JP4455554B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention improves image quality of image data such as still images and moving images, and more specifically, an image processing device that executes processing for correcting gradation and saturation of image data input to the image processing device, and The present invention relates to an image processing method.

  In recent years, in various models that handle image data, it is common to perform image quality enhancement processing in order to make the image data higher quality. For example, a television or the like performs various image processing on input video data. Such image processing is performed to convert input video data into video data that suits the user's preference, or to convert it into video data that matches the characteristics of the display device that displays the video data.

  One example of such image processing is gradation correction processing. The gradation correction processing is intended to improve the image quality of the video by applying conversion characteristics based on a predetermined rule to the input video data and generating optimal output video data. . Such gradation correction processing may be referred to as gamma correction processing (γ correction processing).

  As a gradation correction process in a television, a black expansion process, which is a process for reducing the input video signal on the black side in the luminance value as compared with the input level, is known. Also known is a white expansion process, which is a process for increasing the input video signal on the white side in the luminance value as compared with the input level. In addition, tone correction processing using a generally well-known γ curve is also performed. More recently, a histogram is created for each frame of video based on the luminance values of pixels in one frame. Then, the histogram is divided stepwise according to the luminance level, and the frequency at each level is obtained. Finally, a method of using a γ curve having a complicated shape from the frequency distribution at each level is also performed. That is, complicated gradation correction processing is performed such that the low luminance side performs black expansion processing according to the characteristics of the video signal and the high luminance side similarly performs white expansion processing according to the signal characteristics.

However, there is a problem that there is a color whose saturation changes depending on the gradation correction processing. Japanese Patent Application Laid-Open No. H10-228561 also discloses that when a gradation correction process for correcting a difference in black level in color matching is performed, a change in saturation is caused depending on the process. In order to solve these problems, Patent Document 1 describes that a saturation correction unit that corrects a change in saturation caused by the γ correction process is included (particularly, refer to paragraphs [0051] and thereafter).
JP 2002-152530 A

  The method of Patent Document 1 performs saturation correction associated with the γ value used there, in addition to device range correction and brightness direction range γ correction. Here, in the saturation correction in this method, the power coefficient of the function that should be used in the range γ correction is used. That is, saturation correction is performed based on a single γ value. However, it is often difficult in practice to express the conversion characteristics used in the tone correction processing of a television with a single γ value. For example, it is difficult to represent a complex curve (conversion characteristic) created using a histogram with one γ value. Therefore, it is difficult to say that the method of Patent Document 1 that performs saturation correction according to a single γ value is effective for video processing on a television that has been generalized recently.

  In the case of a general television, a saturation enhancement process for enhancing the saturation is performed on a video signal subjected to the gradation correction process. However, if normal saturation enhancement processing is applied to image data whose saturation in the dark portion (low luminance portion) of the image has been particularly increased by performing black extension processing, the dark portion becomes too saturated and contains noise. It may become video data. In Patent Document 1, no technical reference is made in this regard.

  In JP 2003-235055 A, a saturation modulation characteristic (a characteristic in which a gain of a saturation level is determined with respect to a luminance level) is accurately determined in order to reduce color noise in a low luminance region. The intended technique is disclosed. Japanese Patent Application Laid-Open No. 2003-235055 discloses a process for reducing the correction amount of saturation correction when a process of raising a low luminance area (black portion) of an input video signal (a correction process for increasing luminance) is performed. That is, the image data is corrected to a low-saturated image data when a process opposite to the black expansion process is performed. This is effective for the problem to be solved by Japanese Patent Laid-Open No. 2003-235055. However, the problem to be solved by the present invention may be counterproductive.

  The present invention provides an image processing apparatus and an image processing method capable of preventing a change in saturation of an output signal finally obtained with a simple configuration even when black expansion processing is performed in gradation correction processing. With the goal.

In order to achieve the above object, an image processing apparatus according to the present invention converts image data composed of a plurality of pixels expressed by a color signal including a luminance signal indicating brightness and a color component signal indicating color. On the other hand, gradation correction means for performing gradation correction processing for correcting the luminance signal of each pixel constituting the image data according to a predetermined γ correction characteristic, and the color component signal of each pixel constituting the image data And a saturation correction unit that executes a saturation correction process for correcting the saturation according to a predetermined saturation correction amount, wherein the γ correction characteristic is a luminance signal of each pixel constituting the image data. A black expansion characteristic that corrects a luminance signal having a luminance lower than a predetermined luminance value so that the luminance signal has a lower luminance; and the image processing apparatus has a luminance lower than the predetermined luminance value in the black expansion characteristic. The luminance signal that is Generating means for generating a black modulation amount signal indicating a correction degree when gradation correction is performed so as to achieve low luminance in units of image data subjected to gradation correction processing by the gradation correction means; and generating by the generation means Determination means for determining a saturation correction amount for correcting the color component signal of each pixel constituting the image data subjected to the gradation correction processing by the gradation correction means using the black modulation amount signal thus obtained. The determining means increases the saturation correction amount when the black modulation amount signal indicates that the correction level for lowering the luminance signal is small, and increases the correction level for decreasing the luminance signal. The saturation correction amount is determined so as to reduce the saturation correction amount.

