JPH09214752A - Image processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To process an image at a high speed even by the combination processing of the expansion processing and the compression processing and to obtain the reduction result with less deterioration in image quality by conducting the processing combining expansion processing and reduction processing under a specific condition. SOLUTION: A change point coordinate string after reduction in response to a reduction magnification is obtained from change point coordinate strings to white level picture elements and black level picture elements in the main scanning direction of image data. As to a change point coordinate in the obtained change point coordinate string, a distance of each with a just preceding change point coordinate is obtained. Then each change point coordinate of the change point coordinate string is used for a change point coordinate object to a white level picture element when the obtained distance is 1 or over and the change point coordinate is a change point coordinate to the white level picture element, and when the distance is 0, the change point coordinate object to the white level picture element corresponding to the just preceding change point coordinate is invalidated. When the obtained distance is 1 or over and the change point object to the white level picture element is in existence and the change point is a change point coordinate to the black level picture element, the change point coordinate is used for a change point coordinate to the black picture element and the change point coordinate object to the white level picture element is used for the change point coordinate to the white level picture element.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method for performing processing such as image reduction in a system that handles binary images by software.

[0002]

2. Description of the Related Art Conventionally, in a system in which dedicated hardware for image processing is not provided, the function of image processing is realized by software. These
In the system, when the reduced image data obtained by reducing the original image is generated from the compressed data obtained by compressing the image data, the image editing process in which the decompression process and the reduction process are combined is necessary.

That is, the image data of the entire image is expanded from the compressed data in the working memory by the decompression process, and then the image data expanded in the working memory is reduced by the reduction process according to a predetermined reduction ratio to obtain a reduced image. Generating data.

Further, in the conventional reduction processing, an OR processing method is adopted for image data including characters and figures in order to prevent disappearance and blurring of lines forming characters and figures.

For example, as shown in FIG. 18, in the OR reduction method for obtaining 7-dot reduced image data for 22-dot image data (input image),
A carry pattern is generated using a DDA (Differential Digital Analyzer), the pixels from the pixel on the right of the carry generation position to the next carry are ORed, and the number of carry pixels (7) is output. To do.

Here, the operation principle of the DDA will be briefly described. FIG. 19 shows a process of generating a carry pattern of DDA. When the magnification is N / M (N <M), the initial value and scale value are obtained as follows. However, div indicates integer division, and mod indicates the remainder of integer division.

Scale value = (10000H × N) div M Initial value = (10000H × N) mod M In an example of reducing a 22-dot image to a 7-dot image, scale value = 5174H, according to the above calculation.
The initial value is 0008H.

In the DDA, as shown in FIG. 19, first, the initial value 0008H is set to the current value, and the scale value 5
174H is added. The bit width of the adder is 16 bits, and the scale value is added to the output of the adder. Hereinafter, the scale value is sequentially added to the output of the adder until the number of data is reached, and the carry pattern generated during this time is used to determine the output pixel. That is, the range of pixels to be the target of the OR operation is obtained. As a result, as shown in FIG. 18, all pixels are black in the result of the OR reduction according to the carry occurrence position with respect to the input image.

[0009]

As described above, in the conventional image processing method, when the image is edited by combining the decompression process and the reduction process in order to generate the reduced image data based on the compressed data, the decompression time is increased. And reduction processing must be executed sequentially. Therefore, it takes a decompression time and a demagnification processing time to obtain a final reduced image, resulting in poor work responsiveness.

Further, in the conventional reduction processing, by adopting the OR processing method, it is possible to prevent disappearance or blurring of lines forming a character or a figure. Occasionally, a blackout occurs in a portion, and the image quality is significantly deteriorated.

The present invention has been made in consideration of the above-mentioned circumstances, and can provide a reduction result in which image processing can be performed at high speed even with a combination of decompression processing and reduction processing and image quality deterioration is small. It is an object to provide a possible image processing method.

[0012]

According to the present invention, a change point coordinate sequence after reduction according to a reduction magnification is obtained for a change point coordinate sequence of white pixels and black pixels in the main scanning direction of image data. Regarding the change point coordinates in the obtained change point coordinate sequence, the distances from the immediately preceding change point coordinates are obtained, and the respective change point coordinates in the change point coordinate sequence are obtained, and the obtained distance is 1 or more, and white. If it is the change point coordinate to the pixel, it is set as the change point candidate to the white pixel, and if the obtained distance is 0, the change point candidate to the white pixel corresponding to the immediately previous change point coordinate is invalidated and obtained. When the determined distance is 1 or more, there is a change point candidate for the white pixel, and the change point coordinate is for the black pixel, the change point coordinate for the black pixel is set and the change point for the white pixel is set. The candidate is the coordinates of the change point to the white pixel,
When the obtained distance exceeds 1, the change point candidate for the white pixel does not exist, and the change point coordinates are for the black pixel,
It is characterized in that it is set as a change point to a black pixel, and a change point to a white pixel is obtained by adding a predetermined coordinate value to the coordinates of the change point immediately before.

Further, according to the present invention, a first change point table composed of a change point coordinate sequence of white pixels and black pixels in the main scanning direction of image data, and a line next to the first change table in the sub scanning direction. With respect to the second change point table consisting of the change point coordinate sequence, the change point coordinates having a small change point position to the black pixel in any of the change point tables are set as the noticeable change point and the change point to the converted black pixel As the coordinates, the change point to the first white pixel on the right side of the noticed change point is the white change point candidate, and the change point coordinate value is the white change point in the change point table on the opposite side of the change point table with the noticed change point. Change points that are more than the candidate are detected, and if the detected change point is at the same coordinate position as the white change point candidate and the color is the change point to the white pixel, or at the coordinate position that exceeds the white change point candidate. Yes, the point where the color changes to a black pixel If there is, the white change point candidate is set as the change point coordinates to the converted white pixel, the detected change point is at the same coordinate position as the white change point candidate, and the color of the change point is the change point to the black pixel. If this is the change point of interest, the change point of interest is the change point to the first white pixel on the right of the change point, and the detected change point is at a coordinate position that exceeds the white change point candidate. When the color of the change point is the change point to the white pixel, the position of the white change point candidate is set as the noticed change point, the detected change point is set as the white change point candidate, and the noticed change point or the white change point candidate is set as the line. It is determined whether it is the end, and if it is the end of the line, the white change point candidate is used as the change point coordinates to the converted white pixel.

