JP2006270592A - Image processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To eliminate a troublesome process wherein a three primary color coordinates system is converted to a certain coordinates system, saturation is corrected, and then the certain coordinates system is converted again to the three primary color coordinates system, and to alleviate a color shift in an achromatic color even when a shift or a noise occurs in a white balance. <P>SOLUTION: The three primary color coordinates system is converted to a hue coordinates space, and two or more regions (Z1 to Z9) are set in the hue coordinates space. A distance between the region Z1 and the origin is short, the region Z1 is considered as an achromatic color region, and other regions (Z2 to Z9) are considered as chromatic color regions. It is judged which region pixels are located, a luminance conversion coefficient and a color difference conversion coefficient are calculated, and a luminance signal conversion and a color difference signal conversion are carried out. By this setup, the image data can be subjected to saturation correction processing when color reproduction processing is carried out, a coordinates system re-conversion process can be dispensed with, and a color shift in an achromatic color can be alleviated. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus suitable for performing saturation correction processing on input image data when performing color reproduction processing.

Color saturation correction processing in a conventional image processing apparatus has been performed on color image data (for example, a JPEG image) after color reproduction processing has been performed.
Specifically, the conventional saturation correction processing corrects a color image (such as a JPEG image) that has undergone color reproduction processing so that the saturation decreases as the saturation decreases (patented). References 1, 2, and 3) are corrected so as to decrease when the saturation is equal to or lower than a predetermined value, and are corrected so as to increase when the saturation is equal to or higher than the predetermined value (see Patent Document 4). The saturation correction was appropriately performed in accordance with the degree and brightness (see Patent Document 5).

The color image data after the color reproduction processing is in a data format (for example, luminance signal Y, color difference signals Cb, Cr) displayed by a luminance signal and a color difference signal. In such a data format, Each information of hue, saturation and luminance can be easily obtained. Therefore, it is relatively easy to perform saturation correction using these pieces of information.
JP 07-46623 A JP 2000-224607 A JP 2002-33931 A Japanese Patent Laid-Open No. 2000-152018 JP 2003-85545 A

As described above, it is easy to perform saturation correction using color image data in a data format in which color reproduction processing is expressed by a luminance signal and a color difference signal.
However, if the saturation correction is performed on the color image data after the color reproduction processing of the three primary color output formats called Tiff and BMP (bitmap) is performed, the following problems occur. .

  First, in the case of a data format using the three primary colors R, G, and B, a coordinate system (luminance signal Y, color difference signals Cb, Cr) that once represents the hue and saturation is used to obtain information on hue, saturation, and luminance. Is converted into the original coordinate system (R, G, B) again to correct the saturation in the converted coordinate system and return the corrected saturation to the data format of the three primary colors R, G, B. There is annoyance that must be done.

  Secondly, consider the three primary color data (R, G, B) before obtaining color image data by color reproduction processing. For example, when the hue is determined by the hue coordinate space of (RG) and (BG) and the color difference signal is obtained by the conversion coefficient for each hue, the achromatic color (monochrome) is basically R = G = B. Therefore, (R−G) = (B−G) = 0 should be obtained. However, due to white balance deviation and noise, (RG) ≠ (BG) ≠ 0 is not satisfied, and hue determination varies, and the color difference is determined by a conversion coefficient optimized for each hue. Even if the signal is obtained, subtle coloring occurs in the achromatic region.

  The present invention has been made in view of the above-described problems, eliminates the trouble of performing saturation correction processing again after color reproduction processing, the trouble of saturation correction in the three primary color output format, and shifts to white balance. An object of the present invention is to provide an image processing apparatus capable of reducing a color shift with respect to an achromatic color even when noise occurs.

  The image processing apparatus according to claim 1, based on input image data including a first color component, a second color component, and a third color component, (first color component−third color component) and (second color component− Coordinate conversion means for obtaining a hue coordinate system having the third color component) as a unit system, area setting means for determining a plurality of areas according to the coordinates of the hue coordinate system, and coordinates in the hue coordinate system for each pixel of the input image data Area classification means for classifying into areas based on the obtained coordinates, and for each pixel of the input image data, color to be reproduced using a luminance conversion coefficient and a color difference conversion coefficient predetermined for each classified area And reproduction means.

