JPH04312183A - Picture processor - Google Patents

Abstract

PURPOSE:To obtain satisfactory thinning data or thickenning data. CONSTITUTION:A line memory part 201 which stores input data for (N+2) lines (here, for 19 lines) and simultaneously outputs data of the (N+2) lines, an edge processing part 202 which inputs (N+2) line data, forms the matrix of NXM (here, 17X17) and edge-processes as against binary data of a specified position and a thinning processing part 203 which executes an thinning processing as against binary data outputted from the edge processing part 202 are provided.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention is an image processing apparatus that converts image data input through a predetermined image input means into binary data, and performs thinning processing or thickening processing using a pattern matching method. More specifically, the present invention relates to an image processing apparatus that can perform thinning processing or thickening processing favorably.

0002

[Prior Art] Image processing after binary processing in an image processing device includes, for example, mid-shade processing, contour processing, flat shadow processing, three-dimensional shadow processing, mid-shade/three-dimensional shadow processing, outline/three-dimensional shadow processing, and contour processing.・There is flat shadow processing, etc. Since these processes are basically executed on the basis of thinning and fattening of binary data, it is important that the thinning and fattening processes are performed well in the image processing apparatus.

[0003] Generally, when performing thinning processing and thickening processing, a pattern matching method is used. For example, a square pattern as shown in FIG. 16 or a circular pattern as shown in FIG. 17 is used. (The circle in the figure indicates the center pixel). When the thinning process is executed using the square pattern shown in FIG. 16, as shown in FIG.
Original data 900 becomes thinned data (data after thinning processing) 901. Furthermore, when the thinning process is executed using the circular pattern shown in FIG. 17, the original data 900 is changed to the thinning data 9 as shown in FIG.
It will look like 02. On the other hand, when the fattening process is executed using the square pattern shown in FIG. 16, the original data 903 becomes fattened data (data after the fattening process) 904, as shown in FIG. Furthermore, when the thickening process is executed using the circular pattern shown in FIG.
As shown in , original data 903 becomes fattened data 905. In reality, the closed loops of the original data, thinned data, thickened data, etc. shown in the above figure are all black pixel data (solid black), but for the sake of clarity, only the outlines are shown. show.

0004

[Problems to be Solved by the Invention] However, according to an image processing apparatus that performs thinning processing or thickening processing using a conventional pattern matching method, thinning processing or thickening processing using a specific pattern is not possible. To perform the process,
When compared with the original data, there is a problem in that unnatural parts appear in the thinned data or the thickened data. Specifically, for example, since a square pattern has polarity, when thinning or thickening processing is performed using a square pattern, as shown in FIGS. 18 and 20, the thinning in the vertical and horizontal directions There is a problem in that the amount of thinning or thickening Δβ in the diagonal direction is larger than the amount of thinning or thickening Δα. In addition, when thinning processing or thickening processing is performed using a circular pattern, FIG.
As shown in FIG. 21, there was a problem in that the edges were rounded.

The present invention has been made in view of the above, and its first object is to obtain good attenuation data.

[0006] The present invention has been made in view of the above, and a second object thereof is to obtain good fattening data.

[0007]

[Means for Solving the Problems] In order to achieve the above-mentioned first object, the present invention converts image data input through a predetermined image input means into binary data, and finely analyzes the image data by a pattern matching method. In an image processing device that performs image processing, multiple lines of binary data are simultaneously input to form an N×M matrix, and
edge processing means for performing edge processing on value data;
The present invention provides an image processing apparatus including a thinning processing means for performing thinning processing on binary data outputted via an edge processing means.

Further, in order to achieve the above-mentioned second object, the present invention converts image data input through a predetermined image input means into binary data, and performs an image fattening process using a pattern matching method. The processing device inputs multiple lines of binary data at the same time to form an N×M matrix, and includes an edge processing means for performing edge processing on the binary data at a predetermined position in the matrix; An object of the present invention is to provide an image processing apparatus including a fattening processing means for carrying out fattening processing on binary data outputted through the image processing apparatus.

[0009] Furthermore, in order to achieve the above first object, the present invention converts image data input through a predetermined image input means into binary data, and performs thinning processing using a pattern matching method. In the image processing device, a plurality of thinning processing means that perform thinning processing using a thinning processing pattern, and a plurality of thinning processing means arranged immediately before at least one thinning processing means of the plurality of thinning processing means. , input multiple lines of binary data simultaneously to form an N×M matrix,
The present invention provides an image processing device including edge processing means for performing edge processing on binary data at a predetermined position within a matrix.

