data with a large memory capacity of the three-dimensional LUT. Since a certain amount of rewrite processing time is required, there is a limitation because real-time performance is required.

  For example, Patent Document 1 shows a specific example. Inverse correction circuit that reversely corrects the γ correction applied to the input video signal, and color conversion of the video signal reversely corrected by the reverse correction means according to the color reproduction characteristics or reference white color of this input video signal by linear processing And at least one of the input video signal and the video signal reversely corrected by the reverse correction circuit. The color matrix circuit has a three-dimensional lookup table storing correction data for the color conversion amount by the linear processing. One is supplied, and the correction data output unit outputs correction data corresponding to the supplied video signal, and the correction data output from the correction data output unit is converted into a video signal after color conversion by the color matrix circuit. Adder to add, statically set according to the characteristics of the input signal and display device, inverse γ characteristics (inverse correction), matrix coefficient, A real-time nonlinear color conversion process is performed on the moving image based on the correction data and the γ characteristic.

In Patent Document 2, an input image signal is converted into variables of hue, brightness, and saturation by using predetermined formulas in a first image signal conversion unit, and is created in advance and held in a two-dimensional LUT storage unit. Using a two-dimensional LUT in which two variables of hue, brightness, and saturation are combined, the color correction means performs color correction on the hue, brightness, and saturation variables converted from the input image signal; The output signal is generated using the second predetermined formula for hue, lightness, and saturation after color correction by the image signal conversion means, and real-time processing is performed without reducing the processing speed. It is also suitable for use.
Japanese Unexamined Patent Publication No. 2006-325035 (FIG. 1) JP 2007-19970 (FIG. 1)

  However, although the color conversion device disclosed in Patent Document 1 has a three-dimensional LUT, parameters of color management processing such as inverse γ characteristics, matrix coefficients, correction data, and γ characteristics are used for input signals and display devices. It is set statically (fixed) with respect to the characteristics. Even if the setting is changed when the input signal source is switched or the image display mode is changed, the moving image Color management processing parameters cannot be changed during display.

  In addition, in the case of the color conversion device and the color conversion method disclosed in Patent Document 2, real-time processing is possible, but the lookup table to be used is two-dimensional, and the lookup table is not appropriately updated. Therefore, high-precision color conversion processing cannot be performed.

  An object of the present invention is to make it possible to control highly accurate color management (color conversion, color correction, color gamut conversion, etc.) processing for an input image signal with improved responsiveness.

  The color management module of the present invention includes a color conversion unit that converts color information of an input image signal using a multidimensional lookup table, an image signal converted by the color conversion unit, and an arbitrary input image signal. And an interpolation unit that outputs a composite image signal and a lookup table rewrite unit that changes data of a multidimensional lookup table in the color conversion unit.

  This makes it possible to perform more accurate color management processing with improved responsiveness.

  The present invention also includes a color conversion unit that converts color information of an input image signal using a multidimensional lookup table, an approximate calculation unit that performs linear processing on the input image signal by matrix operation, and the color conversion unit. The converted image signal and the image signal linearly processed by the approximation calculation unit are combined at an arbitrary ratio, and an interpolation unit that outputs a combined image signal, and data of the multidimensional lookup table in the color conversion unit A lookup table rewriting unit to be changed may be provided.

  As a result, the look-up table rewrite unit can replace the process of approximating the color conversion unit with the approximate calculation unit even while the data of the color conversion unit is being changed. Can be realized.

  In the above invention, an image feature detection unit that detects an image feature from an input image signal, and an image that controls at least one of the interpolation unit, the look-up table rewriting unit, and the approximate calculation unit based on the image feature An adaptive interlocking control unit may be provided.

  This makes it possible to perform color conversion processing adapted to the input image signal.

  In the above invention, the image adaptive interlocking control unit controls a time constant of a composite ratio of the composite image signal based on the image feature when controlling the interpolation unit. Also good.

  As a result, a method suitable for the input image signal can be used as the interpolation method, and color management processing with higher accuracy and higher responsiveness can be performed.

  In the above invention, the image feature detection unit may include a color frequency detection unit that detects an appearance frequency of a predetermined color of the input image signal as an image feature.

  As a result, it is possible to perform more accurate color conversion processing based on the color information of the input image signal.

  In the above invention, the image feature detection unit includes a vector detection unit that detects the magnitude of an image change between a plurality of images from the input image signal as an image feature using a motion vector. May be.

  Thereby, it is possible to perform color conversion processing adapted to the image change of the input image signal.

  In the above invention, the image feature detection unit may include a color gamut detection unit that detects a color gamut range of the image signal as an image feature from the input image signal. .

  As a result, the color gamut of the input image signal can be detected, and color conversion processing according to the color gamut of the input image signal and the color gamut required for the output signal can be performed.

