JP3434720B2 - Binary image scaling conversion processing method and black pixel area detection method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for enlarging / reducing an image formed from image data, and particularly to a binary value formed by binary data of 1 or 0 handled by a device such as a facsimile or a personal computer. The present invention relates to a processing method for performing enlargement / reduction conversion of an image at an arbitrary magnification at high speed.

[0002]

2. Description of the Related Art Conventionally, as shown in FIG. 13, an algorithm for enlarging / reducing image data calculates a corresponding pixel in the original image before conversion for each pixel in the converted image, The density (0 or 1) was written as the density of the converted pixel in the converted image.

As a method for performing such processing at high speed, there is a table reference method in which density patterns of converted horizontal lines are stored. This is because when the enlargement / reduction ratio is m / n as shown in FIG.
A horizontal line after conversion having a length m for all combinations (there are 2 ⁿ patterns) of each pixel density is held in advance as a table, and the pre-conversion image is divided into n-bit units in the horizontal direction. In this method, the m-bit pixel after conversion is written by referring to the table for each division unit. FIG. 14 shows an example of 3/2 times enlargement processing, where n = 2 and m = 3, and there are four types of conversion tables in total.

[0004]

However, in the former method of calculating the density of the pixel of the original image corresponding to each pixel of the converted image, when the converted image becomes large, the number of pixels forming the image becomes large. Since the number of times increases, it takes a lot of time for the coordinate conversion process, and the processing speed decreases. Further, in the latter method of referring to the table storing the density patterns, the conversion table is stored only for the values determined in advance, and the enlargement / reduction processing cannot be performed at an arbitrary magnification. Since there are 2 ⁿ combinations of all n pixel densities, for example, 1.41 = (141 /
When the denominator n becomes large when expressed by a fraction like 100), the table size becomes astronomically large and processing cannot be performed in reality.

Therefore, the present invention has been made in view of the above-mentioned points, and enlarges / multiplies at any magnification in both horizontal and vertical directions.
The purpose is to perform reduction processing at high speed.

[0006]

[0007]

According to the conversion processing method for enlarging or reducing a binary image composed of white pixels and black pixels formed of binary image data according to the present invention, white pixels are arranged in advance in the converted image. White pixel arranging means, calculating means for calculating the vertical coordinates of the original image corresponding to the vertical coordinates forming the converted image, and an area where black pixels exist in the horizontal line of the original image having the calculated vertical coordinates. A black pixel detecting means for detecting a horizontal coordinate after conversion processing corresponding to the horizontal coordinate of the detected black pixel, and the converted image position defined by the horizontal coordinate and the vertical coordinate forming the converted image. And a black pixel arranging unit for arranging black pixels, and the image conversion process for each vertical coordinate of the converted image is repeated for all vertical coordinates.

According to the conversion processing method for enlarging or reducing a binary image composed of white pixels and black pixels formed of binary image data according to the present invention, the black pixel detecting means is
Regarding the black pixels continuously detected from the original image, only the minimum value and the maximum value of the horizontal coordinates of the continuous black pixels are stored, and the black pixel arrangement means corresponds to the stored minimum and maximum coordinates. The horizontal coordinate after conversion is calculated, and black pixels are continuously arranged from the minimum horizontal coordinate after conversion to the maximum horizontal coordinate after conversion.

The binary image enlargement / reduction conversion processing method according to the present invention has a conversion table for converting the horizontal coordinates of the original image into the horizontal coordinates of the converted image, and the conversion processing is performed using the conversion table. It is characterized in that the horizontal coordinates of the subsequent black pixels are calculated.

[0010]

The binary image enlargement / reduction conversion processing method according to the present invention detects the original image detected in the vertical coordinate detection of the original image corresponding to the vertical coordinate Y of the converted image in the vertical enlargement processing. Of the original image corresponding to the vertical image Y-1 of the converted image which has already been subjected to the conversion process, the horizontal line of the converted image having the vertical coordinate Y-1 is converted to the vertical coordinate of It is characterized by having a copying means for copying on the horizontal line of Y.

[0012]

The black pixel region detecting method in the binary image scaling conversion processing method according to the present invention is a collective detecting means for collectively detecting black pixels in a horizontal line of an original image in units of words. All bits are 1 in detected word value
And a skip means for skipping the horizontal coordinates until the next word when it is determined that the detected word does not include white pixels. , Are included.
In other words, when detecting how many consecutive black pixels, word units (8, 16, 32 bits, etc.) are compared together and the word value is $ FF, $ FFFF, $ FFFFFFFF, etc. In some cases all pixels in the word are considered black,
Then, the next word is searched for.

