JPH08234723A - Image processing device and method - Google Patents

Abstract

PURPOSE: To restore and form multi-valued dither images, fixed threshold images and their mingled images from preserved data by preserving the many valued dither images and fixed threshold images formed from multi-gradation images with a small memory capacity. CONSTITUTION: A gradation conversion section 20 forms first limited gradation images which are not the multi-valued dither images by the number of the gradations limited from the multiple gradation images stored in a memory 10 and second limited gradation images which are the multi-valued dither images. A bit plane data forming section 30 for synthesized images develops the first and second limited gradation images on the bit planes and stores a plurality of the first limited gradation images by every bit plane into a memory 40 and stores one sheet of the synthesized bit plane formed by synthesizing the bit planes of the first and second limited gradation images into a memory 50. An image display section 60 restores the first and second limited gradation images from the bit planes of the first limited gradation images stored in the memory 40 and the synthesized bit plane stored in the memory 50.

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, and more particularly, to an image processing apparatus used for switching an image to be displayed in an interactive image communication system such as an image database and an image therefor. It relates to a processing method.

[0002]

2. Description of the Related Art Presently, personal computers are often used for document editing, and a binary image of black and white or several colors is sufficient for displaying a document image. Therefore, as a data format of an image, a bit plane format in which 1 bit is extracted from a bit string representing gradation is often used. If you have the image data as a bit plane, the binary image is 1
An image represented by one bit plane and having gradation of m bits (the number of gradations n = 2 to the m-th power) is represented by m bit planes. Further, an m-bit color image of R (red), G (green), and B (blue) is represented by m bit planes of R, G, and B, respectively. However, with 3 or 4 bit planes, even a grayscale image (black and white image) produces gradations of only 8 or 16 values, which is not sufficient for displaying a natural image.

Therefore, when displaying a natural image in gray scale, a multi-valued dither method is used in which gradation is multiplied by a gradation display capability of a display and a dither method that expresses shading depending on pixel density. The explanation of the dither method including the multi-valued dither method is described in “New Edition of Image Electronics Handbook (edited by The Institute of Image Electronics Engineers of Japan, Corona Publishing Co., Ltd.), 5. Dither Expression of Image”, but here, the understanding of the present invention is understood. In order to facilitate the above, the threshold property and the generated image of the multi-valued dither method will be described in detail.

Now, consider conversion of a multi-tone image having a tone number N and a normalized density f _xy into a limited tone image having a tone number n and a normalized density g _xy . The number of thresholds for conversion is n−1, and when the target pixel f _xy of the original image satisfies the expression (1), k / (n−1) ≦ f _xy ≦ (k + 1) / (n−1) , (K = 0, 1, 2, ..., N−2) (1) It is appropriate that the conversion pixel g _xy be k / (n−1) or (k + 1) / (n−1). is there. At this time, the threshold T (k) _xy in the pixel of _interest satisfies the expressions (2) and (3), and the conversion is expressed by the expression (4).

K / (n−1) ≦ T (k) _xy ≦ (k + 1) / (n−1) (2) T (k) _xy <T (k + 1) _xy (3) f _xy <T ( k) _xy → g _xy = k / (n-1), T (k) _xy ≤f _xy → g _xy = (k + 1) / (n-1) (4) As the conversion pixel g _xy , the original pixel f _xy Employing the upper bits of is the same as processing with a threshold value. For example, when a 3-bit converted image is created from an 8-bit original image, it is shown in Table 1 in binary notation.

[0006]

[Table 1]

From Table 1, the threshold value of the original image is 00100000.
0 (2), 01000000 (2), 01100000
(2), 10000000 (2), 10100000
(2), 11,000,000 (2), 11100000
There are seven in (2). However, (2) represents binary notation. In general, the number of thresholds when the upper m bits of the original image represented by M bits is the converted image is (2 to the m-th power) -1, and is constant regardless of the position of the image. Below, here,
An image subjected to gradation conversion with a constant threshold will be referred to as a constant threshold image.

On the other hand, in multivalued systematic dither,
The size of the threshold depends on the position of the pixel, and when the length of one side of the dither matrix is D, the threshold is represented by the equation (5).
And (x mod D) and (y mod D). T (k) _xy = Td (k, x mod D, y mod D) (5) where (x mod D) represents the remainder when x is divided by D. At this time, the pseudo gradation number n _eq is represented by the number of thresholds having different sizes, and n _eq = (n-1) D ² +1 (6).

