JP2005354585A - Device, method and program of image processing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce decoloring also about a high chroma area existing in an low chroma area and to effectively suppress generation of a false color. <P>SOLUTION: An edge component Yedge of a luminance signal is extracted by a luminance edge extraction part 13. An edge component Cedge of a chroma signal is extracted by a chroma edge extraction part 18. A chroma signal level Clpf is extracted by a chroma level detection part 19. A suppression amount of chroma suppression is determined from the edge component Yedge of the luminance signal, the edge component Cedge of the chroma signal and the chroma signal level Clpf and color difference signals Cr and Cb are suppressed by a chroma suppression part 15. No decoloring is generated even when a high choroma thin line exists in the low chroma area by adjusting the suppression amount using the edge component of the chroma signal. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program suitable for use in, for example, a digital camera.

  When a single-plate image sensor is used as an image sensor of a digital camera or digital video camera, since each pixel has only single color information, the missing color information is interpolated from the neighboring same color pixels. However, when such missing color information is obtained by interpolation, a color different from the original color (referred to as a false color) may occur, particularly at the edge portion of the luminance signal.

  For example, as shown in FIG. 12B, a red (R) color filter, a green (G) color filter, and a blue (B) color filter are disposed on the front surface of the image sensor. And Then, it is assumed that a signal to be imaged as originally shown in FIG. The arrangement described in each cell indicates that the color image signals R, G, and B included in one pixel are (R signal, G signal, B signal).

  In this case, as shown in FIG. 12C, only the R signal is obtained from the pixels P1, P3, P7, and P9 of the red filter, and the G signal is obtained from the pixels P2, P4, P6, and P8 of the green filter. Only the B signal is obtained from the pixel P5 of the blue filter.

  Now, assuming that the pixel of interest is a pixel P5, since the pixel of interest P5 is a blue filter pixel, only the information of the B signal is obtained, and the information of the R signal and the G signal is missing. Therefore, the R signal is obtained by interpolation from information on the pixels of the surrounding red filter, and the G signal is obtained by interpolation from information on the pixels of the surrounding green filter.

That is, if the R signals of the pixels P1, P3, P7, and P9 are R1, R3, R7, and R9, and the G signals of the pixel pixels P2, P4, P6, and P8 are G2, G4, G6, and G8, R signal is
(R1 + R3 + R7 + R9) / 4
And the G signal of the pixel of interest P5 is
(G2 + G4 + G6 + G8) / 4
As required.

  FIG. 12D shows the result of obtaining the R signal and G signal of the target pixel P5 in this way by interpolation.

  Each color signal obtained by interpolation should be close to the original color signal, but as can be understood by comparing FIG. 12 (A) and FIG. 12 (D), it can be obtained by interpolation with the original color signal. The color signal may be different. That is, the R signal of the pixel P5 is originally “0” as shown in FIG. 12A, but the R signal obtained by interpolation is “50” as shown in FIG. In this way, if the color signal obtained by interpolation is different from the original color signal, a false color is generated.

  In particular, in a portion where high-frequency components such as a contour portion, a line, and a fine pattern of a subject are present, a signal different from the original color signal is likely to be generated, causing a false color.

  Conventionally, false colors are reduced by performing a process called chroma suppression or color suppression for the above problems. Chroma suppression is a process that makes false colors inconspicuous by suppressing the values of the color difference signals Cr and Cb according to the level of the high-frequency component (luminance edge) extracted from the luminance signal Y. In general, a process for suppressing the saturation of the entire image is performed on the portion where the luminance edge is extracted.

  In the conventional chroma suppress described above, since all the saturation of the portion where the high-frequency component (luminance edge) of the subject is strong is suppressed, the generation of false color at the edge portion can be prevented, but on the other hand, it is called “color loss” A problem occurs. “Color loss” is a phenomenon in which, when a luminance edge is strongly included in an originally saturated portion, the saturation is suppressed by chroma suppression and the color is lost.

