JPH05328141A - Color picture coding decoding method - Google Patents

Abstract

PURPOSE:To obtain the coding and decoding method capable of highly efficient compression coding while ensuring the compatibility with a conventional monochromatic facsimile equipment by dividing a color picture having each of color signals RGB into a luminance signal and color difference signals and coding them respectively. CONSTITUTION:An input picture 100 is converted into a luminance signal 102a and color difference signals 102b, 102c by a color conversion processing section 102. The luminance signal 102a is binarized by a luminance binarizing processing section 103 and coded by a luminance coding processing section 105. The color difference signals 102b, 102c are subjected to ternary processing by a ternary processing section 104 and coded by a color difference coding processing section 106. Since the RGB signals are coded after being converted into the color difference signals and the luminance signal, highly efficient coding is implemented. Furthermore, since the luminance signal is independently extracted, a normal monochromatic picture is obtained even by a conventional facsimile equipment through the use of the luminance signal.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION The present invention encodes a color image.
The present invention relates to a decoding method. In particular, the present invention relates to a method for achieving highly efficient encoding by utilizing the correlation between planes having red (R), blue (B), and green (G) color components.

[0002]

2. Description of the Related Art An example of a conventional color image encoding / decoding method is shown, for example, in "Study of Color Image Encoding Method for Color Facsimile" of the 1991 Conference of the Institute of Image Electronics of Japan. This method relates to a compression coding method of a color image using arithmetic coding. This method is shown in FIG.
This will be described with reference to FIGS. 6 and 7.

FIG. 5 is a block diagram showing the structure of an image coding / decoding device to which this conventional coding / decoding method is applied. FIG. 6 shows a plane in the conventional coding / decoding method. It is explanatory drawing which shows a mode that the correlation between is utilized. It is an explanatory view showing a state of a reference pixel template (described later) particularly in each plane. FIG. 7 is a diagram showing a compression rate of image data when a conventional encoding method is used.

In FIG. 5, the binarization processing unit 50 is R,
The input image 1 composed of G and B color components (hereinafter referred to as planes) is binarized for each plane, and a binary signal 50a is generated.
To occur. The coding prediction processing unit 51 uses the binary signal 50.
The predicted state 51a of the coded pixel and the coded pixel value 51b are generated by using the correlation between the planes of a.

In arithmetic coding, coding is performed by predicting the state of a coded pixel from the value of a reference pixel composed of a plurality of pixels around a pixel to be coded (hereinafter referred to as a coded pixel). Is done. The reference pixel template represents the positions and weights of the plurality of pixels.

The prediction state 51a is a state of coded pixels predicted with a certain probability based on the pixel pattern in the reference pixel template. The coding processing unit 52 generates a code 52a based on the predicted state 51a of the coded pixel and the (actual) coded pixel value 51b. This encoding processing unit performs so-called arithmetic encoding. Then, the decryption processing unit 5
3 decodes the output image 53a from the code 52a.

FIG. 6 is an explanatory diagram for explaining the reference pixel template. R plane reference pixel template 5
Reference numeral 5 is a reference pixel template used when encoding the encoded pixel 55a, and the G plane reference pixel template 56 is a reference pixel template for encoding the encoding target pixel 56a. The B plane reference pixel template 57 is a reference pixel template for encoding the encoding target pixel 57a.

The detailed operation will be described below. R,
The input image 1 composed of the G and B planes is binarized by the binarization processing unit 50 for each plane using the average error minimum method (for example). As a result,
The binarization processing unit 50 binarizes the signal of each plane to generate a binary signal 50a.

The coding prediction processing unit 51 uses the binary signal 50a.
Based on the prediction state 51 of the encoded pixel for each plane.
a and the encoded pixel value 51b are generated. Encoding processing unit 5
In 2, the arithmetic coding is performed as described above.