  In the above invention, the image feature detection unit may include a face recognition unit that detects that an image includes a human face from the input image signal as an image feature. .

  This makes it possible to perform color conversion processing suitable for a person's image.

  In the above invention, the look-up table rewriting unit may rewrite only a part of the multi-dimensional look-up table of the color conversion unit.

  In this case, the time required for rewriting the lookup table can be performed in a relatively short time.

  In addition to the color management module, the above-described invention can be realized as a color management device incorporating the module, an integrated circuit in which the module is mounted, and a display device including a display unit that displays an output signal.

  According to the present invention, it is possible to control highly accurate color management (color conversion, color correction, color gamut conversion, etc.) processing with high responsiveness for an input image signal.

The color management module according to the present invention will be described below with reference to the drawings. The module is a group of elements collected for realizing a certain function, and in the present invention, means a set of components for realizing a target color management process. Actually, in an electronic circuit or an integrated circuit, it corresponds to a functional circuit portion, a circuit block, or the like that realizes the configuration of the present invention. In software, this corresponds to a program block (such as a library) that implements the function.
(First embodiment)
FIG. 1 is a block diagram showing the configuration of the color management module according to the first embodiment of the present invention. The color management module of FIG. 1 includes a color conversion unit 11, an interpolation unit 12, an image feature detection unit 13, an image adaptive interlocking control unit 14, and a lookup table rewriting unit 15, and an RGB format video as an input image signal. An input signal is input, and an RGB image output signal subjected to color management processing is output as an output signal.

  Here, the look-up table (LUT) is a correspondence table used for reference or conversion used to calculate an output value for an input value, and a calculated value for an input value in a domain. It is obtained. The color management module uses this calculated value to determine the final output value.

  Here, the input image signal is considered to include not only a natural image moving image such as a TV or video, but also a CG moving image or a still image video signal.

  The operation of the color conversion unit 11 will be described with reference to FIGS. The color conversion unit 11 converts and corrects color information of the input image signal. In the present embodiment, an example will be described in which the color conversion unit 11 performs color conversion processing using a three-dimensional LUT. However, the present invention is not limited to a three-dimensional LUT and can be similarly applied to a color conversion process using another multi-dimensional LUT. If color conversion processing using a higher-order LUT is performed, generally more accurate color conversion processing can be performed.

  The color conversion unit 11 using the three-dimensional LUT defines a predetermined three-dimensional color space as shown in FIG. 2 for the RGB input image signal, and each axis of the color space is equally divided at a predetermined interval. Each converted value (profile data) is set on each grid point of the unit cube.

  For example, when the RGB input image signal is indicated by digital data consisting of a plurality of bits, the unit cube of X cubes obtained by dividing each axis into X is selected from the upper bits of the input image signal. The operation is performed to calculate the converted value from the lower bits of the input image signal by three-dimensional interpolation (described later).

  FIG. 3 is an enlarged view of an example of the R signal of the 1 unit cube of FIG. As shown in FIG. 3, one-dimensional interpolation, two-dimensional interpolation, three-dimensional interpolation and interpolation are performed in order from the eight lattice point data R000 to R111 and the lower bits lsbR, lsbG, and lsbB of the input image signal. An operation is performed to calculate the value of Rabcd as a value after conversion by performing an insertion operation.

Specifically, when the low-order bit is 8 bits, the primary interpolation operation is
Ra = ((R001-R000) × lsbR) / 256 + R000
Rb = ((R011-R010) × lsbR) / 256 + R010
Rc = ((R101−R100) × lsbR) / 256 + R100
Rd = ((R111−R110) × lsbR) / 256 + R110
The following four calculations are performed.

Next, the secondary interpolation operation uses the result of the primary interpolation operation,
Rab = ((Rb−Ra) × lsbG) / 256 + Ra
Rcd = ((Rd−Rc) × lsbG) / 256 + Rc
Perform the calculation.

Finally, as a cubic interpolation operation,
Rabcd = ((Rcd−Rab) × lsbB) / 256 + Rab
To calculate Rabcd.

  FIG. 4 shows an example of a basic configuration of the color conversion unit 11 that performs the above-described calculation. In this example, a 14-bit RGB image signal is input as an input signal, and pre-processing such as a limiter, normalization processing, and bit division processing is performed, and the 16-bit data is divided into upper 5 bits and lower 11 bits. The The reading control circuit reads the upper 5 bits for each color of RGB, determines the reading address of the corresponding three-dimensional LUT, and reads data (profile data) corresponding to the reading address.

  On the other hand, the lower 11 bits (lsbR, lsbG, lsbB) are input to the three-dimensional interpolation process, where lsbR is a linear interpolation coefficient, lsbG is a quadratic interpolation coefficient, and lsbB is a cubic interpolation coefficient. The interpolation processing as described above is performed between the grid point data read from each LUT of RGB as the interpolation coefficient, and the conversion processing is performed.

