JPH05189553A - Image processing system - Google Patents

Abstract

PURPOSE:To process the image at a high speed and to attain the effective development in an image processing system where the images are moved, tilted, reduced, rotated, etc. CONSTITUTION:An original image storing means 1 is provided to store plural picture elements forming an original image together with an original image reading means 2 which reads out the picture elements of the original image, a thinning out pattern storing means 3 which stores a thinning out pattern to decide whether the picture elements of the original image should be thinned out or not in 1:1 correspondence, a pattern reading means 4 which reads the thinning out pattern out of the means 3, a picture element thinning out means 5 which compares the picture elements of the original image with those of the thinning out pattern and then thins out the picture elements from the original image, and a processed image storing means 7 which stores the processing result of the means 5. In such a constitution, the picture elements of the original image data are compared with those of the thinning out pattern and the original image data are thinned out based on the thinning out pattern. Thus the original image can be processed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention cuts out, moves, and moves two-dimensional, three-dimensional or higher multi-dimensional images.
The present invention relates to an image processing method for processing an image such as italicization, reduction, rotation, and deformation.

Two-dimensional and three-dimensional images can be variously processed by changing the arrangement of pixels.
Since the number of pixels that make up a screen is enormous, it is necessary to perform the processing efficiently.

According to the present invention, binary, multi-valued, color still images and moving images can be processed at high speed with a simple structure.

[0004]

2. Description of the Related Art FIG. 7 shows a conventional technique. In the figure,
Reference numeral 50 denotes an original image storage means, which comprises a storage device such as a semiconductor memory and stores pixels of the original image. 5
Reference numeral 1 is a reading means for reading the data stored in the original image storage means 50. Reference numeral 52 denotes a writing unit that stores the pixels read by the reading unit 51 in the processed image storage unit 53. Reference numeral 53 denotes a processed image storage means, which has the same size as the original image storage means 50 and stores pixels after image processing. Reference numeral 54 is a pixel address conversion means for designating an address for reading a pixel from the original image storage means 50,
An address for storing a pixel in the post-processing image storage means 53 determined according to the processing is calculated.

The processed image storage means 53 may be shared by the same storage device as the original image storage means 50. The operation of the configuration shown in the figure will be described.

The pixel address conversion means 54 designates the address of the pixel read from the original image storage means 50. The reading means 51 reads the pixel data of the specified address from the original image storage means 50. 1 by reading once
A pixel or multiple pixels are read.

The pixel address conversion means 54 calculates where to move the read pixel in accordance with processing such as movement and rotation, and obtains an address to write. Then, the converted address is designated in the writing means 52.

The writing means 52 writes the read pixel data at the address designated by the pixel address converting means 54.

[0009]

In the conventional image processing method, it is necessary to prepare different address converting means for various image processing such as image rotation and reduction, and a means for properly using each type is required. Met.

Further, since the conventional address conversion means calculates the pixel position after processing for each pixel, it takes a long time for processing. Also, the device test needs to be performed for each type of image processing, and the development efficiency is poor.

It is an object of the present invention to provide an image processing system which performs image processing at high speed and has a high development efficiency.

[0012]

According to the present invention, a thinning pattern for thinning out pixels of an original image is prepared, the original image is thinned according to the thinning pattern, and the remaining pixels that have not been thinned move toward the origin on one axis. By moving the pixel,
Image processing such as reduction, enlargement, italicization, and rotation was performed.

FIG. 1 shows the basic configuration of the present invention. In the figure, reference numeral 1 is an original image storage means, which is constituted by a semiconductor memory or the like. Reference numeral 2 is an original image reading means, and 3 is a thinning pattern storage means, which is composed of a semiconductor memory and stores pixels of a binary thinning pattern corresponding to pixel positions of the original image. Reference numeral 4 denotes a thinning-out pattern reading means for reading out the pixels stored in the thinning-out pattern storing means 3. The pixels of the thinning pattern are binary, and for example, "0" is not thinned and "1" is thinned. On the contrary, "1" is not thinned out, and "0" is thinned out. A pixel thinning means 5 thins out the original image data based on the original image and the thinning pattern, and performs non-thinning pixel moving processing toward the origin in the uniaxial direction by the thinned pixels. A post-processing image writing means 6 stores the pixel data of the image processing result of the pixel thinning means 5 in the post-processing image storage means 7. Reference numeral 7 denotes a processed image storage means for storing pixels resulting from the processing performed by the pixel thinning means 5.

