JP2002150283A - Image processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device that can suppress emphasis on an unwanted part such as color noise and color gradation and execute appreciable edge enhancement processing as a whole. SOLUTION: An input color image is subjected to filter processing for each component R, G, B in an enhancement level computing part 11, which computes enhancement levels of each component from results of the filter processing. The computed enhancement levels of each color component are input into a minimum value selecting part 12, which selects the enhancement level that has a minimum absolute value from the enhancement levels of each color component. An enhancement level applying part 13 uses an adding part 23r, 23g, 23b to add the enhancement level with the minimum absolute value selected by the minimum value selecting part 12 to each color component of the color image input into an edge enhancing part 3, and outputs a color image with an enhanced edge.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for performing an edge enhancement process on a color image signal composed of a plurality of color component signals.

[0002]

2. Description of the Related Art FIG. 6 is a block diagram showing an example of a conventional image processing apparatus for performing edge enhancement processing. In the figure, 21r,
21g and 21b are filter processing units, 22r is an R enhancement amount calculation unit, 22g is a G enhancement amount calculation unit, 22b is a B enhancement amount calculation unit, and 23r, 23g and 23b are addition units. Here, the input color image is red (R), green (G), blue (B)
And three color components. Hereinafter, this case will be described as an example. In a conventional image processing apparatus that performs an edge enhancement process, an enhancement amount is calculated by a filter process for each color component and added to each color component to perform the edge enhancement process.

That is, with respect to the R component, the input R color component is subjected to a filtering process by, for example, a Laplacian filter or the like in a filter processing unit 21r, and an enhancement amount for the R color component is calculated in an R enhancement amount calculation unit 22r. hand,
The adder 23r adds the enhancement amount to the original R color component.
Similarly, for the G component, the input G color component is subjected to filter processing by, for example, a Laplacian filter or the like in the filter processing unit 21g, and the G enhancement amount calculation unit 22g
, The enhancement amount for the G color component is calculated, and the enhancement amount is added to the original G color component by the adding unit 23g. As for the B component, the filter processing unit 2 performs processing on the input B color component.
In 1b, a filtering process is performed using, for example, a Laplacian filter or the like, a B enhancement amount calculating unit 22b calculates an enhancement amount for the B color component, and the enhancement amount is added to the original B color component in an adding unit 23b.

As described above, in the conventional image processing apparatus, the edge enhancement processing is performed for each color component signal.
Therefore, when a certain color component signal greatly changes, the edge of the color component signal is emphasized. For this reason, there has been a problem that the color reproducibility is deteriorated and the image quality is deteriorated, for example, the hue is changed by the edge emphasis processing in the colored edge portion, and the achromatic portion is colored.

When a color image signal mixed with color noise is input, any of the color component signals greatly changes in the color noise portion, so that the mixed color noise is emphasized, and the color noise is emphasized. There was also a problem that noise was conspicuous. FIG. 7 is an explanatory diagram of a problem when color noise is input in a conventional image processing apparatus that performs edge enhancement processing. The example shown in FIG. 7A shows a case where yellow noise is mixed in a dark image, and the R component and the G component are increased. In this case, when the edge enhancement processing in the conventional image processing apparatus shown in FIG. 6 is performed, the R component and the G component are largely changed because the changes are large. As shown in FIG. Color noise will be further enhanced. In particular, color image signals acquired by a digital camera or the like are liable to be mixed with color noise in dark areas due to variation in CCD sensitivity, and are susceptible to image quality degradation due to the enhancement of color noise as described above. was there.

Further, in an image compressed for each block such as JPEG, a color step may be generated at the boundary of the block when expanded. If such a color difference occurs, there is a problem that the step is further enlarged by the edge enhancement processing as described above, and the boundary of the block becomes conspicuous.

On the other hand, for example, Japanese Patent Application Laid-Open No. 9-2333
In the contour enhancement device described in JP-A-43-43, R,
An edge component is calculated for each of the G and B color signals, and an edge enhancement process is performed by applying the maximum value of the edge signal to the target pixel. Although the change in hue is reduced by this technique, the above-described color noise and color steps at block boundaries are directly emphasized, so that the image quality still deteriorates.

In an image processing apparatus described in, for example, JP-A-4-14383, a filter process is performed based on at least one image signal among image signals input for each color to obtain an enhancement amount. I have. As an image signal to be used, an example using, for example, a G color signal or an average value of R, G, and B is shown. According to this technique, the change in hue is reduced as in the above-mentioned document. However, in this technique, when a signal of a specific color such as a G color signal is used, the color noise and the color step of the specific color are still emphasized. Also, when the average value of R, G, and B is used, if the signal value of one or more components increases due to color noise, the average value also increases, and as a result, color noise and the like are also emphasized.