Further, according to the present invention, a quotient and a remainder value obtained by dividing a change point coordinate sequentially read from a change point coordinate sequence to a white pixel and a black pixel in the main scanning direction of image data by a predetermined value are obtained. , The difference between the obtained quotient and the quotient obtained for the coordinates of the immediately preceding change point, and when the obtained difference value is 0,
The image pattern for the change point coordinate sequence that does not reach the predetermined value is obtained according to the remainder value, and when the obtained difference value is greater than 1, the image pattern obtained for the change point coordinate sequence that does not reach the predetermined value is obtained. An image pattern of the conversion result, and image data in which white pixels or black pixels corresponding to the obtained difference are continuous is output as the image pattern of the conversion result.

Further, according to the present invention, in an image processing method for generating an image subjected to reduction processing based on compressed data obtained by encoding image data, coordinates of change points from compressed data to white pixels and black pixels. A column is generated, and an image reduction method based on the generated change point coordinate sequence is used to generate a change point coordinate sequence that is reduced in the sub-scanning direction, and an image is reduced based on the change point coordinate sequence. A change point coordinate sequence reduced in the main scanning direction is generated by the method, and image data is generated by the data conversion method based on the generated change point coordinate sequence.
The image data after reduction is generated from the compressed data by executing for the screen.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a system that realizes the image processing method according to this embodiment. As shown in Fig. 1, this system uses CP
The U10, the RAM 12, the VRAM 14, the ROM 16, the input device 18, the input control unit 20, the display device 22, the display control unit 24, the storage device 26, and the storage control unit 28.

The CPU 10 controls the entire apparatus and realizes the function of the image processing method according to the present invention by accessing the RAM 12, the VRAM 14 and the ROM 16 and starting a program according to an input instruction.

The RAM 12 is a program area 12 in which various processing programs for determining the operation of the CPU 10 are stored.
In addition to a, a work area 12b as a work memory for storing various variables and tables used when performing image processing, a change point coordinate data to be processed, and the like are provided.

The VRAM 14 is an area for expanding display data to be displayed on the display device 22. VR
The display data expanded in the AM 14 is read by the display control unit 24 and provided for display on the display device 22.

The ROM 16 is a program area 16 in which various processing programs that determine the operation of the CPU 10 are stored.
a, a data area 16b for storing various data, and the like are provided.

The input device 18 is composed of a keyboard, a mouse or the like, and inputs an instruction to the system. The display device 22 is composed of, for example, a liquid crystal display, and performs display under the control of the display control unit 24. The display control unit 24 controls the display on the display device 22, reads the display data stored in the VRAM 14, and provides the display data on the display device 22.

The storage device 26 stores compressed data obtained by encoding the image data to be processed, change point coordinate data, image data and the like. Storage control unit 28
Controls input / output of data and the like to / from the storage device 26.

Next, a data reduction method in the image processing method according to the first embodiment of the present invention will be described. 2 and 3 are flowcharts for explaining the data reduction method according to the first embodiment. In the data reduction method of the first embodiment, the reduction processing is performed in the main scanning direction based on the coordinate information of the change points of white and black in the main scanning direction of the image data.

Step A1 is an initialization process which is performed before starting the process of the first line of the image data, and initializes the variable for setting the reduction ratio (SCALEX) and the change point coordinates after the scaling. It is a step.

Step A2 is an initialization process for retrieving the changing point after reduction, and the changing point coordinates (R
EG_SCALE_POS) and a change point detection flag, that is, a flag (BLACK_FOUND) indicating that a change point from a white pixel to a black pixel has been detected, and a flag indicating that a change point from a black pixel to a white pixel has been detected ( WHITE_FOUND) is invalidated.

In step A3, the change point coordinates (CRNT_POS) sequentially read in the raster scanning direction are multiplied by the reduction ratio (SCALEX), and the resulting value is used as the current change point calculation value (SCALE_POS). .

In step A4, the difference (DIST_P) between the current change point calculation value (SCALE_POS) and the change point coordinate (REG_SCALE_POS) detected immediately before is calculated.
OS).

Step A5 is a step of judging whether or not the difference (DIST_POS) obtained in step A4 is larger than zero. Step A6 is a step of determining whether or not the color (COLOR) of the current change point calculation value is the change point to the black pixel.

Step A7 is a step of determining whether or not the change point detection flag (WHITE_FOUND) for white pixels is valid. Step A8 is the difference (DIS
This is a step of determining whether or not (T_POS) is greater than 1.

Step A9 is a step when the change point from the white pixel to the black pixel is detected.
That is, the current change point calculation result (SCALE_PO
S) is the coordinates of the change point from white to black (AVL_SCALE_
POSB), and the change point coordinates (REG_
Change SCALE_POS from black to white change point coordinates (AV
L_SCALE_POSW) is executed. In addition, each change point detection flag (WHITE_FOUND), (BLAC) indicating that a change point has been detected.
K_FOUND) is enabled (TRUE).

Step A10 is a process of detecting a change point to a black pixel and restoring the change point to a white pixel which has disappeared in the immediately previous reduction calculation process. The current change point calculation result (SCALE_POS) is changed from white to white. Change point coordinates to black (A
VL_SCALE_POSB), and the change point coordinates (REG_SCALE_POS) +1 determined immediately before are determined as the change point coordinates (AVL_SCALE_POSW) to the white pixel. Also, each change point detection flag (WHITE_FOUND) indicating that a change point has been detected,
Make (BLACK_FOUND) valid (TRUE).

Step A11 is a step when a change point candidate from a black pixel to a white pixel is detected, and the current change point calculation result (SCALE_POS) is used as the change point coordinates (REG_SCALE_POS) determined immediately before.
Also, a change point detection flag (WHITE_FO) for white pixels
UND) is enabled, and the process returns to step A3 to detect the change point to the next black pixel.

Step A12 is a step of invalidating the white change point detection result, and the white change point detection flag (WHITE_FOUND) is invalidated (FALSE). Step A13 is a step of detecting the end of the line.

