JPH08125843A - Image processor - Google Patents

Abstract

PURPOSE: To provide an image processor with which an entire hardware amount is reduced by making most of a hardware into a single common circuit not depending on resolution. CONSTITUTION: An input image is fetched into shift registers 63 to 79 and arithmetic parts 81-95 calculate the correcting amount of dot addition/erasure for performing smoothing processing to the edges of lateral lines and longitudinal lines in the input image. While output resolution 300dpi is selected, that image is transmitted to an output deciding arithmetic part 103 as it is, and the output image of horizontal 1200dpi x vertical 300dpi is provided. While output resolution 600dpi is selected, among the correcting amounts for output resolution 300dpi, the correcting amount on the longitudinal lines is transmitted to the output deciding arithmetic part 103 as it is but the correcting amount on the lateral lines is inputted to a resolution converting part 97 and converted to the correcting amount for output resolution 600dpi, and the output image of horizontal 1200dpi × vertical 600dpi is provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for converting an input image having a certain resolution into a plurality of output images having different resolutions in which edges of a pattern are smoothed.

[0002]

2. Description of the Related Art As this type of conventional technique, for example, an image processing apparatus disclosed in Japanese Patent Laid-Open No. 177867/1993 is known. This conventional device has a resolution of an output image of 300 dpi (dot / inch) and 6 in a laser beam printer, for example.
It is used for switching to 00 dpi, and includes two interpolation logic circuits connected between the printer controller and the printer engine, and one of the two interpolation logic circuits is selected by a signal for selecting the resolution of the printer engine. It is designed to work selectively. And
One of the two interpolation logic circuits inputs a 300 × 300 dpi image from the printer controller and smoothes the edges of the pattern of the input image to 1200 ×
It has a function of converting to an output image of 300 dpi. The other interpolation logic circuit doubles the resolution of the same input image and smooths the pattern edge to 1200
It has a function of converting to an output image of × 600 dpi.

[0003]

One of the problems of this conventional apparatus is that the number of interpolation logic circuits corresponding to the number of types of resolutions that can be selected is required, so that the amount of hardware increases and the cost increases. Is.

Therefore, an object of the present invention is to provide an image processing apparatus for converting an input image of a certain resolution into an output image of a plurality of different resolutions, and most of the hardware thereof is a single common circuit independent of the resolution. By
The goal is to reduce the total amount of hardware.

[0005]

According to the present invention, in an image processing apparatus for converting an input image of a certain resolution into a plurality of smoothed output images of different resolutions, the input image is received and smoothed. A correction amount calculation means for calculating a first edge correction amount to be applied to the input image to obtain an output image of the first resolution, and a second resolution smoothed by receiving the first edge correction amount. Resolution conversion means for converting into a second edge correction amount to be applied to the input image to obtain the output image, and one of the first edge correction amount and the second edge correction amount is selectively input and input. And an output image creating means for creating an output image having a corresponding resolution by performing a process for applying the edge correction amount to the input image.

[0006]

The image processing apparatus of the present invention has the first resolution and the second resolution.
It is possible to selectively output an output image having at least two types of resolutions. This apparatus includes three components, a correction amount calculation means, a resolution conversion means, and an output image creation means.

The correction amount calculation means performs a smoothing process for converting an input image into an output image having a first resolution, and calculates a first edge correction amount for smoothing edges of a pattern of the input image. calculate.

On the other hand, the resolution conversion means inputs the first edge correction amount and converts it into a second edge correction amount for obtaining an output image of the second resolution.

One of the first edge correction amount and the second edge correction amount is selectively input to the output image creating means. The output image creating means performs a process for applying the input edge correction amount to the input image. Therefore, when the first edge correction amount is input, the output image of the first resolution is created, and when the second edge correction amount is input, the second edge correction amount is input.
An output image with a resolution of is created.

