JP2001229373A - Reduction processing method for binary image, and image forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a binary image at a comparatively high speed and low cost without erasing the image nor causing unevenness of density. SOLUTION: A given scale factor is applied to a dither matrix for generating a basic scale matrix (S31-S33). The rows or columns are rearranged in the basic scale matrix for generating plural basic scale matrixes which are different from each other and these matrixes are combined together for generating a large scale matrix (S34). Then the large scale matrix is applied to the binary image data to delete the pixel corresponding to the off-bit of the large sale matrix (S36). The basic scale matrix is rearranged again as a matrix element in each row and column so at to eliminate the spatial regularity of on-dots from the large scale matrix (S35). The matrix thus obtained can be used as a new large scale matrix.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method of scale processing, particularly a reduction processing, for a binary image and an image forming apparatus using the method.

[0002]

2. Description of the Related Art Generally, a Weiman scaling algorithm is known as a scaling (reduction) process for a binary image. The outline of this algorithm is p /
Given a scale ratio of q (p <q, p and q are positive integers), q rows and p columns for obtaining scale pattern data by taking the q value on the horizontal axis and the p value on the vertical axis And a diagonal line is drawn between the lower left and upper right diagonal points. The slope of this diagonal is p / q. It is determined whether a straight line having this inclination intersects any horizontal line of the matrix in each of the successive columns (first column, second column,..., Q-th column). Bit "1" when crossed, bit "1" when not crossed
By setting 0, q-bit thinning-out pattern data is obtained.The bits in the first column are always fixed to "1" and the bits in the p-th column are always fixed to "0." (AND operation is performed) to reduce the image, that is, the image dots to which the bit “1” is allocated are left, and the image dots to which the bit “0” is allocated are deleted.

FIG. 3 shows a matrix for explaining the Weiman scaling algorithm when the scale ratio p / q = 5/8. In this case, the diagonal lines of the matrix of 5 rows and 8 columns intersect the horizontal lines in the second column, the fourth column, the fifth column, and the seventh column. Therefore, the bits of the pattern data corresponding to those columns and the first column are “1”. Bits in other columns are "0". Thus, "1101"
1010 ″ is obtained as 8-bit pattern data. This data is represented by hexadecimal notation 0xda (0x is 16
Indicate the base number). Thereby, the 8 dots (8
The dot at the position of bit "0" of the pattern data is deleted for each pixel.

FIGS. 4 and 5 similarly show pattern data (0xa9, 0x5) with a scale ratio p / q = 3/7, respectively.
2,0xa5,0x4a, 0x95,0x2a, 0x5
4) and p / q = 1/2 (0xaa).

[0005]

When a binary image is reduced using the above-described pattern data for thinning, whether or not dots can be thinned out is determined uniformly in the input order of input data, and thinning is performed. The system cannot determine whether a pixel is a valid image dot (for example, a black dot or a dot of a color other than the background color). For this reason, an image loss may occur unintentionally. For example, when linear data having a line width of 1 dot is input as a binary image, image loss (that is, disappearance) of all the linear lines may occur.

[0006] To solve such data loss, 2
There is also known a method of determining whether or not the content of the value image is a continuous dot and permitting the deletion of the dot only for the continuous dot. However, if the isolated dots are not deleted, the density becomes high, which may cause uneven density in the reduced image. Although a method of suppressing the occurrence of the density unevenness by advanced dot determination is also known, the processing speed is extremely slow, and the cost is increased by using hardware for high speed. There's a problem.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and has as its object to prevent the occurrence of image loss, prevent the occurrence of density unevenness, and achieve a relatively high speed and low speed. An object of the present invention is to provide a method for performing a reduction process of a binary image at a low cost and an image forming apparatus using the method.

[0008]

A method for reducing a binary image according to the present invention is a method for reducing a binary image based on a given scale factor (less than 100%). Applying a rate to the dither matrix to generate a base scale matrix having on / off dots in each cell; and rearranging the rows or columns of the cells in the generated base scale matrix to generate a plurality of different base scale matrices. Generating a large scale matrix by combining the plurality of basic scale matrices, and applying the large scale matrix to the binary image data to correspond to off bits of the large scale matrix. Deleting the pixel.

According to the present invention, a plurality of different basic scale matrices are generated by using a scale matrix having two-dimensional on / off dots obtained by a dither matrix and by rearranging the rows or columns of the cells. Are combined to generate a larger scale matrix, and this large scale matrix is applied to the binary input image data.
In addition, the disappearance of a straight line or the like due to the image reduction processing can be more reliably prevented. Further, once a large-scale matrix is generated, pixels can be deleted based on the cell value of the off-bit, so that reduction processing can be performed relatively easily and at high speed.