Step A14 is a step when the end of the line is reached during the search for the change point, and the change point coordinate (AVL_SCALE_POSW) to the white pixel is placed on the right side of the reduced line end coordinate.

By repeating the processing from step A2 to step A14 until the line of one screen is completed, the change point coordinates after image reduction for one screen are obtained.

Next, the operation of the data reduction processing in the first embodiment will be described using a specific example. FIG.
Is the change point coordinates (AVL_SCALE_POS) after reduction with respect to the change point coordinate (CRNT_POS) sequence used as input data when the reduction ratio (SCALEX) is set to 2/3.
B) shows a process of calculating (AVL_SCALE_POSW).

Target of image reduction processing (input data)
Is the change point coordinates (CRNT_POS) shown in FIG.
Columns, that is, compressed data of the image pattern shown in FIG. The image pattern represents a bit pattern indicated by the change point coordinates (coordinate positions 0, 1, ...).

The change point coordinates (CRNT_POS) start from the change point coordinates to the black pixel, are followed by the change point coordinates from the black pixel to the white pixel, and are alternately arranged until the end of the line.
However, the line end value of the change point coordinates (CRNT_POS) is the value of the virtual change point provided on the right side of the coordinates of the line end of the image pattern.

First, the change point coordinates (CRNT--
POS) is the reduction ratio (2 /
3) and multiply by the current change point calculation value (SCALE_
POS). (Step A3). For example, CRN
For T_POS = 3, COLOR = black change point coordinates, current change point calculation value (SCALE_POS) =
2 is required (see FIG. 4 (c)).

Next, in step A4, the difference (DI) between the current change point calculation value (SCALE_POS) and the change point coordinate (REG_SCALE_POS) detected immediately before is calculated.
ST_POS). For example, the current change point calculation value is 4 (second of SCALE_POS, COLOR
= (White), the change point coordinate detected immediately before is 2, so the difference (DIST_POS) is 2.

Next, the current change point calculation value (SCAL
E_POS) and the change point coordinates (REG_S) detected immediately before.
It is determined whether the difference (DIST_POS) from CALE_POS) is greater than 0 (step A5). FIG.
When the second (SCALE_POS) in (c) is targeted, DIST_POS> 0.

In this case, it is determined whether or not the color (COLOR) of the current change point calculation value is black (step A6). In this example, since COLOR = white, the process proceeds to step A11.

At step A11, the change point (SCALE
_POS) as the change point candidate from the black pixel to the white pixel, the change point coordinates (REG_SCALE_PO) detected immediately before.
S) and change point detection flag (WHITE_FOU)
Enable ND). At this point, it is a candidate for a change point, and is determined by the next change point calculation result.

If the processing has not reached the end of the line, the process returns to step A3 to detect the change point to the next black pixel (step A13). In this example, the current change point calculation value (SCALE_POS) for the next change point coordinate (CRNT_POS) = 9 is 6, DIST_POS> 0 and COLOR = black, and a change check is performed by processing the previous change point coordinate. Since the output flag (WHITE_FOUND) is valid (steps A5, A6, A7), step A9 is executed.

Here, in step A9, since the change point from the white pixel to the black pixel is detected, the change point detection processing is executed. That is, the current change point calculation result (SCA
LE_POS) is the coordinates of the change point (AVL_S) from white to black
CALE_POSB), and the change point coordinates (REG_SCALE_POS) from the black pixel to the white pixel detected immediately before, which are set in step A11, are the change point coordinates (AVL_SCALE_POSW) from black to white.
Confirm as. Furthermore, each change point detection flag (WHI
TE_FOUND) (BLACK_FOUND) is also enabled (TRUE).

Next, the case where the difference (DIST_POS) = 0 between the current change point calculation value (SCALE_POS) obtained in step A4 and the change point coordinate (REG_SCALE_POS) detected immediately before is described.

In the example shown in FIG. 4B, the change point coordinates (C
When RNT_POS) changes from 12 (fourth) to 13 (fifth), as shown in FIG.
Both _POS values become 8, and the condition of DIST_POS = 0 is satisfied.

When the change point coordinate (CRNT_POS) is 13, DIST_POS> 0, that is, DIST_P
It is determined that OS = 0. Here, WHIT
E_FOUND = TRUE and COLO
Since R = black, the change point to white detected immediately before is invalidated (step A12).

By this processing, the difference (DIST_P) between the current change point calculation value (SCALE_POS) and the change point coordinate (REG_SCALE_POS) detected immediately before is calculated.
OS) = 0, that is, white at the same pixel position,
When the calculation result indicating both black is obtained, the black pixel is preferentially left in the reduced image pattern.

Next, the operation when the end of the line is reached during the search for the change point will be described. In the example of FIG. 4, when the value of the change point coordinate (CRNT_POS) is 22, DIST_P
Since OS> 0 and COLOR = white (steps A5, A
6) The process (step A11) when the change point candidate to the white pixel is detected is executed.

At this time, the change point coordinates (CRNT_POS)
Is the line end (step A13), the line end processing is executed (step A14) without proceeding to the processing for searching for the change point to the next black pixel. That is, the white change point coordinates (AVL_SCALE_POSW) are placed to the right of the reduced line end coordinates.

Thus, the change point coordinates (C
RNT_POS) change point coordinates after reduction (change point coordinates from white to black (AVL_SCALE_POSB)
And black to white change point coordinates (AVL_SCALE_PO
SW)) is obtained as shown in FIG. An image pattern corresponding to the reduced change point coordinate sequence shown in FIG. 4D is as shown in FIG.

Next, with reference to FIG. 5, a description will be given of the process of restoring the change point to the lost white pixel in the reduction calculation process of step A10. FIG. 5 shows the reduction ratio (SCALE
Change point coordinates (AVL_SCALE_POSB) (AVL_SCAL) after reduction with respect to the change point coordinate (CRNT_POS) sequence as input data when (X) is ⅓
It shows a process of calculating E_POSW).

When the reduction ratio is set to 1/3,
When the process is sequentially performed on the change point coordinate (CRNT_POS) column shown in (b) as described with reference to FIG. 4, the change point coordinate (CRNT_POS) = 12 is as follows. Processing is performed.