[0014]

The operation of the basic configuration of FIG. 1 will be described. Hereinafter, it is assumed that the value of the pixel of the thinning pattern (hereinafter referred to as the thinning pixel) is "0", and that the value thereof is "1".

The original image reading means 2 reads pixel data from the original image storage means 1. The thinning-out pattern reading unit 4 reads out thinned-out pixels at positions corresponding to the pixel positions read out from the original image storage unit 1.

The pixel thinning means 5 thins out the pixels of the original image based on the thinned pixels read by the thinning pattern reading means 4, and moves so as to close the thinned pixels at the positions of the thinned pixels.

The processed image writing means 6 stores the moved pixels of the pixel thinning means 5 in the processed image storage means 7. A pixel moving method by the pixel thinning means 5 will be described in the case of a two-dimensional image.

The size of the original image is assumed to be X pixels arranged in the X direction and Y pixels arranged in the Y direction. Similarly, in the thinning pattern, X pixels are arranged in the X direction and Y pixels are arranged in the Y direction. The pixel value of the original image is expressed as follows.

A (n, m) (where 0 <m ≦ X, 0 <
n ≦ Y) and A (n, m) are data of a type corresponding to the type of pixel (binary, shade, color, etc.).

Similarly, the processed pattern is expressed as follows. B (n, m) (where 0 <m ≦ X, 0 <n ≦ Y), B
(N, m) is data of a type corresponding to the type of pixel (binary, shade, color, etc.).

The thinning pattern is expressed as follows. I (n, m) (where 0 <m ≦ X, 0 <n ≦ Y), I
(N, m) is binary data. For example, “0” represents a pixel for thinning out, and 1 represents a pixel for not thinning out.

The relationship between the position of the pixel before the movement and the position after the movement is as follows. B (n, k) = A (n, m) The above is the case of the two-dimensional image, but in the case of the three-dimensional image, the original image A (p, n, m), the processed image B (p,
n (k), B (p, n, k) = A (p, n, m) A general n-dimensional image can be similarly defined.

According to the present invention, by simply changing the value of the thinning pattern I (n, m) in this calculation, the cutting, moving, italic transformation, reduction and the like can be performed by the same calculation method. .. Further, processing such as rotation can be realized by performing processing in the other direction after processing in the X direction.

[0024]

FIG. 2 shows an embodiment of the present invention. In the figure,
Reference numeral 20 is an original image, 21 is a thinning pattern, and 22 is a processed image.

As an example, the figure shows a case where one screen has 8 pixels in the X-axis direction (horizontal direction) and 8 pixels in the Y-axis direction (vertical direction). The thinning-out pattern 21 is an example of thinning out the columns 2, 4, and 7.

The pixels of each row of column 1 of the original image 20 are moved to the respective rows of column 1 of the processed image 22 according to the thinning pattern 21 without changing the address of each pixel. Pixels in each row of column 2, column 4, and column 7 of the original image 20 are thinned out and deleted.

The pixels in each row of column 3 of the original image 20 are moved to the left by one pixel in the X direction in the processed image. (The Y coordinate in the address indicating the pixel position is 2 from the origin.
(K = 1 + 1 = 2 in the previous formula (1)).

The pixels in each row of column 5 of the original image 20 are moved two pixels to the left in the X direction in the processed image. (The Y coordinate in the address indicating the pixel position is 3 counting from the origin.
(In the previous formula (1), k = 1 + 1 + 1 =
3)).

The pixels in each row of column 6 of the original image 20 are moved two pixels to the left in the X direction in the processed image. (The Y coordinate at the address indicating the pixel position is 4 counting from the origin.
(In the previous formula (1), k = 1 + 1 + 1 + 1 =
4)).

The pixels in each row of column 8 of the original image 20 are moved to the left by 3 pixels in the X direction in the processed image. (The Y coordinate at the address indicating the pixel position is 5 counting from the origin.
(In the previous formula (1), k = 1 + 1 + 1 + 1 + 1 + 1
= 5)).

FIG. 3 shows an embodiment (2) of the present invention. In the figure, 30 is an original image, 31 is a thinning pattern, and 32 is a processed image.

Row 1 of the original image 30 has a thinning pattern 31
In accordance with the thinning-out pattern of row 1, the pixels in columns 4 to 8 are moved to the left by 3 pixels and are respectively moved to columns 1 to 5 in row 1 of the processed image 32.

Line 2 of the original image 30 has a thinning pattern 31
Each pixel in columns 4 to 8 is moved to the left by 3 pixels in accordance with the thinning-out pattern of row 2 of the above, and is moved to columns 1 to 5 of row 2 of the processed image 32, respectively.