[0009]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and it is possible to suppress the emphasis of unnecessary portions such as color noise and color steps, and to perform good edge emphasis processing as a whole. It is an object of the present invention to provide an image processing apparatus capable of performing such processing.

[0010]

According to the present invention, there is provided an image processing apparatus for performing an edge enhancement process on a color image signal composed of a plurality of input color component signals, wherein an enhancement amount is calculated for each color component signal. Amount calculation means, minimum value selection means for selecting an enhancement amount having a minimum absolute value from the enhancement amounts of the respective color components, and enhancement for applying the enhancement amount selected by the minimum value selection means to each color component signal It is characterized by having an amount application means. In this way, by applying the smallest absolute value of the emphasis amounts of the respective color components to the respective color component signals, the color noise in which the specific color component becomes stronger is hardly emphasized. Also, with respect to the color steps occurring at the block boundaries when the block-coded image is expanded, all the color components do not greatly differ and are hardly emphasized. On the other hand, with respect to black characters, black line drawings, and the like, all color components change greatly, so that the edge emphasis processing is favorably performed.
Therefore, it is possible to solve the problem that the image quality is deteriorated due to the enhancement of the color noise and the color boundary as in the related art, and it is possible to improve the image quality by performing better edge enhancement processing.

Further, since the emphasis amount having the minimum absolute value is applied equally to each color component signal, no hue change occurs, and disadvantages such as color change due to edge emphasis processing and coloring of achromatic portions occur. do not do. Also, since the hue does not change, the contrast, brightness,
It is possible to perform color adjustment such as saturation. It is desirable to perform a smoothing process before the edge enhancement process.

[0012]

FIG. 1 is a block diagram showing an embodiment of an image processing apparatus according to the present invention. In the figure, 1 is a smoothing unit, 2 is a color adjustment unit, and 3 is an edge enhancement unit. The smoothing unit 1 performs a smoothing process on an input image for each color component. The smoothing process can be performed using, for example, a 4-neighbor reference simple average filter having a size of about 3 × 3. Of course, the method of the smoothing process is arbitrary.
This smoothing processing can reduce noise and the like mixed in the input color image.

The color adjustment unit 2 performs color adjustment such as contrast adjustment, saturation adjustment, and brightness adjustment on the input color image (here, the image after the smoothing process). Of course, various adjustments such as hue adjustment can be performed. This color adjustment may be performed in accordance with the color characteristics of the image input device or the image output device in addition to performing the color adjustment according to the user's request.

The edge enhancing section 3 performs an edge enhancing process on an input color image (here, an image after color adjustment). Here, particularly, edges such as black characters and line drawings are emphasized. A black character, a line drawing part, and the like have a high edge part contrast and a sharp image tends to be preferred, and the edge part including the blurred part by the smoothing unit 1 is emphasized.

FIG. 2 is a block diagram showing an example of the edge enhancing section. In the figure, the same parts as those in FIG. 6 are denoted by the same reference numerals. 11 is an enhancement amount calculation unit, 12 is a minimum value selection unit,
Reference numeral 13 denotes an enhancement amount application unit. The enhancement amount calculation unit 11 calculates an enhancement amount for each color component of the color image input to the edge enhancement unit 3. The enhancement amount calculation unit 11 includes a filter processing unit 21
r, 21g, and 21b, an R enhancement amount calculation unit 22r, a G enhancement amount calculation unit 22g, and a B enhancement amount calculation unit 22b.

The filter processing units 21r, 21g and 21b are generally used emphasis filters, and can be constituted by, for example, a Laplacian filter or the like. FIG.
FIG. 3 is an explanatory diagram of an example of a Laplacian filter. In this example, as shown in FIG. 3 (A), four neighboring pixels (a, b, c, d) on the left, right, up and down around the target pixel p are referred to, and the values shown in FIG. Multiplication is performed as a coefficient, and the results of the multiplication are added to obtain a filter result. That is, in the example shown in FIG. 3, a + b + c + d-4
p will be calculated. Of course, the position and the number of the pixels to be referred to and the coefficient value may be determined at the time of design. Further, the filter is not limited to the Laplacian filter. Note that the filter processing unit 21r performs a filter process on the R component of the input color image. Similarly, the filter processing unit 21g performs filter processing on the G component,
The filter processing unit 21b performs filter processing on the B component.

