JPH08256259A - Image processor and its method - Google Patents

Abstract

PURPOSE: To attain efficient smoothing processing with a small matrix even when interpolation is made for resolution conversion processing or magnification processing by interpolating smoothed image data and smoothing the interpolated data. CONSTITUTION: Image data outputted from a read section are given to a 1st matrix 42 via an input terminal 41 of a resolution conversion smoothing processing section 4, and every time image data of one picture element are received, data of a notice picture element 42e are subject to smoothing processing and the result is given to a main scanning direction interpolation processing section 43. Then the image data outputted from the main scanning interpolation processing section 43 are given to a 2nd matrix 45 via a switch 44, and the 2nd matrix 45 comprising 3-picture element × 3-picture element similarly to the case with the 1st matrix 42 applies smoothing processing to data of the notice picture element 45e based on data of surrounding 8-picture elements 45a-45d, 45f-45j and the result is outputted from an output terminal 46. Thus an excellent smoothing processing result is obtained even by a small matrix of 3-picture element × 3-picture element.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method. More specifically, the present invention relates to an image processing apparatus and method having a resolution (pixel density) conversion function or a magnification function.

[0002]

2. Description of the Related Art In image processing apparatuses such as facsimiles and printers, resolution conversion processing is performed when the resolution of a recording apparatus is larger than the resolution of a received original document. For example, the received original is in normal mode (main scanning direction 8 dot / mm, sub-scanning direction 3.85 line / mm), and the recording device is in fine mode (main scanning direction 8 dot / mm, sub-scanning direction 7.7 line / mm). If it has the recording capability of, the resolution conversion processing in the sub-scanning direction is performed. The resolution conversion process in this case is performed by recording the received 1-line data twice.

Recently, with the improvement of the capacity of recording devices,
A high-resolution recording device of 0 dpi (dot per inch) or more may be used. When a document received in the normal mode is printed by a printing device having a printing capability of 400 dpi in both the main scanning direction and the sub scanning direction, a resolution conversion process of about 2 times in the main scanning direction and about 4 times in the sub scanning direction is performed. There is a need to.
The resolution conversion process in this case is performed by recording the data of each pixel twice in the main scanning direction and recording the received data of one line four times in the sub scanning direction.

Image data can be interpolated by recording the data of each pixel and each line a plurality of times to achieve resolution conversion, but it cannot be said that the capability of the recording apparatus is sufficiently brought out. Therefore, the image data that has undergone the resolution conversion process is subjected to a smoothing process to improve the quality recorded by the recording device. This smoothing is performed by determining the state (white or black) of the pixel of interest from the states (white or black) of surrounding pixels surrounding a certain pixel (hereinafter referred to as “pixel of interest”).

Further, in a facsimile machine, a digital copying machine or the like, the size of a document can be enlarged or magnified by a factor of more than 1 and recorded or copied. Although the resolution does not change, it is necessary to record the same data repeatedly in order to enlarge the image, which may substantially reduce the resolution. That is, also in the case of the enlargement processing, the image data is interpolated as in the resolution conversion processing.

[0006]

In the resolution conversion processing and the enlargement processing, since the smoothing processing is performed after the interpolation processing, it is necessary to provide a large matrix for the smoothing processing. In other words, if it is an example of the above resolution conversion processing,
In the sub-scanning direction, the same line is repeated four times, so to perform smoothing processing including received data before and after,
A matrix larger than 9 pixels x 9 pixels is required. If the matrix for the smoothing process becomes large, the number of processing patterns will dramatically increase, and the load on the hardware and software will increase.

Therefore, an object of the present invention is to provide an image processing apparatus and method capable of performing effective smoothing processing with a small matrix even when interpolation processing is performed for resolution conversion processing or enlargement processing. .

[0008]

In order to solve the above problems, according to the present invention, first smoothing means for smoothing image data, means for interpolating image data smoothed by the first smoothing means, A second smoothing means for further smoothing the image data interpolated by the interpolating means.

[0009]

According to the present invention, when image data obtained by scanning an original or received from a partner station by facsimile communication or the like requires interpolation processing for resolution conversion processing and enlargement processing. First, the first smoothing process is performed. Then, interpolation processing is performed on the smoothed image data, and second smoothing processing is further performed on the interpolated image data.