First, change point coordinates (CRNT_POS) =
12 is multiplied by the reduction ratio (1/3), and the current change point calculation value (SCALE_POS) = 4 (COLOR
= White) is calculated (step A3).

In this case, the current change point calculation value (SC
ALE_POS) = 4 and change point coordinates (R
EG_SCALE_POS) = 3 difference (DIST_P
Since OS) = 1, the process proceeds to step A11 by steps A5 and A6.

At step A11, the change point (SCALE
_POS) = 4 as a change point candidate from a black pixel to a white pixel, the change point coordinates (REG_SCALE_P) detected immediately before.
Saved in OS), change point detection flag (WHITE_FO
UND) is enabled.

Next, change point coordinates (CRNT_POS) =
In the process for 13, the current change point calculation value (SC
ALE_POS) = 4 (COLOR = black). Therefore, at step A4, the difference (DIS) between the current change point calculation value (SCALE_POS) = 4 and the change point coordinate (REG_SCALE_POS) = 4 detected immediately before is calculated.
Since T_POS) = 0, the change point detection flag (WHITE_FOUND) validated in the previous step A11 is invalidated in step A12.

The change point coordinates (CRNT_POS) =
Also in the process for 14, the current change point calculation value (SCALE_POS) = 4 (COLOR = white), and thus, in the same manner as the process for the change point coordinate (CRNT_POS) = 13, the current change point calculation value is calculated in step A4. The difference (DIST_POS) = 0 between (SCALE_POS) = 4 and the change point coordinates (REG_SCALE_POS) = 4 detected immediately before is (DIST_POS) = 0.

Next, change point coordinates (CRNT_POS) =
In the process for 18, the current change point calculation value (SC
ALE_POS) = 6 (COLOR = black). Therefore, in step A4, the difference (DIS) between the current change point calculation value (SCALE_POS) = 6 and the change point coordinate (REG_SCALE_POS) = 4 detected immediately before is calculated.
Since T_POS) = 2, the conditions of DIST_POS> 1 and COLOR = black are satisfied (steps A5 and A).
6). Further, during the processing for the previous change point coordinates (CRNT_POS), the change point detection flag (WHITE_FOUND) is invalidated in step A12, and the current change point calculation value (SCALE_PO) is set.
S) and the change point coordinates detected immediately before (REG_SCALE
_POS) (DIST_POS) = 2, the process proceeds to step A10 (step A7,
A8).

At step A10, the change point to the white pixel which has disappeared in the previous reduction calculation process is restored. That is,
The current change point calculation result (SCALE_POS) is changed from white to black change point coordinates (AVL_SCALE_POS).
B), and the change point coordinates (REG_SCA) determined immediately before.
LE_POS) +1 is the coordinate of the change point to the white pixel (AVL
_SCALE_POSW).

Here, the current change point calculation result (S
Since CALE_POS) = 6, the change point coordinates from white to black (AVL_SCALE_POSB) = 6, while the change point coordinates (REG_SCALE_) determined immediately before are set.
Since POS) = 4, the change point coordinates (AV
L_SCALE_POSW) = 5 is set to restore the change point to the white pixel. In addition, a change point detection flag (WHITE
_FOUND) (BLACK_FOUND) is enabled (T
RUE).

Below, the remaining change point coordinates (CRNT_PO
S) is processed in the same manner as described above. Thus, the change point coordinates (change point coordinates from white to black (AVL_SCALE_POSB) and change point coordinates from black to white (AVL_SCALE_POSW)) after shrinking with respect to the change point coordinate (CRNT_POS) column shown in FIG. 5
It is obtained as shown in (d). In FIG. 5D, the coordinates of the change point from black to white (AVL_SCALE_P
OSW) = 5 is the point of change that was revived. An image pattern corresponding to the changed change point coordinate sequence shown in FIG. 5D is as shown in FIG.

In this way, when the reduction processing of the image data in the main scanning direction is performed based on the coordinate information of the changing points to white and black, it is not necessary to develop the image data on the working memory. The data transfer amount can be reduced, and high-speed image processing can be performed. In particular, in an image including figures, characters, etc., when the white (blank) area is small, the calculation amount and memory access amount in the reduction process are reduced, so that the image processing speed can be significantly increased.

Furthermore, change point coordinates (AVL_SCALE_POSB) (AVL_SCALE_POSB) (A) in which black pixels are preferentially left in the reduced image pattern and the loss of white pixels is minimized.
Since VL_SCALE_POSW) is generated, even if an image includes a drawing or a character, it is possible to prevent blurring or disappearance of a line forming a figure, a character, or the like, and high-quality reduction processing can be performed.

Next, the data reduction method in the image processing method according to the second embodiment of the present invention will be described with reference to FIGS.
This will be described with reference to the flowchart shown in FIG. The data reduction method according to the second embodiment is as shown in FIG.
Reduction processing is performed in the sub-scanning line direction on the basis of an input change point table in which coordinate information of the change points of the image data in the main scanning direction to white and black is stored.

In FIG. 8A, a change point table for 1 line (change point coordinate sequence after reduction) is obtained by OR reduction based on a change point table for 2 lines (change point coordinate sequence for each line). ) Is generated. FIG. 8B is a diagram in which the changing point coordinates shown in FIG. 8A are replaced with an image pattern so as to be intuitively understood.

In this embodiment, the change point coordinates after the OR processing are extracted according to the procedure of the flowcharts shown in FIGS. 6 and 7 based on the relative positional relationship of the change points in the change point tables A / B.

Step B1 is an initialization process in which a memory area for storing the changed points after the reduction is secured for two lines (a read bank and a write bank), and this reduced change point table and the changes after the reduction are made. This is a step of initializing variables used for point detection.

Step B2 is a step of judging whether or not the noticeable change point C1 and the white change point candidate C2 used in the subsequent steps are detected. Step B3 is a step of detecting the change points C1 and C2, and one of the input change point table and the read-out reduced change point table having the smaller change point position to the black pixel is written as the noticed change point (C1). The change point to the first white pixel on the right side of the noticeable change point is registered as the white change point candidate (C
2).

Step B4 is a step of determining whether or not the change point in the change point table has reached the end of the line. Step B5 is a step of searching for a change point (CX) whose coordinate value is equal to or more than the white change point candidate (C2) on the line opposite to the noticed change point C1.