In order to achieve the above object, the image processing apparatus of the present invention, for image data composed of a plurality of pixels expressed by a plurality of primary color signals, brightness of each pixel constituting the image data. Histogram generation means for generating a gradation value indicating a component, generating a gradation histogram indicating a pixel number distribution using the gradation value, and a γ correction characteristic determined based on the pixel number distribution of the gradation histogram And a gradation correction unit that executes gradation correction processing of the image data, and a saturation correction processing that corrects the image data that has been gradation corrected by the gradation correction unit according to a predetermined saturation correction amount. An image processing apparatus having a saturation correction unit, wherein the γ correction characteristic has a lower gradation value that is lower than a predetermined luminance value among gradation values of each pixel constituting the image data. Correct for brightness The image processing apparatus performs gradation correction so that a gradation value that is lower in luminance than the predetermined luminance value in the black expansion characteristic is further reduced. Using the generation unit that generates a black modulation amount signal indicating a correction degree in units of image data subjected to gradation correction processing by the gradation correction unit, and using the black modulation amount signal generated by the generation unit, the gradation correction Determining means for determining a saturation correction amount of the image data subjected to the gradation correction processing by the means, wherein the generation means is a gradation having a luminance lower than the predetermined luminance value in the gradation histogram. The black modulation amount signal is generated so that the degree of correction increases as the number of pixels of the value decreases, and the determining means determines that the color saturation amount signal indicates that the degree of correction is small. Increase the degree of correction and Degree and determining a saturation correction amount so as to reduce the saturation correction amount as large.

In order to achieve the above object, the image processing method of the present invention is an image composed of a plurality of pixels represented by color signals including a luminance signal indicating brightness and a color component signal indicating color. A gradation correction step for executing a gradation correction process for correcting the luminance signal of each pixel constituting the image data according to predetermined γ correction characteristics, and a color tone of each pixel constituting the image data A saturation correction step of executing a saturation correction process for correcting the component signal according to a predetermined saturation correction amount, wherein the γ correction characteristic is a luminance signal of each pixel constituting the image data. Among them, the image processing method has a black expansion characteristic that corrects the luminance signal having a luminance lower than a predetermined luminance value to be further lower than the predetermined luminance value in the black expansion characteristic. Brightness with low brightness A generation step of generating a black modulation amount signal indicating a correction degree when performing gradation correction so that the luminance signal is further reduced in brightness in units of image data subjected to gradation correction processing in the gradation correction step; Determination step for determining a saturation correction amount for correcting the color component signal of each pixel constituting the image data subjected to the gradation correction process in the gradation correction step, using the black modulation amount signal generated in the generation step. In the determining step, when the black modulation amount signal indicates that the correction level for reducing the luminance signal to low luminance is small, the correction level for increasing the saturation correction amount and reducing the luminance signal The saturation correction amount is determined such that the saturation correction amount is decreased as the value of the saturation correction amount increases.

In order to achieve the above object, the image processing method according to the present invention is configured such that the brightness of each pixel constituting the image data with respect to image data composed of a plurality of pixels expressed by a plurality of primary color signals. A histogram generation step for generating a gradation value indicating a component, generating a gradation histogram indicating a pixel number distribution using the gradation value, and a γ correction characteristic determined based on the pixel number distribution of the gradation histogram And a gradation correction step for performing gradation correction processing of the image data, and a saturation correction processing for correcting the image data subjected to gradation correction in the gradation correction step according to a predetermined saturation correction amount. An image processing method including a saturation correction step, wherein the γ correction characteristic is such that a gradation value lower than a predetermined luminance value among gradation values of each pixel constituting image data is further reduced. Brightness and The image processing method performs gradation correction so that a gradation value that is lower than the predetermined luminance value in the black expansion characteristic is further reduced. A generation step of generating a black modulation amount signal indicating a correction degree at the time of execution in units of image data subjected to gradation correction processing in the gradation correction step, and using the black modulation amount signal generated in the generation step, A determination step for determining a saturation correction amount of the image data subjected to the gradation correction process in the gradation correction step, wherein the generation step includes a luminance lower than the predetermined luminance value in the gradation histogram. The black modulation amount signal is generated so that the correction degree increases as the number of pixels of the gradation value decreases. In the determining step, the black modulation amount signal has a smaller correction degree. The larger the chroma correction amount in the case shown, the correction degree and determines the saturation correction amount so as to reduce the saturation correction amount as large.

  As described above, according to the present invention, even when black expansion processing is performed in tone correction processing, it is possible to prevent a change in the saturation of the finally obtained output signal with a simple configuration.

Example 1
The present invention will be described in detail with reference to some embodiments which are the best mode for carrying out the present invention.

  FIG. 1 is a circuit block diagram of an image processing apparatus to which the present invention can be applied. Reference numeral 1 denotes a gradation correction processing unit that performs gradation correction processing on an input signal. A saturation correction unit 2 corrects the saturation of the input signal. Symbol b is a signal generated by the gradation correction processing unit 1 and is a black modulation amount signal (modulation degree) indicating the modulation degree of black (low luminance pixel). Details of the black modulation amount signal will be described later.

  The input signal shown in FIG. 1 is a luminance color difference signal composed of a luminance signal Y and color difference signals U and V (hereinafter referred to as YUV signal). A YUV signal is input to the image processing apparatus. A Y′U′V ′ signal obtained by performing image processing to be described later on the YUV signal is output. However, the input / output signals in the present invention are not limited to YUV signals. In the present embodiment, a description will be given using a color signal composed of luminance Y which is a brightness component signal indicating brightness and U and V which are color component signals indicating color. However, as long as the color signal includes a brightness component signal indicating brightness and a color component signal indicating color, a signal in another color space may be used.