In this apparatus, the correction amount calculating means and the output image creating means are commonly used through the first and second output resolutions. Only the resolution conversion means is dedicated to the second resolution. Therefore, the increase in the amount of hardware due to the increase in the selectable resolution is only for the resolution converting means, and therefore the total amount of hardware is smaller than that in the conventional technique.

The preferred embodiment has an output resolution of 300.
Although dpi and 600 dpi can be selected, the ratio of the resolution conversion unit corresponding to the above-mentioned resolution conversion unit in the entire device is very small. Therefore, the hardware amount of the device of this embodiment is not much different from the hardware amount of the conventional device which outputs only 300 dpi.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows a hardware configuration of an embodiment of a control device for a laser printer to which the present invention is applied.

The control device 1 receives print data from an external device 3 such as a personal computer, creates bitmap data of an output image based on the print data, and uses the bitmap data for actual image formation. It outputs to the laser beam modulation circuit 5 for performing.
Here, the resolution of the laser beam modulation circuit 5 (hereinafter referred to as output resolution) can be selected from, for example, two types of 300 dpi and 600 dpi, and a bit created by the control device 1 corresponding to the selected output resolution. The map data resolution is also changed. One feature of the present invention is embodied in a configuration for changing the resolution of the bitmap data, specifically in a bitmap data generation circuit 27 described later.

The control unit 1 includes a central processing unit (CPU) 11 for processing print data, a RAM 13 used as a main memory, a work area, a reception buffer, etc.
The ROM 15 that stores the operation program of the PU 11, the character generator 17 that incorporates the font data for converting the character code in the print data into the bit map data of the character, and the image of the page unit generated based on the print data A page frame buffer 19 for storing bitmap data is provided. C
The PU 11 operates in synchronization with the clock signal from the clock signal generation circuit 21.

The print data received from the external device 3 is
It is taken into the bus 25 via the interface circuit 23, processed by the CPU 11, converted into basic bitmap data for modulating the laser beam, and stored in the page frame buffer 19. The resolution of this basic bitmap data is 300 dpi in the horizontal direction.
The vertical direction is 300 dpi.

In the following description, the resolution is, for example, "12".
It is expressed as “00 dpi × 300 dpi”, where the number before “x” indicates the horizontal resolution and the number after the “×” indicates the vertical resolution.

The basic bitmap data of 300 dpi × 300 dpi is read into the bitmap data generating circuit 27, where it is subjected to processing for smoothing the edges of the pattern (hereinafter referred to as smoothing processing) and finally. Is converted to simple bitmap data. This final bitmap data is created at the resolution corresponding to the selected output resolution, as described above.
That is, when the output resolution is 300 dpi, the final bitmap data resolution is 1200 dpi × 300.
If the output resolution is 600 dpi and the output resolution is 600 dpi, it is 1200 dpi × 600 dpi. The horizontal resolution of 1200 dpi is the result of the smoothing process. That is, in the smoothing process, a dot having a width corresponding to one-fourth of one pixel of basic bitmap data is added to or deleted from the step-shaped portion of the edge of the pattern to smooth the edge without any step. It is to be modified so that it can be seen.

The final bit map data is output to the laser beam modulator 5 to modulate the laser beam on / off. Further, the CPU 11 controls the drive of the photosensitive drum 7 via the bit shift circuit 29 and the address circuit 31, and controls the rotation of the photosensitive drum 1 so as to be suitable for writing the bitmap data. As a result, the laser beam draws an electrostatic latent image corresponding to the final bitmap data on the surface of the photosensitive drum 7.

Further, the CPU 11 has an I / O control circuit 33.
Various control signals are sent to the bit map data generating circuit 27 via. Further, the laser beam modulation circuit 5 also sends necessary control signals to the bitmap data generation circuit 27. The control signals are, for example, a video clock signal synchronized with a laser beam on / off modulation period, a horizontal synchronization signal synchronized with a laser beam horizontal scanning period, and a resolution selection signal for selecting an output resolution. and so on.

