JPH05298442A - Picture outline emphasizing device - Google Patents

Abstract

PURPOSE:To prevent the color of a picture from getting out of balance at the time of the emphasis of an outline in the case that outline emphasis processing is executed about the picture represented by three primary colors R, G, B. CONSTITUTION:In an outline emphasizing device for the picture whose color pixel is represented by the three primary colors of red, green, and blue, an HSI converting means 2 to convert the picture whose each pixel is the three primary colors into hue (H), saturation (S), and lightness (I), an outline emphasis processing means 3 to outline-emphasis-process only the lightness (I), and a three primary colors converting means 4 to obtain the three primary colors from the hue and the satuation obtained from the HSI converting means are provided, and the outline emphasis is executed by the lightness.

Description

Detailed Description of the Invention

[0001]

The present invention relates to three primary colors R (red) G.
The present invention relates to a contour enhancement device that does not disturb the color balance of an image when performing contour enhancement processing on pixels represented by (green) B (blue).

When an original image expressed in RGB is input to the image input device, deterioration due to the image input device occurs. Therefore, in particular, in order to compensate for the deterioration of the contour, an image contour enhancement process for enhancing the contour is generally performed.

[0003]

2. Description of the Related Art A conventional outline emphasis processing apparatus for an original image represented by RGB will be described with reference to FIG. In FIG. 6, 1
Reference numeral 0 is an RGB original image holding unit that holds an original image expressed in RGB, 11 is a contour emphasis filter, and 12 is an RGB contour emphasis image holding unit that holds a contour-enhanced RGB image.

An image represented by R, G, and B as shown in FIG. 6 is held in the original image holding section 10. The contour emphasis spatial filter 11 has a spatial filter coefficient as shown by in FIG. 6, whereby R ′, B ′, G ′ as shown in the calculation of the pixel value after the contour emphasis in FIG. R ′, G ′, and B ′ of the calculation result performed are the RGB contour emphasized image holding unit 1
Held at 2.

When the above calculation result is negative, it is calculated as zero. That is, as shown in FIG.
When (i, j) <0, R ′ (i, j) = 0 G ′ (i, j) <0, G ′ (i, j) = 0 B ′ (i, j) <0 In this case, B '(i, j) = 0.

[0006]

By the way, in the conventional contour enhancement processing as shown in FIG. 6, the RGB components of a pixel are made only from the components of the same kind in the surrounding pixels (R is only R, G is only G). , B only B) are affected, the ratio of RGB in the pixel collapses, resulting in a problem that a color not present in the original image appears.

When the contour of an image input by RGB is emphasized, the contour is generally emphasized for each of the R, G, and B planes, as shown in FIG. Pixels of a color that is not in the original image frequently appear in a monotonous area where the color change of adjacent pixels is small.

For example, when the R, G, and B pixels shown in FIG. 7 are subjected to contour enhancement processing by the spatial filter as described above, the central pixel is (50, 50, 50) to (58, 42, 58). Change. That is, first R, G,
What was in a gray state with B being balanced is broken in color balance and changed to a color in which R and B are emphasized.

Therefore, as the contour is emphasized many times, pixels having a color which is not in the original image are further emphasized, and as a result, these pixels appear as noise in the image.
Therefore, there is a need to provide a contour enhancement device that prevents the appearance of pixels with colors that are not present in the original image.

[0010]

Therefore, according to the present invention,
As shown in FIG. 1, the RGB original image is HSI-converted by the HSI conversion unit 2 into each pixel having a hue (H), a saturation (S), and a lightness (I) represented by Munsell's color system. Then, for the lightness (I) only, the spatial filter conversion for enhancing the contour is performed by the contour enhancing spatial filter 3, and the H, S
And the brightness (I ′) calculated by the spatial filter
RGB conversion is performed by the B conversion unit 4 to obtain RGB of three primary colors.

[0011]

Operation When the RGB original image output from the RGB original image holding unit 1 is converted into the HSI intermediate image, I has a value representing the lightness. Since the lightness I does not include information on the ratio (hue) of RGB, only the lightness I is edge-emphasized, and the lightness I is converted to RGB again to change the hue and vividness of the color. There is no such thing. In this way, it is possible to perform the edge enhancement processing in which pixels having a color that does not appear in the original image do not appear.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIGS. FIG. 2 is a block diagram of an embodiment of the present invention, and FIG.
FIG. 4 is an explanatory diagram of the SI conversion unit, FIG. 4 is an explanatory diagram of the RGB conversion unit thereof, and FIG. 5 is an explanatory diagram of a contour emphasis state in the present invention.

In the drawings, the same symbol parts as in the other drawings indicate the same parts,
Reference numeral 1 is an RGB original image holding unit, 2 is an HSI conversion unit, 3 is a contour emphasis spatial filter, 4 is an RGB conversion unit, 5 is an HSI image holding unit, and 6 is an RGB contour emphasis image holding unit.