For example, the basic scale matrix is N
× N (N is a positive integer of 2 or more).
N × N−1 reordering of the cells in the × N basic scale matrix in units of rows and columns is performed, and the generated N × N−1 N × N basic scale matrices are converted to the original N × N−1.
In combination with × N basic scale matrix (N × N)
Generate a large scale matrix of size (N × N).

Preferably, the (N × N) × (N × N)
Of the large scale matrix is further rearranged in rows and columns using the basic scale matrix as matrix elements so as to eliminate the spatial regularity of the on-dots, and the resulting new (N × N ) × (N × N)
May be applied to the binary image data.

This makes it possible to more reliably prevent a straight line or the like from disappearing.

An image forming apparatus employing the above-described method for reducing a binary image according to the present invention comprises: a data analyzing means for analyzing binary image data to generate print data; a frame memory for expanding the print data; When developing an image in the frame memory according to the analysis result of the data analyzing means, a large scale matrix generated or prepared in advance based on a given scale ratio is applied to a binary image, Image reduction processing means for deleting pixels corresponding to off dots of the scale matrix, and printing means for performing printing based on print data developed in the frame memory, the large scale matrix,
By applying the given scale factor to the dither matrix, a basic scale matrix having on / off dots in each cell is generated, and the rows or columns of the generated cells in the basic scale matrix are rearranged and generated. It is characterized by being generated by combining a plurality of different basic scale matrices.

[0014]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an image forming apparatus such as a plotter or a printer to which the present invention is applied. In FIG. 1, reference numeral 11 denotes a CPU for controlling the operation of the entire apparatus,
Reference numeral 12 denotes a RAM used as a work area for the CPU 11 and a temporary storage area for data. A ROM 13 stores programs and data for driving the image forming apparatus, and is used by the CPU 11. Reference numeral 14 denotes an interface unit for connecting to an external computer terminal device or the like (not shown), through which print target data such as a binary image (plotter description language data in this example) is transferred. Reference numeral 15 denotes an LCD display device for performing display for a man-machine interface, and reference numeral 16 denotes a key operation unit for selecting various settings of the image forming apparatus. 1
Reference numeral 7 denotes a printing unit of the image forming apparatus, and reference numeral 18 denotes a system bus that connects the CPU 11 to other components.

FIG. 2 shows a schematic processing flow from input data reception to printing in the image forming apparatus of FIG.

First, input data (binary image data) is received from the outside (S21), and the received data is subjected to data analysis (language analysis) according to a format of a plotter description language (S22). This data analysis means is generally called an interpreter. Next, resolution conversion based on the difference between the resolutions of the input image and the output image, and scaling processing according to a scale ratio determined based on an input instruction from the key operation unit 16 or an instruction based on received data are performed (S23). Thereafter, the scaled image is developed in a frame memory (in the RAM 12) used for printing (S24). Until the print start command is confirmed by the data analysis, the processes of steps S22 to S24 are repeated (S25). When the print start command is received, the actual print operation is started based on the data developed in the frame memory (S26).

However, if the determination in S25 is not "reception of print start command?" But "has data for one band accumulated?", The frame memory is not for all images but for one band (for example, print head). Scan width). In the present embodiment, any of them may be used.

FIG. 7 shows a dither matrix (Bayer Dither) for creating a basic scale matrix in the embodiment of the present invention. The dither matrix is generally used to convert multivalued image data into binary image data in order to reproduce shades (gradations) of the multivalued image with binary dots. FIG. 7 shows a dither matrix of 8 rows and 8 columns (8 × 8). In this dither matrix, numerical values from "0" to "252" are assigned to 64 cell positions in four steps, respectively, and each numerical value is used as a threshold for binarizing multi-valued data. Thus, 8 × 8 + 1 (= 65)
It is possible to reproduce a multi-value image of gradation.

Normally, line segment data in vector format is represented by the line width, tip shape, density, etc., in addition to the coordinates of its start point and end point. In order to express the shading of a vector, a matrix (mask matrix) having an on / off pattern corresponding to the density value of the vector as shown in FIG. 8 is generated based on the dither matrix of FIG. This matrix corresponds to a case where the density value is 50%. That is, if the density value 100% is 255, the density value of the density value 50% is 127. Therefore, the matrix 12 shown in FIG.
The cell part having a numerical value of 7 or less is turned on, and the cell part having a numerical value larger than 127 is turned off. In the present invention, this mask matrix is used as a thinning matrix (scale matrix) used for reducing a binary image. By thinning out (deleting) the image dots (pixels) corresponding to the OFF cell portion, the target image can be reduced at a scale ratio corresponding to the density value.