In order to solve this “color loss”, Patent Literature 1 detects a high-frequency component (luminance edge) of the luminance signal Y, adds the absolute values of the color difference signals Cr and Cb, and obtains a saturation signal. A low-saturation area is extracted from the degree signal using a spatial filter, and the saturation in the pixels in the low-saturation area is suppressed according to the intensity of the luminance edge, effectively suppressing false colors. What to do has been proposed.
JP 2003-224859 A

  However, in the method disclosed in Patent Document 1, when the high saturation thin line 121 exists in the low saturation area 120 as shown in FIG. 13, the saturation of the thin line may be lost. In other words, in the method of Patent Document 1, “color loss” can be prevented in the high saturation area, but when a thin line with high saturation exists in the low saturation area, the thin line portion of the luminance signal is in the low saturation area. Detected as an edge, chroma suppression processing is performed, and “color loss” occurs.

  In view of the above-described conventional problems, the present invention reduces an image loss even in a high-saturation area existing in a low-saturation area, and can effectively suppress the occurrence of false colors. It is an object to provide a processing method and an image processing program.

  In order to solve the above-described problem, the present invention according to claim 1 is an image processing apparatus that performs predetermined image processing on a color image signal input from an image sensor, wherein the color image signal is a luminance signal. Color signal conversion means for converting to a color difference signal, luminance edge extraction means for extracting an edge component of the luminance signal, saturation signal calculation means for obtaining a saturation signal from the color difference signal, and extraction of the edge component of the saturation signal Saturation edge extraction means, saturation level determination means for determining the level of the saturation signal, an edge component of the luminance signal, an edge component of the saturation signal, and a suppression amount set based on the saturation signal level, It is characterized by comprising chroma suppression means for suppressing the color difference signal.

  The present invention according to claim 2 is characterized in that the chroma suppression means changes the suppression amount of the color difference signal in accordance with the intensity of the edge component of the luminance signal.

  The present invention according to claim 3 is characterized in that the chroma suppression means changes the suppression amount of the color difference signal in accordance with the intensity of the edge component of the saturation signal.

  The present invention according to claim 4 is characterized in that the chroma suppress means suppresses the color difference signal in accordance with the intensity of the saturation signal.

  The present invention according to claim 5 is characterized in that the luminance edge extraction means includes a Laplacian filter.

  The present invention according to claim 6 is characterized in that the luminance edge extraction means includes a Sobel filter.

  The present invention according to claim 7 is characterized in that the saturation edge extraction means includes a Laplacian filter.

  The present invention according to claim 8 is characterized in that the saturation edge extraction means includes a Sobel filter.

  The present invention according to claim 9 is characterized in that the saturation level determination means includes a low-pass filter.

  The present invention according to claim 10 is characterized in that the saturation level determination means includes a median filter.

  According to an eleventh aspect of the present invention, there is provided an image processing method for performing predetermined image processing on a color image signal input from an image pickup device, wherein the color image signal is converted into a luminance signal and a color difference signal. A luminance edge extracting step for extracting an edge component of the luminance signal, a suppression parameter calculating step for calculating a suppression parameter for suppressing the color difference signal from the intensity of the edge component of the luminance signal, and a saturation signal from the color difference signal Saturation signal calculation step for obtaining the saturation signal, a saturation edge extraction step for extracting the edge component of the saturation signal, and an adjustment parameter calculation for calculating an adjustment parameter for adjusting the suppression amount from the intensity of the edge component of the saturation signal A saturation level extraction step for extracting the saturation level of the saturation signal, and a saturation level determination step for determining the intensity of the saturation level. Based on the suppression parameter and the adjustment parameter, a new suppression parameter for suppressing the saturation signal is calculated, and if the saturation level determination result is low saturation, the saturation signal is suppressed by the calculated suppression parameter. And a saturation suppression step.

  According to a twelfth aspect of the present invention, there is provided an image processing program for causing a computer to execute predetermined image processing on a color image signal input from an image pickup device. A color signal conversion step for converting into a luminance signal, a luminance edge extraction step for extracting an edge component of the luminance signal, a suppression parameter calculation step for calculating a suppression parameter for suppressing the color difference signal from the intensity of the edge component of the luminance signal, Saturation signal calculation step for obtaining a saturation signal from the color difference signal, a saturation edge extraction step for extracting the edge component of the saturation signal, and an adjustment parameter for adjusting the suppression amount from the intensity of the edge component of the saturation signal Adjustment parameter calculation step for calculating the saturation level, a saturation level extraction step for extracting the saturation level of the saturation signal, and saturation Based on the saturation level determination step for determining the intensity of the bell, the suppression parameter, and the adjustment parameter, a new suppression parameter for suppressing the saturation signal is calculated. If the saturation level determination result is low saturation, the calculation is performed. And a saturation suppression step of suppressing the saturation signal by the suppressed suppression parameter.