The R emphasis amount calculation section 22r receives the result of the filtering process by the filter processing section 21r and calculates the emphasis amount for the R component. Similarly, the G enhancement amount calculation unit 22g
Receives the filter processing result by the filter processing unit 21g and calculates the enhancement amount in the G component. The B enhancement amount calculation unit 22b receives the result of the filtering process by the filter processing unit 21b, and calculates the enhancement amount of the B component. The method of calculating the enhancement amount from the filter processing result is arbitrary, and for example, an optimally set weighting operation or the like can be performed.

The minimum value selection unit 12 calculates an enhancement amount for each color component by the R enhancement amount calculation unit 22r, the G enhancement amount calculation unit 22g, and the B enhancement amount calculation unit 22b of the enhancement amount calculation unit 11. Select the enhancement amount with the smallest absolute value. And
The selected enhancement amount having the smallest absolute value is output to the enhancement amount application unit 13 as the enhancement amount of each color component.

The enhancement amount application unit 13 applies the enhancement amount having the minimum absolute value selected by the minimum value selection unit 12 to each color component of the color image input to the edge enhancement unit 3. The emphasis amount applying unit 13 is configured by adding units 23r, 23g, and 23b provided for each color component. The adder 23r outputs the R of the input color image.
The absolute value selected by the minimum value selection unit 12 is added to the component and the enhancement amount of the minimum value is added to the R component of the color image signal to be output. Further, the adding unit 23g adds the enhancement amount having the minimum absolute value selected by the minimum value selecting unit 12 to the G component of the input color image to obtain a G component of the color image signal to be output. The addition unit 23b adds the enhancement amount having the minimum absolute value selected by the minimum value selection unit 12 to the B component of the input color image to obtain a B component of the color image signal to be output.

The operation of the edge emphasizing section 3 will be described. The color image input to the edge enhancement unit 3 is input to the enhancement amount application unit 13 and the enhancement amount calculation unit 11
Is input to The enhancement amount calculator 11 performs a filtering process for each of the R, G, and B components to calculate an enhancement amount.
That is, the filtering process is performed on the R component by the filter processing unit 21r, and the enhancement amount for the R component is calculated by the R enhancement amount calculation unit 22r. Similarly, filter processing is performed on the G component by the filter processing unit 21g, and the G enhancement amount calculation unit 2
The enhancement amount for the G color component is calculated using 2 g. In addition, the filter processing unit 21b performs a filter process on the B component,
The B enhancement amount calculator 22b calculates the enhancement amount for the B color component.

The filtering units 21r, 21g, 2
When performing the filter processing in 1b, the value of the reference pixel existing around the target pixel is used in addition to the target pixel to be processed. For example, in the example of the Laplacian filter shown in FIG. 3, reference pixels a, b, c, and d are referred to in addition to the target pixel p shown in FIG. The values of these reference pixels are input to, for example, the enhancement amount calculation unit 11 together with the value of the pixel of interest, or the values of the pixel of interest are input to the enhancement amount application unit 13 while holding a plurality of lines of color images inside. The timing should be adjusted in such a manner.

The enhancement amount for each color component calculated by the enhancement amount calculation unit 11 is input to the minimum value selection unit 12. The minimum value selection unit 12 selects the enhancement amount having the smallest absolute value from the input enhancement amounts of the respective color components, and uses the selected enhancement amount having the minimum absolute value as the enhancement amount of each color component. 13 is output.

In the emphasis amount applying section 13, the minimum value selecting section 12
The emphasis amount having the smallest absolute value selected by the
Adder 23 for each color component of the color image input to
r, 23g, and 23b are added and output.

The above operation will be described using a specific example.
FIG. 4 is an explanatory diagram of a first specific example of the operation of the edge enhancement unit. The specific example illustrated in FIG. 4 illustrates an example in which a color image in which yellow color noise is mixed in a dark image is input. This specific example is the same as that in FIG.

Each color component of the color image shown in FIG. 4A is input to the enhancement amount calculator 11, and the enhancement amount is calculated for each color component. FIG. 4 shows the calculated enhancement amount.
This is shown in FIG. The minimum value selection unit 12 selects the enhancement amount having the smallest absolute value from the enhancement amounts of the respective color components calculated by the enhancement amount calculation unit 11. In this example, since the absolute value of the enhancement amount of the B component is the smallest, the enhancement amount of the B component is selected.
Then, the enhancement amount of the B component is input to the enhancement amount application unit 13 and applied to each of the R, G, and B components. As a result, an edge enhancement processing result as shown in FIG. 4C is obtained.

