JP3354651B2 - Image scaling method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image scaling method for obtaining an enlarged or reduced image from an original image in a digital image forming apparatus, and more particularly, to a method of reading an original image for each pixel by an image scanner to obtain a time series image. The present invention relates to an image scaling method for converting an electric signal into a binary electric signal, and processing the data to enlarge or reduce an original image.

[0002]

2. Description of the Related Art In a digital image forming apparatus such as a digital copying machine or a digital printer, an original image is read pixelwise by an image scanner such as a CCD scanner, and is converted into a time-series binary electric signal. The light is emitted as a light output of an LED or a laser to form an image. In such an apparatus, when an original image is formed by enlarging or reducing (hereinafter, referred to as “magnification”), data processing is performed for main scanning (scanning of the original image in the width direction) in association with a magnification operation by a scanner. Of read signal by
It is necessary to adjust the document feed speed according to the magnification for the adjustment by compression and the sub-scan (scan in the moving direction of the original image).
Generally, adjustment in the sub-scanning direction is performed by controlling a driving pulse motor, and in the main scanning direction, high-speed data processing of input (reading) data by software or hardware is performed.

As a data processing in the main scanning direction, a method is known in which an input signal from an original image is frequency-divided by a predetermined frequency divider according to a set magnification to obtain a necessary signal.

[0004]

However, in such a conventional method, an arbitrary magnification is set, in particular, 0.1.
It is difficult to sufficiently cope with a minute change in magnification such as% increment.

[0005]

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and an image scaling method according to the present invention is a method of inputting a time-series binary electric signal by main-scanning an original image in pixels. In an image scaling method of forming scaled image output data by converting the data into data and processing the input data, the frequency (FWR) of the write clock of the input data can be adjusted so that the frequency of the read clock of the output data (FWR) can be adjusted. the FRD) is fixed and with a × (m / 100) × (fBS / fRD) ( where, a = 1, 2, 3 ... integer, m = magnification, fBS = the frequency of the fundamental wave of the write clock, FRD = Frequency of the read clock) to obtain a sequence of integers (if the calculated value is a decimal value, round the first decimal place to an integer) and calculate this number
Based on the final sequence consisting of the difference between each number in the column,
Count the pulses of the fundamental wave of the write clock and input
The frequency of the data write clock is adjusted .

[0006]

According to the above-mentioned means, the input data is converted into output data of a set magnification by thinning out or overlapping based on the sequence calculated by the above equation.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. Figure 1 shows an overview of an image scaling method according to the present invention, the binary electrical signals of input data inputted from the image scanner (DATE_WR) is FIFO (F ir
The st I n F irst O ut) memory output is the data processing as output data for the scaled image (DATE_RD). F
The IFO memory performs data write and read operations,
Clocks completely asynchronous, that is, independent clocks (write clock CLK_WR, read clock CLD_RD)
When the frequency of the write clock (CLK_WR) is f _WR and the frequency of the read clock (CLK_RD) is f _RD , (1) When f _WR = f _RD , input data (DATA_WR) and output data (DATA_RD) is the same, and forms an image of the same size. (2) When f _WR > f _{RD Since} the write data becomes excessive for the data read, the data is thinned out to form a reduced image. (3) When f _WR <f _RD , since data to be written becomes insufficient for data reading, data is duplicated and an enlarged image is formed.

The read clock (CLK_RD) is fixed in relation to the transfer of data to the printer, and the write clock is adjusted (changed) according to the set magnification. Basically, the adjustment of the write clock may cause a synchronization deviation between lines when the frequency is changed in an analog manner,
The frequency of the read clock (CLK_WR) is 10 MHz.
For example, when an image is reduced to 43%, a frequency close to 30 MHz is required as a fundamental wave (pulse) of the write clock, and there is a problem that it is difficult to adjust the analog frequency. In view of such a problem, in the present invention, the adjustment of the write clock is performed by digital processing as described in detail below.

As described above, the fundamental frequency of the write clock is 30 MHz, and the frequency of the read clock is 1
In the case of 0 MHz, in order to obtain an image with a magnification of 100%, the frequency of the write clock is changed to the frequency of the fundamental wave (30 MHz).
z) is adjusted to be 1/3 (10 MHz). FIG. 2 shows 100% magnification and 2
This shows the frequency of the write clock in the case of 00%. The fundamental wave (pulse) is counted as a shift register, and a pulse signal with a magnification of 100% is formed by outputting one pulse for three pulses of the fundamental wave, and the pulse signal is generated once every six times. A pulse signal of 200% is obtained.