  In addition, the image processing apparatus shown in FIG. 1 includes only the gradation correction processing unit 1 and the saturation correction unit 2, but can be applied to an image processing apparatus having other functional units. . Furthermore, the image processing apparatus of the present invention can be mounted on a digital video recorder (DVR) such as a digital television (DTV) or an HDD recorder. It can also be installed in color copiers, printers, digital still cameras, etc. that handle still images.

  Next, the gradation correction characteristics in the gradation correction processing unit 1 will be described using the graph shown in FIG.

  In FIG. 2, the horizontal axis represents the input luminance signal value (Y) of the input video signal, and the vertical axis represents the output luminance signal value obtained by applying a predetermined gradation correction characteristic to the input luminance signal value Y ( Y ′) is a graph set. That is, it is a diagram conceptually showing conversion characteristics for converting an input luminance signal value into an output luminance signal value. In this example, the input luminance level is 8 bits (0 to 255) and the output luminance level is also 8 bits, but this value may be any value.

  The solid line in the graph of FIG. 2 indicates the tone conversion characteristics and is referred to as a so-called γ correction curve or γ curve. In this example, the solid line conversion characteristic is configured such that the input luminance signal value and the output luminance signal value have the same value, but the present invention is not limited to this.

  The dotted line in the graph of FIG. 2 indicates the gradation conversion characteristics in the section where the input luminance signal value is from 0 to Y0. Here, the slope of this dotted line is k. Therefore, the gradation correction processing unit 1 obtains the output luminance signal value Y ′ using the gradation conversion characteristic defined by the dotted line of the slope value k if the input luminance signal value is from 0 to Y0. On the other hand, when the input luminance signal value exceeds Y0, the output luminance signal value Y 'is obtained by using the gradation conversion characteristic indicated by the solid line.

  That is, the gradation correction process shown in FIG. 2 is intended to obtain an output signal in which the luminance value of the pixel on the low luminance side is further reduced. That is, it corresponds to a black expansion process. In this example, the above-described slope value k is used as the black modulation amount signal b shown in FIG. The black modulation amount signal b corresponds to the degree of modulation in the present invention, and is generally called a γ value. It can be considered that the greater the slope value k, the greater the degree of black expansion, that is, the lower the brightness value. One black modulation amount signal b exists for one image. In the case of a still image, one black modulation amount signal corresponds to one still image. In the case of a moving image, one black modulation amount signal corresponds to one frame.

  In this example, the reference luminance value Y0 serving as an inflection point may be determined by an arbitrary method such as manual setting by user adjustment or automatic setting according to the characteristics of the input image. In any case, the reference luminance value Y0 is a value that defines a low luminance section. Further, the slope value k can take any value within a predetermined range.

  Next, details of the gradation correction processing unit 1 will be described with reference to FIG. An input luminance signal Y is input to the gradation correction processing unit 1. The input luminance signal Y is branched and one is input to the selector 12 and the other is input to the arithmetic unit 10.

The arithmetic unit 10 performs arithmetic processing for executing the gradation correction shown in the graph of FIG. Therefore, the slope value k, which is the gradation conversion characteristic of the black expansion process, is input to the calculation unit 10. The gradation correction processing in the calculation unit 10 is expressed as a mathematical expression as follows.
Y ′ = k (Y−Y0) + Y0 (1)
In this example, the inclination value k is input from the outside of the gradation correction processing unit 1, but the present invention is not limited to this configuration. In addition, the slope value k does not necessarily have to be input to the calculation unit 10. The calculation unit 10 may calculate the inclination value k, or may be configured to hold a predetermined inclination value k. Further, the calculation unit 10 does not need to obtain the output luminance signal Y ′ by calculation every time the luminance signal Y is input. For example, a similar result can be obtained by a luminance signal conversion process using a predetermined look-up table (LUT). In the present invention, the output luminance signal value Y ′ is not necessarily limited to only the calculation of a value using a mathematical expression. In other words, the arithmetic processing includes conversion processing using an LUT.

  When the calculation unit 10 calculates the expression (1), Y ′ may indicate a negative value. Therefore, when the output of the arithmetic unit 10 becomes a negative value, a limiter 11 that limits Y ′ to zero is arranged at the subsequent stage of the arithmetic unit 10.

  The output of the limiter 11 is input to the selector 12. The selector 12 receives an input luminance signal Y that has not been subjected to gradation correction processing and a corrected luminance signal Y ′ that has been subjected to gradation correction processing by the arithmetic unit 10. The selector 12 compares the input luminance signal Y with the value of the inflection point Y0 shown in FIG. When the input luminance signal is equal to or lower than the inflection point Y0 (Y ≦ Y0), the selector 12 outputs the corrected luminance signal Y ′ as an output luminance signal as it is. Further, when the input luminance signal Y is larger than the inflection point Y0, the selector 12 outputs the input luminance signal Y as the output luminance signal Y ′. With such a configuration, it is possible to execute black expansion processing on the input luminance signal.

  In this example, since the correction is not performed when the input luminance level is higher than the inflection point Y0, the gradation correction processing unit 1 adopts the above-described configuration. However, the present invention is not limited to the size of the inflection point, and may be different from the present configuration when correction using some gradation conversion characteristics is performed on the input luminance signal. The present invention is not limited by the internal configuration of the gradation correction processing unit. Further, the present invention is not limited to the gradation conversion characteristics shown in FIG.