Step B6 is a step of determining whether the change point CX detected in step B5 satisfies the condition a or the condition c (details of the conditions a and c will be described later with reference to FIGS. 9 and 10). is there.

Step B7 is a process when the condition a or the condition c is satisfied, and the white change point candidate C2 is registered in the reduced change point table. Further, in order to detect the next attention change point C1 and the white change point candidate C2, (C1C2_FOUN
D) Disable the flag.

Step B8 is a step of determining whether or not the change point CX detected in step B5 satisfies the condition b (details of the condition b will be described later with reference to FIG. 11).

Step B9 is a process when the condition b is satisfied. The change point CX is set as the noticed change point C1, and the change point to the first white pixel on the right side thereof is the white change point candidate (C2).
And

Step B10 is a process other than the conditions a, b, and c. The change point CX is located at the coordinate position exceeding the white change point candidate (C2), and the color of the change point becomes a white pixel. It corresponds to the case where it was the change point of. in this case,
The position of the white change point candidate (C2) is set as the attention change point (C1), and the detected change point (CX) is set as the white change point candidate (C1).
2).

Steps B11, B12, and B13 determine whether the target change point (C1) or the white change point candidate (C2) is the end of the line, and if it is the end of the line, the white change point candidate C2 is reduced and changed. When the line is registered in the point table and the line end is not reached, the (C1C2_FOUND) flag is validated in order to detect a change point whose coordinate value is the white change point candidate C2 or more.

Steps B14 and B15 are steps for selecting the reduced change point table to be output from the line number information of the input / output image. When outputting, step B
1 so that the change point table is initialized (CL
R_TBL) flag enabled.

Steps B16 and B17 are steps for judging whether or not the processing for one screen has been completed, and if not completed, switching the bank of the reduction change point table and repeating the processing of step B1.

Next, the operation of the data reduction processing in the second embodiment will be described. First, the area for storing the changed points after reduction is equivalent to 2 lines (for reading and writing)
The reserved variable point table and the variables used for detecting the changed point after reduction are secured and initialized (step B1).

Next, it is judged whether or not the noticeable change point (C1) and the white change point candidate (C2) used in the subsequent steps are detected, that is, whether the (C1C2_FOUND) flag is valid (TRUE). Yes (Step B
2). The noticed change point C1 and the white change point candidate C2 are extracted from the change point table A (first line), and if not extracted here, the process proceeds to step B3.

In step B3, of the input change point table and the read reduction change point table, the one having a smaller change point position to the black pixel is registered as the noticed change point (C1) in the write reduction change point table, The change point to the first white pixel on the right side of the noticed change point is defined as a white change point candidate (C2).

In the example shown in FIG. 8, when the input change point table is the change point table A and the read change point table is the change point table B, the noticed change point (C1) is 1 in the change point table A. (First change point),
The white change point candidate (C2) is 7 (second change point) on the right side. Regarding the change point of interest (C1),
Since it is decided as the OR reduction result of two lines, it is registered in the write reduction change point table (the first change point 1 in the change point table A shown in FIG. 8).

Next, if the processing has not reached the end of the line, the coordinates are changed from the line on the opposite side of the line in which the noticeable change point (C1) is detected (change point table B which is a read change point table). A change point (CX) whose value is greater than or equal to the white change point candidate (C2) is searched (step B5). In the example shown in FIG. 8, 8 in the change point table B (5th change point)
Is the change point (CX).

Here, the detected change point (CX) is the condition a.
Alternatively, it is determined whether or not the condition c is satisfied (step B
6). Condition a is that the change point (CX) is a white change point candidate (C
It means the case where it is at the same position as 2) and the color is the change point of white. FIG. 9 shows a case where the condition a is satisfied (the fourth change point 7 on the second line is the change point (C
X)). In addition, although the processing is actually performed on the change point table, the image pattern is shown here for the sake of intuitive understanding.

Condition c means that the change point (CX) is at a coordinate position exceeding the white change point candidate (C2) and the color is the change point to the black pixel. In FIG. 10, condition c
Is satisfied (the fifth change point 8 on the second line is the change point (CX)).

If either condition is satisfied, step B7
Move to the processing of. In step B7, white change point candidate C
2 is registered in the write reduction change point table. In the example shown in FIG. 8, since the condition c is satisfied, the change point of the white change point candidate C2, that is, the second change point 7 is registered in the write reduction change point table. Also, the (C1C2_FOUND) flag is disabled (F1) to detect the next C1 and C2.
ALSE) and return to step B2.

At step B2, (C1C2_FOUN
D) Since the flag is invalid (FALSE), the process proceeds to step B3 for detecting the next change points (C1) and (C2). In the example shown in FIG. 8, the fifth change point 8 of the change point table B, which is a read-out reduced change point table, is the change point of interest (C1), and the sixth change point 9 of the change point table B is white change. It is detected as a point candidate C2 (step B3). (Note that in FIGS. 9 and 10, the next change points (C1) and (C2) are represented by (C1 ′) and (C2 ′).) Next, in step B5, the opposite side of the noticeable change point C1. Line 4, the fourth change point 10 in the change point table A of the input change point table is detected as the change point (CX) whose coordinate value is equal to or greater than the white change point candidate (C2).

In this case, since the change point (CX) does not satisfy either condition a or condition c (step B6), the process shifts to step B8 to determine whether condition b is satisfied. The condition b is that the change point (CX) is a white change point candidate (C2).
It means that it is in the same position as, and the color is a change point to black.

FIG. 11 shows a case where the condition b is satisfied. In condition b, the change point (CX) is a white change point candidate (C
It means the case where it is at the same position as 2) and the color is the change point to black.

In the example shown in FIG. 8, the white change point candidate (C2)
Since the (7th change point 9 in the change point table B) and the change point (CX) (the 3rd change point 9 in the change point table A) are at the same position, the condition b is satisfied.

When the condition b is satisfied, the change point (CX) is set as the noticed change point (C1), and the change point to the first white pixel on the right side is set as the white change point candidate (C2) (step B).
9).

In the example shown in FIG. 11, the new change point of interest is (C1 ') and the white change point candidate is (C2').
In this case, the white change point candidate (C2) is the attention change point (C1).
It is not registered in the reduced change point table because it is crushed by the black change point of ´).