  According to the present invention, the edge component of the luminance signal is extracted, the saturation signal is obtained from the color difference signal, the edge component of the saturation signal is extracted, and the level of the saturation signal is detected. The suppression amount of chroma suppression is set based on the edge component amount of the luminance signal, the edge component amount of the saturation signal, and the saturation signal level.

  As described above, in the present invention, not only the saturation level but also the edge component of the saturation signal is extracted and the level of the component is determined, so even when a thin line with high saturation exists in the low saturation area. It can be processed as a fine line that is not a false color. That is, it is possible to accurately determine a low saturation false color and prevent color loss regardless of the saturation level.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows an example of a digital camera to which the present invention is applied. In FIG. 1, reference numeral 1 denotes an image sensor. As the image sensor 1, a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary MOS) image sensor is used. The image pickup device 1 is, for example, a single plate type, and R (red), G (green), and B (blue) color filters are arranged, for example, in a Bayer array on the light receiving surface as shown in FIG. ing.

  A subject image is formed on the light receiving surface of the image sensor 1 via the optical system 2. The optical system 2 includes a lens system that forms an image of subject image light on the light receiving surface of the image sensor 1, an iris that controls exposure, and the like. The subject image is photoelectrically converted by the image sensor 1.

  Note that the arrangement of the color filters is not limited to the Bayer arrangement. Further, a complementary color filter of Cy (cyan), Mg (magenta), or Ye (yellow) may be provided.

  An imaging signal from the imaging device 1 is supplied to an A / D converter 4 via a sample hold and AGC (Automatic Gain Control) circuit 3. An A / D (Analog to Digital) converter 4 digitizes the image signal from the image sensor 1. The output of the A / D converter 4 is supplied to the single-plate / three-plate interpolation unit 5.

  When a single-plate type is used as the image sensor 1, only information of any primary color can be obtained from each pixel. That is, only red information can be obtained from the signal of the red filter pixel, only blue information can be obtained from the signal of the blue filter pixel, and only green information can be obtained from the signal of the green filter pixel. I can't. The single-plate / three-plate interpolation unit 5 forms color signals R, G, and B of each pixel by interpolation processing using information on peripheral pixels.

  The output of the single-plate / three-plate interpolation unit 5 is supplied to the image processing unit 6. The image processing unit 6 performs image processing such as edge enhancement on the RGB color image signal sent from the single-plate / three-plate interpolation unit 5. Then, the RGB color image signal is converted into a luminance signal Y, a color difference signal Cr (R−Y), and a color difference signal Cb (B−Y) and output. Further, the image processing unit 6 performs a chroma suppression process in order to eliminate the occurrence of false colors when the single plate / three-plate interpolation unit 5 performs the interpolation process.

  The present invention is applied to the chroma suppression processing of the image processing unit 6 in the digital camera configured as described above.

  FIG. 3 shows an embodiment of the present invention. In FIG. 3, the color signal conversion unit 11 is supplied with RGB color image signals output from the image sensor 1 of FIG. 1 and interpolated with the color signals of each pixel by the single-plate / three-plate interpolation unit 5. The color signal converter 11 performs the following matrix operation on the RGB color image signal, and converts the RGB color image signal into the luminance signal Y and the color difference signals Cr and Cb.

Y = 0.3R + 0.59G + 0.1B (1a)
Cr = 0.730 (R−Y) (1b)
Cb = 0.564 (B−Y) (1c)

  The luminance signal Y from the color signal conversion unit 11 is supplied to the luminance edge enhancement unit 12 and also to the luminance edge extraction unit 13.

  The luminance edge extraction unit 13 extracts an edge component (high frequency component) of the luminance signal Y. As the luminance edge extraction unit 13, a Laplacian filter is used. A Laplacian filter is also called a secondary differential (difference) filter, and is a filter that extracts high-frequency components of luminance. Specifically, for example, a 3 × 3 spatial filter as shown in FIG. 4 that extracts four vertical, horizontal, and horizontal edges (FIG. 4A), and in addition to the four directions, the diagonal 8 Some of them extract a directional edge (FIG. 4B).