  The image processing apparatus according to claim 2, based on input image data including a first color component, a second color component, and a third color component, (first color component−second color component) and (first color component− Coordinate conversion means for obtaining a hue coordinate system having the third color component) as a unit system, area setting means for determining a plurality of areas according to the coordinates of the hue coordinate system, and coordinates in the hue coordinate system for each pixel of the input image data Area classification means for classifying into areas based on the obtained coordinates, and for each pixel of the input image data, color to be reproduced using a luminance conversion coefficient and a color difference conversion coefficient predetermined for each classified area And reproduction means.

  According to a third aspect of the present invention, an image processing apparatus includes (first color component-third color component) and (second color component-) from input image data including a first color component, a second color component, and a third color component. A coordinate conversion means for obtaining a hue coordinate system having the third color component) as a unit system, a coordinate in the hue coordinate system for each pixel of the input image data, and a saturation based on the origin of the hue coordinate system and the obtained coordinates And saturation correction means for performing correction.

  According to a fourth aspect of the present invention, there is provided an image processing apparatus comprising: (first color component-second color component) and (first color component-) from input image data including a first color component, a second color component, and a third color component. A coordinate conversion means for obtaining a hue coordinate system having the third color component) as a unit system, a coordinate in the hue coordinate system for each pixel of the input image data, and a saturation based on the origin of the hue coordinate system and the obtained coordinates And saturation correction means for performing correction.

The image processing device according to claim 5 is the image processing device according to claim 3 or 4, wherein a distance between the origin of the hue coordinate system and the obtained coordinate is obtained, and saturation correction is performed according to the obtained distance. It has a look-up table storing quantities.
The image processing device according to claim 6 is the image processing device according to claim 3 or 4, further comprising a look-up table storing a saturation correction amount corresponding to the obtained coordinates. To do.

  The image processing device according to claim 7 is the image processing device according to claim 3 or 4, wherein the saturation correction unit determines a saturation correction amount by a distance between an origin of the hue coordinate system and the obtained coordinates. Based on the above, it is obtained by an interpolation method.

According to the present invention, the trouble of performing the saturation correction process again after the color reproduction process and the trouble of the saturation correction in the three primary color output format are eliminated.
Further, according to the present invention, even when a white balance shift occurs, a color shift with respect to an achromatic color can be reduced.

Hereinafter, an embodiment of the present invention will be described.
FIG. 1 is a block diagram showing an embodiment of the present invention. This embodiment applies the present invention to an electronic camera, and corresponds to all claims.
In FIG. 1, 1 is an imaging lens, an image sensor such as 2CCD, 3 is an A / D converter, 4 is a buffer memory, 5 is a CPU, 6 is an electronic camera operation system, and 7 is an ASIC (Application Specific Integrated Circuits). Application integrated circuit), 8 is ROM, 9 is a memory card, 10 is an interface (for example, USB), and 11 is a bus.

Note that the coordinate conversion means, area setting means, area classification means, color reproduction means, and saturation correction means described in the claims correspond to the CPU 5, ROM 8, ASIC 7, and the like. In some cases, the ASIC 7 performs the role of the CPU 5.
(First operation of the embodiment)
Next, the first operation of the embodiment shown in FIG. 1 will be described. Since the photographing system of the electronic camera is well known, only the part related to the processing in this embodiment will be described. Here, the CPU 5 of the electronic camera performs various controls including preparation control before photographing based on a control program stored in the ROM 8, and the ASIC 7 performs a saturation correction process described below.

That is, the image sensor 2 converts an optical image obtained through the photographing lens 1 into an analog signal, and this analog signal is converted into, for example, a 12-bit digital signal (R, G, B) via the A / D converter 3. Converted. In this way, image data (R, G, B) that has not undergone color reproduction processing is input from the image sensor 2 to the electronic camera.
Image data (R, G, B) input to the electronic camera is temporarily stored in the buffer memory 4 and then input to the ASIC 7.

Since the ASIC 7 performs color temperature correction on the image data (R, G, B: RAW data) that has not been subjected to color reproduction processing for each of the R, G, and B image data based on the environmental color temperature, white balance is performed. Processing (processing of multiplying a predetermined gain so that R = G = B) is performed.
Further, the ASIC 7 performs gamma conversion processing corresponding to the gradation characteristics (conversion to 8-bit data easy to handle in the subsequent processing) on the 12-bit image signal (R, G, B) as shown in FIG. I do. A curve γ shown in FIG. 2 is a gradation characteristic conversion curve.