When the condition b is not satisfied, that is, the condition other than the conditions a, b, and c, the position of the white change point candidate (C2) is set as the noticed change point (C1), and the detected change point ( CX) is set as a white change point candidate (C2) (step B10).

FIG. 12 shows another case where none of the conditions a, b and c is satisfied. In the example of FIG. 12, the new noticeable change point is (C1 ′) and the white change point candidate is (C2 ′). Similar to the case of the condition b, the white change point candidate (C2) is also crushed into the change point (C1 ′) to the black change point here, and thus is not registered in the reduced change point table.

When the processing of step B9 (or B10) is completed, it is judged whether the noticeable change point (C1) or the white change point candidate (C2) is the end of the line (step B
11). If it is not the end of the line, the (C1C2_FOUND) flag is set to be valid (TRUE) to detect a change point whose coordinate value is the white change point candidate (C2) or more, and the process returns to step B2.

On the other hand, if it is the end of the line, the white change point candidate C2 is registered in the write reduction change point table (step B12). Even when it is determined in step B4 that the change point in the change point table has reached the end of the line, step B12 is executed.

Next, the reduction change point table to be output is selected from the line number information of the input / output image (step B).
14). For example, it can be realized by the carry generation condition of DDA. That is, the DDA is activated for each line, and the DDA is
It suffices to output the write reduction change point table when a carry occurs.

In addition, when outputting the write reduction change point table, the (CLR_TBL) flag indicating that the change point table is initialized is set to valid (TRUE) (step B15). As a result, the reduction change point table is initialized in step B1.

Here, it is judged whether or not the processing for one screen has been completed (step B16). If not completed, the bank of the reduction change point table is switched, the processing returns to step B1, and the above-mentioned processing is repeated. .

In this way, when the reduction processing of the image data in the sub-scanning direction is performed, the OR reduction change point information is obtained based on the change point table storing the change point coordinate information of each line to white and black. Therefore, similarly to the case of the first embodiment, it is not necessary to expand the image data on the working memory, the data transfer amount can be reduced, and high-speed image processing can be performed.

Next, a data conversion method in the image processing method according to the third embodiment of the present invention will be described with reference to the flowchart shown in FIG. In the data conversion method according to the third embodiment, as shown in FIG. 16, image data is generated based on a change point data string including change point coordinate information of white and black of image data.

Step C1 is a step for executing the initialization processing, and initializes the immediately preceding change point calculation result.
Step C2 is a step of storing the calculation result for the coordinates of the immediately preceding change point in (BYTELEN_OLD) and obtaining the quotient and the remainder by dividing the coordinates of the input change point by a predetermined value.

In step C3, the difference between the byte length (BYTELEN) of the change point coordinates obtained in step C2 and the immediately preceding value (BYTELEN_OLD) is obtained and held in (FILL_NO).

Step C4 is a step of determining whether or not (FILL_NO) obtained in step C3 is 1 or more. Step C5 is a step of determining whether or not the color of the current change point is black.

Steps C6 and C7 are the byte length fractional value (BITPAT_NO) and the current color (COLO).
R) as an address and referring to the image conversion table,
This is the step of obtaining the image pattern (BITPAT_REG) of the fractional part, and when the current color is black, (BI
TPAT_REG) and output of image conversion table
R calculation is performed and the result is substituted into (BITPAT_REG).

On the other hand, if the current color is white, (BITP
AND the output of the image conversion table
The result obtained is substituted into (BITPAT_REG). Step C8 is the determination of step C4 (FILL_NO).
Is a step to be executed when 1 is 1 or more, the image pattern of the fractional part generated in steps C6 and C7 (BI
TPAT_REG) is transferred to the output image area. Further, (BITPAT_REG) is initialized with the image pattern corresponding to (BITPAT_NO) obtained in step C2.

In step C9, a white or black image composed of 1 byte (8 bits in this case) is displayed as (FILL).
_NO) -1 transfer to the output image area for one time.

Steps C10 and C11 are for terminating the line,
This is a step of determining the end of one screen, which is an end determination for sequentially executing processing in the raster scanning direction until the change point information for one screen is converted into an image.

Next, the operation of the data conversion method according to the third embodiment will be described with reference to the concrete examples shown in FIGS. 14, 15 and 16. The input change point data shown in FIG. 14B becomes the input data. Image data corresponding to the change point coordinate sequence of FIG. 14B is shown in FIG.

First, after the initialization of the immediately preceding change point calculation result, the immediately previous calculation result is set to (BYTELEN_OL
D), and obtains the quotient (byte length (BYTELEN)) and the remainder (BITPAT_NO) obtained by dividing the change point coordinates of the input data by a predetermined value (steps C1 and C).
2). In the present embodiment, the predetermined value is set to 8, and the byte length (BYTELEN) of the change point coordinates and the remainder (BI) are set.
TPAT_NO). That is, POSHI
The input change point data (CPOS) is shifted to obtain the quotient with FT = 3, and the remainder is obtained by the logical product of the input change point data (CPOS) with POSMASK = 7.

Next, the byte length (BYTELEN) of the change point coordinates obtained in step C2 and the immediately preceding value (BYT
ELEN_OLD) and calculate this as (FILL_OLD)
NO) (step C3).

If (FILL_NO) is 0 (step C4), the process proceeds to step C5. In step C5, it is determined whether or not the color of the current change point is black. When the color of the current change point is black, the image pattern (BITPAT_REG) of the previous fractional part and FIG.
By ORing the image pattern obtained from the image conversion table for black shown in (1), (B
ITPAT_REG) is generated (step C6).

If the color of the current change point is white, the image pattern of the previous fractional part (BITPAT_R
EG) and the image pattern obtained from the image conversion table for white shown in FIG. 15B are ANDed to generate (BITPAT_REG) (step C7).

In this embodiment, the fraction (BITPA
The image conversion table accessed by (T_NO) is provided with a white change point and a black change point, but two tables are not necessarily required because they are obtained by bit inversion. That is, the image pattern obtained from one of the tables is inverted and used to substitute for the other table. Further, in the present embodiment, the table stores an 8-bit image pattern, but a table configuration of 16 bits / 32 bits or the like is used depending on the unit of expansion in the output image area. The image pattern stored in the image conversion table is, for example, for black, and is a pattern in which after the number of continuous white pixels indicated by the fraction (BITPAT_NO), the black pixels of the remaining pixels are continuous.