The RGB original image holding unit 1 temporarily holds the original images of the three primary colors to be subjected to contour enhancement processing. The HSI conversion unit 2 converts a primary image of three primary colors into a hue (H), a saturation (S), and a lightness (I) represented by Munsell's color system. The HSI conversion unit 2 performs the operations (1) to (3) shown in FIG. 3 according to R, G, and B, and outputs H, S, and I. The H, S, and I are held in the HSI image holding unit 5, for example.

The contour enhancement spatial filter 3 outputs I'which is obtained by performing contour enhancement calculation on I output from the HSI conversion unit 2. For example, the spatial filter as shown in FIG. An operation such as I ′ shown in FIG. 2 is performed by the coefficient and the result is transmitted to the RGB converter 4.

The RGB converter 4 calculates R, G, B from H ', S', I '. As shown in FIG. 2, H', S'from the HSI converter 2 is calculated. The output H and S are input, and for I ', I'which has been subjected to the edge enhancement processing in the edge enhancement spatial filter 3 is input. And shown in FIG. 4 according to these values ~
Is calculated.

Note that H, S, and I in FIG. 4 correspond to H ', S', and I'in FIG. Then, the conversion operation of these FIG. 3 and FIG.
It is based on the definition of Haydn, which is one of the known conversions, and the respective parameter values of RGB space and HSI space are 0 ≦ R, G, B ≦ 255,
It is assumed that the standardization is 0 ≦ H, S, and I ≦ 255.

Therefore, when conversion is performed by another conversion method, the operations of FIGS. 3 and 4 are changed accordingly. The operation of the present invention will be described.

(1) First, in the RGB original image holding unit 1 of FIG.
For example, an original image before contour enhancement as shown in FIG. 5A is input. Then, the HSI conversion unit 2 performs the calculation shown in FIG. 3 to perform the HSI conversion as shown in FIG. Then, the HSI-converted intermediate image is held by the HSI image holding unit 5.

(2) The brightness I of the intermediate image held by the HSI image holding unit 5 is defined by the contour emphasis spatial filter 3
Then, the contour emphasis calculation is performed by the spatial filter coefficient as shown in FIG. Then, I'of the calculation result is RG
It is sent to the B conversion unit 4. At this time, the hue H obtained by HSI conversion from the HSI image holding unit 5 by the HSI conversion unit 2 is performed.
And the saturation S as H ′ and S ′, respectively, are sent to the RGB converter 4. As a result, H ', S', and I'shown in FIG. 2 are sent to the RGB conversion unit 4, and the HSI shown in FIG.
The transformed data is contour-enhanced by I in FIG. 5 (G).
It is transmitted as shown in.

(3) Then, the RGB conversion section 4 performs the calculation shown in FIG. 4 on these, and sends RGB to the RGB contour emphasized image holding section 6. At this time, the data in which the contour of I is emphasized as shown in FIG. 5C is as shown in FIG. 5D.

As is clear from this, the central pixels of the original image shown in FIG.
0, 50), and R, G, and B are 1: 1: 1, respectively.
However, as shown in FIG. 5D, the central pixel subjected to the edge enhancement processing according to the present invention is (58, 5).
8, 58), and the ratio of R, G, B is also 1:
It remains 1: 1 and it is clear that the contour is emphasized without changing the hue.

[0023]

According to the present invention, it is possible to prevent the deterioration of the color tone of an image due to the edge enhancement and to prevent the generation of a pixel having a color tone which is not present in the original image. Even if contour enhancement is performed a plurality of times in order to obtain a stronger contour enhancement effect, the hue is not deteriorated at all, and a high-quality contour enhancement effect can be obtained.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a configuration diagram of an embodiment of the present invention.

FIG. 3 is an explanatory diagram of an HSI conversion unit.

FIG. 4 is an explanatory diagram of an RGB conversion unit.

FIG. 5 is an explanatory diagram of a contour emphasis state according to the present invention.

FIG. 6 is a conventional example.

FIG. 7 is a diagram illustrating a problem of the conventional example.

[Explanation of symbols]

 1 RGB original image storage unit 2 HSI conversion unit 3 Edge enhancement spatial filter 4 RGB conversion unit 5 HSI image storage unit 6 RGB outline enhancement image storage unit

Claims (1)

[Claims] 1. A contour enhancement device for an image in which color pixels of a pixel are represented by three primary colors of red, green, and blue. The image of the three primary colors of each pixel is hue (H), saturation (S),
HSI conversion means (2) for converting to lightness (I), contour emphasis processing means (3) for performing contour emphasis processing only on the lightness (I), lightness subjected to the contour emphasis processing, and the HSI conversion means An image contour emphasizing apparatus comprising: three primary color conversion means (4) for obtaining three primary colors from the obtained hue and saturation.