  In the above description, the case where a three-dimensional LUT is used for the color conversion unit 11 has been described. However, the present invention can also be realized using a multi-dimensional LUT. In this case, more accurate color conversion can be performed.

  Further, in the above description, the case where the LUT is used for the color conversion unit has been described. However, when the color conversion processing calculated from the LUT is realized by a mathematical expression or the like, the calculation is performed by a mathematical expression instead of the LUT. Is also possible. In this case, the content of the color conversion process can be changed by changing the mathematical expression as compared with the rewriting of the LUT described later, so that a color management module with higher responsiveness can be realized.

  On the other hand, the advantage of using the LUT is that, when performing color conversion processing that is difficult to express by mathematical formulas, etc., if the color conversion processing can be expressed by profile data even if it is complicated color conversion processing, the contents of the color conversion processing are generally There are advantages that can be identified.

  In FIG. 2, the LUT profile data space has been described as being equally divided. However, the multidimensional LUT of the color conversion unit 11 according to the present invention may be equally divided. In this case, it is possible to provide a portion for performing color conversion processing with relatively high accuracy and a portion for performing low accuracy with one LUT. That is, it is also possible to realize as an LUT having a plurality of color conversion accuracies in one LUT. Thereby, the color conversion process can be set with higher accuracy.

  The interpolation unit 12 performs an RGB control signal between the RGB signal of the input image signal and the R′G′B ′ output signal subjected to the conversion process by the three-dimensional LUT of the color conversion unit 11 for each RGB signal. The weighting process is performed by performing the interpolation process. The interpolation unit 12 synchronizes the RGB signal of the input image signal that bypasses the color conversion unit 11 with the R′G′B ′ signal that has been subjected to the three-dimensional LUT process. A process for applying the same delay as the time required for the three-dimensional LUT process to the signal is performed as necessary.

  The interpolation unit 12 performs the interpolation based on a control signal from the image adaptive interlocking control unit 14 described later. As shown in FIG. 5, various methods may be used to superimpose the input image signal RGB and the R′G′B ′ signal from the color conversion unit 11. FIG. 5A shows a method of completely switching the interpolation ratio of both RGB and R′G′B ′ signals at time t. This is one of the methods by which the interpolation ratio (synthesis ratio) can be simply switched before and after switching. FIG. 5B shows a method of changing the interpolation ratio (interpolation coefficient, synthesis ratio) linearly and continuously over time. In this case, since the switching can be continuously changed, smooth interpolation (synthesis) can be performed. FIG. 5C shows a method of updating the interpolation ratio in a multistage manner with the passage of time. In this case, the interpolation ratio changes at regular time intervals. In this method, when the input signal is a moving image or the like, the interpolation ratio can be changed for each frame, so that relatively smooth interpolation can be performed. FIG. 5D shows a method of changing the interpolation ratio nonlinearly and continuously with time. In this case, since the interpolation ratio is controlled more characteristically, it is possible to perform a suitable interpolation with individual image signals. That is, the interpolation unit 12 can support a plurality of interpolation methods, and can switch between them to use an optimal interpolation method. Which of these interpolation methods is used is controlled by the image adaptive interlocking control unit 14.

  The image feature detection unit 13 extracts features of the image from the input image signal. Image features can be extracted from various viewpoints. An example is shown in FIG. A color frequency detection unit 61 that extracts the appearance frequency of a specific color or all colors included in the input image signal as a feature, a color gamut detection unit 62 that specifies the color gamut of the input image signal, and the input image signal When a human face is depicted, a face recognition unit 63 that detects the face and extracts the position on the screen, etc. When the input signal is a moving image signal, it is continuous or a plurality of frames (images) Examples include a vector detection unit 64 that detects the magnitude of a change between them by a motion vector, a luminance detection unit 65 that extracts only a luminance signal from an input signal, and the like.

  The color frequency detection unit 61 will be described with reference to FIG. The color frequency detection unit 61 detects and calculates the appearance frequency of each color after converting the input RGB color signal into the HSV color space of hue, saturation, and lightness (Hue, Statin, Value). . Further, by designating a specific color as a color to be noticed from the outside (a color of interest) or by registering a predetermined color to be noticed in advance, as a noticed color designated in the HSV color space, The appearance frequency may be detected and calculated. Here, since the circuit configuration is relatively small and easy to specify with high accuracy, the detection is performed in the HSV color space. However, the present invention is not limited to this, and frequency detection is performed in other color signals and color spaces. Also good.

  First, the color space conversion unit 71 converts the color space of the input RGB color signal into each signal component of HSV in the HSV color space. This color space conversion is performed by, for example, the following arithmetic expression.