Therefore, the matrix shown in FIG. 8 is a basic scale matrix when the scale ratio is 50%. In the case of this scale ratio, in the algorithm described in the related art, if the input data is 0x55 for the thinning pattern data (0xaa) corresponding to p / q = 1/2,
When an AND of this data and 0xaa is obtained, the result is all bits 0. This may cause vertical lines to disappear completely after reduction. In the matrix pattern of FIG. 8, such disappearance of the vertical line is eliminated. However, an oblique line such as an inclination of 45 ° of one line width may completely disappear.

A solution to this problem according to the present embodiment will be described below.

FIG. 6 shows a step S23 in the flow of FIG.
2 shows a specific processing flow of the scale processing.

First, a scale ratio is determined (S31).
This is determined by an externally set scale ratio, a scale ratio by resolution conversion in data, an aspect ratio designation, etc. (S3).
1). The scaling involved in the present invention is only reduction, and the scaling ratio is in the range of 0 to 100% (however, 0% and 100% are not included). Then, in this example, the scale ratio SCi of 0 to 100% is set to 0 to 255.
(S32). This conversion is simply realized by the following equation.

SCo = SCi × (255/100)

By applying the obtained scale value SCo to the dither matrix pattern shown in FIG. 7, a basic scale matrix is created (S33). This fitting method is the same as the above-described method of obtaining the density mask matrix. The numerical value in each cell of the dither matrix is compared with the scale value, and the cell having a value smaller than the scale value is set to “1”.
(ON: put a dot) and set the cell with the larger value to “0”
(Off: Delete dots). In this way,
An 8 × 8 basic scale matrix is created.

The basic scale matrix when the given scale ratio is 50% is as shown in FIG.
That is, a scale value of 12 corresponds to a 50% scale rate.
As a result, a basic scale matrix is obtained in which all cells in the dither matrix whose threshold value is smaller than 127 are "1".

Next, rearrangement is performed N × N−1 times on an N × N matrix (8 × 8 in the present embodiment), and
(N × N) × (N × N) size (6 in this embodiment)
A method of creating a large scale matrix of (4 × 64) will be described.

In order to create a large scale matrix of (N × N) × (N × N) size, the rows and columns of the cells in the basic scale matrix are rearranged, that is, rearranged (S34).

FIGS. 9 and 10 are diagrams for explaining the rearrangement. An example of a program for rearranging rows and columns (described in C language) is as follows. Comments are added to the main part of the program to indicate the role of that part. However, this program is an example for rearranging a matrix, and a program language and its description are arbitrary as long as similar processing can be realized. #define N_MATRIX 8 #define BYTE_DOT 8 extern unsigned char inputMatrix [N_MATRIX * N_MATRIX / BYTE_DOT]; extern unsigned char outputMatrix [N_MATRIX * N_MATRIX * N_MATRIX * N_MATRIX / BYTE_DOT]; {char i, j, n; for (i = 0; i <N_MATRIX; ++ i) {/ * Creation of P1 in FIG. 11 * / outputMatrix [i * N_MATRIX * N_MATRIX / BYTE_DOT] = inputMatrix [i];} for (j = 0; j <N_MATRIX-1; ++ j) {/ * Creation of P2, P3, P4, P5, P6, P7, P8 in FIG. 11 * / for (i = 0; i <N_MATRIX; + + i) {outputMatrix [(i * N_MATRIX) + (j + 1)] = outputMatrix [(N_MATRIX-1-i) * N_MATRIX + j];}} for (n = 0; n <N_MATRIX-1; ++ n) {for (i = 0; i <N_MATRIX; ++ i) {/ * Create P9, P17, P25, P33, P41, P49, P57 in Fig. 11 * / outputMatrix [i * N_MATRIX + ((n + 1 ) * N_MATRIX * N_MATRIX)] = (outputMatrix [i * N_MATRIX + (n * N_MATRIX * N_MATRIX)] >> 1) | (outputMatrix [i * N_MATRIX + (n * N_MATRIX * N_MATRIX) <<7);} for (j = 0; j <N_MATRIX-1; ++ j) {/ * Create remaining P in Fig. 11 * / for (i = 0; i <N_MATRIX; ++ i) {outputMatrix [(i * N_MATRIX) + (j + 1) + ((n + 1) * N_MATRIX * N_MATRIX)] = outputMatrix [(N_MATRIX-1-i) * N_MATRIX + j + ((n + 1) * N_MATRIX * N_MATRIX)];}}}}