Comparing the result of the edge enhancement processing shown in FIG. 4C with the result of the edge enhancement processing by the conventional image processing apparatus shown in FIG. 7, for example, the enhancement amount conventionally added to each color component Were different, the hue had changed.
However, in the present invention, since the uniform enhancement amount is added to each color component, the hue hardly changes. Therefore, even if the color adjustment processing is performed by the color adjustment unit 2 before the edge enhancement processing by the edge enhancement unit 3 as shown in FIG. 1, the adjustment processing can be utilized.

Also, the emphasis amount is not so emphasized since the emphasis amount having the smallest absolute value is applied. Therefore,
The color noise as shown in FIG. 4 is not emphasized, and the deterioration of the image quality can be suppressed. If the smoothing process is performed by the smoothing unit 1, the color noise is suppressed to some extent at the time of input to the edge enhancement unit 3, the enhancement amount is reduced, and even if some enhancement process is performed. The effect is small.

Here, the case where color noise is mixed in the input color image has been described. However, as in the case where an image which has been subjected to block coding such as JPEG is decoded, for example, a color difference The same applies to the case where a color image in which is generated is input. The color steps that occur at the boundaries of the blocks also differ in the amount of occurrence of the steps for each color component, and the absolute value of the enhancement amount obtained from the color steps is applied to each color component with the smallest absolute value. Therefore, it is possible to suppress the enlargement of the color step and the deterioration of the image quality.

FIG. 5 is an explanatory diagram of a second specific example of the operation of the edge emphasizing unit. The specific example illustrated in FIG. 5 illustrates an example in which black characters, line drawings, and the like are drawn on a bright background. In the case of black characters and line drawings, as shown in FIG. 5A, the pixel values of all color components change greatly. for that reason,
As shown in FIG. 5B, the emphasis amounts obtained for the respective color components have large absolute values. Therefore,
Even if an absolute value with the smallest absolute value is selected from among these, the absolute value of the applied amount is not so small, but a large amount is applied to all the color components. Edges are emphasized. Therefore, the edge emphasizing process is favorably performed on black characters, line drawings, and the like. Of course, the same applies to a color image in which white characters and line drawings are drawn on a dark background.

In the above description, the color image is R,
Although an example in which three components of G and B are used has been described, the present invention is not limited to this. For example, the present invention can be applied to a color image composed of components such as cyan (C), magenta (M), and yellow (Y).

[0031]

As is apparent from the above description, according to the present invention, it is possible to suppress the emphasis of unnecessary portions such as color noises and color steps with a simple structure, and to perform good edge emphasis processing as a whole. There is an effect that can be.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a block diagram illustrating an example of an edge enhancement unit.

FIG. 3 is an explanatory diagram of an example of a Laplacian filter.

FIG. 4 is an explanatory diagram of a first specific example of the operation of the edge enhancement unit.

FIG. 5 is an explanatory diagram of a second specific example of the operation of the edge enhancement unit.

FIG. 6 is a configuration diagram illustrating an example of a conventional image processing apparatus that performs edge enhancement processing.

FIG. 7 is an explanatory diagram of a problem when color noise is input in a conventional image processing apparatus that performs edge enhancement processing.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Smoothing part, 2 ... Color adjustment part, 3 ... Edge enhancement part, 11
... Enhancement amount calculation unit, 12... Minimum value selection unit, 13, enhancement amount application unit, 21r, 21g, 21b.
r ... R enhancement amount calculation unit, 22g ... G enhancement amount calculation unit, 22b
... B enhancement amount calculation units, 23r, 23g, 23b.

 ──────────────────────────────────────────────────続 き Continued from the front page F term (reference) 5B057 CA01 CA12 CA16 CB01 CB12 CB16 CE03 CE05 CE06 CE17 CH09 CH18 DA17 DB02 DB06 DC16 5C077 LL02 LL19 MP07 MP08 PP02 PP03 PP32 PP37 PP43 PP47 PQ08 PQ12 5C079 HB01 LA14 LA15 LA31

Claims (3)

[Claims] An image processing apparatus for performing an edge enhancement process on a color image composed of a plurality of input color components, comprising: an enhancement amount calculation unit configured to calculate an enhancement amount for each color component; A minimum value selecting unit that selects an enhancement amount having a minimum absolute value; and an enhancement amount applying unit that applies the enhancement amount selected by the minimum value selection unit to each color component. Image processing device. 2. The image processing apparatus according to claim 1, wherein the color image signal is a color image signal that has been subjected to a smoothing process. 3. The image processing apparatus according to claim 1, wherein the color image signal is a color image signal that has been subjected to a color adjustment process.