Furthermore, in order to achieve the above-mentioned second object, the present invention converts image data input through a predetermined image input means into binary data, and performs an image fattening process using a pattern matching method. The processing device includes a plurality of fattening processing means that perform fattening processing using a fattening processing pattern, and is arranged immediately before at least one fattening processing means of the plurality of fattening processing means, and a plurality of binary value input data simultaneously to form an N×M matrix;
The present invention provides an image processing device including edge processing means for performing edge processing on binary data at a predetermined position within a matrix.

[0011]

[Operation] In the image processing apparatus according to claim 1 of the present invention, first, in the edge processing means, binary data of a plurality of lines are simultaneously inputted to form an N×M matrix, and two or more lines of binary data are input at a predetermined position in the matrix. Edge processing is performed on the value data, and then, in the thinning process, the thinning process is performed on the binary data outputted via the edge processing means.

In the image processing apparatus according to claim 2 of the present invention, first, in the edge processing means, binary data of a plurality of lines are simultaneously inputted to form an N×M matrix, and two or more lines at a predetermined position in the matrix are Edge processing is performed on the value data, and then, in the fattening process, the binary data outputted via the edge processing means is fattened.

In the image processing apparatus according to the third aspect of the present invention, a plurality of thinning processing means execute thinning processing using a thinning processing pattern. At this time, the edge processing means disposed immediately before at least one of the plurality of thinning processing means inputs the binary data of the plurality of lines at the same time to form an N×M matrix. , edge processing is performed on binary data at a predetermined position in the matrix.

[0014] In the image processing apparatus according to the fourth aspect of the present invention, a plurality of fattening processing means execute the fattening process using the fattening process pattern. At this time, the edge processing means disposed immediately before at least one of the plurality of thickening processing means simultaneously inputs the binary data of the plurality of lines to form an N×M matrix. Edge processing is performed on the binary data at a predetermined position within.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The image processing apparatus of the present invention will be described in detail below in the order of [Embodiment 1], [Embodiment 2], [Embodiment 3], and [Embodiment 4] based on the drawings.

[Embodiment 1] FIG. 1 shows a block diagram of the basic configuration of an image processing apparatus according to Embodiment 1, which is roughly divided into an image reading section 100 consisting of a CCD sensor, etc.;
It consists of an image processing section 200 that performs predetermined image processing on input image data input through the image processing section 200, and an image recording section 300 that outputs the output image data that has been subjected to image processing by the image processing section 200 via a laser printer or the like. be done.

FIG. 2 is a block diagram showing only the structure of the part related to the present invention extracted from among various image processing in the image processing section 200.
A line memory unit 201 stores (binarized image data input via 0) for (N+2) lines (in this case, 19 lines) and simultaneously outputs (N+2) lines of data; From part 201 (N+2)
Input the line data and enter N×M (here, 17×
17), and an edge processing unit 202 that forms a matrix and performs edge processing on binary data at a predetermined position, and a thinning process that performs thinning processing on the binary data output from the edge processing unit 202. The processing unit 203 includes a processing unit 203.

The operation of the above configuration will be explained. First, the line memory unit 201 stores input data (1-bit data) for 19 lines, and then simultaneously outputs data while synchronizing from the beginning of each line. Therefore, input data is 19-bit data (that is, data for 19 lines) by passing through the line memory section 201.
is output as

When the edge processing unit 202 inputs 19 lines of data, it stores the following in a 19×19 matrix:
A 17×17 matrix is formed as shown in FIG. Here, one square of the matrix corresponds to one pixel (one bit of data), and 0 is the center pixel. Also,
The outside of the 17×17 matrix (peripheral one square area) is an area for holding reference pixels to be referred to in edge processing to be described later.

Next, the edge processing section 202 processes the 17×17
Edge processing is performed as follows, with pixels at edge processing target positions indicated by x on the matrix as edge processing targets. In edge processing, as shown in FIG. 4, when a pixel to be edge processed is indicated by an
With reference to , three of the surrounding pixels A, B, C, and D are black data pixels, and the edge processing target pixel (× pixel)
This process converts the edge processing target pixel to white data only when is white data, and otherwise converts the edge processing target pixel to black data.

