JP2006332732A - Apparatus for suppressing color shift noise - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of applying chroma suppression only to an actually required image part whose color shift noise is conspicuous. <P>SOLUTION: An apparatus for suppressing the color shift noise includes: a chroma arithmetic section 12 for obtaining a chroma C of an image on the basis of color difference signals C<SB>b</SB>, C<SB>r</SB>separated by a signal separation section 11; and a chroma suppression section 13 for varying gains (chroma suppression strength) of the color difference signals C<SB>b</SB>, C<SB>r</SB>on the basis of both a luminance Y prescribed by a luminance signal separated by the signal separation section 11 and the chroma C obtained by the chroma arithmetic section 12, and the apparatus varies the gains of the color difference signals C<SB>b</SB>, C<SB>r</SB>in response to the chroma C in addition to the luminance Y so that the apparatus places a limit on chroma suppression to be applied to images whose luminance Y and chroma C are small. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a color unevenness noise suppression device, and more particularly to a device for suppressing color unevenness caused by mottled colors different from the original color by suppressing the gain of a chroma signal (color difference signal). Is preferable.

  An image captured with high sensitivity by using a single-plate (Bayer array) imaging device includes a lot of noise. In order to suppress this noise, several methods have been established to suppress the unevenness of the luminance component (brightness / darkness component) most sensitive to human eyes. In general, a method using a Laplacian filter is known. This is a method of suppressing noise by extracting unevenness of a luminance component with a Laplacian filter and subtracting the unevenness component from the original image.

  On the other hand, noise is also included in color components other than luminance components. For example, there is uneven color noise caused by mottled colors different from the original color. In order to suppress this color unevenness noise, a method of applying the above-described method using a Laplacian filter to color components can be considered, but the effect of noise suppression is very small. This is thought to be because the color unevenness noise is a very low frequency component unlike the luminance noise.

As another method for suppressing color unevenness noise, it is conceivable to apply so-called chroma suppression processing. Chroma suppression is a method of reducing the color difference component by suppressing the gain of the color difference signal among the luminance signal and color difference signal (chroma signal) converted from the RGB signal. When this chroma suppression is performed, the color is simply lightened, but the color unevenness is less noticeable. Conventionally, there has been proposed a chroma suppression circuit that reduces coloring by suppressing the gains of the chroma signals C _b and C _r according to the gradation of the luminance signal (see, for example, Patent Document 1).
JP 2004-7679 A

  In general, uneven color noise is conspicuous in a dark image having a small luminance component value. Therefore, when applying chroma suppression, it is possible to effectively suppress color unevenness noise in a dark portion by reducing the gain in an image portion having a small luminance component. However, when the uneven color noise is observed strictly, the uneven color noise may not be so noticeable even in a dark image.

  However, in the conventional chroma suppression method, the gain is simply made variable according to the luminance level (brightness / darkness), and therefore, if the image is dark, the chroma suppression is applied even if the color unevenness noise is not conspicuous. When chroma suppression is applied, the original color becomes light, and there is a problem that an undesirable image is formed. Since chroma suppression makes the original color lighter, it is desirable that chroma suppression is not performed as much as possible in places where uneven color noise is not noticeable.

  The present invention has been made to solve such problems, and improves image quality by enabling chroma suppression to be applied only to the part of the image that is really necessary where color unevenness noise is conspicuous. The purpose is to be able to make it.

  In order to solve the above-described problem, the color unevenness noise suppression apparatus of the present invention extracts a luminance signal and a color difference signal from an input image signal, and obtains the color difference signal in addition to the luminance specified by the luminance signal. The gain of the color difference signal is varied according to the saturation.