Lightness signal V = MAX (R, G, B)
Hue signal H = {MID (R, G, B) −MIN (R, G, B)}
/ {MAX (R, G, B) -MIN (R, G, B)}
Saturation signal S = {MAX (R, G, B) −MIN (R, G, B)}
/ MAX (R, G, B)
Here, MAX (R, G, B) is the maximum value among the RGB signals, MIN (R, G, B) is the intermediate value among the RGB signals, and MIN (R, G, B) is This represents the minimum value among the RGB signals.

  The HSV signal obtained in this way is input to the lightness comparison unit 72, the hue comparison unit 73, and the saturation comparison unit 74, and a color space designation range of a plurality of colors of interest that is arbitrarily set from the outside by each HSV parameter value. The color determination unit 75 determines whether each color or the designated attention color is met. The frequency counting unit 76 calculates the frequency of one screen for each color or target color thus detected.

  The color gamut detector 62 detects the color gamut of the input image signal. For example, the color gamut or standard of the input signal is identified, such as whether the input image signal is a signal that conforms to BT709 or is a YCrCb extended standard and a signal that conforms to the xvYCC standard format wider than BT709. Is. By this detection, the image adaptive interlocking control unit 14 can perform a suitable color conversion process (described later).

  The face recognition unit 63 recognizes the face of a person photographed as a subject in the image and calculates a position in the image. By using face recognition as an image feature, it is possible to perform color conversion processing suitable for a human face, for example, high-precision color conversion processing for a color gamut centered on skin color.

  The vector detection unit 64 detects the magnitude of change between successive screens (frame screens, scenes) in a moving image or the like. Furthermore, a feature may be extracted by statistical processing or the like for changes on each screen. By detecting the magnitude of change between screens, it is possible to determine a scene change of a moving image.

  The luminance detection unit 65 is characterized by the luminance of the input image signal, its average value, minimum value, maximum value, and the like such as the luminance level of the entire screen, or the characteristics obtained by changes on these time axes and statistical processing. Etc. are detected. Since the luminance signal is one of the elements having a great influence on the signal constituting the image, it is useful to detect the feature of the entire screen by capturing this feature.

  As described above, the image feature detection unit 13 extracts image features by analyzing the input image signal from various viewpoints, and outputs the result to the subsequent image adaptive interlocking control unit.

  The image adaptive interlocking control unit 14 includes image features obtained from the image feature detecting unit 13 and preset display color patterns or desired display color patterns desired by the user, device characteristics of the output device, output device Suitable color conversion processing and interpolation method by inputting a plurality of control information (not shown) such as external factors affecting the conversion characteristics of the color management module such as the surrounding environment Select. In accordance with this selection, the LUT rewriting unit 15 changes the multidimensional LUT data of the color conversion unit 11. The interpolation unit 12 synthesizes two image signals input by the interpolation method controlled by the image adaptive interlocking control unit 14.

  The LUT rewrite control unit 15 rewrites any part or all of the LUT data of the three-dimensional LUT of the color conversion unit 11. In the three-dimensional LUT of the color conversion unit 11, assuming that each axis of the color space is equally divided into X as shown in FIG. 2, the lattice point for each color has (X + 1) cubed data. Since this is in each of RGB, it has 3 times as many data as the cube of (X + 1). Although depending on the module configuration, for example, it is conceivable to arrange the memory space of the RAM in 8 space units corresponding to 8 vertices of a cube for each color. The LUT write control unit 15 calculates an address search of grid point data corresponding to any specified color portion of interest from the total of 24 memory spaces, performs memory control, and controls writing to the RAM. Is.

  The overall operation of the color management module configured as described above will be described focusing on the adaptive control of the image adaptive interlocking control unit 14.

(Operation example 1 of image adaptive interlocking control)
The color gamut detection unit 62 of the image feature detection unit 13 detects the color gamut or signal standard of the input image signal. The image adaptive interlocking control unit 14 converts the color gamut of the input image signal detected by the color gamut detection unit 62 and an output device (not shown) connected to the output side of the module, for example, the display color gamut of the display device. Based on this, the LUT rewriting unit 15 is controlled to rewrite the profile data of the three-dimensional LUT of the color conversion unit 11.

  By rewriting the data of the three-dimensional LUT of the color conversion unit 11, the input image signal in the xvYCC standard format, which is an extension standard of the YCrCb signal, is made to correspond to the display color gamut of the display device connected to the output side of the color management module. This will be described using an example of a simple color conversion process. When the display device supports BT709 but does not support the xvYCC standard having a wider color gamut than BT709, the display device to which an input signal of the xvYCC standard is input remains as it is. Is not supported, only the color gamut of a part of the input signal can be displayed. Therefore, the color conversion unit 11 performs color conversion processing for converting the color gamut of the xvYCC signal into the range of the color gamut of BT709.