FIGS. 9 and 10 schematically show the threshold processing of the equations (2) and (4) in the figures. FIG. 9 shows a constant threshold and FIG. 10 shows a case of multi-value dither. There is. With the constant threshold value, the value of the converted pixel is determined only by the value of the original pixel as shown in FIG. 9, but with multi-value dither, as shown in FIG.
The value of the converted pixel changes depending on the original pixel and the threshold value of the original pixel, and the converted image expresses pseudo density by the density ratio of the converted pixel. For example, as shown in FIG. 11, when a certain portion of the converted image is made up of pixels of densities 5 and 6,
The density there appears to be pseudo between 5 and 6, and if 6 is large, it looks like a density close to 6, and if 5 is large, it looks like a density close to 5.

The multi-valued error diffusion method can be defined in the same manner. If the threshold for the corrected normalized density _f'xy is T (k) ' _xy = (k + 1/2) / (n-1) (7) ,

[0011]

[Equation 1]

Can be expressed as Where α _ij is a weighting coefficient, E
_{x and y} are errors. Here, the error satisfies the following expression (11). −1/2 (n−1) ≦ Ex _{, y} <1/2 (n−1) (11) Here, when the threshold value for the density f _xy is defined as in Expression (12),

[0013]

[Equation 2]

From equations (7), (8) and (11), the threshold value of the multi-valued error diffusion method also satisfies equation (2), and the threshold value T (k) _xy is
The transformed image is similar to the systematic dither method because it varies spatially due to error propagation. The above is the case of a grayscale image, but to make a color dither image,
The dither method or the multi-valued dither method described above may be used for each of R, G, and B.

[0015]

In an image display device such as a personal computer having a limited memory for images, multi-valued dither is good for expressing a natural image, and a constant threshold is used for expressing a document image. The image is good, but depending on the image, you may not know which one is better, such as when a character is mixed in a natural image. Therefore, as shown in FIG. 12, the gradation conversion unit 2 creates a constant threshold image and a multi-valued dither image from the common multi-gradation image f _xy stored in the memory 1, and sets the constant threshold image and the multi-valued dither image in the bit planes C 1 _xy to C3 _xy and d
To keep decomposes in 1 _xy to D3 _xy in the memory 3 ₁ to 3 ₃ and 41 _{to 3,} which method for synthesizing the image suitable for each parts in the image display section 5 is considered a. However, the bit plane format data storage method requires twice as much memory as one image to store two types of images. Since only one composite image is finally required, it is wasteful to store two types of images in advance.

The present invention has been made in view of the above situation, and stores two limited gradation image data (multi-value dither image, constant threshold image) created from a multi-gradation image with a small memory capacity. It is an object of the present invention to provide an image processing device that creates a multi-valued dither image, a constant threshold image, and a mixed image thereof from stored data at high speed and without color unevenness.

[0017]

In order to solve the above problems, the present invention provides (1) an image which is not a multi-value dither with a limited number of gradations from a multi-gradation image (first limited gradation image). And means for creating a multi-valued dither image (second limited gradation image) with a limited number of gradations; means for developing the first and second limited gradation images on a bit plane; the first
Storing a plurality of limited gradation images for each bit plane
A means for synthesizing the bit planes of the first and second limited gradation images to create one synthetic bit plane; a second memory for storing the synthetic bit plane; Means for restoring the first and second limited gradation images from the bit plane of the first limited gradation image stored in the memory and the combined bit plane stored in the second memory. And an image processing device, further comprising (2) the first and second
In the means for creating a limited gradation image, some of the thresholds for creating the first limited gradation image are equal to the density levels that the limited gradation image can have, and further, (3)
The bit planes of the first and second limited gradation images are gray code bit planes;
(4) The composite bit plane is separable into the bit plane of the second limited gradation image, and further,
(5) It is characterized in that the combined bit plane is an exclusive OR bit plane of the gray code bit planes of the first and second limited gradation images.