Thus, the magnification for enlargement or reduction is m
, 1 for every (m / 100) × (30/10) times
By emitting the pulse twice, a signal of a desired magnification can be obtained. When this is expressed by a general formula, it is expressed as a × (m / 100) × (f _BS / f _RD ). Here, a is an integer of 1, 2, 3,..., For example, when the magnification is variable in 0.1% steps,
It becomes 1000. m is the magnification, f _BS is the frequency of the fundamental wave, f
_RD indicates the frequency of the read clock. When the calculated value includes a decimal in the above formula, the first decimal is rounded off to a sequence of integers.

For example, if the read clock is 10 MHz,
The fundamental frequency of the write clock is 30 MHz and the magnification m is 60%
From the above equation, (60/100) × (30/10) = 1.8, so that 1.8, 3.6, 5.4, 7.2 ... 1798.
A sequence of 2,1800 is obtained, and each number is rounded off to the first decimal place to obtain a sequence of integers of 2,4,5,7..., 1798,1800. The difference between the adjacent numbers in this sequence is calculated as 2,2,1,2... 2 (referred to as the final sequence), whereby the pulses of the fundamental wave are counted, and once every two times, once every two times. Once, once, twice, once ... 2
A write clock having a magnification of 60% is formed by emitting pulses one by one in order. FIG. 3 is a diagram schematically illustrating a waveform in such a case. An error due to rounding at the first decimal place (error when digitizing an analog amount) can be dispersed in 1000 pulses on average, and therefore does not adversely affect the image quality of the scaled image.

FIG. 4 shows an example of a circuit for forming the write clock. Basically, it has a CPU 1, data selectors 2, 3, a RAM 4, a shift register 5, and an address counter 6. The calculation according to the above equation is performed by the CPU 1, and the obtained (final sequence) is coded data of 1, 1 2, 10 3, 100 4, 1000 5, 10000 6, 100000 And stores this data from address 0 to address 999 of the RAM 4 in the order of a numerical sequence. At this time, the address data bus of the RAM 4 is connected to the CPU 1 by the data selector 3. When outputting a target pulse (write clock), the data selector 3 sets the address bus of the RAM 4 to the address counter 6.
Then, the data bus is switched to the shift register 5 by the data selector 2. When the data at address 0 is sent to the shift register 5, the parallel data is converted into serial data by the basic clock and output. When the signal “1” is output, the address counter 6 advances the address, and
The data at the address is sent to the shift register 5. By repeating this, a target pulse is obtained.

FIG. 5 shows a timing chart when the above-mentioned magnification is set to 60%. When the data of address 999 is sequentially advanced as described above and the 1000th pulse is output by the data of address 999, the address is changed. Clear and repeat the operation from address 0 again.

[0014]

According to the present invention, an integer obtained by a magnification
By digital signal adjustment in which the frequency of the fundamental wave of the write clock is adjusted based on the sequence , it is possible to perform fine magnification adjustment without deviation between lines.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of an image scaling method according to the present invention.

FIG. 2 is a magnification of a writing clock with respect to a fundamental wave of 100.
Waveform diagram showing% and 200% pulse extraction

FIG. 3 is a waveform diagram showing a write clock waveform according to the present invention.

FIG. 4 is a block diagram showing a circuit for obtaining a write clock according to the present invention.

FIG. 5 is a timing chart of the circuit in FIG. 4;

[Explanation of symbols]

 1 CPU 2 Data selector 3 Data selector 4 RAM 5 Shift register 6 Address counter

Claims (1)

(57) [Claims] 1. An original image is subjected to main scanning in a pixel manner and a time series
In the image scaling method of converting the input electric signal into input data and processing the input data to form the scaled image output data, the frequency (FWR) of the write clock of the input data can be adjusted, and the output data a fixed frequency (FRD) of the read clock, with a × (m / 100) × (fBS / fRD) ( where, a = 1, 2, 3 ... integer, m = magnification, fBS = basic write clock frequency of the wave, FRD = integer sequence (calculated value consisting of the read frequency of the clock) is an integer by rounding off the first decimal place in the case of a fractional value) is obtained, for each number adjacent of this sequence Based on the final sequence of differences
Next, count the pulses of the fundamental wave of the write clock.
To adjust the frequency of the input data write clock.
And an image scaling method.