FIG. 11 is a diagram showing a configuration of a large-scale matrix obtained by this program processing.
The pattern P1 is an 8 × 8 basic matrix that is the basis of the rearrangement. P2 is obtained by moving the top row of the basic matrix to the bottom row and rearranging the row of P1 as shown in FIG. P3 is obtained by rearranging rows similarly to P2. Similarly, up to P8 are created. P9 is obtained by shifting the rightmost column of the basic matrix to the leftmost column and rearranging the column of P1 as shown in FIG. P10 is obtained by rearranging the same column with respect to P2. Similarly, up to P16 is created. Further, patterns P17 to P17, which are the next pattern rows in FIG.
Similarly, P24 is created by rearranging the corresponding pattern in the immediately preceding pattern row. The same applies to the following pattern rows.

In this way, based on the N × N basic matrix of P1, the (N × N) ×
A large (N × N) matrix can be created.

Hereinafter, a specific example will be described for a case where the most severe condition is a scale ratio of less than 2%. First, a large-scale matrix pattern obtained by simply combining basic scale matrices having a scale ratio of less than 2% as shown in FIG. In this case, the line (line segment) shown in FIG.
Only the line on the on-dot such as A can be expressed without disappearing, and the line not on the on-dot such as line B and line C disappears as a whole line. On the other hand, FIG. 14 shows the result of performing the processing of step S34 described above. In this case, the possibility of the disappearance of the line is reduced as compared with the case of FIG. 13. However, as for the line whose inclination from the horizontal or vertical is slight like the line D and the line E in FIG. That is, disappearance occurs.

In order to solve such a problem, rearrangement is performed on a large scale matrix in units of rows and columns using the basic scale matrix as matrix elements (S35). The large-scale matrix shown in FIG. 12 is obtained by rearranging the large-scale matrix shown in FIG. Thus, the large scale matrix of the example of the scale ratio of 2% in FIG. 14 is deformed as shown in FIG. As can be seen from the matrix pattern of FIG. 15, the spatial regularity of the on-dots shown in FIG. 14 is excluded. As a result, the line segment does not disappear at any inclination.

An example of how to eliminate the spatial regularity of the on-dot for the large scale matrix shown in FIG. 12 will be described. The spatial regularity is eliminated by changing the arrangement of an M × M (here, M = 8) matrix in which an N × N (here, N = 8) basic matrix is one matrix element. In the present embodiment, first, regarding the rows of the M × M matrix, the contents of the eighth row are moved to the third row,
Move the contents of the original third line to the fourth line. Similarly, the contents of the original fourth line are to the sixth line, the contents of the original sixth line are to the seventh line, the contents of the original seventh line are to the fifth line, and the contents of the original fifth line are to the fifth line. Move to line 8. In this way, the rows of the M × M matrix are rearranged. Next, the same reordering is performed on the M × M matrix for that column. By such an operation, a new M × M matrix as shown in FIG. 12 is generated. There are various other ways to eliminate the regularity, and any method may be used.

Using the large scale matrix obtained in this way, this is applied to the input image data.
That is, the pixel corresponding to the off dot (here, “0”) of the large scale matrix is deleted (FIG. 6, S3).
6).

As described above, FIG.
The large scale matrix finally generated enables the image to be reduced with high accuracy by rearranging the basic scale matrix in matrix units and row and column units. Note that the scale processing by the large-scale matrix need not always be performed for all scale ratios, and may be used only when necessary. For example, when the scale ratio is relatively large (for example, exceeds 50%) so that the line segment does not disappear, the above processing is unnecessary.

Although the preferred embodiment of the present invention has been described above, various modifications and changes are possible. For example,
The basic scale matrix for each of the scale rates (and the final large-scale scale matrix) may be obtained in advance for each of the different scale rates and stored in a table (stored in a ROM or the like). This eliminates the need to generate a matrix each time the scale processing is performed, and merely refers to the table of the matrix, thereby further improving the processing speed.

[0039]

According to the present invention, it is possible to perform relatively high-speed and low-cost scale processing of binary image data while preventing density unevenness, and to perform high-precision binary image data processing with high precision that does not cause image loss. An image reduction processing method capable of expressing a value image and an image forming apparatus using the same can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image forming apparatus according to the present invention.