First, in encoding the R pixel 55a,
Pixels around the R pixel 55a in the R plane are referred to as the reference pixel template 55, and the prediction state 51a and the encoded pixel value 51b are generated. In encoding the G pixel 56a, the pixels around the G pixel 56a in the G plane and the R pixel 55a at the same position as the G pixel 56a are referred to as a reference pixel template. Then, in encoding the B pixel 57a, the B pixel 57 in the B plane is
Pixels around a and the G pixel 56a and the R pixel 55a at the same position as the B pixel 57a are referred to as a reference template. As described above, while referring to each reference pixel template for each plane, the prediction state 51a,
Arithmetic encoding is performed by predicting the value of the encoded pixel value 51b and its appearance probability. ..

FIG. 7 shows the result of encoding six SCID images (Standard Color Image Data) provided by the Image Processing Technology Standardization Committee whose secretariat is the Japan Standards Association using this method. Is shown in. As shown in FIG. 7, the average compression rate is 1.86 bits / pixel. In addition, in the figure, “pel” represents “pixel”.

[0012]

The conventional image coding method is configured as described above.
When binarized and coded using the average error minimum method, the correlation between each plane is not completely removed, and the correlation still remains between the codes of each generated plane.
Therefore, high compression rate encoding cannot be performed.

When the destination of the color image is a conventional monochrome facsimile, the conventional monochrome facsimile selects one plane from each of the R, G, B planes to be transmitted, and selects only that one plane. The method of outputting (on paper) was adopted. For this reason, there is a problem that the colored characters are lost depending on the selected plane. That is, the conventional color facsimile has a problem that it is not compatible with the monochrome facsimile and coexistence with the monochrome facsimile is difficult.

The present invention has been made in order to solve the above problems, and a color signal is divided into a luminance signal and a color difference signal and coded to remove the correlation between the planes and to obtain a highly efficient code. The purpose is to obtain a method that can be realized.

It is another object of the present invention to obtain a method capable of achieving compatibility with a conventional monochrome facsimile by binarizing and encoding a luminance signal.

[0016]

In order to solve the above-mentioned problems, the first invention is a method for encoding a color image composed of R (red), G (green) and B (blue) color signals. There
A conversion step of converting the color image into a brightness signal and a color difference signal; a brightness binary conversion step of converting the brightness signal into a binary value to generate a brightness binary value signal; and a color difference signal into a ternary value, A color difference ternary conversion step of generating a color difference ternary signal, and an encoding step of encoding the luminance binary signal and the color difference ternary signal, respectively.
It is a color image coding method characterized by performing the coding after removing the correlation between each color signal of (red), G (green), and B (blue).

Therefore, since the coding is performed after the correlation is removed, highly efficient compression coding can be performed.

In order to solve the above problems, the second aspect of the present invention is the image encoding method according to the first aspect of the present invention, wherein the color difference ternary conversion step includes pixels surrounding a pixel to be encoded. Error mean value calculation step of calculating the weighted error mean value of the quantization error, and subtracting the weighted error mean value calculated in the error mean value calculation step from the color difference signal of the pixel to be encoded. Then, the subtraction step of calculating the subtraction color difference signal and the subtraction color difference signal calculated by the subtraction step are ternarized using two predetermined threshold values.
A color image coding method including a binarizing step and diffusing a quantization error in the ternarization to surrounding pixels.

Therefore, it is possible to diffuse the quantization error of each pixel to the pixels around that pixel.