In order to solve the above problems, the present invention provides (6) an image which is not a multi-value dither (first limited gradation image) with a limited number of gradations from a multi-tone image and is limited. Multi-valued dither image based on the number of gradations (second limited gradation image)
A means for creating the first and second limited gradation images on a bit plane; a first memory for storing a plurality of the first limited gradation images for each bit plane; An image processing apparatus comprising: means for synthesizing bit planes of the first and second limited gradation images to create one synthetic bit plane; and a second memory for storing the synthetic bit plane, or ( 7) Means for creating a constant threshold image with a limited number of gradations and a multi-valued dither image from the multi-tone image, and a bit plane for each of the constant threshold image and the multi-valued dither image with a limited number of gradations. Means for creating and storing in the memory, and means for creating an image of a limited number of gradations from the bit plane stored in the memory, the means for creating the bit plane and storing in the memory, Limited A means for creating a gray code bit plane from a bit plane of a constant constant threshold image and a multivalued dither image, and an exclusive OR bit plane from a gray code bit plane of a constant threshold image and a multivalued dither image And an image processing apparatus having means for synthesizing a plurality of exclusive OR bit planes into one bit plane, or (8) converting a multi-tone image into a constant threshold image and a multi-value dither image, respectively. Gradation conversion means for developing into a bit plane of the multi-valued dither image; means for generating a gray code bit plane of the constant threshold image from the bit plane; Exclusive OR bit of gray code bit plane of threshold image and gray code bit plane of multi-level dither image Synthetic image bit plane data including an exclusive OR bit plane generation unit for generating a lane and a bit plane synthesizing unit for synthesizing the exclusive OR bit plane by taking the logical sum of the exclusive OR bit planes Generating means; first memory for storing a gray code bit plane of the constant threshold image; and a composite bit plane of a gray code bit plane of the constant threshold image and a gray code bit plane of the multi-level dither image And an image processing apparatus having a second memory for storing the image data and a restoring unit for restoring an image corresponding to the data stored in the first and second memories.

Further, in order to solve the above-mentioned problems, the present invention is limited to (9) means for creating a constant threshold image and a multi-valued dither image with a limited number of gradations from a multi-gradation image. Means for creating each bit plane from the constant threshold image of the number of gradations and the multi-valued dither image and storing it in the memory, and creating and displaying the limited number of gradations image from the bit planes stored in the memory Means for creating and displaying an image with a limited number of gradations from the bit plane data stored in the memory, and the means for displaying the synthesized exclusive OR bit plane is a gray code bit of a constant threshold image. A means for decomposing using the information of the plane, and a gray code bit plane of the multilevel dither image is recovered from the exclusive OR bit plane and the gray code bit plane of the constant threshold image. Means, a means for selecting either a constant threshold image or a multi-valued dither image for each pixel from the gray code bit plane, creating a bit plane format composite image, and outputting the composite image, An image processing apparatus comprising means for accumulating bit planes, means for outputting a color having a density corresponding to the accumulated bit plane image, and means for displaying the output color, or
(10) An image that is not multi-valued dither with a limited number of gradations (first limited gradation image) and a multi-valued dither image with a limited number of gradations (second limited gradation image) from the multi-tone image Are created, the first and second limited gradation images are expanded into bit planes, a plurality of the first limited gradation images are stored in a first memory for each bit plane, and the first and second limited gradation images are stored. The bit planes of the limited gradation image are combined to create one combined bit plane, the combined bit plane is stored in the second memory, and the first limited floor stored in the first memory is stored. It is characterized by an image processing method for restoring the first and second limited gradation images from the bit plane of the toned image and the composite bit plane stored in the second memory.

[0020]

[Function] The multi-tone image is tone-converted into two limited-tone images (constant threshold image, multi-valued dither image) represented by bit planes, and the bits of the gray code representation of the constant threshold image are converted from these bit planes. Store the plane in memory. Here, the threshold value at the time of gradation conversion to the constant threshold image is adjusted to the density level that the limited gradation image can take, and the exclusive OR of the bit planes of the gray code representations of the two limited gradation images is taken to separate them. It is enabled and combined into one bit plane and stored in memory. The image in the designated range is determined to be either a multi-valued dither image or a constant threshold image, is combined in bit plane units, and is displayed as a multi-valued dither image or a constant threshold image.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is an overall configuration diagram for explaining an embodiment of an image processing apparatus according to the present invention.
A memory 10 for storing a multi-tone image and a multi-tone image f _xy connected to the memory 10 and stored in the memory 10.
Is converted into a constant threshold image c _xy and a multi-valued dither image d _xy , and is developed into bit planes c1 _{xy to} c3 _xy , d1 _{xy to} d3 _xy , and is connected to the gradation conversion unit 20, Bits that are the logical OR of the gray plane bit planes cg1 _{xy to} cg3 _xy of the constant threshold image for creating the composite image and the bit plane of the exclusive OR of the multi-value dither and the gray code bit planes of the constant threshold image Plane xgs _xy
And a bit plane cg1 _xy , which is connected to the synthesized image bit plane data generation unit 30 and which is output.
Bit plane memories 40 (41, 42, 43) and 50 that store cg3 _xy and xgs _xy , and a composite image that is connected to these bit plane memories 40 and 50 and that displays an image according to the stored data Display unit 60
It consists of