Of the above-mentioned constitutions, one feature of the present invention is embodied, and the bit map data generating circuit 27.
Will be described in detail below.

FIG. 2 shows a bit map data generating circuit 27.
The internal configuration of is shown.

In FIG. 2, from a page frame buffer 19 (not shown), basic bitmap data of 300 dpi × 300 dpi starts from the pixel at the left end of the top horizontal line of the page, and moves each horizontal line from left to right. It is read out at a fixed timing by the raster scan method in which the upper horizontal line is moved to the lower horizontal line. Each read pixel data of one horizontal line is written in the RAM 43 through the latch 41.

The RAM 43 has nine FIFs each having a capacity capable of holding bitmap data for one horizontal line.
The configuration is equivalent to the parallel arrangement of O memories, and the data of one horizontal line from the above-mentioned latch 41 is written in the first FIFO memory. In addition, data of 8 horizontal lines from 8 latches 45 to 59 described later is stored in the RAM.
They are written in the second to ninth FIFO memories 43, respectively. Therefore, the RAM 43 stores the bitmap data for nine consecutive horizontal lines.

From the RAM 43, the stored 9
The data of the nine pixels located at the same horizontal position that is read first in the horizontal line of the book is read in synchronization. The pixel data at the same position on these 9 horizontal lines is 9
It is input to nine nine-stage shift registers 63 to 79 through the individual latches 45 to 61, respectively. The shift registers 63 to 79 sequentially shift the pixel data of the horizontal line input to each to the subsequent stage. Therefore, the shift register 63 to 79 holds the bitmap data of the area of 9 pixels × 9 pixels, and further, the 9 pixels × 9 pixels
The area of 9 pixels is horizontally shifted by one pixel at a constant timing, and when the horizontal shift reaches the end of the horizontal line, the area is vertically shifted by one horizontal line and the same horizontal shift is repeated again. In other words, 300 dpi × 300 dpi stored in the page frame buffer
One page of the basic bitmap data of i is scanned in the horizontal and vertical directions by a window of 9 pixels × 9 pixels made by the shift registers 63 to 79 (hereinafter, the above window is scanned). Windows).

9 pixels in this scanning window ×
The 9-pixel bitmap data is stored in the horizontal edge calculator 81.
And the vertical line edge calculation unit 83. The horizontal line edge calculation unit 81 calculates the vertical differentiation (= black-and-white change) of the input 9 pixel × 9 pixel bitmap data to obtain the horizontal line (= horizontal component of the black pattern) present in this scanning window. ) Edge (= contour) is detected. That is, the black pixel at the position where the vertical differential rises (= changes from white to black) is used as the rising edge (= upper edge) of the horizontal line and the position where it falls (= changes from black to white). The black pixel of is detected as the falling edge (= lower edge) of the horizontal line.

On the other hand, the vertical line edge calculation unit 83 calculates the differential in the horizontal direction of the input 9 pixel × 9 pixel bitmap data to obtain a vertical line (= vertical line) of the black pattern existing in this scanning window. The edge of the (directional component) is detected. That is, the black pixel at the position where the differential in the horizontal direction has risen is represented by the rising edge of the vertical line (=
The black pixel at the falling position is detected as the falling edge (= right edge) of the vertical line.

The data indicating the rising edge and the falling edge of the horizontal line and the vertical line detected by the horizontal line edge calculation unit 81 and the vertical line edge calculation unit 83 are the horizontal line edge feature extraction calculation unit 85 and the vertical line edge feature calculation unit 87. It is input respectively.