In order to solve the above problems, a third aspect of the present invention is an image decoding method for decoding the code obtained by the color image encoding method according to the first and second aspects of the present invention. A decoding process of decoding the code to calculate the luminance binary signal and the color difference ternary signal, and R of the original color image from the luminance binary signal and the color difference ternary signal.
The color inverse conversion step of restoring the respective color signals of (red), G (green), and B (blue), and the respective color signals restored in the color inverse transformation step are normalized within the range of values that can be taken by the original color signals. And an overflow diffusion step for converting the overflow diffusion step, wherein the overflow diffusion step includes a normalization error average value calculation step of calculating a weighting error average value of normalization errors of pixels around a pixel to be decoded, The weighted error average value calculated in the normalization error average value calculation step is subtracted from each color signal of the pixel that is the object of decoding, and R
The normalized subtraction step of calculating the subtracted color signals of (red), G (green), and B (blue) and the subtracted color signals calculated by the normalized subtraction step are divided into two predetermined thresholds. Values A and B
If each of the subtracted color signals is greater than or equal to the threshold value A, then A is calculated, and if each of the subtracted color signals is less than the threshold value B, then B is calculated. If the subtracted color signal is less than the threshold value A and is equal to or more than the threshold value B, a normalization step of directly calculating each subtracted color signal is included and restored in the color inverse conversion step. The color image decoding method is characterized in that the overflow value of each of the color signals is diffused to surrounding pixels.

Therefore, the overflow value generated when the original color signals are restored can be diffused to the surrounding pixels from the luminance binary signal and the color difference ternary signal.

[0022]

According to the first aspect of the present invention, since each color signal is converted into the luminance signal and the color difference signal, the correlation between the signals can be reduced. Further, since the luminance signal is taken out independently, even in the conventional black facsimile apparatus, a normal monochrome image can be obtained by using this luminance signal.

According to the second aspect of the present invention, when the color difference signal is ternarized, the quantization error is diffused to the surrounding pixels, so that more natural quantization is performed.

According to the third aspect of the present invention, it is possible to decode the code by the encoding method according to the first and second aspects of the present invention. Further, according to the overflow diffusion step, it is possible to normalize each color signal in which an overflow has occurred and diffuse the overflow value to surrounding pixels, so that it is possible to restore a more accurate color image. is there.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the functions of an apparatus for realizing the image coding / decoding method of this embodiment. In FIG. 1, a color image 100, which is an input image composed of signals of RGB planes, is converted by a color conversion processing unit 102 into a luminance signal 102a and color difference signals 102b and 102c.

The feature of this embodiment is that the signals of the RGB planes are not encoded as they are, but are once converted into a luminance signal and a color difference signal and then encoded. It is to be decrypted.

The brightness binarization processing unit 103 binarizes the brightness signal 2a and generates a brightness binary signal 3a. The color difference ternarization processing unit 104 ternarizes the color difference signals 102b and 102c,
Color difference ternary signals 104a and 104b are generated. Since the luminance signal has meaning only in its absolute value (there is no such thing as negative luminance), binarization is performed, but the chrominance signal takes positive and negative values, and is therefore ternarized.

The luminance encoding processing unit 105 encodes the luminance binary signal 103a and generates a luminance code 105a. The color difference encoding processing unit 106 encodes the color difference signals 104a and 104b to generate color difference codes 106a and 106b.

The coding process is performed by the coding processing unit having the above-mentioned configuration.

On the other hand, the luminance signal decoding processing unit 107 decodes the luminance code 105a and generates a luminance binary signal 107a. The color difference signal decoding processing unit 108 uses the color difference code 106a.
.106b is decoded and color difference ternary signals 108a and 108b are decoded.
To occur. The color inverse conversion processing unit 109 uses the luminance binary signal 10
7a and the color difference ternary signals 108a and 108b.
G. The B signal 109a is generated.

In the present embodiment, after the signals of each plane are converted into the luminance / color difference signals and coded / decoded in this way, the color inverse conversion processing unit 109 converts each original plane into each plane. The R, G, and B color signals are restored.
Therefore, since encoding and decoding are performed only on the converted luminance / color difference signal, almost no correlation remains in the generated code. Therefore, it is possible to generate a highly efficient code.

Further, in this embodiment, the overflow diffusion processing unit 110 is configured to restore the restored R.D. G. B signal 1
The overflow value of 09a is diffused to the surrounding pixels to restore the color. G. The B signal 110a is created. The overflow diffusion processing unit 110 uses a ternary color difference signal (+
When 1, 0, -1) is inversely converted by the color reverse conversion processing unit 109 into a signal of each plane of RGB, the value may overflow, but a function of diffusing this overflow to surrounding pixels Have.