  After the gamma conversion processing is performed, the ASIC 7 converts the R data, G data, and B data for each pixel in the mosaic arrangement of the image sensor 2 (see FIG. 1) shown in FIG. As shown in FIG. 4, color interpolation processing is performed for processing as image data of three colors per pixel. That is, R, G, and B data are given to the image data of each pixel. That is, FIG. 3A shows image data corresponding to a color filter (any one color of R, G, and B) provided in each pixel of the image sensor 2, and FIG. This is to provide each data of R, G, B by color interpolation processing in consideration of the color of the pixel. In FIG. 3B, as shown in the figure, using the R data, G data, and B data shown in each pixel of FIG. 3A, (RG) is used as new R data, and (BG) is used. New B data. The G data is used as it is without being processed.

  Next, since the ASIC 7 performs color reproduction processing for each pixel, from the (RG) data (new R data) and (BG) data (new B data) shown in FIG. As shown in FIG. 2, a hue coordinate space represented by coordinates (RG) and (BG) is created for each pixel. In FIG. 4, the hue coordinate space is divided into nine regions Z1 to Z9, but the number of divisions may be arbitrary. The division of the hue coordinate space is not limited to the example shown in FIG. 4, and concentric divided areas corresponding to the distance from the origin may be set around the origin of the hue coordinate space.

Here, the region Z1 is a region in which (RG) and (BG) are both close to zero, and is regarded as an achromatic color (see FIG. 4).
FIG. 5 shows the nine regions Z1 to Z9 shown in FIG. 4 that are classified by hue. Each pixel corresponds to one of the regions Z1 to Z9 based on hue coordinates determined by (RG) and (BG).

Further, the pixel corresponding to any one of the areas Z1 to Z9 becomes the hue of the area corresponding to the coordinates converted into the hue coordinates. As hues, as shown in FIG. 4, achromatic colors and chromatic colors (for example, blue series 1, red series 1, red series 2, etc.) are predetermined.
In the areas Z1 to Z9, a luminance conversion coefficient and a color difference conversion coefficient are determined in advance, and luminance conversion and color conversion are performed using the luminance conversion coefficient and the color difference conversion coefficient of the area where the pixel exists. At this time, since it is determined which of the regions Z1 to Z9 corresponds to the pixel, saturation correction is performed together with the color conversion processing.

That is, the saturation is determined according to the distance from the origin in the hue coordinate space of (RG) and (BG). Therefore, color conversion and saturation correction can be performed with higher accuracy by dividing a larger number of regions.
The conversion of the luminance signal Y and the conversion of the color difference signals Cb and Cr can generally be performed by the following formulas (1) to (3) with respect to (R, G, B).

That is, the luminance signal Y is
Y = YGn * G + YBn * (BG) + YRn * (RG) (1)
Converted by
The color difference signals Cb and Cr are
Cb = KBBn × (BG) + KBRn × (RG) + KCb (2)
Cr = KRBn × (BG) + KRRn × (RG) + KCr (3)
Converted by These calculations are performed in the ASIC 7.

  Here, the luminance conversion coefficients YGn, YBn, YRn and the color difference conversion coefficients KBBn, KBRn, KRBn, KRRn are coefficients set for each divided area in the hue coordinate space. For example, the hue coordinate space is expressed as Z1 to Z9 ( If the area is divided into (9 hues), nine types of coefficients can be set as shown in FIG. Here, since the region Z1 is an achromatic color, the color difference conversion coefficients KBB1 to KRR1 of the region Z1 are set to values that make the color difference zero in total. Further, since the region Z1 is an achromatic color, the luminance signal Y need not be converted. Further, in equations (2) and (3), KCb and KCr are color difference signal offset values for making the color difference signals Cb and Cr positive values.

The ASIC 7 performs the calculation with reference to, for example, a lookup table storing the table shown in FIG. In the first operation, the color reproduction process is performed as described above.
As is clear from the above description, according to the first operation of the embodiment, it is possible to perform saturation correction processing on input image data when performing color reproduction processing. Therefore, after the three primary colors (R, G, B) coordinate system is converted into a coordinate system (luminance signal Y, color difference signals Cb, Cr) representing hue and saturation, the saturation is corrected, and then the three primary colors (R, G, B) are again corrected. ) The trouble of converting to the coordinate system is eliminated, and even when a deviation or noise occurs in the white balance, the color difference signals Cb and Cr for the region Z1 are corrected, so that the color deviation for the achromatic color can be reduced.
(Second operation of the embodiment)
Next, the second operation of the embodiment shown in FIG. 1 will be described. The second operation differs from the first operation in that the region Z1 shown in FIG. 4 is not defined, and depends on the distance D from the origin in the hue coordinate space of (RG) and (BG). In other words, the achromatic color and the chromatic color are determined, and the saturation correction coefficient is changed according to the distance D. In the second operation, the description of the same part as the first operation is omitted. That is, also in the second operation, the procedure of the luminance signal conversion and the color difference signal conversion described in the first operation is the same.