  In FIG. 3, the edge component of the luminance signal extracted by the luminance edge extraction unit 13 is supplied to the luminance edge enhancement unit 12. The luminance edge enhancement unit 12 adds the edge component of the luminance signal to the luminance signal Y, and enhances the edge component of the luminance signal Y. Thus, the luminance signal Y ′ with the edge component enhanced is output from the output terminal 21.

  The color difference signals Cr and Cb from the color signal conversion unit 11 are supplied to the chroma suppression unit 15 and also to the saturation signal forming unit 17. As described above, the image pickup device 1 is of a single plate type, and a single plate / three-plate interpolation unit 5 performs an interpolation process so that the color signal of each pixel is interpolated. . The chroma suppress unit 15 suppresses the color difference signals Cr and Cb in order to prevent the occurrence of such false colors.

  The saturation signal forming unit 17 forms the saturation signal Cs by adding absolute values of the color difference signal Cr and the color difference signal Cb. The saturation signal Cs formed by the saturation signal forming unit 17 is supplied to the saturation edge extraction unit 18 and also to the saturation level detection unit 19.

  The saturation edge extraction unit 18 extracts the edge component Cedge of the saturation signal from the high frequency component of the saturation signal Cs. As the saturation edge extraction unit 18, for example, a Laplacian filter is used.

  The saturation level detection unit 19 detects the saturation signal level Clpf by smoothing the saturation signal Cs. The saturation level detection unit 19 is a low-pass filter, and for example, a 3 × 3 spatial filter as shown in FIG. 5 is used.

  The edge component Cedge of the saturation signal extracted by the saturation edge extraction unit 18 and the saturation signal level Clpf detected by the saturation level detection unit 19 are supplied to the suppression amount calculation unit 16. The edge component Yedge of the luminance signal extracted by the luminance edge extraction unit 13 is supplied to the suppression amount calculation unit 16.

  As described later, the suppression amount calculation unit 16 includes an edge component Yedge of the luminance signal from the luminance edge extraction unit 13, an edge component Cedge of the saturation signal from the saturation edge extraction unit 18, and a saturation level detection unit. The chroma suppression amount S in the chroma suppression unit 15 is calculated using the saturation signal level Clpf from 19. The calculated suppression amount S is supplied to the chroma suppression unit 15. In the chroma suppression unit 15, chroma suppression processing is performed on the color difference signals Cr and Cb. The color difference signals Cr ′ and Cb ′ (or Cr and Cb) subjected to the chroma suppression process are output from the output terminals 22 and 23.

  As described above, in this embodiment, the suppression amount of chroma suppression is determined using the edge component Yedge of the luminance signal, the edge component Cedge of the saturation signal, and the saturation signal level Clpf. This will be described in detail below.

  As described above, the image pickup device 1 is of a single plate type, and the single plate / three-plate interpolation unit 5 performs an interpolation process to interpolate the color signal of each pixel. For this reason, a false color may occur. The false color is included more as the change of the luminance signal Y is larger, that is, as the edge component amount of the luminance signal Y is larger. For this reason, it is necessary to increase the amount of suppression of chroma suppression as the edge component amount of the luminance signal Y increases.

  On the other hand, when chroma suppression processing is performed in the case of high saturation, “color loss” may occur. Further, even if the saturation is low, in order to prevent color loss in a high saturation portion in the low saturation area, it is necessary to reduce the amount of suppression of chroma suppression as the edge component of the saturation signal Cs increases. .

  Therefore, in this embodiment, the saturation signal level Clpf extracted by the saturation level extraction unit 19 is compared with the reference saturation signal level Cref so that “color loss” does not occur in the case of high saturation, and the saturation signal level Cref is compared. When the chroma signal level Clpf is larger than the reference chroma signal level Cref, it is assumed that the chroma is high and the chroma suppression process is not performed.

  Based on the edge component Yedge of the luminance signal extracted by the luminance edge extraction unit 13, the suppression amount of chroma suppress is calculated so that the suppression amount increases as the edge component amount of the luminance signal Y increases, and the saturation edge Based on the edge component Cedge of the saturation signal extracted by the extraction unit 18, the suppression amount is adjusted so that the suppression amount of the chroma suppression is reduced as the edge component of the saturation signal is larger.