FIG. 2 is a diagram for explaining the details of the gradation conversion unit 20 shown in FIG. 1. The gradation conversion unit 20 is shown in FIG.
As shown in FIG. 5, the multi-tone image f _xy stored in the memory 10 is read, the constant threshold image c is converted to a constant threshold image c _xy , and the constant threshold image generator 21 is connected to the constant threshold image generator 21. , The constant threshold image c _xy to the bit planes c1 _{xy to} c3 _xy
And a multi-valued dither image generation unit 22 that similarly reads the multi-tone image f _xy from the memory 10 and converts the multi-tone image f _xy into a multi-valued dither image d _xy for output.
And a bit plane development unit 24 which is connected to the multi-valued dither image generation unit 22 and develops the multi-valued dither image d _xy into bit planes d1 _{xy to} d3 _xy and outputs it.

FIG. 3 is a diagram for explaining the details of the composite image bit plane creating unit 30 shown in FIG. 1. The composite image bit plane creating unit 30 has a constant threshold value as shown in FIG. The gray coding unit 31 for converting the bit planes c1 _{xy to} c3 _xy of the image into the bit planes cg1 _{xy to} cg3 _xy of the gray code, and similarly, the bit planes d1 _{xy to} d3 _xy of the multi-value dither image are converted to the multi-value dither image. gray coding section 32 for converting into bit planes d1 _xy to D3 _xy, are connected to these gray coding section 31, the cg1 _xy and dg1 _xy, cg2 _xy and dg2 _xy, c
Exclusive OR of g3 _xy and dg3 _xy xg1 _xy , xg2 _xy ,
An exclusive OR bit plane generation unit 33 that calculates and outputs xg3 _xy , and the exclusive OR bit plane generation unit 33 that is connected to the exclusive OR bit planes xg1 _xy , xg2
and a _xy, XG3 bit plane synthesizing portion 34 for outputting the logical sum XGS _xy of _xy.

Bit plane memory 40 (41, 4)
2, 43) is the gray coding unit 31 as described above.
The bit planes cg1 _{xy to} cg3 _xy output from are stored. The bit plane memory 50 also stores the bit plane xg output from the bit plane combining unit 34.
Store s _xy .

FIG. 4 is a view for explaining the details of the image display section 60 shown in FIG. 1. The image display section 60 is shown in FIG.
, The bit plane memory 40 (41, 4
2, 43) and is connected to the 50 separates the XGS _xy output from the bit plane memory 50 by cg1 _xy ~cg3 _xy output from the bit plane memory 41 to 43, reproduced xg1 _xy ~xg3 _xy And output the bit plane separation unit 61 and the bit plane memory 41.
To 43 and the bit plane separation unit 61, and output from them are cg1 _xy and xg1 _xy , cg2 _xy and xg2.
_{xy, cg3 xy} and xg3 each from _xy dg1 _xy, dg2
A multi-valued dither image bit plane restoration unit 62 for reproducing and outputting _xy , dg3 _xy, and a range for outputting a control signal (pixel selection signal) s _xy for determining whether to use a multi-valued dither image or a constant threshold image for each pixel It is connected to the designation unit 63, the bit plane memories 41 to 43, and the exclusive OR bit plane generation unit 62, and v1 _xy = cg1 _xy or v1 for each pixel by the image selection signal s _xy from the range designation unit 63. _xy =
dg1 _xy ; v2 _xy = cg2 _xy or v2 _xy = dg2 _xy ; v3
_xy = _cg3 select the _xy or _{v3 xy = dg3 xy, v1 xy} , v
Selector 64 that outputs 2 _xy , v3 _xy , and the selector 6
4, VRAMs 65, 66 and 67 for storing v1 _xy , v2 _xy and v3 _xy respectively, and these VRAM6
5, 66, 67 connected to the look-up table 68 to output a color L _xy having a density corresponding to v1 _xy , v2 _xy , v3 _xy;
and a display 69 for displaying _xy .