  In the second operation, as shown in the hue coordinate space of FIG. 7, the achromatic region Z1 shown in FIG. 4 does not exist. Instead, in the second operation, the distance D from the origin in the hue coordinate space shown in FIG.

And the calculated value of the distance D is compared with the determination value TH. As a result of the comparison, if the distance D is smaller than the determination value TH, it is determined as an achromatic color, and if it is larger, it is determined as a chromatic color. The saturation correction coefficient KSn is obtained according to the distance D.
When the ASIC 7 obtains the saturation correction coefficient KSn from the look-up table, the ASIC 7 refers to a value corresponding to the distance D as shown in FIGS. FIG. 8 shows an example in which the distance D is roughly taken, and FIG. 9 shows an example in which the distance D is taken finely. The saturation shown in FIG. 9 can be corrected more finely.

The color difference signals Cb and Cr are corrected as follows using the saturation correction coefficient KSn. As is apparent from FIGS. 8 and 9, the color difference signals Cb and Cr can be corrected also for the region determined to be an achromatic color.
Cb = Cb × KSn
Cr = Cr × KSn
Next, a method for obtaining the saturation correction coefficient KSn by saturation correction coefficient interpolation will be described with reference to FIGS. 10 and 11. That is, as shown in FIG. 10, the distance D from the origin in the hue coordinate space is divided into a plurality of regions (region boundary values KSTn: n = 1, 2,... 6) and applied to each region boundary value KSTn. Using the saturation correction coefficient (KSYn: n = 1, 2,... 6), an accurate saturation correction coefficient KSn is obtained by coefficient interpolation for saturation correction. It is assumed that the relationship between the boundary region value KSTn and the saturation correction coefficient KSYn is obtained in advance.

  FIG. 10 is a diagram showing the relationship between the region boundary value KSTn and each saturation correction coefficient KSYn applied to each boundary region value KSTn, and FIG. 11 obtains the saturation correction coefficient by coefficient interpolation for saturation correction. It is explanatory drawing which shows the specific example in a case. 10 and 11, n is 1-6. Here, the saturation correction coefficient KSY6 is 1.0, which means a case where no correction of the saturation correction coefficient KSn is performed. Note that a point a in FIG. 11 indicates a saturation correction coefficient KSY obtained by coefficient interpolation for saturation correction.

  As is apparent from FIGS. 10 and 11, between KST (n + 1) and KSTn, the saturation correction coefficient KSYn is expressed by the following equation (5) by the saturation correction coefficient interpolation.

Sought by. The ASIC 7 performs coefficient interpolation processing for the saturation correction.
Needless to say, each saturation correction coefficient KSn in each region boundary value KSTn (n = 1, 2,... 6) is KSYn (n = 1, 2,... 6).
As is clear from the above description, according to the second operation of the embodiment, it is possible to perform saturation correction processing on input image data when performing color reproduction processing. Therefore, after the three primary colors (R, G, B) coordinate system is converted into a coordinate system (luminance signal Y, color difference signals Cb, Cr) representing the hue and saturation, the saturation is corrected, and then the three primary colors (R, G, B) are restored. ) The trouble of converting to a coordinate system is eliminated, and even if a deviation or noise occurs in the white balance, the color difference signals Cb and Cr are corrected, so that the color deviation with respect to the achromatic color can be reduced.

In the embodiment described above, the color difference conversion coefficient is set to a different value for each region (Z1, Z2, etc.), but the present invention is not limited to this, and each region (Z1, Z2, etc.) is set. The same value may be set at.
Further, in the embodiment described above, the color reproduction is performed by obtaining the saturation correction coefficient according to the distance D from the origin in the hue coordinate space, but the present invention is not limited to this, and the hue is not limited thereto. The saturation correction coefficient may be obtained based on the coordinate position in the coordinate space. This calculation process is performed in the ASIC 7.

  Further, in the embodiment described above, the hue coordinate space is created from (RG) and (BG), but the present invention is not limited to this, and (RB) and ( (GB), (GR) and (BR), (RG) and (RB), (GR) and (GB), (BR) and (B- The hue coordinate space may be configured using each combination of G). That is, the hue coordinate space is also established when the third color component is subtracted from the first and second color components and when the second and third color components are subtracted from the first color component.