  FIG. 6 is a specific example of the suppression amount calculation unit 16. In FIG. 6, look-up tables 51 and 52 are composed of, for example, a ROM (Read Only Memory).

  The look-up table 51 stores data of chroma suppression suppression amount Sedge for the edge component amount of the luminance signal (absolute value of the edge component level of the luminance signal) | Yedge |. As described above, since the suppression amount of the chroma suppression needs to be increased as the edge component of the luminance signal Y is larger, the suppression amount Sedge for the edge component amount | Yedge | of the luminance signal is shown in FIG. It becomes a function like this. Alternatively, as shown in FIG. 7B, a function may be used in which the suppression amount increases when the edge component amount | Yedge | of the luminance signal is equal to or greater than a preset threshold value. The look-up table 51 stores data on the suppression amount “Syge” for the edge component amount | Yedge | of the luminance signal based on the function shown in FIG. 7A or 7B.

  The look-up table 52 stores adjustment amount sedge data with respect to the edge component amount of the saturation signal (absolute value of the edge component level of the saturation signal) | Cedge |. As described above, since the suppression amount of chroma suppress needs to be reduced as the edge component of the saturation signal Cs increases, the adjustment amount Sedge for the edge component amount | Cedge | of the saturation signal is as shown in FIG. Function.

  The lookup table 51 is supplied with the edge component amount | Yedge | of the luminance signal extracted by the luminance edge extraction unit 13. From the look-up table 51, a suppression amount Syedge corresponding to the edge component amount | Yedge | of the luminance signal is output.

  The look-up table 52 is supplied with the edge component amount | Cedge | of the saturation signal extracted by the saturation edge extraction unit 18. From the look-up table 52, an adjustment amount Sedge corresponding to the edge component amount | Cedge | of the saturation signal is output.

Outputs of the lookup table 51 and the lookup table 52 are supplied to the calculation unit 53. With respect to the suppression amount Syedge for the edge component of the luminance signal read from the lookup table 51 and the adjustment amount Sedge for the edge component of the saturation signal read from the lookup table 52 in the calculation unit 53, The following operation is performed. Thereby, the suppression amount S of chroma suppress is calculated | required.
S = Syedge × Scedge (%) × (1/100) (2)

  That is, in FIG. 8, when the edge component amount | Cedge | of the saturation signal of the target pixel is within the false color region N, it is determined that the target pixel is likely to be a false color, and the adjustment amount Sedge Is 100%, and the suppression amount Syedge for the edge component of the luminance signal is used as it is, and the chroma suppression processing is performed. Further, when the edge component amount | Cedge | of the saturation signal of the pixel of interest is outside the range of the false color region N, it is determined that it is not a false color but an edge portion of the color existing in the subject, and the luminance signal edge component The suppression amount Syedge is adjusted by the adjustment amount Sedge.

  In FIG. 6, the output of the calculation unit 53 is supplied to the masking unit 55. The comparator 54 is supplied with the saturation signal level | Clpf | detected by the saturation level detector 19 and the reference saturation signal level Cref. The comparator 54 compares the saturation signal level | Clpf | detected by the saturation level detector 19 with the reference saturation signal level Cref. This comparison output is supplied to the masking unit 55. When the saturation signal level | Clpf | is higher than the reference saturation signal level Cref, the masking unit 55 masks the output of the suppression amount of the chroma suppress as high saturation.

The suppression amount S obtained by the suppression amount calculation unit 16 as described above is sent to the chroma suppression unit 15 (FIG. 3), and the chroma suppression unit 15 performs the chroma suppression process as follows.
Cr ′ = Cr− (Cr × S) (3a)
Cb ′ = Cb− (Cb × S) (3b)

Here, when the saturation signal level | Clpf | is higher than the reference saturation signal level Cref and the saturation is high, the suppression amount of the chroma suppress is masked by the masking unit 55. For this reason, chroma suppression processing is not performed, and the color difference signals Cr and Cb from the color signal converter 11 are output as they are. In this case, the processing of the chroma suppression unit 15 is as follows.
Cr ′ = Cr (4a)
Cb ′ = Cb (4b)

  Therefore, the chroma suppression unit 15 process is effective only when it is determined that the color saturation is low. Further, even in a portion where the edge component Yedge of the luminance signal Y is not detected, the chroma suppression process is not performed, and the color difference signals Cr and Cb calculated by the above formula are output from the chroma suppress unit 15.