The first smoothing process is performed on the image data for which the interpolation process has not been performed. Since the image data that has not undergone the interpolation process is not subjected to the process of repeating the same pixel and the same line, the matrix used for the smoothing process may have a minimum size. Interpolated,
Image data in which the same pixel and the same line are repeated,
Further, the second smoothing process is performed.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to a facsimile machine will be described with reference to FIG. Since the resolution conversion processing and the enlargement processing are the same in principle in that the number of pixels per line and the number of lines per page are increased, this embodiment will be described as the resolution conversion processing.

FIG. 1 is a block diagram showing the configuration of the facsimile apparatus 1. A reading unit 2 scans an original with a CCD or the like to generate binary image data.
The resolution in the main scanning direction is 8 dot / mm, but the resolution in the sub scanning direction is 3.85 line / mm (normal mode), 7.7 lin
There are three types, e / mm (fine mode) and 15.4 line / mm (super fine mode). The operator can select any one of the three resolutions in the sub-scanning direction when reading the original.

The image data generated by the reading unit 2 is
Coded by the codec 5 and transmitted to the partner station, coded by the codec 5 and stored in the RAM 9, or output from the recording unit 3. A recording unit 3 prints out the image data generated by the reading unit 2 or the image data received from the partner station. Also,
The recording unit 3 is 400 dpi (dot p
er inch) resolution, and records on ordinary recording paper by electrophotography.

Reference numeral 4 is a resolution conversion smoothing processing unit, the details of which will be described later. A codec 5 encodes and decodes image data by the MH system, the MR system and the like. A modem 6 modulates and demodulates the encoded image data so that the image data can be transmitted and received through the telephone line L. 7 is NCU (Network Control Unit)
Yes, closing and opening control of telephone line L and incoming call (call signal)
The detection and connection of the attached telephone 12 are performed.

A ROM 8 stores programs and data necessary for control. A RAM 9 stores temporary control data and coded image data. Reference numeral 10 denotes a CPU, which controls each of the above-mentioned units connected via the bus 11 based on a program stored in the ROM 8.

FIG. 2 shows the resolution conversion smoothing processing unit 4.
Will be described with reference to. The image data output from the reading unit 2 or decoded by the codec 5 passes through the input terminal 41 of the resolution conversion smoothing processing unit 4 and then the first matrix 42.
42a and the first line memory LM1.
The output of the first line memory LM1 is the first matrix 42.
42d and the second line memory LM2.
The output of the second line memory LM2 is input to the pixel 42g of the first matrix 42.

The first line memory LM1 and the second line memory LM2 are 2048-bit FIFO (First In Firs).
t Out) memory, which can store image data corresponding to one line (A3 size) in the main scanning direction. With the above-described configuration, the image data input from the input terminal 41 includes the pixel data of the first matrix 42 together with the image data of one line before and two lines before.
Input to a, d and g.

The first matrix 42 is composed of 3 pixels × 3 pixels, and determines the state of the target pixel 42e based on the data of the surrounding 8 pixels 42a-d, f-i. That is, when two lines of image data are input from the input terminal 41, the first and second line memories LM1 and LM2 are full. After that, when the image data for the third and subsequent lines is input, the image data for three lines is sequentially input to the first matrix 42. Each time one-pixel image data is input, the data of the target pixel 42e is output as smoothed image data.

The image data output from the target pixel 42e of the first matrix 42 is processed by the main scanning direction interpolation processing unit 4.
Input to 3. When the resolution conversion process in the main scanning direction is not performed, the number of pixels input to the main scanning direction interpolation processing unit 43 is the same as the number of pixels output from the main scanning direction interpolation processing unit 43. When the resolution conversion processing unit is performed, thinning or interpolation is performed for each predetermined number of pixels. Normal mode 40
When performing resolution conversion processing to 0 dpi, image data of each pixel is output twice.

The image data output from the main scanning direction interpolation processing unit 43 is input to the switch 44. The output of the switch 44 is the pixel 45a of the second matrix 45 and the third pixel 45a.
It is input to the line memory LM3. Third line memory L
The output of M3 is input to the pixel 45d of the second matrix 45 and the fourth line memory LM4. Fourth line memory LM
The output of No. 4 is input to the pixel 45g of the second matrix 45. Third line memory LM3 and fourth line memory LM
Reference numeral 4 is a FIFO memory capable of storing image data of one line after interpolation processing in the main scanning direction.