  According to the present invention configured as described above, chroma suppression is applied in consideration of not only the luminance of the image signal but also the saturation. The color unevenness noise is most noticeable in an image with low luminance and low saturation (an image that is dark and close to an achromatic color). On the other hand, even if the luminance is small, the color unevenness noise is not so noticeable in an image with high saturation. According to the present invention, it is possible to adjust to reduce the gain of the color difference signal by narrowing down to an image with low luminance and low saturation, and only for an image portion that is really necessary where color unevenness noise is conspicuous. Chroma suppression can be applied. Thereby, the quality of an image can be improved.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram illustrating a configuration example of a main part of the uneven color noise suppression apparatus according to the present embodiment. FIG. 2 is a block diagram illustrating an overall schematic configuration of a digital camera to which the uneven color noise suppression apparatus according to the present embodiment is applied.

  As shown in FIG. 2, the digital camera 20 includes an optical system 21 including a shutter, a lens, an iris, an image sensor 22 such as a charge coupled device (CCD) or a complementary metal oxide semiconductor (CMOS), and an A / D. A converter 23, a digital signal processing unit 24, an AE / AF processing unit 25 that performs automatic exposure adjustment and automatic focus adjustment processing, and a controller 26 that controls the entire digital camera 10 are configured.

  In the digital camera 20 configured as described above, light from the subject is imaged by the image sensor 22 via the optical system 21. The imaging element 22 photoelectrically converts the incident light that has been imaged and generates an analog imaging signal corresponding to the incident light. The imaging signal generated here is supplied to the A / D converter 23 and converted into digital image data.

  Image data obtained by the A / D converter 23 is supplied to the digital signal processing unit 24. The signal processing unit 24 performs color correction processing, color interpolation processing, contour enhancement processing, noise suppression processing, and the like on the input image data. This noise suppression processing includes luminance component noise suppression processing and color difference component chroma suppression processing performed using a Laplacian filter. That is, the digital signal processing unit 24 includes the uneven color noise suppression device 10 according to the present embodiment.

  As shown in FIG. 1, the uneven color noise suppression apparatus 10 of the present embodiment includes a signal separation unit 11, a saturation calculation unit 12, and a chroma suppression unit 13 as its functional configuration. FIG. 1 illustrates only the functional configuration for performing chroma suppression among the functional configurations of the digital signal processing unit 24, and includes color correction processing, color interpolation processing, contour enhancement processing, and luminance components. Since the functional configuration for performing the noise suppression process is not the subject of the present invention, the illustration thereof is omitted.

The signal separation unit 11 performs processing for separating an input image signal into a luminance signal and a color difference signal. That, R · G · B input from the A / D converter 23 (red, green, blue) three primary color signals of the luminance signal Y _{· C} b _· C _r, the color difference signal of blue, red color difference signal Convert to. A known method can be applied to this conversion process, and a detailed description thereof is omitted here.

The saturation calculation unit 12 performs processing for obtaining the saturation C of the image from the color difference signals C _b and C _r separated by the signal separation unit 11. The saturation C is a value determined according to the magnitudes of the color difference components C _b and C _r , and can be obtained, for example, according to (Equation 1) below.
C = | C _b | + | C _r | (Formula 1)

Note that this calculation method is merely an example. For example, the saturation C may be obtained by the following (Expression 2) or (Expression 3).
C = (C _b ² + C _r ² ) ^1/2 (Expression 2)
C = C _b ² + C _r ² (Formula 3)
The above (Equation 1) is easier to calculate than (Equation 2) and (Equation 3), whereas the accuracy of saturation C is better with (Equation 2) and (Equation 3).

The chroma suppression unit 13 is separated by the signal separation unit 11 based on both the luminance Y specified by the luminance signal separated by the signal separation unit 11 and the saturation C obtained by the saturation calculation unit 12. The gain (chroma suppression intensity) of the color difference signals C _b and C _r is varied, and the color difference signals C _b ′ and C _r ′ after gain adjustment are output. Specifically, the color difference signals C _b ′ and C _r ′ whose gains are adjusted according to the following (Expression 4) and (Expression 5) are obtained.
C _b ′ = C _b × α (Formula 4)
C _r ′ = C _r × α (Formula 5)
(However, 0 ≦ α ≦ 1)

In the present embodiment, the value of the gain α in the above (Expression 4) and (Expression 5) is determined based on both the value of the luminance component Y and the value of the saturation component C. That is,
(Condition 1) The value of the luminance component Y is small, and
(Condition 2) In an image region that satisfies the condition that the value of the saturation component C is small, the gain α is decreased, and in other image regions, the gain α is increased (set to “1” or a value close thereto). ).