On the other hand, in step C4, step C
When it is determined that (FILL_NO) obtained in 3 is 1 or more, the fractional image pattern (BITPAT_REG) generated in steps C6 and C7 is set as the final image pattern generated based on the change point data. Transfer to the output image area (step C8). Also,
(BITPAT_REG) is processed in step C2 for processing (steps C6 and C7) for the next change point data.
The image pattern is initialized with the image pattern obtained from the image conversion table corresponding to (BITPAT_NO) obtained in step 1.

Then, the 8-bit white or black image is transferred to the output image area for (FILL_NO) -1 times. That is, if (FILL_NO) is 2 or more, it indicates that the same color of white or black is continuous for 1 byte or more, so that the required number of bytes of white or black data is continuously output. .

If the line end has not been reached,
Returning to the process of step C2, the process is repeated for the remaining part of the target line. If the end of the line has been reached and the processing for one screen has not been completed, the process returns to the initialization processing in step C1 to perform processing on the input change point data of the remaining lines. When the processing for all lines is completed, the processing ends.

In the example shown in FIG. 14, first, (FILL_
The calculation of the fractional part is performed until NO) becomes 1 or more. In this example, while the input change point data (CPOS) is from 5 to 7, as shown in FIG. 14C, the byte length (BY
Since TELEN) is 0, the calculation for the fractional part is performed.

When the input change point data (CPOS) = 5,
Fraction (BITPAT_NO) is 5 (change point to black pixel)
Therefore, the image pattern of the fractional part (BITPAT
_REG) is the image conversion table value for black (0x07) corresponding to (BITPAT_NO) = 5, that is, BIPTAT (BIT corresponding to CPOS shown in FIG.
Initialize with PAT_NO).

When the input change point data (CPOS) = 6,
Since the fraction (BITPAT_NO) is 6, which is the change point to the white pixel, the table value (0xf) of (BITPAT_NO) = 6 obtained from the image conversion table for white.
c) and the image pattern of the previous fractional part (BITPAT
AND with (_REG) (BITPAT_RE
G) is obtained (step C7).

When (CPOS) = 7, since the color of the changing point is black, the image pattern corresponding to (BITPAT_NO) = 7 obtained from the image conversion table for black and the image pattern of the previous fractional part ( BIPTAT_
REG) and OR (BITPAT_REG)
Is calculated (step C6).

When (CPOS) = 9 (change point to white pixel), the byte length (BYTELEN) = 1, so that (FILL_NO) = 1 is obtained in step C3. Therefore, it was calculated immediately before (BITPAT_RE
Transfer the value of G) to the output image area. In addition, (BI
The image conversion table for white is referred to from TPAT_NO) = 1 and is substituted for (BITPAT_REG) (step C8).

Hereinafter, the same calculation can be performed up to (CPOS) = 15. When the last change point (CPOS) = 24,
Since the previous byte length (BYTELEN-OLD) = 1 and the byte length (BYTELEN) = 3, (FILL_NO) = 2. Therefore, in step C8, the immediately preceding (BITPAT_REG) is output, and then (FILL_NO) -1 byte data, that is, a 1-byte black image pattern is output. Depending on the input change point data (image pattern) to be processed, a white image pattern is output here.

In this way, the input change point data consisting of the change point coordinate information sequence of white or black in the main scanning direction of the image pattern can be converted into image data.

Next, an image processing method according to the fourth embodiment of the present invention will be described with reference to the flowchart shown in FIG. The fourth embodiment is a process of generating reduced image data from compressed data of image data without expanding the entire image on a working memory.

Step D1 is a step of generating one line of change point coordinate data consisting of change point coordinate information from the compressed data to white or black. Step D2 is a step of generating change point information reduced in the sub-scanning direction based on the change point coordinate data generated in step D1.

Step D3 is a step of generating change point information reduced in the main scanning direction based on the change point coordinate data generated in step D2. Step D4
Is a step of generating image data based on the change point information generated in step D2.

Next, the operation of data processing in the fourth embodiment will be described. First, from the compressed data of the image data, the change point coordinate data consisting of the change point coordinate information for white or black is generated for one line and stored in the line buffer (step D1).

For example, as a coding method for a binary image, there are generally an MH (Modified Huffman) coding method and an MR (Modified READ) coding method. The compressed data (coded data) obtained by compressing the image data by these coding methods is coded according to a predetermined rule, and the change point coordinates to white or black are extracted based on this rule, Change point coordinate data.

Next, using the change point coordinate data obtained in step D1, change point information reduced in the sub-scanning direction is generated. Here, the data reduction method described in the second embodiment is used. However, since the reduction in the sub-scanning direction is performed on the change point coordinate information sequence for two lines, that is, the change point data for two lines, the change point coordinate data for two lines (change point table) in step D1. ) Is generated and then executed.

Note that step D2 uses the data reduction method described in the second embodiment, and detailed description thereof will be omitted. Next, the change point coordinate data obtained by the reduction process in the sub-scanning direction in step D2 is subjected to the reduction process in the main scanning direction (step D).
3). Note that step D3 uses the data reduction method described in the third embodiment, and detailed description thereof will be omitted.

Next, the change point coordinate data obtained by the reduction processing in the main scanning direction in step D3 is converted into image data. Change point coordinate data is
Since reduction processing is performed in both the main scanning direction and sub-scanning direction, reducing image data can be obtained without expanding the original image data by generating image data from the change point coordinate data. Will be done.
Note that step D4 uses the data conversion method described in the fourth embodiment, and detailed description thereof will be omitted.

Here, if the process for the code data for one screen is not completed (step D5), the process returns to the process of step D1, and the process described above is executed for the next line. After that, when the process for the code data for one screen is completed, the process ends.

In this way, the reduced image data can be generated from the compressed image data without expanding the image data on the working memory. Therefore, the amount of data transfer can be reduced and high-speed image processing can be performed. In particular, in an image including figures, characters, etc., when the white (blank) area is small, the calculation amount and memory access amount in the reduction process are reduced, so that the image processing speed can be significantly increased.