  As another example, when the display device is a device having a display color gamut wider than BT709, processing such as level conversion is performed in the former stage processing of FIG. 4 so as to conform to the profile possessed by the latter stage process (three-dimensional LUT process etc.). I do. In this way, even if the input image signal is a signal having one or more of xvYCC as an extension or a negative region value, color management processing can be performed with a profile corresponding to the color gamut of the display device. Even when a display device having a color gamut other than BT709 is connected, an appropriate image (still image, video) can be displayed.

(Operation example 2 of image adaptive interlocking control)
As another example, when the display device has a display color gamut wider than BT709, when an image signal in the sRGB standard format corresponding to the BT709 color gamut is input, the display color gamut of the display device is not used to the maximum extent. Although a method of strictly reproducing the BT709 color gamut is also conceivable, color conversion processing that virtually expresses a color gamut wider than sRGB by effectively using the display color gamut of the display device can be adopted. For example, although depending on the capability of the display color gamut of the display device, it can be close to a DCI (Digital Cinema Initiatives) color gamut. In this color conversion process, in addition to such color conversion process (color gamut conversion), the adjustment of the overall color tone according to the purpose of color management, the emphasis or suppression of the color of a specific hue, the hue or color of a specific memory color It is also necessary to perform various color adjustments as image quality adjustment, such as adjustment of brightness and brightness.

  If the color conversion process is simply a process that only performs color gamut conversion, the trade-off between the input image signal and the display color gamut of the display device may be overemphasized, and adverse effects due to the color conversion process may occur. . For example, when the input image signal in the sRGB standard format is displayed on a device having a display color gamut wider than BT709, in the example of color conversion processing that virtually approaches the DCI color gamut, the color gamut R of the display device is used. By using a color gamut in which G and G are wider than BT709, red can be reproduced as dark red having higher saturation than BT709. However, in this case, considering the color balance as a whole, there are phenomena such as the original skin color becoming too red.

  Such a phenomenon may occur not only in skin color but also in other colors. In particular, colors called memory colors such as cherry pink, skin color of human skin, green of plants, and blue of blue sky have a sense of incongruity when the color associated with the object is different from the actual color. Because of the characteristics, these reductions can be significant. In addition, the same phenomenon may occur when there is a tradeoff with the color gamut characteristics of the display device or the set profile is biased for the purpose of the color management processing. is there. Of particular note here is that this phenomenon is not only caused by the profile, but also by the input signal format, the color gamut characteristics of the display device, and the video signal processing settings other than the color management module. In other words, the total system of the entire video display apparatus can cause adverse effects and uncomfortable feelings.

  For this reason, in the above example, sufficient optimization cannot be achieved with only the preset color conversion process. If priority is given to the original processing purpose of the color conversion process, on the other hand, another reduction will occur. Will also occur. Therefore, in the present invention, a part of the color conversion process is appropriately changed so as to reduce the occurrence of the above-described phenomenon by optimizing the state of the display image and to make it more effective.

  Therefore, the operation in the case of the above example will be described. First, it is assumed that a three-dimensional LUT for virtually bringing an input image signal in the sRGB standard format close to the DCI color gamut is set in the color conversion unit 11. When an image signal in the sRGB (BT709) format is input to the color management module, the image feature detection unit 13 detects the feature of the input image signal. Specifically, the color gamut detection unit 62 detects that the input image signal is in the BT709 format.

  Since the three-dimensional LUT for converting the BT709 color gamut into the DCI color gamut is already set in the color conversion unit 11, the image adaptation interlocking control unit 14 controls the LUT rewriting unit 15 to change the LUT data. There is no need.

  The color frequency detection unit of the image feature detection unit 13 detects the appearance frequency of the skin color of the input image signal. The image adaptive interlocking control unit 14 determines that the data of the three-dimensional LUT of the color conversion unit 11 should be updated when the appearance frequency of the skin color is equal to or higher than a certain ratio or when it is equal to or less than the certain ratio. . Based on this determination, the LUT rewriting unit 15 is instructed to change the three-dimensional LUT data of the color conversion unit 11, and the LUT rewriting unit 15 changes the processing content (three-dimensional LUT data) of the color conversion unit 11 according to the instruction. The LUT data to be changed in this case is for the purpose of performing color gamut conversion from BT709 to DCI and suppressing redness of the skin color.

  The image adaptive interlocking control unit 14 instructs the LUT rewriting unit 15 to change the three-dimensional LUT data, and changes (weakens) the composition ratio of the image signal from the color conversion unit 11 and the input image signal to the interpolation unit 12. Take control. Since this LUT data change requires a certain processing time, it is performed to suppress the influence of the image signal output from the color conversion unit 11 during the data change.

  When the data change by the LUT rewriting unit 15 is completed, the image adaptive interlocking control unit 15 controls the interpolation unit 12 and again uses the image signal from the color conversion unit 11 as an output signal. At this time, since the color conversion unit 11 is set with a three-dimensional LUT that performs color conversion processing suitable for an input image (in this case, suitable for an image including skin color), the phenomenon that the skin color becomes reddish is improved.