[0014]

[0015]

It is possible to reduce the calculation processing of the horizontal coordinates associated with the enlargement / reduction processing by creating in advance a table for converting the horizontal coordinates of each pixel of the pre-conversion image into the corresponding horizontal coordinates of the post-conversion image. it can. By using the table, the calculation that has to be performed for the number of pixels in the pre-conversion image when the table is not used can be completed by the number of pixels of the horizontal line of the pre-conversion image.

When black pixels are continuous on the horizontal line of the original image, attention is paid only to the minimum value and the maximum value of the continuous area, and the pixels between the corresponding minimum value and maximum value in the converted image are also black pixels. By arranging, it is not necessary to refer to the original image for all the pixels of the pre-conversion image, and the enlargement processing can be performed at high speed.

[0017]

When a binary image is enlarged in the vertical direction, the vertical coordinates of the original image detected corresponding to the vertical coordinates of the converted image are the vertical coordinates of the original image corresponding to the vertical coordinates of the converted image immediately above. If it matches, the horizontal line with the vertical coordinate immediately above is copied. As a result, it is possible to prevent the same horizontal line of the pre-conversion image from being inspected a plurality of times and perform high-speed enlargement processing.

[0019]

When calculating the number of continuous black pixels from the horizontal line in the pre-conversion image, first the pixels are detected in word units, and the word value is compared once with the word in which all bits are 1 The presence of black pixels continuous for words can be detected by the comparison process of 1., and the number of continuous black pixels can be detected at high speed.

[0021]

DETAILED DESCRIPTION OF THE INVENTION Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 1 shows the meanings of variables. IW and IH are the number of pixels (unit: dot) in the horizontal and vertical directions of the image before conversion, and OW and OH are the horizontal and vertical directions of the image after conversion, respectively. The number of pixels (unit: dot). The coordinates of each pixel are 0, 1, 2, ..., Horizontally to the right from the origin at the upper left of the image.
n. From there, it is vertically divided into 0, 1, 2, ..., N. FIG. 2 shows the meanings of the variables a, b, c, d, e, and f, which are reversed from the coordinates (X, Y) in the post-conversion image to the coordinates (x, y) in the pre-conversion image. It is an element of the 3 × 3 matrix to be transformed. For example, when the magnification is doubled in the horizontal and vertical directions, a = e = 0.5 and b = c = d = f = 0.0.
Further, the maximum integer value that does not exceed the value x is represented by int (x).

FIGS. 3 and 4 are flowcharts showing an algorithm for enlarging / reducing a binary image according to the present invention. With reference to this flowchart, the enlargement processing for multiplying the lower right 3 × 3 dot area in the 5 × 5 dot pre-conversion image shown in FIG. 5 by 1.5 times in the horizontal and vertical directions will be described. In this case, the conversion matrix for obtaining the coordinates in the pre-conversion image from the coordinates in the post-conversion image is as shown in FIG. In this case, IW and IH are 5, and OW and OH are 7.

(Creation of conversion table) Step S10
1. In step 1102, first, a conversion table table [x] for converting the horizontal coordinates in the pre-conversion image into the horizontal coordinates in the post-conversion image is defined. An integer array having IW + 1 elements is secured (step S101). In this example,
Since IW is 5, only 5 + 1 = 6 are prepared. Then, for all x satisfying 0 ≦ x ≦ IW, a conversion table having the following values is defined in this example.

Table [x] = int ((x-c) /a+0.5) table [0] = int ((0-2) /0.666+0.5) = -2 table [1] = int ((1- 2) /0.666+0.5) = -1 table [2] = int ((2-2) /0.666+0.5) = 0 table [3] = int ((3-2) /0.666+0.5) = 2 table [ 4] = int ((4-2) /0.666+0.5) = 3 table [5] = int ((5-2) /0.666+0.5) = 5 Here, both variables c and a are converted into 3 × 3. It is an element of the matrix.

In step S103, a portion where the value in the conversion table becomes less than 0 is detected and the value is redefined to 0. Further, the maximum value of the subscript of the table that satisfies this condition is xs. The values in this result table are as follows, xs
← 1. When there is no x satisfying such a condition, xs ← 0.

