JP3219421B2 - Information recording device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information recording apparatus such as a laser beam printer, and more particularly to a method for smoothing bitmap data representing characters and figures to print the contours of the printed characters and figures. The present invention relates to an information recording device that can be smoothed to improve printing quality.

[0002]

2. Description of the Related Art In recent years, laser beam printers using electrophotography have been used in computer output devices, facsimile output units, so-called digital copiers for printing image data read from an image scanner, and the like.

[0003] The laser beam printer used in these printers prints at a resolution of, for example, 300 dots / inch. In this case, the characters and figures are 300, as shown in FIG.
It is drawn and printed by black dots (● marks) and white dots (ド ッ ト marks) printed at positions arranged on a dot / inch grid. This figure shows a dot pattern of the alphabetic character "a". At a resolution of 300 dots / inch, the dot spacing is about 85 microns. It is generally said that human eyes can recognize up to about 20 microns, but in contrast, at the above resolution (about 85 microns), the gaps between characters and figures formed by dots look jagged, and are not necessarily. It cannot be said to be high quality printing.

[0004] To solve this, there are the following approaches.

A first approach is to simply increase the resolution (for example, 1200 dots / inch). However, in this case, 4 × 4 =
There is a disadvantage that a 16-times bitmap memory is required and the device becomes very expensive.

A second approach is to modulate the print data of the target pixel by referring to dot data around the target pixel to be printed by adding a small amount of buffer memory without increasing the capacity of the bit map memory. There is known a method of equivalently increasing the resolution in the main scanning direction or the main scanning direction and the sub-scanning direction.

USP-4,437,122 and USP-
The technique described in US Pat. No. 4,700,201 is a method of correcting a target pixel to be printed by referring to only a target pixel and all eight pixels around the target pixel. In this type of method,
Since the reference area of the surrounding pixels is narrow, it is possible to recognize that the target pixel is a part of the curve, but it is not possible to recognize the degree of curvature of the curve. In particular, it is difficult to detect a contour portion that is close to horizontal or a contour portion that is close to vertical, and it is not possible to perform optimal correction according to the curvature. As a result, it is difficult to optimize the effect of smoothing.

On the other hand, the technique described in US Pat. No. 4,847,641 refers to a wider area as compared with the above two techniques. Can also be recognized. However, in this technique, although the reference target area is certainly large, each of the matching patterns matched individually refers to only a part of the reference target area. Therefore, this technique has the following disadvantages.

First, it is not possible to identify whether the pixel of interest is a part of a binary halftone image such as a dither image or an image obtained by an error diffusion method. For this reason, even if the smoothing effect on the character image can be improved, a part of the dots forming the halftone pixels by the dither image or the error diffusion method may be erroneously smoothed. For example, FIG. 9A is a diagram in which a part of a 4 × 4 dither image is extracted. Referring to the peripheral limited area with respect to the target pixel 5f in the same figure, the target pixel is recognized as being a part of a character or a figure. The pixel is changed to a pixel having the density. Therefore, there is a high possibility that a local change in image density is generated for the halftone image, thereby causing image quality deterioration such as generation of a false contour.

Second, it is not possible to identify whether or not the target pixel belongs to a part of the image where the image is dense (entered). For example, FIG. 9B shows an example of an image composed of a dense line group of one dot line. In this case, the pixels which need to change the dot density in order to make each line smooth are shown in FIG.
Pixels marked with “Δ” or “x” shown in FIG. As can be seen from the figure, the changed pixel is adjacent to or adjacent to the pixel to be changed for the adjacent pixel adjacent thereto.
This results in reduced image resolution. Such a case where pixels are densely arranged in a complicated manner may occur not only when line drawings are densely arranged but also for small-sized characters and kanji. In such a case, the changed pixel of interest performed for smoothing comes close to the changed pixel for the adjacent image, and the pixel (the relevant line or the side of the corresponding character) and the adjacent pixel As a result, the resolution (resolution) of the image in the vicinity of the part is extremely reduced, resulting in a blurred image or "moire" in the image, resulting in a deterioration in image quality. there is a possibility. Furthermore, if a pixel is expressed in halftone within one pixel due to smoothing in a portion where such images are dense, the reproduction of density becomes unstable due to the interaction with neighboring pixels, and fluctuation with respect to the environment (temperature and humidity) The smoothing effect varies depending on the environment, and a problem occurs such that the character shape appears to be a different font each time printing is performed.
Of course, it is only necessary to widen the reference area of each matching pattern sufficiently so that it is possible to identify whether the image belongs to a dither image or a dense portion of the image. Will no longer be effective.