Next, the operation of the image processing apparatus of the present invention and the method of realizing details will be described by taking the case of converting from N gradations to n = 8 (= 2 ³ ) gradations as an example. First, the internal operation of the gradation conversion unit 20 will be described. The multi-valued dither image generation unit 22 shown in FIG. 2 converts the multi-valued image f _xy stored in the multi-tone image memory 10 into a multi-valued dither image d _xy by the above-described systematic dither method or error diffusion method. Convert. At this time, the threshold value T (k) _xy = Td (k) _xy for gradation conversion _satisfies the conditional expressions (2) and (3) for gradation conversion. The constant threshold image generation unit 21 converts the multivalued image f _xy into a limited gradation image (constant threshold image) c _xy with a constant threshold that does not depend on location. However, here, n−1 of the conversion threshold values T _xy (k) = Tc (k) are made to agree with the normalized density of the converted gradation value as shown in Expression (13). Tc (k) = (k + 1) / (n-1) (k = 0,1, ..., (n-2) / 2-1) Tc (k) = 1/2 (k = (n-2) ) / 2) Tc (k) = k / (n-1) (k = (n-2) / 2 + 1, (n-2) / 2 + 2, ..., n-2) (13) Threshold Tc (k) also satisfies the conditional expressions (2) and (3) for gradation conversion.

By thus matching the threshold value with the converted gradation value, the density of the original image can be specified from the density of the converted image. For example, as shown in FIG. 5, Tc (k) =
When k / (n−1), the normalized density f _xy is normalized so that all original pixels satisfying k / (n−1) ≦ f _xy <(k + 1) / (n−1) (14) The density g _xy becomes a conversion pixel of g _xy = (k + 1) / (n-1) (15). Therefore, it can be seen that if the converted pixel satisfies the expression (15), the original pixel satisfies the expression (14). Due to this property, bit plane xgs
xg1 from _{xy xy,} xg2 _{xy, xg3} can restore the _xy. The specific method of separation will be described later.

Since the maximum values of c _xy and d _xy are n-1 and the normalized densities are c _xy / (n-1) and d _xy / (n-1), respectively, k / (n-1) ≦ f _xy ≦ (k + 1) / (n−1) (16), c _xy = k and d _xy = k or c _xy = k and d _xy = k + 1 or c _xy = k + 1 , And d _xy = k (17). Therefore, | c _xy −d _xy | = 0,1 (18), and the difference between c _xy and d _xy is 0 or 1. Where | x
| Represents the absolute value of x.

The constant-threshold image c _xy and the multi-valued dither image d _xy created from the multi-tone image f _xy by the threshold processing are bit plane expansion (decomposition) units 23 and 24, respectively, and bit planes c1 _{xy to} c3 _xy , It is decomposed into d1 _{xy to} d3 _xy ,
These relationships are expressed by the equations (19) and (20). c _xy = 2 ² · c1 _xy +2 ¹ · c2 _xy +2 ⁰ · c3 _xy (19) d _xy = 2 ² · d1 _xy +2 ¹ · d2 _xy +2 ⁰ · d3 _xy (20)

Next, the operation of the composite image bit plane data creation unit 30 will be described. The Gray coding units 31 and 32 shown in FIG.
From _{xy ~c3 xy, d1 xy ~d3 xy} , and calculates and outputs the bit-plane _{cg1 xy ~cg3 xy, dg1 xy ~dg3} xy Gray code. These relationships are (21), (2
It is expressed by equation (2), and the calculation results are summarized in table 2.

[0031]

(Equation 3)

[0032]

[Table 2]

By the way, the relationship between c _xy and d _xy is (18)
Since the expression is satisfied, when | c _xy −d _xy | = 0 and | c _xy −
Bit plane when d _xy | = 1 cg1 _{xy to} cg
Let's look at the relationship of 3 _xy , dg1 _{xy to} dg3 _xy . | c _xy −
Obviously, d _xy | = 0 only when cg1 _xy = dg1 _xy , cg2 _xy = dg2 _xy , cg3 _xy = dg3 _xy (23).

When a number is represented by a Gray code, the Hamming distance between two numbers having a size difference of 1 has the property that | c _xy −d _xy | = 1 is _satisfied . It is when the following expression (24) is satisfied. _{cg1 xy ≠ dg1 xy, cg2 xy} = dg2 xy, cg3 xy = dg3 xy _{or, cg1 xy = dg1 xy, cg2} xy ≠ dg2 xy, cg3 xy = dg3 xy _{or, cg1 xy = dg1 xy, cg2} xy = dg2 xy, cg3 _xy ≠ dg3 _xy (24) For example, when c _xy = 3, d _xy = 4, as can be seen from Table 2, cg1 _xy = 0, dg1 _xy = 1 and cg2 _xy = d
The equation (24) is satisfied with g2 _xy = 1 and cg3 _xy = dg3 _xy = 0.

The exclusive OR bit plane generator 33 generates xg1 _{xy to} xg3 _xy according to the equation (25).