  Furthermore, in the embodiment described above, an example in which the saturation correction coefficient is stored in the lookup table is shown, but the present invention is not limited to this, and the origin of the hue coordinate system, the obtained coordinates, The saturation correction amount corresponding to the obtained distance may be stored in the lookup table. Further, the saturation correction amount corresponding to the obtained coordinates may be stored in the lookup table.

  The present invention can be used greatly in the field of image processing apparatuses mounted on digital cameras and the like.

It is a block diagram which shows one Embodiment of this invention. It is explanatory drawing of a gamma conversion process. It is a figure explaining the procedure which converts the input image data of one color per pixel into the input image data of three colors per pixel. It is a figure which shows the hue coordinate space represented by the coordinate of (RG) and (BG) for every pixel. It is the figure which classified 9 area | regions Z1-Z9 shown in FIG. 4 with the hue. It is a figure which shows the luminance conversion coefficient and color difference conversion coefficient corresponding to nine area | regions Z1-Z9 shown in FIG. It is a figure which shows the hue coordinate space represented by the coordinate of (RG) and (BG) for every pixel. It is a figure which shows the example at the time of calculating | requiring a saturation correction coefficient from a look-up table. It is a figure which shows the example at the time of calculating | requiring a saturation correction coefficient from a look-up table. It is explanatory drawing for calculating | requiring a saturation correction coefficient by coefficient interpolation of saturation correction. It is explanatory drawing which shows the specific example in the case of calculating | requiring a saturation correction coefficient by the coefficient interpolation of saturation correction.

Explanation of symbols

  1 photographing lens, 2 CCD, 3 A / D converter, 4 buffer memory, 5 CPU, 6 operation system, 7 ASIC, 8 ROM, 9 memory card, 10 interface, 11 bus, Cb, Cr color difference signal, D from origin Distance, KSn, KSYn saturation correction coefficient, KSTn region boundary value, Y luminance signal, Z1, Z2-Z9 region

Claims (7)

Hue having unit system of (first color component-third color component) and (second color component-third color component) from input image data composed of the first color component, the second color component, and the third color component Coordinate conversion means for obtaining a coordinate system;
Area setting means for defining a plurality of areas according to the coordinates of the hue coordinate system;
Area classification means for obtaining coordinates in the hue coordinate system for each pixel of the input image data, and classifying the area based on the obtained coordinates;
An image processing apparatus comprising: color reproduction means for reproducing colors using a luminance conversion coefficient and a color difference conversion coefficient that are predetermined for each of the classified areas for each pixel of the input image data. Hue having unit system of (first color component-second color component) and (first color component-third color component) from input image data composed of the first color component, the second color component, and the third color component Coordinate conversion means for obtaining a coordinate system;
Area setting means for defining a plurality of areas according to the coordinates of the hue coordinate system;
Area classification means for obtaining coordinates in the hue coordinate system for each pixel of the input image data, and classifying the area based on the obtained coordinates;
An image processing apparatus comprising: color reproduction means for reproducing colors using a luminance conversion coefficient and a color difference conversion coefficient that are predetermined for each of the classified areas for each pixel of the input image data. Hue having unit system of (first color component-third color component) and (second color component-third color component) from input image data composed of the first color component, the second color component, and the third color component Coordinate conversion means for obtaining a coordinate system;
An image comprising: saturation correction means for obtaining coordinates in the hue coordinate system for each pixel of the input image data, and performing saturation correction based on the origin of the hue coordinate system and the obtained coordinates. Processing equipment. Hue having unit system of (first color component-second color component) and (first color component-third color component) from input image data composed of the first color component, the second color component, and the third color component Coordinate conversion means for obtaining a coordinate system;
An image comprising: saturation correction means for obtaining coordinates in the hue coordinate system for each pixel of the input image data, and performing saturation correction based on the origin of the hue coordinate system and the obtained coordinates. Processing equipment. The image processing apparatus according to claim 3 or 4,
An image processing apparatus comprising: a lookup table that obtains a distance between an origin of the hue coordinate system and the obtained coordinates and stores a saturation correction amount according to the obtained distance. The image processing apparatus according to claim 3 or 4,
An image processing apparatus comprising a lookup table storing a saturation correction amount corresponding to the obtained coordinates. The image processing apparatus according to claim 3 or 4,
The image processing apparatus according to claim 1, wherein the saturation correction means obtains a saturation correction amount by an interpolation method based on a distance between an origin of the hue coordinate system and the obtained coordinates.

Images