FIG. 2 is an explanatory diagram showing weighting coefficients used in error diffusion in encoding or in diffusion of overflow. In error diffusion or overflow diffusion at the time of encoding, an error from surrounding pixels is added to encoded pixels with a predetermined weighting, or an overflow value is diffused to surrounding pixels. The same weighting is used as the predetermined weighting in these cases.
This weighting, or coefficient, is the coefficient shown in FIG.

FIG. 3 is a conversion table for converting the ternary color difference signal into a binary signal.

Next, an outline of the whole operation will be described.

The input image 100 is converted by the color conversion processing unit 102 into a luminance signal 102a and color difference signals 102b and 102c by the conversion formulas shown below.

Luminance signal 102a = 0.299 * R1 + 0.587 * G1 + 0.114 * B1 ... (1) Color difference signal 102b = 0.701 * R1-0.587 * G1-0.114 * B1 ... (2) Color difference signal 102c = -0.299 * R1-0.587 * G1 + 0.886 * B1 (3) The above conversion formula is CCIR Rec. It is a conversion formula based on 601-2.

The luminance signal 102a is supplied to the luminance binarization processing unit 1
At 03, the multilevel signal is converted into a binary signal. Although there are various binarization methods, in the present embodiment, the binarization is performed by using the minimum average error method, and the luminance binary signal 1
03a is generated.

The color difference signals 102b and 102c are once normalized in the color difference ternarization processing unit 104, and I _xy (-1≤I _xy ≤)
1). 3 for this normalized color difference signal I _xy
Value conversion is performed. Color difference ternary signal 104 finally obtained
When a is P _xy , this P _xy is calculated for the color difference signal I ′ _xy in which error signals from surrounding pixels are added. That is, I ′ _xy is a value obtained by adding the weighted average value A _xy of the errors in the peripheral pixels to the above I _xy . _I'xy is represented by the following equation.

I ′ _xy = I _xy + A _xy (4) Then, the average value A _xy is expressed by the following equation.

A _xy = Σα _ij · E _{x + i} , _{y + j} / Σα _ij (5) Here, α _ij is the coefficient shown in FIG. 2, and i, j
Is chosen to refer to all the pixels of the reference pixel template shown in FIG. The I ′ _xy thus calculated is ternarized with two threshold values (= P _xy , and the values are “−1”, “0”, and “1”), and the color difference ternary signal 1
04a is generated. The color difference signal 2c is similarly calculated, and the color difference ternary signal 104b is generated. Even in the pixel for which P _xy is calculated in this way, the error E
_xy occurs. This error is expressed by the following equation.

E _xy = I ′ _xy −P _xy (6) The error E _xy obtained by this equation is used as described above when encoding other encoded pixels.

The luminance binary signal 103a is subjected to prediction processing by the luminance coding processing unit 105 and then arithmetically coded based on the prediction. As a result, the luminance code 105a is generated.

The color difference ternary signal 104a is first assigned a binary signal in the color difference signal encoding processing unit 106. This binary signal allocation is performed based on FIG. That is, when "1" is input as the color difference ternary signal,
When "01" is output as the binary value and "-1" is input, "10" is output as the binary value. The obtained binary signal is expanded into MSB and LSB, that is, two planes of the upper digit and the lower digit, and after prediction processing is performed for each plane, arithmetic coding is performed and a color difference code 106a is generated. The color difference ternary signal 104b is similarly processed, and the color difference code 106b is generated.

In this embodiment, the color difference signals are encoded in this way. The signals of the RGB planes before being converted into the color difference signals have a strong correlation with each other, but the color difference signals generally have a smaller correlation than the signals of the RGB planes. Therefore, it is possible to reduce the amount of codes generated by performing arithmetic coding after conversion into color difference signals.