As a technique for compensating for the above-mentioned disadvantages of the prior art, the present applicant has proposed the following.

That is, the feature of the dot pattern in the peripheral region of the target pixel is extracted for the entire region, and the dot pattern at the boundary of the figure to which the target pixel belongs is determined by a plurality of predetermined features and boundary portions. In this case, the target pixel is changed when the pattern matches, and a wide reference area can be referred to by a simplified logic circuit. or,
According to this method, it is possible to detect a substantially horizontal contour portion and a substantially vertical contour portion, and it is possible to perform optimal smoothing correction according to the curvature of the contour portion of a character or figure. In addition, it is possible to improve the degradation of halftone images by prohibiting smoothing processing on dither images and dense images by adding a function to identify binarized halftone images or figures such as dither images and the like. It is. Also, by performing the pixel change for smoothing only when there is a predetermined white area around the target pixel to be changed, the effect of smoothing is less likely to be affected by the environment. Have.

As described above, many methods have been proposed as methods for examining and detecting what features of an image, such as a figure or a character, belong to a boundary portion having an image. On the other hand, US Pat. No. 4,933,689, JP-A-61-214661, and US Pat. JP-A-61-214666 is known.

[0014]

However, the above technique has the following disadvantages.

The first problem is that the thickness of the line printed when the smoothing process is performed is different from that when the smoothing process is not performed. That is, in the related art, when the target pixel is changed by the smoothing processing, no consideration is given to the correlation regarding the addition or reduction area with the changed neighboring pixel or the correlation regarding the addition or reduction area with respect to the left end and the right end of the line. Therefore, there is a problem that the line thickness after printing changes when the smoothing process is performed, as compared with the case where the smoothing process is not performed.

The second problem is that a smoothing process may be performed on a line that is not desired to be smoothed. That is, there is a problem that the bit data of the line created by the host controller or the like may be smoothed up to the line where the contour portion is intentionally printed jagged.

The third problem is a problem that occurs in smoothing processing for a line having a width of one dot. FIG.
Reference numeral 2 denotes a smoothing process according to the prior application filed by the present applicant for a horizontal line of 1 dot width having a slope of 45 degrees or less. FIG. 5A shows a line of a type in which one dot width is secured regardless of where the data of one dot line is located on the line. FIG. This is a type of line in which one dot width is not locally secured at a place where the dot position changes.
As described above, there are two types of lines even if one dot line is used. For example, when a command for performing line drawing in a printer control language such as a well-known PostScript language is executed, one line is drawn depending on the language. Which type of line is drawn depends on the interpreter being built.
When the two types of one-dot lines are subjected to the smoothing process using the feature detection method of the graphic boundary portion, the printing after the smoothing process of (d) is performed in the processing area compared to (a). In this case, printing is performed with a small pixel area corresponding to one dot in the portion. As a result, as can be seen from FIGS. 7C and 7F, the processing portion for (d) becomes thin or becomes a broken line like a broken line, which is different from the appearance for (a). There is a disadvantage that the image quality deteriorates. Conversely, if the algorithm is changed to eliminate the thinning of (d), there is an inconvenience that the processing portion for (a) becomes fat.

This problem also occurs for a vertical line. FIG. 47 shows a smoothing process according to the prior application technique for a vertical line having a one-dot width with an inclination of 45 degrees or more. FIG. 7A shows a line of a type in which one dot width is secured regardless of where the data of one dot line is located on the line. FIG. 1 locally where the dot position changes
This is a type of line where the dot width is not ensured.