  As described above, in this embodiment, the suppression amount of chroma suppression is obtained using not only the edge component of the luminance signal and the level of the saturation signal but also the edge component of the saturation signal. For this reason, even when a high saturation thin line exists in the low saturation area, no color loss occurs. FIG. 9 shows this.

  For example, as shown in FIG. 9A, there are an area 101 with low saturation and high brightness, and an area 102 with low brightness and high saturation, and a thin line 103 with high saturation is drawn in the area 102 with low brightness and high saturation. Suppose that an image like this is taken. The area 102 is only the R component, and the thin line 103 is the R component thin line.

  When this image is taken, the image pickup signal from the image sensor 1 is sent to the single-plate three-plate interpolation unit 5, and the color signal is interpolated by the single-plate three-plate interpolation unit 5, and the RGB color from the single-plate three-plate interpolation unit 5. The image signal is sent to the color signal converter 11, and the color signal converter 11 converts the image signal into a luminance signal Y and color difference signals Cr and Cb. As a result, for example, the luminance signal Y and the color difference signals Cr and Cb in the portion of the line L1 in FIG. 9 change as shown in FIGS. 9B, 9D, and 9E.

  Here, when the edge component Yedge of the luminance signal is extracted by the luminance edge extraction unit 13, as shown in FIG. 9C, the portion of the thin line 103 (point A) or the boundary between the area 101 and the area 102 (point B) ), An edge signal is output.

  Further, as shown in FIG. 9F, a saturation signal Cs is formed by adding absolute values of the color difference signal Cr shown in FIG. 9D and the color difference signal Cb shown in FIG. Here, when the level of the saturation signal Cs is detected by the saturation level detector 19, the saturation signal level is increased in the area 102 as shown in FIG.

  Further, when the edge component Cedge of the saturation signal Cs is extracted by the saturation edge extraction unit 18, as shown in FIG. 9H, the portion of the thin line 103 (point A) and the boundary between the area 101 and the area 102 ( At point B), an edge signal is output.

  At point B at the boundary between the area 101 and the area 102, as shown in FIG. 9G, the saturation signal level Clpf is equal to or higher than the reference saturation signal level Cref, and it is determined that the saturation is high. For this reason, chroma suppression processing is not performed, and the color difference signals Cr and Cb output from the color signal conversion unit 11 are output as they are.

  At point A of the thin line 103, the saturation signal level Clpf is smaller than the reference saturation signal level Cref, so that it is regarded as low saturation and is subject to chroma suppression processing. In the prior art, such a line is determined to be a false color, and chroma suppression is performed using the suppression amount obtained from the edge level of the luminance signal. Therefore, there is a large amount of suppression for the saturation at point A, and there is a high possibility that the thin line is lost.

  On the other hand, in the present invention, the point A is determined to be an edge portion of a color that originally exists in the subject based on the edge component amount | Cedge | of the saturation signal Cs extracted by the saturation edge extraction unit 18. Therefore, as described above, the suppression amount of chroma suppress is adjusted. That is, based on equation (2), the suppression amount Sydge obtained from the edge component amount | Yedge | level of the luminance signal Y is adjusted by the adjustment amount Sedge obtained from the edge component amount | Cedge | of the saturation signal. The As a result, the amount of suppression of chroma suppression at point A can be reduced, color loss at point A can be prevented, and the saturation of the thin line can be maintained.

  In the above-described embodiment, an example in which the Laplacian filter is used in the luminance edge extraction unit 13 is shown.

  The Sobel filter is a filter that weights pixels near the pixel of interest when performing first-order differentiation. Specifically, for example, there are four types of filters shown in FIG. 10A extracts a line segment that rises to the right, FIG. 10B extracts a line segment that is a vertical line, FIG. 10C extracts a line segment that descends to the right, and FIG. 10D extracts a line segment that is a horizontal line. With these filters, the degree of the line segment in the target pixel can be calculated. The degree of the line segment corresponds to the edge of the luminance signal, and has the same meaning as the Laplacian filter described in the first embodiment, and can be substituted.