The luminance signal decoding processing unit 107 decodes the luminance code 105a and generates a luminance binary signal 107a.
The color difference signal decoding processing unit 108 decodes the color difference code 106a and generates a color difference ternary signal 108a. The luminance signal decoding processing unit 107 also performs the same processing on the color difference code 106b to generate a color difference ternary signal 108b.

The color inverse conversion processing unit 109 uses the luminance binary code 107a and the color difference ternary signals 108a and 108b
Based on the following equation, R. G. The B signal 109a is generated.

R1a = color difference ternary signal 108a + brightness binary signal 107a (7) B1a = color difference ternary signal 108b + brightness binary signal 107a (8) G1a = −0.299 / 0.587 * Color difference ternary signal 108a −0.114 / 0.587 * Color difference ternary signal 108b + Luminance binary signal 107a (9) Finally, in the overflow diffusion processing unit 110,
As described above, the overflow is diffused. That is, the restored R. G. The B signal 109a is −1 ≦ R.
G. Values in the range of B signal ≦ 2 can be taken. Therefore, in order to reproduce the correct color, those values are set to 0 ≦ R. G. A process for suppressing the B signal within the range of 1 is required. Such processing is performed in the overflow diffusion processing unit 110.

R. G. The B signal 109a is set to I _xy (values are “−1”, “0”, “1”, and “2”),
The case of conversion into P _xy (having 0 ≦ P _xy ≦ 1 as a value) will be described. That is, R. G. Since the B signal 109a may take "-1" or "2" as a value, the overflowed amount is diffused to peripheral pixels.

The method is basically the same as the above-mentioned minimum mean error method. P _xy is the final output, as in the method described above, are calculated error weighted average A _xy of the pixels around the pixel to be _I'xy obtained by adding the I _xy. That, _I'xy is represented by the following equation.

I ′ _xy = I _xy + A _xy (10) Here, the weighted average A _xy is expressed by the following equation.

A _xy = Σα _ij · E _{x + i} , _{y + j} / Σα _ij (11) Here, α _ij is the coefficient shown in FIG. 2, and i, j
Is chosen to refer to all the pixels of the reference pixel template shown in FIG. When I ′ _xy thus calculated exceeds “1”, P _xy = “1”
'P _xy = "0" if _xy is less than "0", 0 ≦ _I'xy ≦
In the case of 1, P _xy = I ′ _xy . As a result, the overflow value can be diffused to the surroundings, and R. G. B
The signal 110a can be obtained.

As described above, according to this embodiment, R
Since a color image composed of a GB signal is converted into a luminance signal and a color difference signal and arithmetic coding is performed on each of them, more efficient coding can be performed. FIG. 4 shows the compression ratio of each of the six SCID images encoded according to the present embodiment. As shown in FIG. 4, as described above, according to the present embodiment, the average compression ratio can achieve 1.46 bits / pixel, and thus the coding efficiency is higher than that of the related art. Can be done.

Further, according to the encoding performed in this embodiment, since the luminance signal is encoded independently, when this code is received by the monochrome facsimile, the image is obtained using only the luminance signal. It is possible. In this case, a natural black-and-white image can be obtained, and the problem of disappearing colored characters does not occur as compared with the conventional technique in which only an image for a plane of one color can be obtained. As described above, according to the present embodiment, since the conventional monochrome facsimile has compatibility, it is possible to coexist with the conventional model.

[0056]

As described above, according to the present invention, the color image (R, G, B) is divided into the luminance signal and the color difference signal, thereby reducing the correlation existing between the RGB color planes. be able to. Therefore, if the luminance signal and the color difference signal are encoded, it is possible to perform encoding with a higher compression rate.

Further, the luminance signal can be received by the conventional monochrome facsimile without loss of color characters, etc., by encoding and transmitting by the conventional monochrome facsimile encoding method. That is, there is an effect that a color facsimile apparatus capable of maintaining compatibility with a conventional monochrome facsimile machine can be obtained.