[0021] The thinning processing unit 203 includes the edge processing unit 20
When data is input from 2, the thinning process is executed using an N×N square pattern (here, a 17×17 square pattern) as shown in FIG. When thinning processing is performed on the data that has been subjected to the edge processing described above, Figure 6
As shown in , thinning data similar to that obtained when processing with a square pattern is obtained at the edge part, and thinning data similar to that obtained when processing with a circular pattern is obtained in other parts. . Therefore, as is clear from a comparison with the conventional example shown in FIGS. 18 and 19, good thinning data can be obtained. In other words, the thinning process is performed satisfactorily by switching the pattern used for processing between the edge portion and the non-edge portion of the target data for the thinning process.

Note that the original data is the same as the original data in FIGS. 18 and 19 shown in the conventional example. Furthermore, in reality, the closed loops of the original data, thinning data, etc. shown in the above diagram are all black pixel data (solid black), but are shown only by outlines to make the explanation easier to understand.

[Embodiment 2] FIG. 7 shows a block diagram in which only the configuration of the portion related to Embodiment 2 of the image processing section 200 shown in FIG. 1 is extracted. Note that the basic configuration is the same as that in FIG. 1, so illustration and description will be omitted. Input data (image reading section 1
Data obtained by converting image data input via 00 into binary data)
A line memory unit 204 stores data for (N+2) lines (here, 19 lines) and simultaneously outputs data for the (N+2) lines;
) line data and enter the data of N×M (here, 17
x17) matrix and performs edge processing on the binary data at a predetermined position; and The processing unit 206 is also provided with a processing unit 206.

The operation of the above configuration will be explained. First, the line memory unit 204 stores input data (1-bit data) for 19 lines, and then simultaneously outputs data while synchronizing from the beginning of each line. Therefore, the input data is 19 bits of data (that is, 19 lines of data) by passing through the line memory section 204.
is output as

When the edge processing unit 205 inputs 19 lines of data, it stores the following in a 19×19 matrix:
A 17×17 matrix is formed as shown in FIG. Here, one square of the matrix corresponds to one pixel (one bit of data), and 0 is the center pixel. Also,
The outside of the 17×17 matrix (peripheral one square area) is an area for holding reference pixels to be referred to in edge processing to be described later.

Next, the edge processing unit 205 performs edge processing on the pixels at the edge processing target positions indicated by x on the matrix as edge processing targets. In edge processing, as shown in FIG. 4, when a pixel to be edge processed is indicated by an x, referring to its surrounding pixels A, B, C, and D,
Any one of the surrounding pixels A, B, C, and D is a black data pixel, and the edge processing target pixel (x pixel) is only black data, or the edge processing target pixel is black data, otherwise This process converts pixels to be edge processed into white data.

[0027] The thickening processing unit 206
5, the thickening process is executed using an N×N square pattern (here, a 17×17 square pattern) as shown in FIG. When thickening processing is performed on the data that has been subjected to the edge processing described above, Figure 8
As shown in FIG. 2, thickening data similar to that obtained when processing with a square pattern is obtained in the edge portion, and thickening data similar to that obtained when processing with a circular pattern is obtained in other portions. Therefore, as is clear from the comparison with the conventional example shown in FIGS. 20 and 21, good fattening data can be obtained. In addition, the original data is
This is the same as the original data in FIGS. 20 and 21 shown in the conventional example. Furthermore, in reality, all of the closed loops of the original data, thickened data, etc. shown in the above diagram are black pixel data (solid black), but are shown only as outlines to make the explanation easier to understand.

In the first and second embodiments described above, FIG.
Although edge processing was performed using the position of the x shown in as the edge target position, the present invention is not particularly limited to this. Also,
The reference peripheral pixels shown in FIG. 4 are not limited to A, B, C, and D, and the reference pixels and their combinations may be changed. Furthermore, the matrix of FIG. 3 need not be square either.

[Example 3] Example 3 includes a plurality of thinning processing sections that perform thinning processing using a thinning processing pattern, and at least one thinning processing section of the plurality of thinning processing sections. Just before the processing section, input multiple lines of binary data at the same time,
An N×M matrix is formed, and an edge processing unit is arranged to perform edge processing on binary data at a predetermined position within the matrix.