  Next, the saturation correction unit 2 will be described in detail with reference to FIG. The saturation correction unit 2 receives the luminance signal Y ′ output from the previous gradation correction processing unit 1 and the color difference signals U and V. The black modulation amount signal b output from the gradation correction processing unit 1 is also input to the saturation correction unit 2. A multiplication unit 20 multiplies each of the color difference signals U and V by a gain g described later. A conversion table 21 determines and outputs a gain g based on the black modulation amount signal b. Of the Y′UV video signal, the saturation correction unit 2 outputs the output luminance signal Y ′ output from the gradation correction processing unit 1 as it is. The input color difference signals U and V are multiplied by a gain g determined based on the black modulation amount signal b, and output color difference signals U 'and V' are output, respectively. As described above, the black modulation amount signal b is not determined in units of pixels, but is determined corresponding to one screen (one still image or one frame), and thus the gain g is also one screen (one still image). Or 1 frame). Therefore, the gain g is multiplied to the color difference signals U and V corresponding to all the pixels constituting one screen.

  Determination of the gain g in the conversion table 21 will be described with reference to FIG. FIG. 5 is a graph showing an example of a table for inputting the black modulation amount signal b and outputting the gain g. The horizontal axis indicates the black modulation amount signal b, and the vertical axis indicates the gain g. Note that this graph is conceptual only. Therefore, the present invention is not limited to the numerical values and curves shown in FIG. In the case of a normal television, saturation correction is generally performed in the enhancement direction. Therefore, the conversion characteristics in FIG. 5 are also set so that the gain is always 1 or more. However, the present invention does not necessarily assume that saturation is emphasized, and the characteristics may be set so that the gain becomes a value smaller than 1 (however, zero or more).

  As described as a problem, when the gradation correction processing is performed on the luminance data using the gradation conversion characteristics as shown in FIG. 2, the saturation may be increased depending on the color. For this reason, saturation conversion is conspicuous in the finally obtained image, resulting in an unfavorable image.

  However, as a result of the inventor's image quality evaluation study, even when black expansion processing is performed in gradation correction processing, the saturation change is appropriate by appropriately performing saturation correction processing according to the degree of expansion. It was found that it can be suppressed.

  Therefore, in the present invention, when the luminance value of the pixel on the low luminance side is corrected to a lower luminance value, the gain g which is the correction amount of the color difference signal is determined according to the correction amount (modulation degree). . In FIG. 5, the table is such that the gain g becomes smaller as the black modulation amount signal b is larger, that is, the tendency to lower the luminance is stronger. In FIG. 5, the value of the gain g is greater than 1 no matter what black modulation amount signal is input, but the value of the gain g is a value smaller than 1 depending on the value of the black modulation amount signal b ( However, 0 <g) may be set. Further, although 4 is clearly specified as the slope value k, it is not defined as the upper limit value of the slope value k. The present invention adjusts the saturation correction amount according to the gradation correction degree on the low luminance side (black side), and the saturation level in the saturation correction is increased as the degree of lowering the luminance level on the low luminance side increases. The degree of correction is controlled to be reduced.

  As described above, according to the first embodiment, it is possible to prevent the change in the saturation of the finally obtained output signal with a simple configuration even when the black extension process is performed as the gradation correction process on the video signal. .

(Example 2)
A second embodiment, which is another embodiment to which the present invention can be applied, will be described with reference to the drawings. FIG. 6 is a circuit block diagram of the image processing apparatus according to the second embodiment. Similarly to the first embodiment, a YUV signal is input to this image processing apparatus as an input video signal.

  Reference numeral 30 denotes a color space conversion unit that converts a YUV signal, which is an input video signal, from a YUV color space to an RGB color space. The details of the color space conversion process in the color space conversion unit 30 are not related to the essence of the present invention, and thus the description thereof is omitted. In this embodiment, description is made using RGB signals in the RGB color space, but signals in other color spaces may be used as long as they are color signals including a plurality of primary color signals.

  The input video signal converted into RGB data by the color space conversion unit 30 is then input to the gradation correction processing unit 31. A detailed configuration of the gradation correction processing unit 31 will be described with reference to FIG.

  FIG. 7 is a detailed circuit block diagram of the gradation correction processing unit 31 in the second embodiment. The gradation correction processing unit 31 performs gradation correction processing, which will be described later, on the input RGB signal, and outputs an R′G′B ′ signal.

  First, the input RGB signals are input to the conversion table 43 and the histogram creation unit 40, respectively. The histogram creation unit 40 creates a histogram for one screen. The histogram creation unit 40 obtains an average value obtained by dividing the sum of gradation values of R, G, and B data by 3, and creates a histogram using the average value. FIG. 8 shows an example of a histogram created by the histogram creation unit 40.

  For example, if the input RGB signal is 8-bit data having gradations from 0 to 255, the total value of all signals can be 0 to 765 (= 255 × 3). Then, by dividing this total by 3, the range that the average gradation value can take is from 0 to 255. Here, the average gradation value is a signal representing the brightness component of the RGB signal.