The horizontal line edge feature calculation unit 85 determines the relative positions of the detected rising edge and falling edge of the horizontal line with respect to the pixel (hereinafter referred to as the target pixel) DX located at the center of the scanning window. Focusing on whether or not there is any, the rising edge and the falling edge are classified. That is, regarding the rising edge (= upper edge), first, from among all the rising edges, a predetermined positional relationship (for example, a position substantially having the meaning of the upper edge with respect to the target pixel DX) is given.
Only the edges in the mask area shown in FIG. 3) are extracted,
The pixel forming the extracted upper edge is the target pixel DX.
From the viewpoint of the number of pixels above or below and the number of pixels to the left or right, the data AU, BU, CU, DU, FU, GU characterizing the shape of the upper edge of the data.
Is calculated. Also, falling edge (= lower edge)
Also, first, only the edge having a predetermined positional relationship (for example, in the mask area shown in FIG. 4) substantially having the meaning of the lower edge with respect to the pixel of interest DX is extracted, and the extracted lower edge is extracted. Data AD, BD, and CD characterizing the shape of the lower edge of the pixel from the viewpoint of the number of pixels above or below the pixel of interest DX to be positioned and to the left or right of which pixel. Calculate DD, FD, GD.

Here, the shape feature data A of the upper edge
The meanings of U, BU, CU, DU, FU, GU and the shape feature data AD, BD, CD, DD, FD, GD of the lower edge will be supplementarily described as follows. That is, FIG.
As shown in FIG. 4 and FIG.
A, B, depending on the number of pixels (distance) in the horizontal direction from
It is divided into C, D, F, and G pixel columns. Then, regarding the upper edge, the pixel row A in the mask area of FIG.
Each of the data AU, BU, CU, DU, FU, and GU (4 bit data) is generated according to a predetermined logical expression for shape feature extraction of the upper edge according to each black-and-white pattern of B, C, D, F, G. Can be decided Also for the lower edge,
Data AD, BD, CD, D according to a predetermined logical expression for shape feature extraction of the lower edge according to the black-and-white pattern of each of the pixel rows A, B, C, D, F, G in the mask area of FIG.
Each of D, FD and GD (4 bit data) is determined.

On the other hand, the vertical line edge feature calculation unit 87 pays attention to the relative positions of the rising edge and the falling edge of the detected vertical line with respect to the target pixel DX. The rising edge and the falling edge are classified. That is, regarding the rising edge (= left side edge), first, from among all the rising edges, a predetermined positional relationship substantially having the meaning of a left side edge with respect to the target pixel DX (for example, in the mask area shown in FIG. 5). ), The left side is extracted from the viewpoint that the pixels forming the extracted left side edge are positioned above or below the pixel of interest DX and at which pixel to the left or right. Data AL, BL, CL, DL, EL characterizing the shape of the edge are calculated. As for the falling edge (= right edge), first, only the edge having a predetermined positional relationship (for example, in the mask area shown in FIG. 5) substantially having the meaning of the right edge with respect to the target pixel DX is extracted. , The data AR that characterizes the shape of the extracted right edge from the viewpoint of which pixel is located above or below the pixel of interest and which pixel is located to the left or to the right. , B
Calculate R, CR, DR, ER.

Here, the shape feature data A of the left edge
The meanings of L, BL, CL, DL, EL and the shape feature data AR, BR, CR, DR, ER of the right edge will be supplementarily described as follows. That is, as shown in FIG. 5, the mask area is divided into A, B, C, D, and E pixel rows according to the number of pixels (distance) in the vertical direction from the target pixel DX. As for the left edge, FIG.
Data AL, BL, CL, DL, EL according to a predetermined logical expression for shape feature extraction of the left edge in accordance with the black-and-white pattern of each of the pixel rows A, B, C, D, E in the mask area of
Each (AL and CL are 4-bit data, BL and DL are 6
Bit data and EL are 2-bit data).
As for the right side edge, the data AR, BR are also calculated according to a predetermined logical expression for extracting the shape feature of the right side edge in accordance with the black and white pattern of each of the pixel rows A, B, C, D, E in the mask area of FIG. , CR, DR, ER (AR and C
R is 4-bit data, BR and DR are 6-bit data, E
R is 2-bit data.

The data characterizing the shapes of the rising edges and the falling edges of the horizontal lines and the vertical lines calculated by the edge feature extraction calculation units 85 and 87 are the upper edge correction amount calculation unit 89 and the lower edge correction amount calculation, respectively. It is input to the unit 91, the left edge correction amount calculation unit 93, and the right edge correction amount calculation unit 95.