[0036]

[Equation 4]

From Table 2 and the expressions (24) and (25), the following expression (26) is established. xg1 _xy = 0, xg2 _xy = 0, xg3 _xy = 0 or xg1 _xy = 1 and xg2 _xy = 0, xg3 _xy = 0 or xg1 _xy = 0, xg2 _xy = 1 and xg3 _xy = 0 or xg1 _xy = 0, xg2 _xy = 0, xg3 _xy = 1 (26)

[0038] That is, xg1 _xy, xg2 _xy, xg3 of _xy, bit plane is 1 is one most. 0
If white is represented by 1 and 1 is represented by black, the plane relationship is as shown in FIG. As can be seen in FIG. 6, 1 for each plane
Because the place where is different, even if you take the logical disjunction of the plane,
If you know which plane pixel the pixel of interest originally was,
The composite plane can be restored to the original first to third planes. Further, the relationship between the combination of c _xy and d _{xy and} the plane that can be 1 is summarized in Table 3.

[0039]

[Table 3]

The bit plane synthesizing unit 34 uses xg1 _xy ,
xg2 _xy, XG3 but calculates the XGS _xy superimposed the _xy, XGS _xy is represented by the equation (27). xgs _xy = xg1 _xy + xg2 _xy + xg3 _xy (27) Here, + represents a logical sum. As described above, the operation of each unit of the combined image bit plane data creation unit 30 can be represented by a simple logical operation, and thus can be realized by a simple circuit as shown in FIG. When the data of the constant threshold image and the data of the multi-valued dither image are held in the bit plane format, only the gradation conversion unit 20 has c1 _xy , c2 _xy , c3 _xy , d.
Six bit planes of 1 _xy , d2 _xy , and d3 _xy were required, but the composite image bit plane data creation unit 30 sets the bit planes to cg1 _xy , cg2 _xy , and cg3.
_{It only} needs 4 of _xy and xgs _xy .

Next, the operation of the image display section 60 will be described. In FIG. 4, the bit plane decomposition unit 61 converts xgs _xy into xg1 _xy , xg2 _xy , x according to the following rules.
Decomposes into g3 _xy . From the expressions (13), (16), (17) and (25), the following expression (28) is established. 0 ≤ f _xy <Tc (0). → c _xy = 0, xg3 _xy = 0, 1, xg1 _xy = xg2 _xy = 0. Tc (0) ≤ f _xy <Tc (1). → c _xy = 1, xg2 _xy = 0,1, xg3 _xy = xg1 _xy = 0. Tc (1) ≦ f _xy <Tc (2). → c _xy = 2, xg3 _xy = 0,1, xg1 _xy = xg2 _xy = 0. Tc (2) ≤ f _xy <Tc (4). → c _xy = 3,4, xg1 _xy = 0,1, xg2 _xy = xg3 _xy = 0. Tc (4) ≤ f _xy <Tc (5). → c _xy = 5, xg3 _xy = 0,1, xg1 _xy = xg2 _xy = 0. Tc (5) ≦ f _xy <Tc (6). → c _xy = 6, xg2 _xy = 0,1, xg3 _xy = xg1 _xy = 0. Tc (6) ≦ f _xy <1. → c _xy = 7, xg3 _xy = 0,1, xg1 _xy = xg2 _xy = 0 .... (28)

As shown in equation (13), the threshold T of the constant threshold image
Since c (k) is determined, xg1 _xy , x is obtained by c _xy .
A bit plane that can be 1 can be specified among g2 _xy and xg3 _xy . The relationship between the planes that can be 1 and cg1 _xy , cg2 _xy , and cg3 _xy is as shown in Table 4, so (2
9) is established, and cg1 _xy , cg2 _xy , cg3 _xy and x
from gs _xy, it can be separated _{xg1 xy, xg2 xy, xg3 xy} .

[0043]

(Equation 5)

[0044]

[Table 4]

Multi-level dither image bit plane restoration unit 62
Is cg1 _{xy to} cg3 _xy and xg1 according to the equation (30).
From _{xy ~xg3 xy,} to recover the bit plane dg1 _xy ~dg3 _xy Gray code multivalued dither image.

[0046]

(Equation 6)

As described above, the operation of combining the bit plane decomposition unit 61 and the multi-valued dither image bit plane restoration unit 62 is as follows.
Since the operation is the reverse of the operation of the composite image bit plane data creation unit 30 and can be realized by a logical operation, it can be realized by the circuit of FIG. The selector 64 selects a constant threshold image or a multi-valued dither image according to the value of the image selection signal s _xy from the range designation unit 63. The values of the output bit planes v1 _{xy to} v3 _xy are as shown in equation (31). v1 _xy = cg1 _xy · s _xy + dg1 _xy · s _xy ￣, v2 _xy = cg2 _xy · s _xy + dg2 _xy · s _xy ￣, v3 _xy = cg3 _xy · s _xy + dg3 _xy · s _xy 31 ￣, The created bit planes v1 _{xy to} v3 _xy are stored in the VRAMs 65, 66 and 67, respectively.