  In this example, the histogram 40 is created by classifying the average value into four categories. That is, the input gradation level is divided into four categories (sections). Assuming that the histogram values (frequency) of these four categories are h1, h2, h3, and h4, h1 is assigned a frequency at 0 to 63 gradations, h2 is assigned a frequency at 64 to 127 gradations, and so on. In the created histogram, the frequency h1 of the category that is a gradation close to black and the frequency h4 of the category that is a gradation close to white are output to the gradation conversion curve selection unit 41, respectively. That is, the gradation level is a level indicating the degree of brightness. In categories h1 and h2, category h1 is darker than category h2.

  The gradation conversion curve selection unit 41 holds a plurality of predetermined conversion characteristics (gradation conversion curves) corresponding to the frequency of the category as an LUT. Then, based on the frequencies h1 and h4 output from the histogram creation unit 40, the held gradation conversion curve is selected.

  Here, the gradation conversion curve corresponding to the frequency h1 on the low luminance side is referred to as a black side curve. That is, the gradation value 64 is a predetermined brightness level (threshold value). The gradation conversion curve corresponding to the high luminance side frequency h4 is referred to as a white side curve. FIG. 9 is a graph of gradation conversion characteristics including the black side curve 50, the white side curve 51, and the halftone area curve selected by the gradation conversion curve selection unit 41. Here, it is assumed that there are 256 types of black side curves and white side curves held in the gradation conversion curve selection unit 41. Then, one black side curve is selected from 256 types of black side curves depending on the frequency h1. The same applies to the white curve. Note that the graph shown in FIG. 9 is conceptual only. Therefore, the present invention is not limited to the numerical values and curves shown in FIG.

  FIG. 10 is a diagram schematically showing conversion characteristics of 256 types of black side curves. A curve number as an ID is assigned to each curve (gradation conversion characteristic). The larger the curve number, the stronger the black extension. In this example, when the histogram frequency h1 is small, that is, in the case of an image with few dark gradation pixels in one screen, a curve that increases the degree of black expansion is selected. That is, the black curve with the large curve number in FIG. 10 is selected. By adopting such a gradation conversion curve selection rule, a curve that reduces the degree of black expansion is selected in the case of a scene with many dark pixels, that is, when the frequency h1 shows a large value. As a result, the gradation of the input video can be accurately expressed even in a dark scene. Conversely, in the case of a scene with few dark pixels, that is, when the frequency h1 indicates a small value, a curve that increases the degree of black expansion is selected. As a result, it is possible to suppress the black floating of the image.

  The white side curve corresponding to the high luminance side frequency h4 is also controlled in the same manner as the black side curve. However, the selection control of the white side curve is not important in the present invention, and the description is omitted.

  The black side curve and the white side curve selected as described above are output to the halftone interpolation unit 42. The halftone interpolation unit 42 creates gradation conversion characteristics of a halftone portion between the black side curve and the white side curve. In this example, the halftone interpolation unit 42 interpolates between the black side curve and the white side curve with a straight line. What is indicated by a dotted line in FIG. 9 is a gradation conversion characteristic of the halftone area created by the halftone interpolation unit.

  The gradation conversion curve output from the halftone interpolation unit 42 is a curve that defines conversion characteristics in the entire input gradation (0 to 255 in this example). That is, the tone conversion curve shown in FIG. This curve is set in the gradation conversion table 43. This setting process is performed every time the video changes. For example, in the case of a moving image, it is performed in units of frames. Thereby, the conversion characteristic according to the input image is automatically set, and the gradation correction process is executed. Thus, the tone correction processing unit 31 outputs the R′G′B ′ signal that has been subjected to tone correction.

  Note that the tone conversion curve selection unit 41 in FIG. 7 outputs the curve number of the selected black curve as the black modulation amount signal (modulation degree) b described in the first embodiment. Then, the black modulation amount signal b is input to a saturation correction unit 32 described later. In this example, the curve number is associated with the black modulation amount signal b, but the black modulation amount signal b need not be a curve number. Further, the black modulation amount signal b may be configured to take a value that becomes an arbitrary modulation amount within a predetermined range.

  Next, the configuration of the saturation correction unit 32 in the present embodiment will be described. FIG. 11 is a block diagram showing a detailed configuration of the saturation correction unit 32. The R′G′B ′ signal output from the gradation correction processing unit 31 is input to the average value detection unit 33. Further, it is input to the subtracting unit 34.

The average value detection unit 33 performs processing for obtaining the average value Av of the R′G′B ′ signal. A formula for obtaining Av is as follows.
Av = (R ′ + G ′ + B ′) / 3 (2)
As described above, this average value corresponds to a signal representing the brightness component of the R′G′B ′ signal. The average value Av obtained by the average value detection unit 33 is input to the subtraction unit 34. The subtracting section 34 subtracts the average value Av from each value of the R′G′B ′ signal. As a result, R1, G1, and B1 signals are output from the subtracting unit 34. The absolute values of the R1, G1, and B1 signals output here are obtained, and it can be said that the saturation is higher as the maximum value among the obtained three absolute values is larger. For example, if the signal is an 8-bit signal, the closer the maximum value is to 255, the higher the saturation data.