The upper edge correction amount calculation unit 89 calculates the characteristic data AU, BU, CU, DU, FU of the upper edge of the horizontal line.
Based on the GU, the dot addition amount and the shaving amount for the smoothing process for the upper edge are calculated. Further, the lower edge correction amount calculation unit 91, based on the feature data AD, BD, CD, DD, FD, and GD of the lower edge of the horizontal line, adds the dot addition amount and the dot addition amount for smoothing processing to the lower edge. Calculate the shaving amount. FIG. 6 shows an example of the calculation result of the dot addition amount and the shaving amount for this horizontal line. When the original horizontal line pattern as shown in the upper part of the figure is given, the upper edge and the lower edge of this original pattern are given. Dots of 1/4 size of one pixel (hereinafter referred to as 1/4 dot) are added to the step portion of the side edge,
When it is deleted, the step is corrected so that it looks smooth.

On the other hand, the left-side edge correction amount calculation unit 93 calculates the characteristic data AL, BL, CL, DL of the left-side edge of the vertical line.
Based on EL, the dot correction amount for the smoothing process for the left edge is calculated. Further, the right edge correction amount calculation unit 95 calculates the dot correction amount for the smoothing process on the right edge based on the feature data AR, BR, CR, DR, ER of the right edge of the vertical line. FIG. 7 shows an example of the calculation result of the correction amount for this vertical line. When the original vertical line pattern as shown in the left part of the figure is given, the positions of the pixels forming this original pattern are determined. The left edge and the right edge are corrected so as to be shifted in the unit of 1/4 dot to the left and right, and the step of the original pattern is corrected so that it looks smooth.

As is clear from FIGS. 6 and 7, the correction amounts for the edges of the horizontal and vertical lines calculated as described above are obtained by converting the basic bitmap data of 300 dpi × 300 dpi to the final bitmap of 1200 dpi × 300 dpi. It is a correction amount for converting into a typical bitmap data, that is, a correction amount for smoothing processing when the output resolution is 300 dpi.

It should be noted that the above-mentioned edge calculators 81 and 83 for calculating the correction amount for the output resolution of 300 dpi,
Edge feature extraction calculation units 85 and 87 and correction amount calculation unit 89
Japanese Patent Laid-Open No. 5-6495
No. 23 (see, in particular, FIG. 46 of the publication).
See the description corresponding to). Although the publication describes the processing by software, the content of the processing is NOT.
This is a simple logical operation in which AND and AND are combined, and this is realized by a hard logic circuit in this embodiment.

Now, of the correction amounts for the output resolution 300 dpi calculated by the above configuration, the correction amount for the vertical line is directly input to the output determination calculation unit 103. on the other hand,
The addition amount and the shaving amount regarding the horizontal line are input to the resolution conversion unit 97, which will be described later, on the one hand, and to the selector 101 together with the output data of the resolution conversion unit 97 on the other hand. here,
The resolution conversion unit 97 has an output resolution of 300d regarding horizontal lines.
The addition amount and the shaving amount for pi are converted into those for 600 dpi. When the output resolution 300 dpi is selected, the selector 101 sends the horizontal line correction amount for 300 dpi to the output determination calculation unit 103 when the output resolution 300 dpi is selected, and when the output resolution 600 dpi is selected, , The horizontal line correction amount for the output resolution 600 dpi from the resolution conversion unit 97 is output determination calculation unit 10
Send to 3.

The output determination calculation section 103 creates final smoothed bitmap data based on the input horizontal line correction amount and vertical line correction amount. The specific processing contents of the output determination calculation unit 103 are also described in the above-mentioned Japanese Patent Laid-Open No.
It is disclosed in detail in No. 64923.