The rows 3 and 4 of the original image 30 are moved to the left by 2 pixels in the pixels of columns 3 to 7 in accordance with the thinning patterns of the rows 3 and 4 of the thinning pattern 31, respectively, and the rows of the processed image 32 respectively Moved to columns 1-5 of row 3.

Rows 5 and 6 of the original image 30 are moved to the left by one pixel for each pixel of columns 2 to 6 in accordance with the thinning pattern of rows 5 and 6 of the thinning pattern 31, respectively. Moved to columns 1-5 of rows 5 and 6.

Rows 7 and 8 of the original image 30 have pixels of columns 1 to 5 corresponding to rows 7 and 8 of the thinning pattern 31, respectively, and columns 1 to 1 of rows 7 and 8 of the processed image 32, respectively. Moved to 5.

The pixel position after the movement of each pixel in the direction of the origin on the Y-axis (to the left) is expressed by the above formula.
Calculated according to (1). FIG. 4 shows an embodiment of the pixel thinning means of the present invention.

The figure shows pixel thinning means for an 8-dot image. In the figure, x _1, x ₂ , x ₃ , x ₄ , x
₅ , x ₆ , x ₇ , and x ₈ are horizontal row input pixels, y ₁ ,
_{y 2, y 3, y 4} , y 5, y 6, y 7, y 8 represents an output pixel. I ₁ , I ₂ , I ₃ , I ₄ , I ₅ , I ₆ , I ₇ ,
I ₈ are pixels x ₁ , x ₂ , x ₃ , x ₄ , x ₅ , respectively.
It represents the values of the thinned pixels corresponding to x ₆ , x ₇ , and x ₈ .
A, B, C, and D represent thinning circuits for thinning and moving pixels based on the input pixel and thinning pattern, respectively (see FIG. 5). The operation of the configuration shown in the figure will be described later.

Each thinning circuit "A", "B",
I ₇ is input to the thinning circuit “D” of the first stage with respect to “C” and “D”, and the thinning circuit “A” of the second stage,
I ₅ and I ₆ are input to “B” and “C”, respectively, and 3
Each thinning circuit “A”, “B”, “B”, “B”,
I ₃ and I ₄ are input to “C”, respectively, and the thinning circuits “A”, “B”, “B”, “B”, “B”, and
I ₁ and I ₂ are input to “B” and “C”, respectively.

FIG. 5 shows an embodiment of a pixel thinning circuit, which is the pixel thinning circuits A, B, C and D in FIG. In the figure, x _K represents the input of the Kth pixel, and x _{K + 1} ' _,
x _{K + 2} 'represents the data of the K + 1-th and K + 2-th pixel input lines, respectively. P _K = 0 and P _{K +1} = 0 represent inversions of the pixel values I _K and I _{K + 1} of the thinning pattern corresponding to the pixels x _{K and} x _{K + 1} , respectively. Also, P _K = 1 and P _{K +1}
= 1 represents the value "I" (non-inverted) of the pixel of the thinning pattern corresponding to the pixels x _{K and} x _{K + 1} , respectively.

In the following description, the value of the thinned pixel is I
_K = 0 represents thinning-out, and the value I _K = 1 of thinned-out pixels represents non-thinning-out. In (A), I _K = 1
If (not thinning x _K ), output x _K and I _K = 0
If I _{K + 1} = 1 in (thinning x _K ), x _{K + 1} is output, and if I _K = 0 (thinning x _K ), I _{K + 1} = 0, x _{K + 2} Is output.

(B) shows that I _K = 1 and I _{K + 1} = 1 (x _K ,
If x _{K + 1} is not thinned out), output x _K. I _K =
If I _{K + 1} = 1 at 0, x _{K + 1} 'is output, and I _K = 0 (x _K
If I _{K + 1} = 1 in (thinning), x _{K + 2} is output.

(C) is x _K if I _K = 1 and I _{K + 1} = 1
Is output, and otherwise (at least one of I _K and I _{K + 1} is 0), x _{K + 1} 'is output. (D) outputs x _K if I _K = 1 and outputs x _{K + 1} 'if I _K = 0.