Since the image area satisfying the condition 2 is an area where the absolute values of the color difference components C _b and C _r are originally small, even if chroma suppression is performed when the color unevenness noise is small (even if the value of the gain α is reduced). ) The original color will never be extremely light. On the other hand, in a situation where a large amount of color unevenness noise is included, it can be said that the effect of chroma suppression is great because the color unevenness itself can be reduced by applying chroma suppression.

  For example, by setting table information in advance, the value of the gain α to be determined when the value of the luminance component Y and the value of the saturation component C are determined. Is possible. The table information in this case is, for example, a two-dimensional matrix table having the luminance Y value on the vertical axis and the saturation C value on the horizontal axis, and the value of the gain α is set in each column of the matrix.

When the gain α is determined using such two-dimensional table information, the chroma suppression unit 13 uses the luminance Y value specified by the luminance signal separated by the signal separation unit 11 and the saturation calculation unit 12. A predetermined weight is added to each of the obtained values of saturation C, and the gains of the color difference signals C _b and C _r are obtained by referring to the two-dimensional table information based on both the weighted luminance and saturation. α may be determined.

If the gain α is determined using the two-dimensional table information, a large memory area is consumed for storing the table information. Therefore, one parameter integrated from the luminance Y specified by the luminance signal separated by the signal separation unit 11 and the saturation C obtained by the saturation calculation unit 12 is obtained, and the color difference is determined based on the obtained parameter. The gain α of the signals C _b and C _r may be determined.

As an example of the parameters to be obtained here, it is possible to obtain the luminance V according to the following (Equation 6) and determine the gain α of the color difference signals C _b and C _r based on the luminance V. If the parameters are integrated into one in this way, the table information can be one-dimensional and the memory area for storing the table information can be reduced.
V = (1-k) C + kY (Formula 6)
(However, 0 <k <1)

  FIG. 3 is a diagram for explaining the concept of brightness saturation V. In FIG. FIG. 4 is a diagram showing an example of the stored contents of the one-dimensional table information using the luminance V (the relationship between the luminance V and the gain α). As shown in FIG. 3, the luminance saturation V is represented by a straight line A in a plane space, for example, with the luminance Y on the horizontal axis and the saturation C on the vertical axis. FIG. 3 shows an example when k = 0.25. By setting k ≠ 0.5, it is possible to weight each of luminance Y and saturation C.

  In the planar space of FIG. 3, the color unevenness noise is particularly noticeable in the image of the region B indicated by the oblique lines, that is, the image satisfying both the above-described conditions 1 and 2. In the case of the image of the region B where the luminance Y and the saturation C are small, the luminance saturation V is also reduced from (Equation 6). Therefore, as shown in the example of FIG. 4, the value of the gain α is decreased when the luminance V is small, and the value of the gain α is increased when the luminance V is large. In the example of FIG. 4, α = 1 is set when the saturation V is greater than 0.25, and the gain α is linearly decreased according to the value when the saturation V is 0.25 or less. Accordingly, α = 0.5 is set when V = 0.

  As described above in detail, according to the present embodiment, the luminance component Y and the saturation component C are obtained from the image signal, and the chroma suppression gain α is determined according to both the luminance Y and the saturation C. Therefore, chroma suppression can be applied to an image with low luminance and low saturation, that is, an image that is really necessary where color unevenness noise is conspicuous, and the quality of the image can be improved. In particular, the color unevenness noise can be effectively suppressed by applying the chroma suppression processing of the present embodiment to an image shot with an increased ISO sensitivity.