  Subsequently, the R1, G1, and B1 signals output from the subtractor 34 are input to the multiplier 35. The multiplier 35 multiplies each signal by a gain g. Here, the gain g corresponds to the adjustment amount of the saturation data. The gain g is a value obtained by converting the black modulation amount signal b output from the gradation correction processing unit 31 by the conversion table 37. That is, the gain g is determined based on the black modulation amount signal b. The relationship between the black modulation amount signal b and the gain g in the conversion table 37 is shown in FIG. The relationship between the black modulation amount signal b and the gain g in the conversion table 37 is the same as that of the conversion table 21 in the first embodiment described above. Note that this graph is conceptual only. Therefore, the present invention is not limited to the numerical values and curves shown in FIG.

  Now, the multiplier 35 multiplies the R1, G1, and B1 signals by a gain g that is a saturation adjustment amount. Subsequently, the R2, G2, and B2 signals obtained by multiplication are input to the adder 36. The adder 36 performs a process of adding the average value Av obtained by the average value detector 33 to each of the signals R2, G2, and B2. By adding the average value Av, R ″, G ″, and B ″ signals are obtained.

  That is, the saturation correction unit 32 converts the RGB data into saturation data by the subtraction unit 34, then performs the saturation correction by multiplying the gain g, and then the RGB data subjected to the saturation correction by the addition unit 36. A process of converting back to (R "G" B ") is executed.

  As described above, in this embodiment, when the saturation correction amount is adjusted in accordance with the gradation correction degree on the low luminance side (black side), the degree to which the luminance level on the low luminance side is lowered increases. Thus, control is performed to reduce the saturation correction amount in the saturation correction. By performing such control, it becomes possible to prevent a change in the saturation of the finally obtained output signal with a simple configuration.

(Example 3)
Next, a case where the present invention is applied to a configuration using a three-dimensional lookup table (hereinafter referred to as 3DLUT) for color correction will be described as a third embodiment. The circuit block diagram of the image processing apparatus in this embodiment is the same as the circuit block diagram shown in FIG. Further, since the processing in the color space conversion unit 30 and the gradation correction processing unit 31 is the same as that in the second embodiment, description thereof is omitted. Hereinafter, the saturation correction unit 32 will be described.

  FIG. 13 is a detailed diagram of the saturation correction unit in this embodiment. Reference numeral 61 denotes an interpolation calculation unit that performs an interpolation calculation described later. Reference numeral 62 denotes a RAM, and 63 denotes an adder that adds the result of interpolation calculation to each signal of R ′, G ′, and B ′. A ROM 64 stores a 3DLUT. A conversion table 65 outputs a gain g based on the black modulation amount signal b. A multiplication unit 66 multiplies the output g of the ROM 64 by the gain g.

  The ROM 64 stores a 3DLUT as shown in FIG. In this embodiment, a 3DLUT in which the RGB color space is divided into 5 × 5 × 5 as shown in FIG. 14 is assumed. That is, the ROM 64 holds data of grid points (125 points) when the RGB color space is divided into 5 × 5 × 5. The RGB data of the grid points are as shown in FIG. 16 when each color has an 8-bit gradation level.

  Each grid point stores the input RGB data at the grid point and the difference between the output RGB data corresponding to the RGB data (output RGB data−input RGB data). The difference data is ΔR, ΔG, and ΔB, respectively. That is, the 3DLUT is for performing color correction in the RGB color space, and the difference data is color correction data.

The difference data ΔR, ΔG, ΔB can be obtained as follows, for example. For example, in FIG. 12 used in the second embodiment, the difference between the output data R ″, G ″, B ″ and the input data R ′, G ′, B ′ when the gain g is 1.5 which is the maximum value. That is,
ΔR = R ″ −R ′
ΔG = G ″ −G ′
ΔB = B ″ −B ′
It becomes. At this time, the input data R ′, G ′, and B ′ are calculated assuming that they are RGB data of the above-described 3DLUT lattice points. By performing such processing, difference data of 125 lattice points is obtained. The difference data is stored in the ROM 64. The difference data ΔR, ΔG, ΔB stored in the ROM 64 is data necessary for calculating a correction amount for saturation correction of each grid point. For example, the difference data ΔR, ΔG, ΔB obtained as described above is multiplied by the gain g = 1 and added to the input data R ′, G ′, B ′. Then, the saturation correction data when the gain g = 1.5 in the above-described second embodiment, that is, R ″, G ″, B ″ data is obtained.

  The conversion table 65 receives the black modulation amount signal b generated by the gradation correction processing unit 31 and outputs a gain g. In this embodiment, the conversion table characteristic shown in FIG. 15 is used.

The multiplication unit 66 multiplies the difference data output from the ROM 64 and the gain g. There are 125 points of difference data in the ROM 64 as described above. Accordingly, the multiplication unit 66 executes a process of multiplying all 125 difference data by the gain g as follows.
△ R '= △ R × g
△ G '= △ G × g
△ B '= △ B × g
ΔR ′, ΔG ′, and ΔB ′ output from the multiplier 66 are stored in the RAM 62. The RAM 62 stores a 5 × 5 × 5 3DLUT, like the ROM 64.

  The interpolation calculation unit 61 executes interpolation calculation processing using the input R ′, G ′, B ′ data and the difference data ΔR ′, ΔG ′, ΔB ′ multiplied by the gain stored in the RAM 62. Thus, the difference data corresponding to ΔR ″, ΔG ″, ΔB ″ corresponding to the R ′, G ′, B ′ data is obtained. The interpolation calculation executed by the interpolation calculation unit 61 is, for example, linear interpolation or the like. A known interpolation method can be applied.