[0136]

As described in detail above, according to the present invention, it is possible to provide high-speed image processing even with a combination of decompression processing and reduction processing and to provide a reduction result with little image quality deterioration. .

[Brief description of the drawings]

FIG. 1 is a block diagram showing the configuration of a system that realizes an image processing method according to an embodiment of the present invention.

FIG. 2 is a flowchart for explaining a data reduction method according to the first embodiment.

FIG. 3 is a flowchart for explaining the data reduction method according to the first embodiment.

FIG. 4 is a diagram for explaining a process of processing of detecting change point coordinates after reduction according to the first embodiment.

FIG. 5 is a diagram for explaining a process of restoring a lost change point to a white pixel in the reduction calculation process of step A10 in the first embodiment.

FIG. 6 is a flowchart for explaining a data reduction method in the image processing method according to the second embodiment.

FIG. 7 is a flowchart for explaining a data reduction method in the image processing method according to the second embodiment.

FIG. 8 is a diagram showing an example of an input change point table which is a processing target of the second embodiment.

FIG. 9 is a diagram for explaining a condition a in the second embodiment.

FIG. 10 is a diagram for explaining a condition c in the second embodiment.

FIG. 11 is a diagram for explaining a condition b in the second embodiment.

FIG. 12 is a diagram for explaining another case in which none of the conditions a, b, and c in the second embodiment is satisfied.

FIG. 13 is a flowchart for explaining a data conversion method in the image processing method according to the third embodiment.

FIG. 14 is a diagram showing an image generation process for explaining the operation of the data conversion method according to the third embodiment.

FIG. 15 is a diagram showing an example of an image conversion table according to the third embodiment.

FIG. 16 is a diagram showing an image generation process for explaining the operation of the data conversion method according to the third embodiment.

FIG. 17 is a flowchart for explaining an image processing method according to the fourth embodiment.

FIG. 18 is a diagram for explaining conventional OR reduction processing.

FIG. 19 is a diagram showing a process of generating a carry pattern by DDA.

[Explanation of symbols]

 10 ... CPU 10 12 ... RAM 14 ... VRAM 16 ... ROM 18 ... Input device 20 ... Input control unit 22 ... Display device 24 ... Display control unit 26 ... Storage device 28 ... Storage control unit

Claims (4)

[Claims] 1. A change point coordinate sequence after reduction according to a reduction ratio is calculated for a change point coordinate sequence for white pixels and black pixels in the main scanning direction of image data, and the calculated change point coordinate sequence is included in the obtained change point coordinate sequence. Regarding the change point coordinates of
The distance from the immediately preceding change point coordinate is calculated, and each change point coordinate in the change point coordinate sequence is set to a white pixel if the calculated distance is 1 or more and the change point coordinate is to a white pixel. If the calculated distance is 0, the change point candidate to the white pixel corresponding to the immediately previous change point coordinate is invalidated, and the calculated distance is 1 or more and the white pixel If there is a change point candidate and it is the change point coordinate to the black pixel, the change point coordinate to the black pixel is set, and the change point candidate of the white pixel is set to the change point coordinate to the white pixel. If the distance exceeds 1, there is no change point candidate for the white pixel, and the change point coordinates are for the black pixel,
An image reduction method characterized in that the coordinates of a change point to a black pixel are set, and the coordinates of a change point to a white pixel are added after a predetermined coordinate value is added to the coordinates of a change point immediately before. 2. A first changing point table comprising a changing point coordinate sequence of white pixels and black pixels in the main scanning direction of image data, and changing point coordinates of the next line in the sub scanning direction of the first changing table. With respect to the second change point table composed of columns, the change point coordinates having a small change point position to the black pixel in any of the change point tables are set as the change point of interest and the change point coordinates to the converted black pixel are set, The change point to the first white pixel on the right side of the noticed change point is set as the white change point candidate, and the change point coordinate value on the opposite side of the change point table with the noticed change point is equal to or greater than the white change point candidate. , The detected change point is at the same coordinate position as the white change point candidate,
If the color is a change point to a white pixel, or if the color is at a coordinate position beyond the white change point candidate and the color is a change point to a black pixel, the white change point candidate is converted to a white after conversion. As the change point coordinates to the pixel, the detected change point is at the same coordinate position as the white change point candidate,
When the color of the change point is the change point to the black pixel, this change point is set as the noticed change point, and the change point to the first white pixel on the right side is set as the white change point candidate, and the detected change point is If the color is at a coordinate position that exceeds the white change point candidate and the color of that change point is a change point to a white pixel,
The position of the white change point candidate is set as the noticed change point, the detected change point is set as the white change point candidate, and it is determined whether the noticed change point or the white change point candidate is the end of the line. An image reduction method characterized in that a point candidate is used as coordinates of a change point to a white pixel after conversion. 3. A quotient and a remainder value obtained by dividing a change point coordinate sequentially read from a change point coordinate sequence to a white pixel and a black pixel in the main scanning direction of image data by a predetermined value are obtained. The difference between the quotient and the quotient obtained for the immediately preceding change point coordinates is obtained, and when the obtained difference value is 0, the image pattern for the change point coordinate sequence that does not reach the predetermined value is obtained according to the remainder value. , When the obtained difference value is 1 or more, the image pattern obtained for the change point coordinate sequence that does not reach the predetermined value is used as the conversion result image pattern, and the white pixel or the black pixel corresponding to the obtained difference is A data conversion method characterized in that a continuous image pattern is output as an image pattern of a conversion result. 4. An image processing method for generating a reduced image based on compressed data obtained by encoding image data, wherein a sequence of change point coordinates from the compressed data to white pixels and black pixels is generated. Then, based on the generated change point coordinate sequence, a change point coordinate sequence reduced in the sub-scanning direction is generated by the image reduction method according to claim 2, and the generated change point coordinate sequence is also generated. And, according to the image reduction method described in claim 1, a change point coordinate sequence reduced in the main scanning direction is generated, and based on the generated change point coordinate sequence, the data conversion described in claim 3. An image processing method, wherein image data is generated by a method, and the above-described processing is performed for one screen to generate reduced image data from compressed data.