The lookup table 68 is the VRAM 6
The color L _xy having the density corresponding to the values of v1 _{xy to} v3 _xy output from 5, 66 and 67 is selected and output. In this case, v
Since 1 _{xy to} v3 _xy represent the Gray code, the relationship with L _xy is shown in Table 5. The display 69 displays the color L _xy output from the look-up table 68.

[0049]

[Table 5]

[0050]

EFFECTS OF THE INVENTION Two limited gradation image data (multi-valued dither image, constant threshold image) created from a multi-tone image are stored with a small memory capacity, and the multi-valued dither image and constant threshold are stored from the stored data. Images and mixed images thereof can be restored at high speed. Since a gray code bit plane is created from the bit planes of the multi-value dither image and the constant threshold image, the bit plane of the exclusive OR of the bit planes of the two images can be combined by OR to form one bit plane, and thus, Image data can be reduced to save memory. Moreover, since these conversions can be realized by a simple circuit, the conversions can be performed at high speed. When converting a multi-gradation image into a constant threshold image, some of the thresholds are equal to the gradation of the constant threshold image. The bit plane of the sum can be restored. Since this restoration can also be realized by a simple circuit, it can be converted at high speed. Since the threshold value at the time of gradation conversion into the constant threshold image is adapted to the density level that the limited gradation image can have, an image without color unevenness can be restored. Converts a multi-tone image into two limited-tone images (constant threshold image, multi-valued dither image) represented by bit planes, and stores the bit plane of the gray code representation of the constant threshold image from these bit planes in a memory. Since it is stored in, it is possible to perform gradation conversion with good correlation. Created from a multi-tone image with less memory, because the exclusive OR of the bit planes of the gray code representations of the two limited tone images is taken to be separable and combined into one bit plane. The two limited gradation images (multi-valued dither image and constant threshold image) that have been created can be saved and can be reproduced at high speed.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of an image processing apparatus of the present invention.

FIG. 2 is a block diagram showing details of a gradation conversion unit of the image processing apparatus of the present invention.

FIG. 3 is a block diagram showing details of a composite image bit plane creation unit of the image processing apparatus of the present invention.

FIG. 4 is a block diagram showing details of an image display unit of the image processing apparatus of the present invention.

FIG. 5 is a diagram for explaining the principle of matching a threshold value with a converted gradation value.

FIG. 6 is a diagram illustrating superposition of bit planes of exclusive OR images.

FIG. 7 is a circuit diagram for realizing a composite image bit plane creation unit.

FIG. 8 is a circuit diagram that realizes a bit plane decomposition unit and a multi-level dither image bit plane restoration unit.

FIG. 9 is a diagram for explaining an example of constant threshold gradation conversion.

FIG. 10 is a diagram for explaining an example of multi-value dither gradation conversion.

FIG. 11 is a diagram showing a part of conversion pixels by multi-value dither.

FIG. 12 is a diagram for explaining that a memory for two images is required in an image processing apparatus that does not have a combined image bit plane creation unit.

[Explanation of symbols]

1 ... Gradation image memory, 2 ... Gradation converter, 3 _{1 to} 3 ₃ , 4 ₁
... 4 ₃ ... Memory, 5 ... Image display unit, 10 ... Multi-gradation image memory, 20 ... Gradation conversion unit, 21 ... Constant threshold image generation unit,
22 ... Multi-valued dither image generation unit, 23, 24 ... Bit plane expansion / decomposition unit, 30 ... Synthetic image bit plane data generation unit, 31, 32 ... Gray coding unit, 33 ... Exclusive OR bit plane generation unit, 34 ... Bit plane synthesis unit, 40 (41, 42, 43), 50 ... Bit plane memory, 60 ... Image display unit, 61 ... Bit plane separation unit, 62 ... Multi-value dither image bit plane restoration unit, 63 ... Range Designating unit, 64 ... Selector, 65, 66,
67 ... VRAM, 68 ... Look-up table, 69 ...
display.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G09G 5/02 9377-5H G09G 5/02 C H04N 1/38 H04N 1/38 7/14 1 / 40 7/18 K 7/14 G06F 15/68 320A 7/18 H04N 1/40 103B

Claims (10)