  That is, when the interpolation calculation unit 61 divides the RGB color space as shown in FIG. 14, the difference data ΔR ′, ΔG ′ of the eight vertices of one lattice block to which the input R ′, G ′, B ′ data belongs. , ΔB ′. Then, difference data ΔR ′, ΔG ′, and ΔB ′ are obtained by calculation according to the distances between the input R ′, G ′, and B ′ data and the eight vertices. The difference data ΔR ″, ΔG ″, ΔB ″ obtained in this way are added to the R ′, G ′, B ′ data input to the saturation correction unit, respectively. R ″, G ″, B ″ data that has undergone degree correction processing is output.

  Subsequently, a processing flow of the present embodiment will be described. As described above, the ROM 64 stores the difference data ΔR, ΔG, and ΔB when the gain in the second embodiment is g = 1.5.

For example, when the curve number of the black side curve is 0, the black modulation amount signal b is 0. At this time, gain g = 1 from FIG. Therefore, the input / output of the multiplication unit 66 has the following relationship.
△ R '= △ R
△ G '= △ G
△ B '= △ B
That is, the ΔR ′, ΔG ′, and ΔB ′ data are the saturation correction output data R ″ G ″ B ″ and the input data R′G′B ′ when the gain g = 1.5 in the second embodiment. It represents the difference.

  Difference data ΔR ′, ΔG ′, ΔB ′, which are outputs of the multiplier 66, are stored in the RAM 62. At this time, there are only difference data at 125 lattice points.

  Thereafter, the interpolation calculation unit 61 refers to the difference data ΔR ′, ΔG ′, ΔB ′ stored in the RAM 62, and the difference data ΔR ″, ΔG ″ corresponding to the input data R′G′B ′. , ΔB ″. At this time, the difference data ΔR ″, ΔG ″, ΔB ″ corresponding to the input data R′G′B ′ is also obtained when the gain g = 1.5 in the second embodiment. This corresponds to difference data.

  The difference data ΔR ″, ΔG ″, ΔB ″ are added to the input data R′G′B ′ by the adder 63. Thus, when the black modulation amount signal b = 0, the input data R′G 'B' is output as R "G" B "data subjected to saturation correction corresponding to the gain g = 1.5 in the second embodiment.

  As described above, in this embodiment, even when the saturation correction process is performed using the 3DLUT, the saturation correction process can be appropriately performed. Therefore, when adjusting the saturation correction amount according to the gradation correction degree on the low luminance side (black side), the saturation correction in the saturation correction becomes stronger as the degree of lowering the luminance level on the low luminance side increases. The amount is controlled to be reduced. By performing such control, it becomes possible to prevent a change in the saturation of the finally obtained output signal with a simple configuration.

1 is a circuit block diagram of an image processing apparatus in Embodiment 1. FIG. FIG. 3 is a diagram conceptually showing tone conversion characteristics in a tone correction processing unit 1 of Example 1. 3 is a block diagram of a gradation correction processing unit 1 in Embodiment 1. FIG. 4 is a block diagram of a saturation correction unit 2 in Embodiment 1. FIG. It is the figure which showed notionally the conversion characteristic of the conversion table 21 in the saturation correction | amendment part 2 of Example 1. FIG. 6 is a circuit block diagram of an image processing apparatus according to Embodiment 2. FIG. FIG. 10 is a block diagram of a gradation correction processing unit 31 according to the second embodiment. It is a figure which shows an example of the histogram produced in the histogram creation part 41 of Example 2. FIG. 10 is a graph showing gradation conversion characteristics set in a gradation change table 43 of Example 2. FIG. 6 is a diagram schematically illustrating black-side curve conversion characteristics in Example 2. FIG. 10 is a block diagram of a saturation correction unit 32 according to the second embodiment. It is the figure which showed notionally the conversion characteristic of the conversion table 37 in the saturation correction | amendment part 32 of Example 2. FIG. FIG. 10 is a block diagram of a saturation correction unit according to a third embodiment. It is the figure which showed the concept of the three-dimensional lookup table in the saturation correction | amendment part of Example 3. FIG. It is the figure which showed notionally the conversion characteristic of the conversion table 65 in the saturation correction | amendment part of Example 3. FIG. 10 is a table for explaining lattice point information of a three-dimensional lookup table in a saturation correction unit according to the third embodiment.

Claims (8)