  Through the operation as described above, the color management module can perform highly accurate and highly responsive color conversion control according to the image signal of the input image.

  In the above example, the detection of skin color is detected by the color frequency detection 61. However, by detecting the face of a person by the face recognition unit 63, the same LUT data rewrite control is performed by the image adaptive interlock control unit 14. May do. Furthermore, control may be performed using the results of the plurality of feature detections. Even in this case, it is possible to perform color conversion processing with higher accuracy.

(Operation example 3 of image adaptive interlocking control)
An operation example of the image adaptive interlock control unit 14 when the vector detection unit 64 of the image feature detection unit 13 detects that a change between a plurality of images (frames) is a certain value or more by motion vector calculation will be described.

  When the vector detection unit 64 detects the magnitude of the change between images (scene change) above a certain level, the image adaptive interlocking control unit 14 also acquires the color information of the image after the scene change from the color frequency detection unit 61. . The image adaptive interlocking control unit 14 determines the content of a suitable color conversion process after a scene change based on the color information acquired from the color frequency detection unit 61. For example, when the scene generally changes from a dark scene to a bright scene, color conversion processing corresponding to the change is performed.

  The image adaptive interlocking control unit 14 selects suitable three-dimensional LUT data based on the color information after the scene change, and controls data update of the color conversion unit 11 via the LUT rewriting unit 15. At this time, the degree of influence of the output signal from the color conversion unit 11 is controlled during the data change or after the data update by controlling the synthesis ratio also in the interpolation unit 12 as in the above example.

  For the control of the composition ratio, a method suitable for the content of data update of the color conversion unit 11 can be taken. For example, if the correlation between before and after the scene change is low (such as when the data of the three-dimensional LUT is changed significantly), the composition ratio is greatly changed with the time t interposed therebetween as shown in FIG. May be good. On the other hand, when the appearance frequency of a specific color (for example, skin color) decreases with time in the color frequency detection unit 61, the composition ratio is changed step by step as shown in FIG. A process for suppressing the phenomenon of reddish skin color can be applied together with a scene change.

  As another example, an example in which the three-dimensional LUT processing by the color conversion unit 11 is adaptively controlled will be described. For example, when the vector detection unit 64 of the image feature detection unit 13 detects that the video scene has changed significantly, the interpolation control for the interpolation unit 12 performs interpolation after the scene change, for example, as shown in FIG. The coefficient is controlled to temporarily weaken the influence of the processing by the color conversion unit 11 (reducing the signal ratio from the color conversion unit 11). As a result, when the color correction by the three-dimensional LUT of the color conversion unit 11 has a bias, when the scene changes to an image in which the bias greatly affects, the processing effect greatly affects the appearance. It is possible to reduce the feeling of discomfort at the time of scene change by reducing the possibility of becoming too much.

  More specifically, the degree of influence in the profile data of the currently set three-dimensional LUT conversion is determined according to the signal level and appearance frequency of each color component of RGB detected by the color frequency detection unit 61 of the image feature detection unit 13. judge. When the profile is set with a profile that is strongly biased with respect to a specific color component, the degree of influence on the processing effect can be determined using this method. In the case of a change, the degree of weakening the influence of the three-dimensional LUT processing by the color conversion unit 11 in FIG. 8 is strengthened (more greatly weakened). Should be less (not weakened too much).

  Next, an example in which the control state is changed according to control information input to the image adaptive interlocking control unit 14 will be described. As an example, illuminance information around the display device connected to the output side of the color management module of the present invention as control information is detected by a sensor or the like (not shown), and the inside of the interpolator 12 according to the illuminance. The control content of the insertion coefficient can be adaptively controlled. For example, when the illuminance is low and dark, in the case of a profile in which the color is controlled in the direction of darkening in the three-dimensional LUT processing by the color conversion unit 11, the interpolation coefficient is lowered to reduce the three-dimensional LUT effect. Weaken. Thereby, the color management processing of the input video is performed with higher accuracy, so that the visibility on the display device can be improved and the difference in color can be easily recognized.

  On the contrary, when the illuminance is high, the interpolation coefficient is increased to further enhance the three-dimensional LUT processing effect so that the effect is obtained even under bright conditions.

  In addition to the illuminance information, the control content of the interpolation coefficient can be adaptively controlled based on other light information. For example, the color of the ambient light where the display device is installed (for example, the illumination color) is detected by a sensor, etc., and control is performed taking into account that the color displayed on the display device is affected by the illumination color It is. As a result, it is possible to perform color management processing in consideration of the state that the user actually sees. In addition, color management processing can be similarly performed using the color temperature of light.