Table [0] ← 0 table [1] ← 0 table [2] = 0 table [3] = 2 table [4] = 3 table [5] = 5 Here, it is said that the table value is displayed negatively. This means that the image is mapped outside the converted image, and since it does not affect the converted image, it is excluded in advance.

In step S104, a value in the conversion table is detected as OW, that is, a portion where the number of pixels in the converted image is larger than the number of pixels in the horizontal direction is detected, and the value is redefined as OW. Further, the minimum value of the subscript of the table that satisfies this condition is xe. In this example, since there is no element satisfying this condition in the table, the value in the table is unchanged and xe ← IW-1 = 4. The above is the procedure for creating the conversion table. Here, the fact that the table value is larger than OW means that the table value is mapped outside the converted image, and since it does not affect the converted image, it is excluded in advance.

Subsequently, step S105 and step S10.
In 6, the value of the vertical coordinate Y of interest in the converted image is set to 0.
, And the variable prevY that stores the vertical coordinate of the horizontal line in the pre-conversion image referred to in the previous loop is initialized to -1.

(Calculation of Vertical Coordinates of Original Image) Step S1
In 07, the vertical coordinate in the pre-conversion image corresponding to the vertical coordinate Y in the post-conversion image is calculated, and this is set as iy. The calculation formula is

[Formula 1] iy ← int (eY + f).

Here, the variables e and f are both elements of a 3 × 3 conversion matrix. In this example, since Y = 0 has just been initialized, iy ← 2. Since iy is 2, the vertical coordinates of 0 and 1 in the pre-conversion image are not referred to in the enlargement conversion process.

In step S108, it is confirmed whether or not the vertical coordinate iy in the pre-conversion image calculated in step S107 indicates a pixel in the pre-conversion image. In this example, 0 ≦ 2 <5 (IH), so the conditional expression becomes true, and the process of step S111 is continued.

In step S111, it is confirmed whether the horizontal line in the pre-conversion image to be referred to below is the same as the horizontal line already inspected in the previous loop.
In this example, this is the first loop, iy is 0, and prevY is initialized to -1 in step S106, so this conditional expression becomes false, and the process of step S113 is subsequently performed.

(Arrangement of White Pixels) In step S113, all pixels in the converted image whose vertical coordinates are Y are filled with white. In this example, white pixels are arranged for all the pixels (0, 0) to (6, 0) whose vertical coordinates are 0 in the converted image.

In step S114, the vertical coordinate of the horizontal line in the pre-conversion image referred to below is stored in the variable prevY. In this example, prevY ← 2. This work
It is necessary for the determination in step S111.

(Coordinate detection) In step S115, the densities of pixels (xs, iy) to pixels (xe, iy) of the pre-conversion image are inspected in order from the left. Here, (1, 2) to (4,
2). A block area having a single black pixel and a block area having continuous black pixels are detected, and these are detected as B1 and B
2, ..., Bn. In this example, the pixel (2, 2) and the pixel (3, 2) are a block region in which black pixels are continuous, and n
= 1 and B1 = {(2,2), (3,2)}.

In step S116, the value of the variable i used in the loop for the number of blocks in the black pixel area is initialized to 1.

In step S117, it is determined whether or not the value of the variable i is less than or equal to the number n of blocks in the black pixel area.
In this example, i is 1 and is the same as n, so that it becomes true and the process of step S118 is continued.

In step S118, the minimum and maximum values of the horizontal coordinates of the pixels included in Bi are set to Si and Ei, respectively. In this example, B1 = {(2,2), (3,2)}, so Si ← 2 and Ei ← 3.

(Arrangement of Black Pixels) In step S119, pixels (table [Si], Y) to (tab) of the converted image are displayed.
Fill le [Ei + 1] -1, Y) with black. In this example, (table [2], 2) to (table [3 + 1]
-1,2), and (0,2) to (2,2 of the converted image
Fill 2) with black. FIG. 7 shows a state in which black pixels are arranged.

At step S120, the value of the variable i is incremented by 1, and attention is paid to the block Bi in the next black pixel area. In this example, since n = 1, there is only one block, and the conditional expression in step S117 is false, so the processing in step S121 is continued.

In step S121, the value of Y is incremented by 1 in order to process the next vertical coordinate of the converted image. S122
Then, it is determined whether or not the value of Y exceeds OH (the height of the converted image data). If it exceeds, all the vertical coordinates in the converted image have been examined, and the process ends. If it does not exceed as in this example, the process of step S107 is continued.