For image data composed of a plurality of pixels represented by a color signal including a brightness signal indicating brightness and a color component signal indicating color, the brightness signal of each pixel constituting the image data is A tone correction unit that executes a tone correction process that corrects according to a predetermined γ correction characteristic, and a saturation correction process that corrects the color component signal of each pixel constituting the image data according to a predetermined saturation correction amount An image processing apparatus having saturation correction means for performing
The γ correction characteristic has a black expansion characteristic that corrects a luminance signal having a luminance lower than a predetermined luminance value out of luminance signals of each pixel constituting the image data so as to have a lower luminance.
The image processing apparatus includes:
In the black expansion characteristic, a black modulation amount signal indicating a degree of correction when performing a gradation correction so that a luminance signal having a luminance lower than the predetermined luminance value is further reduced by the gradation correction means. Generating means for generating image data to be subjected to gradation correction processing;
Determination to determine a saturation correction amount for correcting the color component signal of each pixel constituting the image data subjected to the gradation correction processing by the gradation correction unit using the black modulation amount signal generated by the generation unit Means,
Have
The determining means increases the saturation correction amount when the black modulation amount signal indicates that the correction level for reducing the luminance signal to a low luminance level is small, and increases the saturation level as the correction level for decreasing the luminance signal increases. An image processing apparatus that determines a saturation correction amount so as to reduce the degree correction amount. For image data composed of a plurality of pixels represented by a plurality of primary color signals, a gradation value indicating a brightness component of each pixel constituting the image data is generated, and the pixel is generated using the gradation value Histogram generation means for generating a gradation histogram showing a number distribution, and gradation correction means for executing gradation correction processing of the image data using γ correction characteristics determined based on the pixel number distribution of the gradation histogram And a saturation correction unit that executes a saturation correction process for correcting the image data subjected to gradation correction by the gradation correction unit according to a predetermined saturation correction amount,
The γ correction characteristic has a black expansion characteristic that corrects a gradation value that is lower than a predetermined luminance value among gradation values of each pixel constituting the image data so that the luminance value is further reduced.
The image processing apparatus includes:
In the black expansion characteristic, a black modulation amount signal indicating a correction degree when performing gradation correction so that a gradation value that is lower than the predetermined luminance value is further reduced to the gradation correction means Generating means for generating image data in units of gradation correction processing;
Determining means for determining a saturation correction amount of the image data subjected to gradation correction processing by the gradation correction means, using the black modulation amount signal generated by the generation means;
Have
The generating means generates the black modulation amount signal so that the degree of correction increases as the number of pixels of the gradation value having lower luminance than the predetermined luminance value decreases in the gradation histogram,
The determining means increases the saturation correction amount when the black modulation amount signal indicates that the correction degree is small, and reduces the saturation correction amount as the correction degree increases. An image processing apparatus for determining a degree of correction. The saturation correction unit calculates a gradation value of each primary color signal of the image data subjected to gradation correction by the gradation correction unit, and determines a value obtained by subtracting the gradation value from each primary color signal. the image processing apparatus according to claim 2, characterized in that multiplying the chroma correction amount. The image processing apparatus according to claim 3, wherein the saturation correction unit adds the gradation value to a value of each primary color signal multiplied by the saturation correction amount. For image data composed of a plurality of pixels represented by a color signal including a brightness signal indicating brightness and a color component signal indicating color, the brightness signal of each pixel constituting the image data is A gradation correction step for executing a gradation correction process for correcting according to a predetermined γ correction characteristic, and a saturation correction process for correcting the color component signal of each pixel constituting the image data according to a predetermined saturation correction amount An image processing method including a saturation correction step,
The γ correction characteristic has a black expansion characteristic that corrects a luminance signal having a luminance lower than a predetermined luminance value out of luminance signals of each pixel constituting the image data so as to have a lower luminance.
The image processing method includes:
In the gradation correction step, a black modulation amount signal indicating a correction degree when gradation correction is performed so that a luminance signal having a luminance lower than the predetermined luminance value in the black expansion characteristic is further reduced. A generation step for generating image data in units of gradation correction processing;
Determination to determine a saturation correction amount for correcting the color component signal of each pixel constituting the image data subjected to the gradation correction process in the gradation correction step using the black modulation amount signal generated in the generation step Steps,
Have
In the determining step, the saturation correction amount is increased when the black modulation amount signal indicates that the correction level for reducing the luminance signal to be low is low, and the saturation level is increased as the correction level for decreasing the luminance signal is increased. An image processing method, wherein the saturation correction amount is determined so as to reduce the degree correction amount. For image data composed of a plurality of pixels represented by a plurality of primary color signals, a gradation value indicating a brightness component of each pixel constituting the image data is generated, and the pixel is generated using the gradation value A histogram generation step for generating a gradation histogram indicating a number distribution, and a gradation correction step for executing a gradation correction process on the image data using a γ correction characteristic determined based on the pixel number distribution of the gradation histogram And a saturation correction step for executing a saturation correction process for correcting the image data subjected to the gradation correction in the gradation correction step according to a predetermined saturation correction amount,
The γ correction characteristic has a black expansion characteristic that corrects a gradation value that is lower than a predetermined luminance value among gradation values of each pixel constituting the image data so that the luminance value is further reduced.
The image processing method includes:
In the gradation correction step, a black modulation amount signal indicating a degree of correction when gradation correction is performed so that a gradation value that is lower in luminance than the predetermined luminance value in the black expansion characteristic has a lower luminance. A generation step for generating image data in units of gradation correction processing in
A determination step of determining a saturation correction amount of the image data subjected to the gradation correction process in the gradation correction step using the black modulation amount signal generated in the generation step;
Have
In the generation step, in the gradation histogram, the black modulation amount signal is generated so that the degree of correction increases as the number of pixels of gradation values that are lower than the predetermined luminance value decreases.
In the determining step, the saturation correction amount is increased when the black modulation amount signal indicates that the correction degree is small, and the saturation correction amount is decreased as the correction degree increases. An image processing method characterized by determining a degree correction amount. In the saturation correction step, a gradation value of each primary color signal of the image data subjected to gradation correction in the gradation correction step is calculated and determined for a value obtained by subtracting the gradation value from each primary color signal. The image processing method according to claim 6, wherein the saturation correction amount is multiplied. The image processing method according to claim 7, wherein in the saturation correction step, the gradation value is added to a value of each primary color signal multiplied by the saturation correction amount.

Images