FIG. 6 is a diagram for explaining the operation of the thinning means of FIG. The drawing is for the case of 4 pixels. In the figure, (A), (B), (C) and (D) are the same thinning circuits as in FIG. The operation will be described for the case of the thinning pattern 49 shown in FIG.
Although the circuit up to the stage, each input is x ₁ , x ₂ , x ₃ ,
x ₄ , I ₁ , I ₂ , and the outputs are y ₁ , y ₂ , y ₃ , y _4. )

The thinning circuit (D) has input pixels x _3, x ₄
And inversion of I _3, I ₃ are inputted as shown, the output of the AND circuit 35 is x _3, and the output of the OR circuit 37 becomes x _3.

The output of the OR circuit (D), x ₄ and the inversion of I _1, I ₂ , I ₁ , and the logical value based on the inversion of I ₂ are input to the decimation circuit (C) as shown. As a result, the output of the AND circuit 39 becomes x ₄ , the output of the OR circuit 40 becomes x ₄ , and y ₄ = x ₄ .

To the decimation circuit (B), x ₂ , the output of the decimation circuit (D), the inversion of x ₄ , I ₁ , I ₂ , I _{1 and} the inversion of I ₂ are input. Then, the output of the AND circuit 42 becomes x ₃ , and the OR circuit 44 outputs x ₃ (y ₂ = x ₃ ).

The decimation circuit (A) inputs x ₁ , x ₂ , the output of the decimation circuit (D), the inversion of I ₁ , I ₂ , I ₁ , and the logical value based on the inversion of I ₂ as shown in the figure. To be done. The output of the AND circuit 46 becomes x ₂ and the OR circuit 48
To output x ₂ (y ₁ = x ₂ ).

By the above operation, the input pixel columns (x ₁ , x
₂ , x ₃ , x ₄ ), output pixel columns (x ₁ , x ₂ ,
x ₃ , x ₄ ) is obtained.

In this embodiment, x at the right end
₄ (x _{8 in} FIG. 4) is output as it is as y ₄ (y _{8 in} FIG. 4), but in the subsequent circuit, only the pixels that have moved without being thinned out from the left of the output are taken out.
It doesn't matter. If you want to extract the previous pixel without thinning out,
The y ₄ (y _{8 in} FIG. 4) should be taken out. Similarly, x ₄ (x _{8 in} FIG. 4) is written in the position while left-shifting, but in the subsequent circuit, only the pixels that have moved to the left are taken into consideration in consideration of the number of thinned pixels, which is a problem. Absent.

[0051]

According to the present invention, since pixels are thinned out in accordance with the thinning pattern and a simple movement process on one axis is performed, image processing can be performed at high speed. In addition, since various types of image processing can be easily performed only by changing the thinning pattern, the device cost can be reduced and the device test can be easily performed.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of the present invention.

FIG. 2 is a diagram showing an embodiment (1) of the present invention.

FIG. 3 is a diagram showing an embodiment (2) of the present invention.

FIG. 4 is a diagram showing an embodiment (reduction) of pixel thinning means of the present invention.

FIG. 5 is a diagram showing an embodiment (reduction) of a pixel thinning circuit of the present invention.

FIG. 6 is a diagram showing an operation explanatory view (in the case of 4 pixels) of the pixel thinning means of the present invention.

FIG. 7 is a diagram showing a conventional technique.

[Explanation of symbols]

 1: Original image storage means 2: Original image reading means 3: Thinning pattern storage means 4: Thinning pattern reading means 5: Pixel thinning means 6: Post-processing image writing means 7: Post-processing image storage means

Claims (3)

[Claims] 1. An original image storage means (1) for storing a plurality of pixels forming an original image, an original image reading means (2) for reading out pixels of the original image, and one-to-one correspondence with the pixels of the original image. Decimation pattern storage means for storing a decimation pattern for determining whether or not to decimate the pixels of the original image (3)
A thinning pattern reading means (4) for reading the thinning pattern from the thinning pattern storage means (3), a pixel thinning means (5) for thinning pixels from the original image by comparing the pixels of the original image with the pixels of the thinning pattern, A post-processing image storage means (7) for storing the processing result of the pixel thinning means (5),
An image processing method characterized by processing the original image by comparing the pixels of the original image data with the pixels of the thinning pattern and thinning the original image data according to the thinning pattern. 2. The pixel thinning means (5) according to claim 1.
Is an image processing method characterized by moving pixels that are not thinned out toward the origin on one axis according to the number of thinned pixels, and moving, reducing, or italicizing the original image. 3. The method according to claim 1 or 2, wherein, when one direction of the original image is the X direction and the other direction is the Y direction, the original image is processed in the X direction, and the result of the processing in the X direction is Y. An image processing method characterized by performing rotation processing by performing processing processing on a direction.