FIG. 9 shows a block diagram in which only the configuration of the portion related to the third embodiment of the image processing section 200 shown in FIG. 1 is extracted. Note that the basic configuration is the same as that in FIG. 1, so illustration and description will be omitted. A line memory unit 207 stores nine lines of input data (binarized image data input via the image reading unit 100) and outputs nine lines of data simultaneously; The first thinning processing unit 208 inputs data and executes thinning processing, and the thinning data (here, 1-bit data) of the first thinning processing unit 208 is processed for 11 lines. A line memory unit 209 stores and simultaneously outputs 11 lines of data, and inputs 11 lines of data from the line memory unit 209 to form an N×M (here, 9×9) matrix, and stores it at a predetermined position. an edge processing unit 210 that performs edge processing on the binary data, and a second thinning processing unit 211 that performs thinning processing on the binary data output from the edge processing unit 210. ing.

The operation of the above configuration will be explained. First, the line memory unit 207 stores nine lines of input data (1-bit data), and then simultaneously outputs data while synchronizing from the beginning of each line. Therefore, by passing the input data through the line memory section 207,
It is output as 9-bit data (ie, data for 9 lines).

The first thinning processing unit 208 executes thinning processing using a 9×9 square pattern as shown in FIG. data).

The line memory section 209 stores input data (1
When 11 lines of data (bit data) are stored, data is output simultaneously while synchronizing from the beginning of each line. Therefore, the input data is output as 11-bit data (that is, data for 11 lines) via the line memory section 209.

When the edge processing unit 210 inputs 11 lines of data, it stores the following in an 11×11 matrix:
A 9×9 matrix is formed as shown in FIG. Here, one square of the matrix corresponds to one pixel (one bit of data), and 0 is the center pixel. Also, 9
The area outside the x9 matrix (one square area around it) is an area for holding reference pixels to be referred to in edge processing, which will be described later.

Next, the edge processing unit 210 performs edge processing on the pixels at the edge processing target positions indicated by x on the matrix as edge processing targets. In edge processing, as shown in FIG.
In the case shown by , referring to the surrounding pixels A, B, C, and D, three of the surrounding pixels A, B, C, and D are black data pixels, and the edge processing target pixel (× pixel) This process converts the edge processing target pixel to white data only when is white data, and otherwise converts the edge processing target pixel to black data.

When the second thinning processing section 211 receives data from the edge processing section 210, it executes thinning processing using a 9×9 square pattern as shown in FIG. When thinning processing is performed on the data that has been subjected to the edge processing described above, thinning data similar to that obtained when processing with a square pattern is obtained at the edge part, and the square pattern is reduced to 45% in other parts. Pattern rotated by 1 degree (Fig. 1
The same thinning data as in the case of processing with the pattern formed other than the x portion shown in 1) is obtained. Therefore, since the processing is performed using a square pattern and a pattern obtained by rotating the square by 45 degrees, narrowed data with good vertical, horizontal, and diagonal lines and edge processing can be obtained. Figure 13
Shows the reduced data. Note that the intermediate data indicates thinning data output from the first thinning processing section 208. In addition, the original data is shown in Figures 18 and 1 in the conventional example.
It is the same as the original data of 9. Figures 18 and 19
As is clear from the comparison with the conventional example shown in , it is possible to obtain good thinning data. Furthermore, in reality, the closed loops of the original data, thinning data, etc. shown in the above diagram are all black pixel data (solid black), but are shown only by outlines to make the explanation easier to understand.

[Embodiment 4] In Embodiment 4, there are a plurality of fattening processing sections that perform fattening processing using a fattening processing pattern, and immediately before at least one fattening processing section of the plurality of fattening processing sections. Input multiple lines of binary data at the same time,
An N×M matrix is formed, and an edge processing unit is arranged to perform edge processing on binary data at a predetermined position within the matrix.

FIG. 14 shows the image processing section 200 shown in FIG.
A block diagram is shown in which only the configuration of a portion related to the fourth embodiment is extracted. Note that the basic configuration is the same as that in FIG. 1, so illustration and description will be omitted. A line memory unit 212 stores nine lines of input data (binarized image data input via the image reading unit 100) and outputs nine lines of data simultaneously; The first fattening processing unit 213 inputs data and executes fattening processing, and the fattening data (here, 1-bit data) of the first fattening processing unit 213 is stored for 11 lines. A line memory section 214 simultaneously outputs line data, and 11 lines of data are input from the line memory section 214 to form an N×M (here, 9×9) matrix, and binary data at a predetermined position is input. an edge processing unit 215 that performs edge processing on the edge processing unit 215;
and a second fattening processing section 216 that executes fattening processing on the binary data output from the .