  In the saturation edge extraction unit 18, as in the case of the luminance edge extraction unit 13, an example using a Laplacian filter has been shown. Specifically, for example, there are four types of filters shown in FIG. 10A extracts a line segment that rises to the right, FIG. 10B extracts a line segment that is a vertical line, FIG. 10C extracts a line segment that descends to the right, and FIG. 10D extracts a line segment that is a horizontal line. With these filters, the degree of the line segment in the target pixel can be calculated. Moreover, since a line segment is extracted, a thin line can also be extracted accurately. The degree of the line segment corresponds to the edge component of the saturation signal, and has the same meaning as the Laplacian filter described in the first embodiment, and can be substituted.

  Although an example using a low-pass filter has been shown as the saturation level detection unit 19, the present invention is not limited to this, and a median filter can also be used. The median filter replaces the median of neighboring pixel values with the pixel of interest. Specifically, in the 3 × 3 region, the nine pixel values are arranged in the order of low (or high), and the fifth (center) pixel value is a filter of interest pixel value. When there is a value that does not exist in the surrounding pixels (isolated point), processing is performed in which the surrounding pixels are replaced. In other words, it is possible to extract the average saturation signal level in the low saturation area, which has the same meaning as the low-pass filter described in the first embodiment, and can be substituted.

  The present invention can also be realized by software on a computer. FIG. 11 is a flowchart when the processing of the image processing unit is realized by software in the processing shown in FIG.

  In FIG. 11, the RGB color image signal input to the image processing block is converted into a luminance signal Y and color difference signals Cr and Cb (step S1).

  Next, the edge component amount | Yedge | of the luminance signal Y is extracted (step S2), the edge portion of the luminance signal is added to the luminance signal, and the edge of the luminance signal is emphasized (step S3).

  On the other hand, the absolute values of the color difference signals Cr and Cb are added to calculate the saturation signal Cs (step S4), and the saturation signal level Clpf is extracted (step S5).

  The extracted saturation signal level Clpf is compared with a preset reference saturation signal level Cref to determine whether the saturation is high. If the extracted saturation signal level Clpf is greater than the reference saturation signal level Cref, it is determined that the saturation is high, and the process proceeds to step S7.

  If the extracted saturation signal level Clpf is smaller than the reference saturation signal level Cref in step S6, it is determined that the saturation is low. In this case, the suppression amount Syedge of chroma suppression is calculated from the edge component amount | Yedge | of the luminance signal extracted in Step S2 (Step 8). Then, the edge component amount | Cedge | of the saturation signal is extracted (step S9), and the edge component amount | Cedge | of the saturation signal is compared with a preset threshold value N (step S10).

  If the edge component amount | Cedge | of the saturation signal is greater than or equal to the threshold value N, the adjustment amount Sedge is calculated from the edge amount of the saturation signal (step S11), and the process proceeds to step S12. On the other hand, if it is less than or equal to the threshold value N, the process advances to step S12 without calculating the adjustment amount Scope.

  In step S12, the suppression amount S of chroma suppression is calculated using the suppression amount Syedge and the adjustment amount Sedge, and using this suppression amount S, the chroma suppression process is performed on the color difference signals Cr and Cb. Then, the process proceeds to step S7.

  In step S7, the color difference signals Cr and Cb, or the chroma-suppressed color difference signals Cr 'and Cb' (or Cr and Cb), and the luminance signal Y 'with enhanced edge components are output.

  As described above, in this embodiment, the suppression amount of chroma suppression is determined using the edge component of the luminance signal, the edge component of the saturation signal, and the saturation signal level. For this reason, even when a high saturation thin line exists in the low saturation area, no color loss occurs.

  The present invention is not limited to the above-described embodiments, and various modifications and applications can be made without departing from the gist of the present invention.

It is a block diagram which shows the structure of an example of the digital camera which can apply this invention. It is explanatory drawing of the arrangement | sequence of the color filter arrange | positioned at an image sensor. It is a block diagram which shows the structure of one Embodiment of this invention. It is explanatory drawing of a Laplacian filter. It is explanatory drawing of a low-pass filter. It is a block diagram which shows the structure of the suppression amount calculating part in one Embodiment of this invention. It is a graph which shows the relationship between the amount of edge components of a luminance signal, and the amount of suppression of chroma suppress. It is a graph which shows the relationship between the amount of edge components of a saturation signal, and the adjustment amount of chroma suppress. It is each part waveform diagram used for description of one Embodiment of this invention. It is explanatory drawing of a Sobel filter. It is a flowchart used for description at the time of implement | achieving this invention by software. It is a figure used for description of generation | occurrence | production of a false color. It is a figure used for description of the subject of a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image pick-up element 2 Optical system 4 A / D converter 5 Single board | plate 3 plate interpolation part 6 Image processing part 11 Color signal conversion part 12 Luminance edge emphasis part 13 Luminance edge extraction part 15 Chroma suppression part 16 Suppression amount calculation part 17 Saturation signal Formation unit 18 Saturation edge extraction unit 19 Saturation level detection unit