[Brief description of drawings]

FIG. 1 is a configuration block diagram showing functions of an apparatus using a preferred embodiment of an image encoding / decoding method of the present invention.

FIG. 2 is an explanatory diagram showing a weighting coefficient of a quantization error of a peripheral pixel and an overflow value in the device shown in FIG.

FIG. 3 is an explanatory diagram showing how a binary signal is assigned to a ternarized color difference signal in the apparatus shown in FIG.

FIG. 4 is an explanatory diagram showing an image compression rate in the apparatus shown in FIG.

FIG. 5 is a configuration block diagram showing functions of an apparatus using a conventional image encoding / decoding method.

FIG. 6 is an explanatory diagram showing how the correlation between planes in the device shown in FIG. 5 is used.

FIG. 7 is an explanatory diagram showing an image compression rate in the apparatus shown in FIG.

[Explanation of symbols]

 102 color conversion processing unit 103 luminance binarization processing unit 104 color difference ternary processing unit 105 luminance encoding processing unit 106 color difference signal encoding processing unit 107 luminance signal decoding processing unit 108 color difference signal decoding processing unit 109 inverse color conversion Processing unit 110 Overflow diffusion processing unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Koho Kawabata, 1-1 1-1 Ofuna, Kamakura-shi, Kanagawa Mitsubishi Electric Corporation Communication Systems Laboratory (72) Inventor Fumitaka Ono 5-1-1, Ofuna, Kamakura-shi, Kanagawa No. Mitsubishi Electric Corporation Communication Systems Laboratory

Claims (3)

[Claims] 1. A method of encoding a color image composed of R (red), G (green), and B (blue) color signals, comprising a conversion step of converting the color image into a luminance signal and a color difference signal. A brightness binary conversion step of converting the brightness signal into a binary value to generate a brightness binary signal; a color difference ternary conversion step of converting the color difference signal into a ternary value to generate a color difference ternary signal; An encoding process for encoding the luminance binary signal and the color difference ternary signal, respectively, and R (red), G (green), B of the color image.
A color image coding method characterized by performing the coding after removing the correlation between the (blue) color signals. 2. The color image encoding method according to claim 1, wherein the color difference ternary conversion step calculates a weighted error average value of quantization errors of pixels around a pixel to be encoded. An error average value calculation step, a subtraction step of subtracting the weighted error average value calculated in the error average value calculation step from the color difference signal of the pixel to be encoded, and calculating a subtraction color difference signal, A ternarization step of ternarizing the subtracted color difference signal calculated in the subtraction step using two predetermined threshold values, and a quantization error in ternarization to surrounding pixels. A color image coding method characterized by spreading. 3. An image decoding method for decoding a code obtained by the color image coding method according to claim 1, wherein the code is decoded, and the luminance binary signal and the color difference are obtained. A decoding step of calculating a ternary signal, and a color for restoring each color signal of R (red), G (green), and B (blue) of the original color image from the luminance binary signal and the color difference ternary signal. Inverse conversion step, the color signal restored in the color inverse conversion step,
An overflow diffusion step of normalizing to a range of possible values of each original color signal, and the overflow diffusion step, wherein the overflow diffusion step is to calculate a weighted error average value of normalization errors of pixels around a pixel to be decoded. The normalization error average value calculating step of calculating and the weighting error average value calculated in the normalization error average value calculating step are subtracted from each color signal of the pixel to be decoded, and R (red ), G (green), B (blue)
Of the subtracted color signals, and the subtracted color signals calculated by the normalized subtraction step, using the two predetermined threshold values A and B, Is greater than or equal to the threshold A, then
A is calculated, and if each of the subtracted color signals is less than the threshold value B,
B, and if each subtracted color signal is less than the threshold value A and is greater than or equal to the threshold value B, a normalization step of directly calculating each subtracted color signal is included. Of the respective color signals restored in the inverse conversion step,
A color image decoding method, characterized in that an overflow value is diffused to surrounding pixels.