In step S107, iy for Y = 1
As a result of calculating, iy = 2. The conditional expression in step S108 is true, and the process proceeds to step S111. Since iy is the same as when Y = 0 in the previous loop, step S
The conditional expression 111 is also true. That is, prev of S114
Since Y and iy match, the process of S112 is performed.
In step S112, the horizontal line immediately above the horizontal line of Y = 1, that is, the horizontal line of Y = 0 is copied as it is onto the horizontal line of Y = 1 in the converted image. Therefore, as shown in FIG. 8, horizontal lines having vertical coordinates of 0 and 1 have the same black pixel arrangement. In this way, if the vertical coordinates of the converted image and the vertical coordinates of the original image referred to by the vertical coordinates directly above the converted image match, the horizontal line of the vertical coordinates immediately above is copied, and therefore S113 to S12.
Processing up to 0 can be omitted. Then step S1
In step 21, Y ← 2 is set, and the process of step S107 is performed again for the next vertical coordinate. Below, Y = 2, 3, 4
In the loop, the same processing as the above processing is performed (see FIG. 9).

The processing when Y = 5 and step S107 is reached will be described. Iy in the formula of step S107
← 5, which is not less than IH, so step S10
Conditional expression 8 is false. This means that in the converted image, the horizontal line with the vertical coordinate Y = 5 corresponds to the area outside the unconverted image. Therefore, in step S109, all the pixels of the horizontal line of the vertical coordinate Y = 5 of the converted image are filled with white.

After the above processing is completed, Y ← in step S121.
Since this is 6, the conditional expression in S122 is true, and therefore the process again reaches step S107. In the expression of step S107, iy ← 6, and the conditional expression of step S108 becomes false. In step S109, the vertical coordinate of the converted image Y =
Fill all pixels in the horizontal line of 6 with white.

Subsequently, 1 is added in step S121, and Y is added.
← 7 and this value makes the conditional expression in S122 false, so
All processing ends.

FIG. 12 is a flow chart showing the overall processing procedure of FIGS. 3 and 4 in a simplified manner. 12 and 3
And the corresponding relationship between the steps of FIG. 4 is as follows. S1: S101 to S104, S2: S105 to S106, S3: S107, S4: S108 to S114, S5: S115, S6: S116 to S120, S7: S121, S8: S122.

Next, the densities of pixels (xs, iy) to pixels (xe, iy) of the pre-conversion image in step S115 of FIGS. 3 and 4 are inspected in order from the left, and black pixel areas are detected. The processing of B1, B2, ..., Bn will be described in detail. 10 and 11 are flowcharts showing the process of S115 in more detail. The word length in this device is W (= 8, 16, 32, etc.).

In step S201, xs is set to the work variable X.
Is set, and the value of the variable n representing the number of blocks in the black pixel area is initialized to 0 in step S202.

In step S203, the value of X is checked and X>
If it is xe, it means that it is outside the given area, so this detection processing is terminated. If X ≦ xe, the processing from step S204 is performed to detect a black pixel.

In step S204, it is checked whether xe-X + 1≥W. If it is W or more, the process of step S205 is performed in order to inspect whether or not a black pixel is included in the series of W dots starting from the horizontal coordinate X by one word comparison. If it is W or less, the process proceeds to step S207, and it is checked whether or not each pixel (x, iy) is black.

In step S205, W dots from (X, iy), that is, pixels (X, iy), (X + 1, i).
It is checked whether y), ..., (X + W-1, iy) includes one or more black pixels by comparing the value of the word with 0. For example, if the word value is 0, all W pixels are white pixels. Therefore, the word is skipped and W is added to the value of X in step S206, and the processing from step S203 is repeated. On the other hand, if the word value is not 0, W
Since it means that one pixel includes one or more black pixels, the number of pixels is set to 1 in step S207 and subsequent steps without skipping.
The density of each pixel is inspected pixel by pixel to detect black pixels.

In step S207, it is checked whether or not each pixel (X, iy) is black. If it is not black, the value of X is incremented by 1 in step S208 to continue the inspection of the density of the pixel on the right. This process is repeated until black pixels are found. If black pixels are found here, the number of black pixels in succession is checked by continuing the processing from step S209.

In step S209, the value of X at which a black pixel is found is set as the minimum horizontal coordinate S of the black pixel area. In step S210, the value of X is increased by 1 to continue the density comparison of the pixel on the right side.