The operation of the above configuration will be explained. First, the line memory section 212 stores nine lines of input data (1-bit data), and then simultaneously outputs data while synchronizing from the beginning of each line. Therefore, by passing the input data through the line memory section 212,
It is output as 9-bit data (ie, data for 9 lines).

The first fattening processing unit 213 executes fattening processing using a 9×9 square pattern as shown in FIG. Output.

The line memory section 214 stores input data (1
When 11 lines of data (bit data) are stored, data is output simultaneously while synchronizing from the beginning of each line. Therefore, the input data is outputted as 11-bit data (that is, data for 11 lines) via the line memory section 214.

When the edge processing unit 215 inputs data for 11 lines, the edge processing unit 215 stores the following data in an 11×11 matrix.
A 9×9 matrix is formed as shown in FIG. Here, one square of the matrix corresponds to one pixel (one bit of data), and 0 is the center pixel. Also, 9
The area outside the x9 matrix (one square area around it) is an area for holding reference pixels to be referred to in edge processing, which will be described later.

Next, the edge processing unit 215 performs edge processing on the pixels at the edge processing target positions indicated by x on the matrix as edge processing targets. In edge processing, as shown in FIG.
In the case indicated by If the pixel (pixel) is only black data, the pixel to be edge processed is black data, and in other cases, the pixel to be edge processed is converted to white data.

When the second fattening processing unit 216 receives data from the edge processing unit 215, it executes fattening processing using a 9×9 square pattern as shown in FIG. When the data that has been subjected to the edge processing described above is thickened, the same fattening data is obtained at the edges as when processing with a square pattern, and in other parts, the square is rotated by 45 degrees. pattern (Figure 1
The same fattening data as in the case of processing using the pattern formed other than the x portion shown in 1) is obtained. Therefore, since processing is performed using a square pattern and a pattern obtained by rotating the square by 45 degrees, thickened data with good vertical, horizontal, and diagonal lines and edge processing can be obtained. Figure 15
shows the fattening data. Note that the intermediate data indicates fattening data output from the first fattening processing section 213. In addition, the original data is shown in Figures 18 and 1 in the conventional example.
It is the same as the original data of 9. Figures 18 and 19
As is clear from the comparison with the conventional example shown in , it is possible to obtain good fattening data. Furthermore, in reality, all of the closed loops of the original data, thickened data, etc. shown in the above diagram are black pixel data (solid black), but are shown only as outlines to make the explanation easier to understand.

In the third and fourth embodiments described above, FIG.
Although edge processing was performed using the position of the x shown in 1 as the edge target position, the present invention is not particularly limited to this. Furthermore, the reference peripheral pixels shown in FIG. 12 are not limited to A, B, C, and D, and the reference pixels and their combinations may be changed. Furthermore, the matrix of FIG. 11 need not be square either.

Furthermore, in the first to fourth embodiments described above, since the edge processing section uses a method of referring to peripheral pixels, hardware processing can be easily implemented.

[0047]

As explained above, the image processing apparatus of the present invention performs image processing in which image data input through a predetermined image input means is converted into binary data and thinned by a pattern matching method. In the device, multiple lines of binary data are input simultaneously to form an N×M matrix, and edge processing is performed on the binary data at a predetermined position in the matrix through an edge processing means and an edge processing means. Since the present invention is provided with a thinning processing means for executing thinning processing on the binary data outputted from the filter, it is possible to obtain good thinning data.

The image processing apparatus of the present invention is an image processing apparatus that converts image data input through a predetermined image input means into binary data and performs thickening processing using a pattern matching method. An edge processing means that simultaneously inputs binary data to form an N×M matrix and performs edge processing on the binary data at a predetermined position in the matrix, and a binary value outputted through the edge processing means. Since the present invention includes a fattening processing means for carrying out fattening processing on data, good fattening data can be obtained.

The image processing apparatus of the present invention is an image processing apparatus that converts image data input through a predetermined image input means into binary data and performs thinning processing using a pattern matching method. A plurality of thinning processing means that perform thinning processing using a thinning processing pattern, and a plurality of thinning processing means arranged immediately before at least one thinning processing means of the plurality of thinning processing means, and a plurality of lines of binary data. are input simultaneously to form an N×M matrix, and edge processing means is provided for performing edge processing on binary data at a predetermined position in the matrix, so that good thinning data can be obtained. .