One of the new configurations in the present embodiment is that the resolution conversion unit 97 causes the above-mentioned output resolution of 300 dp.
The point is that the horizontal line correction amount for i is converted into the correction amount for output resolution 600 dpi.

FIG. 8 shows an example of the conversion. If a correction amount for the horizontal line for output resolution 300 dpi as shown in the upper part of FIG. 8 is given, this is converted into a correction amount for output resolution 600 dpi as shown in the lower part of FIG. The law of this conversion is as follows. That is,
As shown in FIG. 8, one horizontal line at 300 dpi is divided into two upper and lower horizontal lines at 600 dpi. Then, the 1/4 dot added at 300 dpi is invalid (= not added) in the upper line of the upper edge, valid (= added) in the lower line of the upper edge, and in the upper line of the lower edge at 600 dpi. Effectiveness,
And, it is invalid in the lower line of the lower edge. Also, 3
The 1/4 dot deleted at 00 dpi is 600 dpi.
, The upper line of the upper edge is valid (= delete), the lower line of the upper edge is invalid (= not deleted), the upper line of the lower edge is invalid, and the lower line of the lower edge is valid. .

The resolution conversion unit 97 calculates the addition amount and the deletion amount of the horizontal line for the output resolution of 600 dpi according to this law. The basic logic is shown below.

ADD = addU.LIN2 + addD.LIN1 (1) DEL = delU.LIN1 + delD.LIN2 (2) PIC = pic + delU.LIN2 + delD.LIN1 (3) where ADD, DEL and PIC are , 1/4 dot addition at 600 dpi, 1/4 dot shaving, and 1 pixel perfect printing, respectively. Also, addU, addD, d
elU, delD, and pic are 1/4 dot addition of the upper edge, 1/4 dot addition of the lower edge, 1/4 dot scraping of the upper edge, and 1 / of the lower edge at 300 dpi.
4 dot scraping and 1 pixel complete printing are meant respectively.
Further, LIN1 and LIN2 are line signals that mean an upper line and a lower line at 600 dpi, respectively. The line signals LIN1 and LIN2 are 600d
It is generated by a 1-bit counter 99 which counts the horizontal synchronizing signal at pi.

The logic of the above equation (1) is that at 300 dpi, only when 1/4 dots are added on the lower line of the upper edge and when 1/4 dots are added on the upper line of the lower edge, 600 dpi is applied. This means that 1/4 dot addition is performed. Further, the logic of the equation (2) is 1/600 at 600 dpi only when 1/4 dot is deleted on the upper line of the upper edge and at 1/4 dot on the lower line of the lower edge at 300 dpi. This means that 4 dots are deleted. Further, the logic of the equation (3) is 1 / when the one pixel is completely printed at 300 dpi and the lower line of the upper edge.
1/4 when deleting 4 dots and on the upper line of the lower edge
This means that 1-pixel complete printing at 600 dpi is performed only when dots are deleted. According to this logic, the horizontal line correction amount for 300 dpi as shown in FIG.
Conversion to a horizontal line correction amount for 0 dpi is possible.

Incidentally, 300 input to the resolution conversion unit 97
The 1 / 4-dot addition amount and deletion amount data for dpi indicates not only information indicating simply addition or deletion but also the position and number of 1/4 dots to be added or deleted in the pixel. Information is also included. This position and number information is not considered in the basic logic described above,
This is also taken into consideration in the actual logical operation in the resolution conversion unit 97. That is, for example, as in the example of FIG. 8, the number and position of addition or deletion dots at 300 dpi are reflected as they are at pixels where addition or deletion is effective even at 600 dpi.

As described above, the resolution conversion unit 97 converts the horizontal line correction amount for the output resolution 300 dpi to the output resolution 6
By converting to the horizontal line correction amount for 00 dpi, it is possible to support output resolution of 600 dpi by utilizing the hardware resource for output resolution of 300 dpi. Therefore, the configuration is such that a separate correction circuit is provided according to the conventional resolution. The total amount of hardware is greatly reduced in comparison.