FIG. 2 is a flowchart illustrating a processing flow from reception of input data to printing of the image forming apparatus of FIG. 1;

FIG. 3 is a diagram illustrating a matrix in the case of a scale ratio p / q = 5/8 for explaining a Weiman scaling algorithm.

FIG. 4 is a diagram illustrating a matrix for explaining a Weiman scaling algorithm when a scale ratio p / q = 3/7.

FIG. 5 is a diagram illustrating a matrix for explaining a Weiman scaling algorithm when a scale ratio p / q = 1/2.

FIG. 6 is a flowchart showing a specific processing flow of a scale processing in step S23 in the flow of FIG. 2;

FIG. 7 shows a dither matrix (bayer) for creating a basic scale matrix in the embodiment of the present invention.
FIG.

8 is a diagram showing a mask matrix created based on the dither matrix of FIG. 7 and a basic scale matrix having a scale ratio of 50% used in the embodiment of the present invention.

FIG. 9 is a diagram showing rearrangement of rows of a basic scale matrix in the embodiment of the present invention.

FIG. 10 is a diagram showing rearrangement of columns of a basic scale matrix according to the embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a large-scale scale matrix obtained by simple rearrangement of a basic scale matrix in the embodiment of the present invention.

FIG. 12 is a diagram illustrating a final form of rearrangement of a large-scale matrix according to the embodiment of the present invention.

FIG. 13 shows a scale factor of 2 according to the embodiment of the present invention.
It is a figure which shows the large scale matrix at the time of%.

FIG. 14 shows a scale factor 2 according to the embodiment of the present invention.
It is a figure which shows the large-scale matrix after simple rearrangement of the large-scale matrix at%.

FIG. 15 shows a scale factor 2 according to the embodiment of the present invention.
It is a figure which shows the final form of rearrangement of the large scale matrix at%.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 CPU 12 RAM 13 ROM 14 Interface 15 Liquid crystal display device 16 Key operation part 17 Printing part 18 System bus

Claims (6)

[Claims] 1. A method for reducing a binary image based on a given scale factor (less than 100%), wherein the given scale factor is applied to a dither matrix to form on / off dots in each cell. Generating a basic scale matrix having: and generating a plurality of different basic scale matrices by rearranging the rows or columns of the cells in the generated basic scale matrix, and combining these plurality of basic scale matrices to produce a large scale Generating a scale matrix; and applying the large scale matrix to the binary image data to delete pixels corresponding to off bits of the large scale matrix. Method. 2. The basic scale matrix is N × N (N
Is a positive integer greater than or equal to 2), and N × N in rows and columns of cells in this N × N basic scale matrix.
-1 permutation, and the resulting N × N
-1 N × N basic scale matrix is combined with the original N × N basic scale matrix to obtain (N × N) × (N
2. A method according to claim 1, wherein a large-scale matrix having a size of (N) is generated. 3. The method according to claim 1, wherein said basic scale matrix is used as a matrix element in row and column units so as to eliminate spatial regularity of on-dots for said large scale matrix of (N × N) × (N × N). 3. The binary image according to claim 2, wherein the reordering is further performed, and a new large scale matrix (N × N) × (N × N) obtained as a result is applied to the binary image data. Reduction processing method. 4. An image forming apparatus for printing binary image data, wherein the data analyzing means analyzes the binary image data to generate print data; a frame memory for expanding the print data; According to the analysis result of the means, when the image is developed in the frame memory, a large scale matrix generated or prepared in advance based on a given scale ratio is applied to the binary image, and the large scale matrix Image reduction processing means for deleting pixels corresponding to off dots, and printing means for printing based on print data developed in the frame memory, the large scale matrix, the large scale matrix, the given scale rate Generate basic scale matrix with on / off dots in each cell by applying to dither matrix An image forming apparatus characterized by being produced by combining a plurality of basic scale matrix mutually different generated by rearranging the rows or columns of the generated basic scale matrix in the cell. 5. The basic scale matrix has a size of N × N, and the large scale matrix has a size of (N × N) × (N × N), where (N × N) × ( N × N) large scale matrix is composed of N × N−1 N × N basic scales generated by N × N−1 permutations of cells and rows in the N × N basic scale matrix. Matrix to the original N
The image forming apparatus according to claim 4, wherein the image forming apparatus is generated in combination with a × N basic scale matrix. 6. The (N.times.N) .times. (N.times.N) large scale matrix is further rearranged on a row and column basis using a basic scale matrix as a matrix element so as to eliminate spatial regularity of on-dots. The image forming apparatus according to claim 5, wherein the image forming is performed.