In step S211, it is checked whether the value of X is larger than xe. If X> xe, it means that the maximum horizontal coordinate E of the black pixel area is xe, and therefore the value xe is set to the maximum horizontal coordinate value E in step S212. If X ≦ xe, step S21
The processes after 3 are continuously performed.

In step S213, it is checked whether xe-X + 1≥W. If it is W or more, the process of step S214 is performed to inspect whether or not a series of black pixels for W dots starting from the horizontal coordinate X are continuous by one word comparison. If it is W or less, it is checked in step S216 whether each pixel is a white pixel.

In step S214, W dots from (X, iy), that is, pixels (X, iy), (X + 1, i).
y), ..., (X + w-1, iy) includes one or more white pixels, whether the value of the word and all the bits in the word are 1, that is, $ FF, $ FFFF, $ FFFFFFFF, etc. Check by comparing. If all bits in the word are 1, then W pixels are all black pixels. Therefore, the word is skipped and W is added to the value of X in step S215, and the processing from step S211 is repeated. On the other hand, the fact that all the bits in the word do not become 1 means that one or more white pixels are included in the W number of pixels, and therefore the pixel density is inspected pixel by pixel in the processing from step S216.

In step S216, it is checked whether or not the pixel (X, iy) is white. If it is not white, the density of the pixel on the right is further inspected by increasing the value of X by 1 in step S220. If the pixel (X, iy) is a white pixel, it can be seen that the maximum horizontal coordinate of the black pixel area is X-1, so in step S217, the maximum horizontal coordinate value E is set to X-1, and subsequently step S218. Is processed.

In step S218, the value of the variable n representing the number of blocks in the black pixel area is incremented by 1, and in step S219, the minimum and maximum values of the horizontal coordinates of the pixels contained in Bn are set to Sn and En. And step S2
Returning to 03, the next black pixel area is inspected. S
When the black pixels obtained here are not continuous, n and En have matching values.

[0059]

As described above, according to the conversion processing method of the present invention for enlarging or reducing, the converted image is filled with white pixels in advance, black pixels are detected in the image before conversion, and black pixels are found. Since the corresponding pixels of the converted image are filled with black only when the conversion is performed, efficient enlargement / reduction processing can be performed on binary image data having a small number of black pixels, such as CAD drawing or document data. In fact, in the actual image data as described above, the ratio of black pixels to the whole pixels is almost several percent or less.

When black pixels are continuous on the horizontal line of the original image, attention is paid only to the minimum value and the maximum value of the continuous area, and the pixels between the corresponding minimum value and maximum value in the converted image are also black pixels. By arranging, it is not necessary to refer to the original image for all the pixels of the pre-conversion image, and the enlargement processing can be performed at high speed.

Since the conversion table for converting the horizontal coordinate value in the pre-conversion image into the horizontal coordinate value in the post-conversion image is used, it is possible to reduce the calculation amount associated with the conversion of the horizontal coordinate value.

The storage capacity required for the conversion table is equal to the capacity for storing an integer value of +1 for the number of pixels in the horizontal direction of the pre-conversion image, which is a storage capacity that can be practically sufficiently secured.

In the present invention, of the pixels in the image before conversion, those that do not affect the image after conversion are not inspected for their densities.
As a result, useless density reference processing can be avoided and processing speed performance can be improved.

When the image is enlarged in the vertical direction, one horizontal line in the pre-conversion image may be enlarged in a plurality of horizontal lines in the post-conversion image. In this case, in the present invention, when the horizontal line of the post-conversion image is constructed, when the same horizontal line as the immediately preceding horizontal line is referred to,
It was decided to copy the horizontal line directly above, avoiding reference again. This makes it possible to avoid referring to the densities of pixels on the same horizontal line of the pre-conversion image multiple times.
It is possible to improve the speed performance by eliminating unnecessary references.

In the present invention, when the black pixel area is detected from the pre-conversion image, it is inspected whether all the plurality of pixels are white by comparing the value of the word with 0. This allows
As described above, the number of black pixels in the CAD drawing or the document data is relatively small, and the black pixels are not randomly arranged in many cases but are often arranged locally. It is possible to skip the area at high speed.

According to the present invention, when a black area is searched for from a pre-conversion image, it is determined whether or not a plurality of pixels are all black by comparing the value of the word with the value of the word in which all bits are 1. Was inspected. Since black pixels are often arranged locally as described above, a plurality of black pixels can be recognized by one comparison operation, and the number of consecutive black pixels can be calculated at high speed. Is possible.