  In the above embodiment, an example in which the color unevenness noise suppression apparatus 10 of the present embodiment is implemented in the digital camera 20 has been described. However, an electronic device that can be implemented is not limited to the digital camera 20. The present invention can be widely applied to electronic devices that handle image signals.

  In the above-described embodiment, a linear function expression such as (Expression 6) is given as an example of an arithmetic expression for obtaining the saturation V, but the present invention is not limited to this example. For example, the luminance saturation V may be obtained according to an nth-order function equation (n ≧ 2) using the luminance Y and the saturation C as parameters.

  In the above embodiment, the relationship as shown in FIG. 4 is shown as an example of a one-dimensional table that specifies the relationship between the brightness V and the gain α, but this is merely an example. That is, as long as the value of the gain α is set to be small in the region where the luminance V is small, the value and shape are not limited to those shown in FIG.

  Further, in the above-described embodiment, an example in which the chroma suppression gain α is determined based on two parameters of luminance Y and saturation C or one parameter of brightness saturation V obtained from the luminance Y and saturation C is described. However, the present invention is not limited to this. For example, the gain α may be determined in consideration of the ISO sensitivity in addition to the parameter and the ISO sensitivity. For example, when the ISO sensitivity is high, the value of the gain α is small, and when the ISO sensitivity is low, the value of the gain α is large.

In the above embodiment, as shown in (Expression 4) and (Expression 5), an example in which the gain is adjusted by multiplying both the color difference signals C _b and C _r by a common gain α will be described. However, the present invention is not limited to this. For example, the gain alpha _b multiplies the color difference signal C _b of the blue, as such multiplies the gain alpha _r for red color difference signal C _r, the color difference signals C _b, independent with respect to C _r The gain may be adjusted by multiplying the gains α _b and α _r .

  In addition, each of the above-described embodiments is merely an example of the embodiment for carrying out the present invention, and the technical scope of the present invention should not be construed in a limited manner. In other words, the present invention can be implemented in various forms without departing from the spirit or main features thereof.

  The uneven color noise suppression device of the present invention can be widely used in image processing devices that need to process image signals. For example, it is useful for a digital camera, a video camera, a television receiver, a DVD (Digital Versatile Disk) player, and the like. Further, the present invention can be applied not only to digital signal processing but also to an image processing apparatus that performs analog signal processing.

It is a block diagram which shows the principal part structural example of the color nonuniformity noise suppression apparatus by this embodiment. 1 is a block diagram illustrating an overall schematic configuration of a digital camera to which an uneven color noise suppression device according to an embodiment is applied. It is a figure for demonstrating the concept of the brightness saturation by this embodiment. FIG. 6 is a diagram illustrating an image of a relationship between brightness and chroma suppression gain according to the present embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Color unevenness noise suppression apparatus 11 Signal separation part 12 Saturation calculating part 13 Chroma suppression part 24 Digital signal processing part

Claims (4)

Signal separating means for separating an input image signal into a luminance signal and a color difference signal;
Saturation calculation means for obtaining saturation from the color difference signal separated by the signal separation means;
Chroma suppression means for varying the gain of the color difference signal based on both the luminance specified by the luminance signal separated by the signal separation means and the saturation obtained by the saturation calculation means. Color unevenness noise suppression device. 2. The color unevenness according to claim 1, wherein the chroma suppression unit weights each of the luminance and the saturation, and varies the gain of the color difference signal based on both of the weighted luminance and saturation. Noise suppression device. The chroma suppression unit obtains one parameter obtained by integrating the luminance specified by the luminance signal separated by the signal separation unit and the saturation obtained by the saturation calculation unit, and based on the obtained parameter. The uneven color noise suppression apparatus according to claim 1, wherein the gain of the color difference signal is varied. When the luminance specified by the luminance signal separated by the signal separation unit is Y and the saturation obtained by the saturation calculation unit is C,
The chroma suppression unit obtains the luminance V by calculating V = (1-k) C + kY (0 <k <1), and varies the gain of the color difference signal based on the luminance V. The uneven color noise suppression device according to claim 3.

Images