[Claims] 1. An image which is not multi-valued dither with a limited number of gradations from a multi-tone image (first limited gradation image) and a multi-valued dither image with a limited number of gradations (second limited gradation). An image); a means for developing the first and second limited gradation images on a bit plane; a first memory for storing a plurality of the first limited gradation images for each bit plane; A means for synthesizing the bit planes of the first and second limited gradation images to create one synthetic bit plane; a second memory for storing the synthetic bit plane;
A means for restoring first and second limited gradation images from the bit plane of the first limited gradation image stored in the first memory and the combined bit plane stored in the second memory; An image processing device comprising: 2. In the means for creating the first and second limited gradation images, some of the threshold values when the first limited gradation image is created are set to the density levels that the limited gradation image can have. The image processing apparatus according to claim 1, wherein they are equal. 3. The image processing apparatus according to claim 1, wherein the bit planes of the first and second limited gradation images are gray code bit planes. 4. The image processing apparatus according to claim 1, wherein the composite bit plane is separable into a bit plane of the second limited gradation image. 5. The combination bit plane is an exclusive OR bit plane of gray code bit planes of the first and second limited gradation images, according to any one of claims 1 to 4. The image processing device according to item 1. 6. An image which is not multi-valued dither with a limited number of gradations from a multi-tone image (first limited gradation image) and a multi-valued dither image with a limited number of gradations (second limited gradation). An image); a means for developing the first and second limited gradation images on a bit plane; a first memory for storing a plurality of the first limited gradation images for each bit plane; A means for synthesizing the bit planes of the first and second limited gradation images to create one synthetic bit plane; and a second memory for storing the synthetic bit plane. Image processing device. 7. A means for creating a constant threshold image and a multi-valued dither image with a limited number of gradations from a multi-tone image, and a means for creating a constant threshold image and a multi-valued dither image with a limited number of gradations, respectively. The bit plane is created and stored in the memory, and the bit plane is stored in the memory. The bit plane is created and stored in the memory. The means includes means for creating a gray code bit plane from the bit planes of the constant threshold image and the multi-valued dither image having a limited number of gradations, and an exclusive OR from the gray code bit planes of the constant threshold image and the multi-valued dither image. An image processing apparatus comprising: means for creating a bit plane; and means for synthesizing a plurality of exclusive OR bit planes into one bit plane. 8. A gradation conversion means for converting a multi-gradation image into a constant threshold image and a multi-valued dither image and developing them into respective bit planes; and a gray code bit plane of the constant threshold image from the bit planes. Generating means, means for generating the gray code bit plane of the multi-level dither image, and bit plane of exclusive OR of the gray code bit plane of the constant threshold image and the gray code bit plane of the multi-level dither image Bit plane data generation for a composite image, which includes an exclusive OR bit plane generation unit that generates a logical sum of the exclusive OR bit planes and a bit plane synthesizing unit that synthesizes the exclusive OR bit planes. A first memory for storing a gray code bit plane of the constant threshold image;
A second memory for storing a composite bit plane of a gray code bit plane of the constant threshold image and a gray code bit plane of the multi-valued dither image;
And an image restoration device that restores an image corresponding to the data stored in the second memory. 9. A means for creating a constant threshold image and a multi-valued dither image with a limited number of gradations from a multi-tone image, and a means for creating a constant threshold image and a multi-valued dither image with a limited number of gradations, respectively. It comprises means for creating a bit plane and storing it in a memory, and means for creating and displaying an image with a limited number of gradations from the bit plane stored in the memory. A means for creating and displaying an image with a limited number of gradations is a means for decomposing a synthesized exclusive OR bit plane using information of a gray code bit plane of a constant threshold image, and the exclusive OR. Means for recovering a gray code bit plane of a multi-valued dither image from the bit plane and the gray code bit plane of the constant threshold image; A means for selecting either a constant threshold image or a multi-valued dither image for each pixel to create and output a bit plane format composite image, a means for accumulating a bit plane of the output composite image, and an accumulated bit An image processing apparatus comprising: a unit that outputs a color having a density corresponding to a plain image; and a unit that displays the output color. 10. An image which is not multi-valued dither with a limited number of gradations (first limited gradation image) and a multi-valued dither image with a limited number of gradations (second limited gradation) from the multi-tone image. Image), develops the first and second limited gradation images in a bit plane, stores a plurality of the first limited gradation images in a first memory for each bit plane, The bit planes of the second limited gradation image are combined to create one combined bit plane, the combined bit plane is stored in the second memory, and the first bit stored in the first memory is stored. Bit plane of limited gradation image and the second
From the composite bitplane stored in the memory of
And an image processing method characterized by restoring the second limited gradation image.