[Brief description of drawings]

FIG. 1 shows variables IW and I according to an embodiment of the present invention.
It is a figure showing the meaning of H, OW, and OH.

FIG. 2 shows variables a, b, and
It is a figure showing the meaning of c, d, e, and f.

FIG. 3 is a flowchart of enlargement processing according to the embodiment of the present invention.

FIG. 4 is a flowchart of enlargement processing according to the embodiment of the present invention.

FIG. 5 illustrates an example of an original image before enlarging processing according to the embodiment of the present invention.

FIG. 6 shows a transformation matrix used in an example of enlargement processing in the embodiment of the present invention.

FIG. 7 shows an example of enlargement processing in the embodiment of the present invention, which shows a converted image after processing up to a horizontal line with a vertical coordinate of 0.

FIG. 8 illustrates an example of enlargement processing according to an embodiment of the present invention, which shows a converted image after processing up to a horizontal line with a vertical coordinate of 1.

FIG. 9 shows an example of enlargement processing in the embodiment of the present invention, which shows a converted image after processing up to a horizontal line of vertical coordinates 4.

FIG. 10 is a flowchart of processing for detecting a black pixel area in a certain horizontal line in the embodiment of the present invention.

FIG. 11 is a flowchart of a process of detecting a black pixel area in a horizontal line according to the embodiment of the present invention.

FIG. 12 is a flowchart showing the processing of FIGS. 3 and 4 according to the embodiment of the present invention in a simplified manner.

FIG. 13 is a diagram showing mapping of pixels of a post-conversion image and pixels of a pre-conversion image in a conventional technique.

FIG. 14 is a diagram for explaining a process of performing 1.5 times enlargement in the conventional technique.

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 1/38-1/393

Claims (5)

(57) [Claims] 1. A white pixel arrangement means for arranging white pixels in a converted image in advance, a calculation means for calculating vertical coordinates of an original image corresponding to vertical coordinates forming a converted image, and the calculated vertical coordinates. Black pixel detection means for detecting an area in which a black pixel exists in the horizontal line of the original image, and horizontal coordinates after conversion processing corresponding to the horizontal coordinates of the detected black pixel are calculated. A black pixel arranging means for arranging black pixels at the post-conversion image position defined by the vertical coordinates forming the image, and repeating the image conversion process for each vertical coordinate of the post-conversion image for all the vertical coordinates. A binary image enlarging / reducing conversion processing method consisting of white pixels and black pixels formed of characteristic binary image data. 2. The black pixel detecting means stores only the minimum value and the maximum value among the horizontal coordinates of the continuous black pixels for the black pixels continuously detected from the original image, and the black pixel arranging means 2. The converted horizontal coordinates corresponding to the stored minimum and maximum coordinates are calculated, and black pixels are continuously arranged from the converted minimum horizontal coordinate to the maximum horizontal coordinate. A binary image enlargement / reduction conversion processing method consisting of white pixels and black pixels formed of binary image data. 3. A conversion table for converting the horizontal coordinate of the original image into the horizontal coordinate of the converted image, wherein the horizontal coordinate of the black pixel after the conversion processing is calculated using the conversion table. 3. A scaling conversion processing method for a binary image formed of the binary image data according to claim 1, comprising a white pixel and a black pixel. 4. In the detection of the vertical coordinates of the original image corresponding to the vertical coordinates Y of the converted image in the enlargement processing in the vertical direction, the detected vertical coordinates of the original image correspond to those of the converted image which has already been subjected to the conversion processing. The image forming apparatus further comprises copying means for copying a horizontal line of the converted image having the vertical coordinate Y-1 to a horizontal line of the vertical coordinate Y when the vertical coordinate of the original image corresponds to the vertical coordinate Y-1. 6. A scaling conversion processing method for a binary image formed of the binary image data according to claim 1, comprising a white pixel and a black pixel. 5. A conversion processing method for enlarging or reducing a binary image composed of white pixels and black pixels formed of binary image data, wherein detection of black pixels in a horizontal line of an original image is performed in units of words. Collectively detecting means, and a judging means for judging the presence or absence of a white pixel in the word by comparing the detected word value with a word in which all the bits are 1, and the detected word contains a white pixel. A black pixel area detection method comprising: a skip means for skipping the horizontal coordinate to the next word when it is determined that the black pixel area is not present.