[0046]

As described above, according to the present invention,
In an image processing apparatus for converting an input image having a certain resolution into output images having a plurality of different resolutions, it is possible to reduce the total amount of hardware.

[Brief description of drawings]

FIG. 1 is a block diagram showing a hardware configuration of an embodiment of a control circuit of a laser printer to which the present invention is applied.

FIG. 2 is a block diagram showing a detailed hardware configuration of a bitmap data generation circuit.

FIG. 3 is a diagram showing an example of a region mask used when extracting a feature of an upper edge of a horizontal line.

FIG. 4 is a diagram showing an example of a region mask used when extracting a feature of a lower edge of a horizontal line.

FIG. 5 is a diagram showing an example of a region mask used when extracting a feature of an edge of a vertical line.

FIG. 6 is a diagram showing an example of horizontal line correction.

FIG. 7 is a diagram showing an example of vertical line correction.

FIG. 8 is a diagram showing an example in which a horizontal line correction amount for 300 dpi is converted to a horizontal line correction amount for 600 dpi.

[Explanation of symbols]

27 Bit map data generation circuit 41, 45, 47, 49, 51, 53, 55, 57, 5
9, 61 Latch 43 RAM 63, 65, 67, 69, 71, 73, 75, 79 Shift register 81 Horizontal line edge calculation unit 83 Vertical line edge calculation unit 85 Horizontal line edge feature extraction calculation unit 87 Vertical line edge feature extraction calculation unit 89 Upper edge correction amount calculation unit 91 Lower edge correction amount calculation unit 93 Left edge correction amount calculation unit 95 Right edge correction amount calculation unit 97 Resolution conversion unit 99 1-bit counter 101 Selector 103 Output determination calculation unit

Claims (3)

[Claims] 1. An image processing device for converting an input image of a certain resolution into a plurality of smoothed output images of different resolutions, wherein the input image is received and an output of a smoothed first resolution is received. A correction amount calculation means for calculating a first edge correction amount to be applied to the input image to obtain an image, and an output image of a second resolution that has been smoothed by receiving the first edge correction amount. A resolution conversion means for converting into a second edge correction amount to be applied to the input image in order to selectively input one of the first correction amount and the second correction amount, and input the edge correction amount. And an output image creating unit that creates an output image of a corresponding resolution by performing a process for applying the above-mentioned input image to the image processing apparatus. 2. The image processing apparatus according to claim 1, wherein the correction amount calculation means detects a horizontal line edge and a vertical line edge from the input image, and increases the horizontal resolution of the input image. To obtain the output image of the first resolution, the first horizontal line correction amount and the vertical line correction amount to be applied to the horizontal line edge and the vertical line edge, respectively. Calculating means based on the edge of a vertical line, wherein the resolution converting means obtains the output image of the second resolution by increasing the vertical resolution of the first output image, The second horizontal line correction amount to be applied to the edge of the horizontal line is the first horizontal line from the correction amount calculation means.
Is calculated based on the horizontal line correction amount, and the output image creating means selectively receives one of the first and second horizontal line correction amounts and receives the vertical line correction amount to obtain the first horizontal line correction amount. An output image of the first resolution is created from the amount and the vertical line correction amount, and an output image of the second resolution is created from the second horizontal line correction amount and the vertical line correction amount. A characteristic image processing device. 3. The image processing device according to claim 2, wherein the output image of the second resolution is an integral multiple higher than the first resolution in the vertical resolution, and therefore,
A plurality of horizontal lines corresponding to the respective horizontal lines of the output image of the first resolution, wherein the resolution conversion means has a plurality of horizontal lines for each horizontal line of the output image of the first resolution. A means for generating a signal for distinguishing lines and the first horizontal line correction amount for each horizontal line of the output image of the first resolution are made effective according to the signal for distinguishing the plurality of horizontal lines. An image processing apparatus comprising: means for obtaining the second horizontal line correction amount by selecting whether to invalidate or to invalidate.