  Note that the interpolation control described above performs weighted interpolation between the processing result of the three-dimensional LUT processing of the color conversion unit 11 and the signal without processing, and thus can be used as an intensity adjustment of the processing effect. . In a TV or the like using this color management module, the user can arbitrarily adjust the interpolation coefficient, so that it can be used as a simple intensity adjustment function of the three-dimensional LUT process.

  As described above, the processing effect of the color conversion processing (three-dimensional LUT processing) is realized in real time by adopting a configuration in which the interpolation coefficient for the interpolation unit 12 of the three-dimensional LUT processing and a system bypassing the three-dimensional LUT processing is adaptively controlled. Can be seamlessly controlled by linking color adjustment and color image quality adjustment.

In the present embodiment, the image signal between the color conversion unit 11 and the interpolation unit 12 has been described as an RGB format signal, but the present invention is not limited to this. For example, if processing is performed in a hue, saturation, and lightness (HSV) color space at all stages, it is also possible to perform processing using an HSV color space signal.
(Second Embodiment)
FIG. 9 is a block diagram showing the configuration of the color management module according to the second embodiment of the present invention. The color management module shown in FIG. 9 differs from the color management module shown in FIG. 1 in that an approximate calculation unit 16 is added in the present embodiment and the internal operation of the image adaptive interlocking control unit 14 is changed. It is a point. Since the other points are the same as those of the color management module shown in FIG. 1, the same portions are denoted by the same reference numerals, and detailed description thereof is omitted.

  The approximate calculation unit 16 performs color conversion processing on the input image signal by matrix calculation. Since the color conversion process in the approximate calculation unit 16 is a linear conversion using a matrix, the interpolation unit 12 uses both the color conversion process based on the three-dimensional LUT in the color conversion unit 11 and the matrix process in the approximate calculation unit 16. An image signal can be input.

  The interlock control operation by the image adaptive interlock controller 14 will be described with reference to FIG. For example, when the illuminance data of the control information input from the outside greatly changes during image display, the 3D LUT data of the color conversion unit 11 may be rewritten to perform color conversion processing suitable for the illuminance. desirable. However, as described in the first embodiment, since the LUT data has a relatively large amount of data, rewriting of the LUT data during operation is interrupted, delayed, missing, or the output image signal. There is a possibility of causing disturbances.

  Therefore, as shown in FIG. 10, the interpolation coefficient k is controlled to 0 from the state where the influence level of the output signal of the color conversion unit 11 is 100 (the interpolation coefficient k is 100). Conversely, the image signal that has been subjected to the color conversion processing by the approximate calculation unit 16 is 100 because it is an interpolation coefficient of (100−k). In this state, the LUT rewriting unit 15 rewrites the three-dimensional LUT data. When the rewriting is completed, the weight of the output signal of the color conversion unit 11 is returned to 100 by the reverse operation. In this way, the three-dimensional LUT data can be rewritten without stopping the image display, and the color conversion process by the approximate calculation unit 16 can be performed on the image signal being rewritten. Therefore, it is possible to perform both LUT data update and color conversion processing simultaneously. At this time, the approximate calculation unit 16 sets the content of the approximation process that is the substitute process of the color conversion unit 11 in advance.

  In the example of FIG. 10, the control method for the interpolation coefficient k from 100 to 0 and the control from 0 to 100 are substantially the same, but the present invention is not limited to this. For example, even if the rate of change of the interpolation coefficient k from 0 to 100 is made slower than the rate of change from 100 to 0, or the rate of change of the interpolation coefficient is changed depending on the type of feature detected by the image feature detection unit 13. Good. Furthermore, it is not always necessary to set the interpolation ratio to one (0 or 100), and an intermediate value, for example, an interpolation coefficient can be changed in the range of 100 to 50.

  The LUT rewriting unit 15 may not only rewrite all the three-dimensional LUT data (lattice point data), but may rewrite only an arbitrary part. In this case, since it is possible to cope with data rewriting of only a part of the three-dimensional LUT data, the data rewriting time is relatively short, and the color conversion process can be changed at a higher speed. The part of the three-dimensional LUT data may be fixed, or the part may be changed as necessary. When the rewrite portion of the LUT data is changed as necessary, the optimal rewrite portion (the portion that needs to be changed) is specified by the control from the image adaptive interlock control unit 14, and the LUT rewrite unit rewrites the data. Become.

  Further, in the first or second embodiment, the processing by the color conversion unit 11 by the three-dimensional LUT has been shown, but this is not limited to the three-dimensional LUT, and can be similarly realized by other LUT conversions. It is. For example, an input signal is once converted into an HSV color space, and an arbitrary two variable, for example, a two-dimensional LUT conversion of any one of hue and saturation, hue and illuminance, and saturation and illuminance may be used. Is possible.

  In the first or second embodiment, the input signal and the output signal have been described using RGB signals, but the present invention is not limited to this. The present invention can also be realized with color signals other than RGB signals.