The switch 44 can selectively close the contacts A and B. While the contact A is closed, the image data output from the main scanning direction interpolation processing unit 43 is
The pixels 45a of the second matrix 45 and the third line memory L
Input to M3. Further, while the contact B is closed, the input of image data from the input terminal 41 and the output of image data from the main scanning direction interpolation processing unit 43 are temporarily stopped. Then, the image data of the third line memory LM3 and the pixels 45d of the second matrix 45 and the fourth
The pixel 45 is input again to the line memory LM4 and
a and the third line memory LM3.

The second matrix 45 is the first matrix 42.
Similar to the above, it is composed of 3 pixels x 3 pixels, and the surrounding 8 pixels 4
The pixel of interest 45e based on the data of 5a to d and f to i
Determine the state of. The data of the pixel of interest 45e is output from the output terminal 46 as the smoothed image data.

Reference numeral 47 is a magnification setting unit having a register and a counter for setting the resolution conversion magnification in the sub-scanning direction.
When the enlargement ratio is 1, the contact A of the switch 44 is closed at all times (a period for processing one page of the original). When the enlargement ratio is 2, the contact A of the switch 44 is closed during the period of processing the image data for one line, and then the contact B of the switch 44 is closed during the period of processing the image data for the next one line. During this period, instead of stopping the output of the image data from the main scanning direction interpolation processing unit 43, the image data of the third line memory LM3 is also output to the pixel 45a and the third line memory LM3. After that, the contact A of the switch 44 is closed again during the next one line period, and the image data from the main scanning direction interpolation processing unit 43 is processed.

Thereafter, the switches 44 are alternately arranged for each line.
By switching the contact point of, it becomes possible to process the data of the same line twice, and the number of lines in the sub-scanning direction can be doubled. The first matrix 42 performs a smoothing process on the image data before the resolution conversion process, and the second matrix 45 performs a smoothing process on the smoothed image data. The case where the resolution conversion magnification in the sub-scanning direction is 4 corresponds to the case where the image data received in the normal mode is recorded at 400 dpi.

The operation of the first and second matrices 42 and 45 will be described with reference to FIG. First of the present embodiment
The operations of the second and second matrices 42 and 45 are the same. The state of the target pixel e shown in FIG. 3A is the surrounding 8 pixels a to
It is determined by the states of d and f to i. As shown in FIG. 3B, when the pixels a, d, g, h, and i are black pixels (B) and the pixels b, c, and f are white pixels (W), the pixel of interest e
The state of the target pixel e is made black regardless of the state of. On the contrary, as shown in FIG. 3C, the pixels a, d, g,
h and i are white pixels (W), and pixels b, c, and f are black pixels (B)
If, then, the state of the target pixel e is set to white. FIG.
Patterns (2) and (3) are 90 ° and 180, respectively.
The same applies when rotated by 270 ° or 270 °. When the pattern does not correspond to these patterns, the target pixel e is output while maintaining the original state.

Details of the operation of the present invention will be described with reference to FIG. FIG. 4A shows a part of the image data in the normal mode read by the reading unit 2, and such image data occurs when a diagonal line is read. The resolution of this image data in the main scanning direction is 8 do.
The resolution in t / line and sub-scanning direction is 3.85 line / mm.
The resolution of the recording unit 3 is 400 dp in both the main scanning direction and the sub scanning direction.
Since i ≈ 15.7 dots / line, it is necessary to perform resolution conversion processing of 2 times in the main scanning direction and 4 times in the sub scanning direction.

When the image data shown in FIG. 4 (1) is input from the input terminal 41 and the first smoothing process is performed by the first matrix 42, the image data shown in FIG. 4 (2) becomes
It is output to the main scanning direction interpolation processing unit 43. Since the processing is performed according to the matrix processing pattern shown in FIG. 3, the corners and stepped portions are smooth. The main scanning direction interpolation processing unit 43 performs double resolution conversion in the main scanning direction.