Furthermore, the image processing apparatus of the present invention is an image processing apparatus that converts image data input through a predetermined image input means into binary data, and performs fattening processing using a pattern matching method. a plurality of fattening processing means for carrying out fattening processing using a pattern for fattening; Since an N×M matrix is formed and edge processing means is provided for performing edge processing on binary data at a predetermined position within the matrix, good fattened data can be obtained.

[Brief explanation of drawings]

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

FIG. 2 is a block diagram in which only the configuration of a portion related to the present invention is extracted from among various image processing in the image processing section of the first embodiment.

FIG. 3 is an explanatory diagram showing a matrix formed by an edge processing section.

FIG. 4 is an explanatory diagram for explaining edge processing.

FIG. 5 is an explanatory diagram showing a 17×17 square pattern used in Example 1.

FIG. 6 is an explanatory diagram showing thinning data output by the first embodiment.

FIG. 7 is a block diagram in which only the configuration of a portion related to the present invention is extracted from among various image processing in the image processing section of the second embodiment.

FIG. 8 is an explanatory diagram showing fattening data output according to the second embodiment.

FIG. 9 is a block diagram in which only the configuration of a portion related to the present invention is extracted from among various image processing in the image processing unit of the third embodiment.

10 is an explanatory diagram showing a 9×9 square pattern used in Example 3. FIG.

FIG. 11 is an explanatory diagram showing a matrix formed by an edge processing section.

FIG. 12 is an explanatory diagram for explaining edge processing.

FIG. 13 is an explanatory diagram showing thinning data output by the third embodiment.

FIG. 14 is a block diagram in which only the configuration of a portion related to the present invention is extracted from among various image processing in the image processing section of the fourth embodiment.

FIG. 15 is an explanatory diagram showing fattening data output according to the fourth embodiment.

FIG. 16 is an explanatory diagram showing a square pattern used in a conventional pattern matching method.

FIG. 17 is an explanatory diagram showing a circular pattern used in a conventional pattern matching method.

FIG. 18 is an explanatory diagram showing a conventional example of thinning processing when a square pattern is used.

FIG. 19 is an explanatory diagram showing a conventional example of thinning processing when a circular pattern is used.

FIG. 20 is an explanatory diagram showing a conventional example of thickening processing when a square pattern is used.

FIG. 21 is an explanatory diagram showing a conventional example of thickening processing when a circular pattern is used.

[Explanation of symbols]

100 Image reading section
200 Image processing section 201 Line memory section
202 Edge processing section 203 Thinning processing section
204 Line memory section 205 Edge processing section
206 Thickening processing section 207 Line memory section
208 First thinning processing section 209 Line memory section
210 Edge processing section 211 Second thinning processing section
212 Line memory section 213 First thickening processing section
214 Line memory section 215 Edge processing section
216 Second thickening processing section 300 Image recording section

Claims (4)

[Claims] Claim 1: An image processing device that converts image data input through a predetermined image input means into binary data and performs thinning processing using a pattern matching method, which simultaneously inputs multiple lines of binary data. to form an N×M matrix, and 2
edge processing means for performing edge processing on value data;
An image processing apparatus comprising: thinning processing means for performing thinning processing on the binary data outputted via the edge processing means. [Claim 2] An image processing device that converts image data input through a predetermined image input means into binary data and performs fattening processing using a pattern matching method, which simultaneously inputs multiple lines of binary data. , form an N×M matrix, and 2 at a given position in said matrix.
edge processing means for performing edge processing on value data;
An image processing apparatus comprising: fattening processing means for carrying out fattening processing on the binary data outputted via the edge processing means. 3. An image processing apparatus that converts image data input through a predetermined image input means into binary data and performs thinning processing using a pattern matching method. A plurality of thinning processing means for performing thinning processing are arranged immediately before at least one thinning processing means of the plurality of thinning processing means, and binary data of a plurality of lines are simultaneously inputted. An image processing device comprising: an edge processing unit that forms a xM matrix and performs edge processing on binary data at a predetermined position within the matrix. 4. An image processing apparatus that converts image data input through a predetermined image input means into binary data and performs fattening processing using a pattern matching method, which performs fattening processing using a pattern for fattening processing. a plurality of fattening processing means for performing the above-mentioned fattening processing means; and a plurality of fattening processing means arranged immediately before at least one fattening processing means of the plurality of fattening processing means, and simultaneously inputting binary data of a plurality of lines to form an N×M matrix. and edge processing means for performing edge processing on binary data at a predetermined position within the matrix.