  In the first or second embodiment, the color management module has been described. However, the present invention can be realized by a method other than the module. For example, each processing unit shown in FIG. 1 or FIG. 8 or the like can be realized by an electronic circuit of a functional unit and realized as an integrated circuit or device as hardware.

  Similarly, each processing means shown in FIG. 1 or FIG. 8 is realized as a software program that operates on hardware using a CPU or DSP and a memory, etc., by making the function unit of software (for example, processing function). Is possible.

  Furthermore, a display device such as a liquid crystal display, a plasma display, an EL display, or an LED display is provided on the output side of the first or second embodiment, and can be realized as a display device. In this case, it is possible to perform more appropriate display by performing color management processing according to the characteristics of the connected display device.

  A color correction processing apparatus and a processing method thereof according to the present invention are a color management apparatus using a multi-dimensional LUT for performing color correction processing in a video display device for displaying a moving image such as a TV, and more particularly an image of an input image signal. The present invention is useful when applied to multidimensional color management processing that adaptively controls color correction processing in real time according to features and the like.

1 is a block diagram showing a configuration of a color management module according to a first embodiment of the present invention. Conceptual diagram explaining the color space and grid points of the three-dimensional LUT conversion means Conceptual diagram for explaining the three-dimensional interpolation operation in the three-dimensional LUT conversion means Block diagram showing the configuration of the three-dimensional LUT conversion means The figure which shows the interpolation method in an interpolation part The figure which shows the structure of an image feature detection part Block diagram showing the configuration of HSV color correction means The figure which shows an example of the interpolation control method when a video scene changes large The block diagram which shows the structure of the color management module by 2nd Embodiment of this invention. The figure explaining an example of the interpolation control method at the time of LUT rewriting

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Color conversion part 12 Interpolation part 13 Image feature detection part 14 Image adaptive interlocking control part 15 LUT rewriting part 16 Approximate calculation part 61 Color frequency detection part 62 Color gamut detection part 63 Face recognition part 64 Vector detection part 65 Luminance detection part 71 Color space conversion unit 72 Lightness comparison unit 73 Hue comparison unit 74 Saturation comparison unit 75 Color determination unit 76 Frequency coefficient unit

Claims (13)

A color conversion unit that converts color information of an input image signal using a multi-dimensional lookup table, an image signal converted by the color conversion unit, and the input image signal are combined at an arbitrary ratio to generate a combined image An interpolation unit for outputting a signal;
A color management module comprising: a look-up table rewriting unit for changing data of a multidimensional look-up table in the color conversion unit. The color management module according to claim 1.
An image feature detector for detecting an image feature from the input image signal;
A color management module comprising: an image adaptive interlocking control unit that controls at least one of the interpolation unit or the lookup table rewriting unit based on the image feature. A color conversion unit that converts color information of the input image signal using a multidimensional lookup table;
An approximate calculation unit for linearly processing the input image signal by matrix calculation;
An interpolation unit that combines the image signal converted by the color conversion unit and the image signal linearly processed by the approximate calculation unit at an arbitrary ratio, and outputs a combined image signal;
A color management module comprising: a look-up table rewriting unit for changing data of a multidimensional look-up table in the color conversion unit. The color management module according to claim 3.
An image feature detector for detecting an image feature from the input image signal;
A color management module comprising: an image adaptive interlocking control unit that controls at least one of the interpolation unit, the lookup table rewriting unit, or the approximate calculation unit based on the image feature. The color management module according to claim 2 or 4,
The color management module, wherein the image adaptive interlock control unit controls a time constant of a synthesis ratio of the synthesized image signal based on the image feature when controlling the interpolation unit. The color management module according to claim 2 or 4,
The color management module, wherein the image feature detection unit includes a color frequency detection unit that detects an appearance frequency of a predetermined color of the input image signal as an image feature. The color management module according to claim 2 or 4,
The color feature module, wherein the image feature detection unit includes a vector detection unit that detects a magnitude of an image change between a plurality of images as an image feature from a motion vector from the input image signal. The color management module according to claim 2 or 4,
The color management module, wherein the image feature detection unit includes a color gamut detection unit that detects a color gamut range of the image signal as an image feature from the input image signal. The color management module according to claim 2 or 4,
The color management module, wherein the image feature detection unit includes a face recognition unit that detects, as an image feature, that a human face is included in an image from the input image signal. The color management module according to claim 2 or 4,
The color management module, wherein the lookup table rewriting unit rewrites only a part of the multidimensional lookup table of the color conversion unit. A color management apparatus incorporating the color management module according to claim 1. An integrated circuit in which the color management module according to claim 1 is incorporated. 11. A display device comprising: the color management module according to claim 1; and a display unit that inputs an output signal from the color management module and outputs an image.

Images