In the magnification setting section 47, a magnification of 4 in the sub-scanning direction is set. The switch 44 closes the contact A for a period of one line, and the main scanning direction interpolation processing unit 43 to 1
The image data for the lines is supplied to the pixels 45a of the second matrix 45 and the third line memory LM3. During the following three lines, the switch 44 closes the contact B,
The image data of the third line memory LM3 is converted into pixels 45
a, 45d, the third line memory LM3, and the 44th line memory LM4. In this way, the image data of the same line is output four times, and the resolution conversion processing in the sub-scanning direction is performed. In the second matrix 45, the second smoothing process is performed in parallel with the resolution conversion process in the sub-scanning direction.

In the present embodiment, the resolution conversion process and the second smoothing process are performed at the same time, but here, in order to make the description easier to understand, they will be described separately. FIG. 4 (3) is the image data of FIG. 4 (2) that has been subjected to a resolution conversion process of 2 times in the main scanning direction and 4 times in the sub scanning direction. FIG.
When the image data of (3) is smoothed by the second matrix 45, the image data of FIG. 4 (4) is obtained.
It can be confirmed that the corners and stairs are smoother.

In this embodiment, the smoothing process is performed before and after the interpolation process for resolution conversion, but it is also possible to combine the interpolation process and the smoothing process. That is, the original image data → smoothing process → interpolation process → smoothing process → interpolation process → smoothing process → recording unit increases the matrix for smoothing even with a larger interpolation magnification. The smoothing process can be performed without the need.

FIGS. 3 (5) and 3 (6) are examples of the case where the conventional processing is performed. The processing results of the image data of FIG. 3A used in this embodiment will be compared. Conventionally, the smoothing process is performed after the resolution conversion process. First, the image data of FIG. 3A is doubled in the main scanning direction,
When quadruple resolution conversion processing is performed in the sub-scanning direction, FIG.
The image data shown in (5) is obtained. Next, when smoothing processing is performed using the matrix of 3 pixels × 3 pixels described in FIG. 3, the image data shown in FIG. 4 (6) is obtained. FIG.
The difference is clear when compared with the result of this example shown in (4). In order to perform the smoothing processing on the image data of FIG. 4 (5) and obtain the processing result equivalent to that of FIG. 4 (4), a smoothing circuit having a very large matrix is required.

[0032]

Since the smoothing process is performed before and after the interpolation process, even if the matrix is about 3 pixels × 3 pixels,
An excellent smoothing processing result can be obtained. Since the matrix for the smoothing process may be small, the burden on hardware and software can be greatly reduced. Further, when the data obtained by repeatedly outputting the image data of the same pixel and the same line is used for the interpolation processing, the following effects are further obtained. By performing the smoothing process before the interpolation process, the number of times the same image data is repeated by the interpolation process is reduced. By performing the smoothing process after the interpolation process, the smoothing process can be performed on the image data repeated by the interpolation process with a small matrix. INDUSTRIAL APPLICABILITY Since the present invention can be applied to the case of interpolating image data, not only the resolution conversion processing (changing the number of pixels per unit length or unit area), but also the enlargement processing (constant resolution, length or area Change).

[Brief description of drawings]

FIG. 1 is a block diagram of a facsimile apparatus using the present invention.

FIG. 2 is a block diagram showing a main part of the present invention.

FIG. 3 shows a matrix used for smoothing according to the present invention.

FIG. 4 is a diagram illustrating a smoothing process of the present invention.

[Explanation of symbols]

 42, 45 First and second matrices 43 Main scanning direction interpolation processing unit 44 Switches LM1, LM2 First and second line memories LM3, LM4 Third and fourth line memories

Claims (5)

[Claims] 1. A first smoothing means for smoothing image data, a means for interpolating the image data smoothed by the first smoothing means, and a further smoothing for the image data interpolated by the interpolation means. 2. An image processing device comprising: two smoothing means. 2. The image processing apparatus according to claim 1, wherein the first smoothing means is composed of a matrix of 3 pixels × 3 pixels. 3. The image processing apparatus according to claim 1, wherein the second smoothing means is composed of a matrix of 3 pixels × 3 pixels. 4. The interpolation means performs the same image data by repeating the same image data.
The image processing device described in each of. 5. An image processing method, wherein image data is smoothed, interpolated, and further smoothed.