Claims (12)

An image processing apparatus that performs predetermined image processing on a color image signal input from an image sensor,
Color signal conversion means for converting the color image signal into a luminance signal and a color difference signal;
Luminance edge extraction means for extracting edge components of the luminance signal;
A saturation signal calculating means for obtaining a saturation signal from the color difference signal;
Saturation edge extraction means for extracting edge components of the saturation signal;
Saturation level determination means for determining the level of the saturation signal;
An image comprising: chroma suppression means for suppressing the color difference signal with a suppression amount set based on an edge component of the luminance signal, an edge component of the saturation signal, and a saturation signal level. Processing equipment.   The image processing apparatus according to claim 1, wherein the chroma suppression unit changes a suppression amount of the color difference signal in accordance with an intensity of an edge component of the luminance signal.   The image processing apparatus according to claim 1, wherein the chroma suppression unit changes a suppression amount of the color difference signal according to an intensity of an edge component of the saturation signal.   The image processing apparatus according to claim 1, wherein the chroma suppression unit suppresses the color difference signal in accordance with an intensity of the saturation signal.   The image processing apparatus according to claim 1, wherein the luminance edge extraction unit includes a Laplacian filter.   The image processing apparatus according to claim 1, wherein the luminance edge extraction unit includes a Sobel filter.   The image processing apparatus according to claim 1, wherein the saturation edge extraction unit includes a Laplacian filter.   The image processing apparatus according to claim 1, wherein the saturation edge extraction unit includes a Sobel filter.   The image processing apparatus according to claim 1, wherein the saturation level determination unit includes a low-pass filter.   The image processing apparatus according to claim 1, wherein the saturation level determination unit includes a median filter. An image processing method for performing predetermined image processing on a color image signal input from an image sensor,
A color signal conversion step for converting the color image signal into a luminance signal and a color difference signal;
A luminance edge extraction step of extracting an edge component of the luminance signal;
A suppression parameter calculation step of calculating a suppression parameter for suppressing a color difference signal from the intensity of an edge component of the luminance signal;
A saturation signal calculation step for obtaining a saturation signal from the color difference signal;
A saturation edge extraction step of extracting an edge component of the saturation signal;
An adjustment parameter calculating step for calculating an adjustment parameter for adjusting the amount of suppression from the intensity of the edge component of the saturation signal;
A saturation level extracting step of extracting a saturation level of the saturation signal;
A saturation level determination step for determining the intensity of the saturation level;
Based on the suppression parameter and the adjustment parameter, a new suppression parameter for suppressing the saturation signal is calculated. If the saturation level determination result is low saturation, the saturation signal is calculated by the calculated suppression parameter. And a saturation suppression step for suppressing the image. An image processing program for causing a computer to execute predetermined image processing on a color image signal input from an imaging device,
On the computer,
A color signal conversion step for converting the color image signal into a luminance signal and a color difference signal;
A luminance edge extraction step of extracting an edge component of the luminance signal;
A suppression parameter calculation step of calculating a suppression parameter for suppressing a color difference signal from the intensity of an edge component of the luminance signal;
A saturation signal calculation step for obtaining a saturation signal from the color difference signal;
A saturation edge extraction step of extracting an edge component of the saturation signal;
An adjustment parameter calculating step for calculating an adjustment parameter for adjusting the amount of suppression from the intensity of the edge component of the saturation signal;
A saturation level extracting step of extracting a saturation level of the saturation signal;
A saturation level determination step for determining the intensity of the saturation level;
Based on the suppression parameter and the adjustment parameter, a new suppression parameter for suppressing the saturation signal is calculated. An image processing program that executes a saturation suppression step for suppressing.

Images