When the two types of one-dot lines are subjected to the smoothing process using the feature detection method of the graphic boundary portion, the printing after the smoothing process of (a) is performed in comparison with (c). Thus, a large pixel area corresponding to one dot is printed in the processing area. As a result, as can be seen from FIGS. 6D and 6B, the processed portion corresponding to FIG. When the processing algorithm is changed to eliminate the fatness of (a), (c)
Is thinly printed. As a result, (a)
And (c) are different from each other, and the image quality is degraded in any one.

As described above, in either case of the horizontal line or the vertical line, it is difficult to print the two types of lines without causing any thickening or thinning.

The fourth problem is a problem that occurs in smoothing processing for a one-dot white line. FIG.
Reference numeral 1 denotes a horizontal line having a width of one dot with a white inclination of 45 degrees or less: the smoothing process shown in FIG.
As can be seen from FIG. 3C, when smoothing processing is performed on such a white one-dot line, the white line in the portion where the smoothing processing has been performed becomes thinner or a broken line like a broken line. There was an inconvenience of becoming. In addition, such a portion is easily affected by the influence of the electrophotographic process and the environment such as temperature and humidity, and the appearance of smoothing is also unstable.

The fifth problem is a problem of smoothing processing particularly occurring for kanji characters. FIG. 54 is a diagram showing a dot pattern of 300 dpi of the kanji character “day”. When this dot pattern is subjected to smoothing processing, it becomes as shown in FIG. As can be seen from the figure, unnecessary pulses (: must not be attached) such as (A), (B), (C), and (D) are added to the local portion of the character "day". As a result, there is a disadvantage that the image quality of the character after the smoothing processing is deteriorated.

[0023]

SUMMARY OF THE INVENTION The present invention is to eliminate the above-mentioned third to fifth disadvantages of the prior art by the following constitution.

[0024]

[0025]

With respect to the third problem, a line width of a one-bit width is ensured for a line formed with a line width of one bit or more at a variation portion of the line. Is not secured, and the algorithm for changing the pixel of interest is changed accordingly.Thus, regardless of the type of line, the line becomes thicker or the image quality deteriorates due to broken lines, etc. It is an object of the present invention to provide an information recording device capable of performing smoothing without any problem. In order to achieve this object, an information recording apparatus according to the present invention comprises: an information signal generating means for generating an information signal; a storage means for temporarily storing a part of a bit information signal emitted from the information signal generating means; Feature detecting means for detecting whether or not the feature of the bit information thus obtained matches one of a plurality of predetermined features to be subjected to smoothing processing; and a plurality of predetermined smoothing processes performed by the feature detecting means. When it is detected that one of the features to be matched, the print information of a specific pixel of interest in the information group stored in the temporary storage means is subjected to a smoothing process to subdivide the small picture. Information changing means for changing to a section signal; and output means for outputting a change signal from the information changing means to an output unit. The information changing means identifies whether the line width in the eccentric portion of the line is 1 bit width or more or not, and the information changing means determines the attention corresponding to each case. It is characterized in that the pixel changing algorithm is different.

With respect to the fourth problem, when it is detected that a white line is formed with a line width of 1 bit width, the pixel of interest is not changed by smoothing. It is an object of the present invention to provide an information recording device capable of preventing the thinning of the information. To achieve this object, according to the information recording apparatus of the present invention, an information signal generating means for generating an information signal, a storage means for temporarily storing a part of a bit information signal emitted from the information signal generating means, Whether or not the feature of the stored bit information matches one of a plurality of predetermined features to be smoothed, and prohibit the feature of the stored bit information from a predetermined smoothing process. A feature detecting means for detecting whether or not the pattern matches a one-bit-width white line pattern; and the feature detecting means matches one of a plurality of predetermined features to be subjected to smoothing processing. When it is detected, the printing information of the specific target pixel in the information group stored in the temporary storage unit is smoothed and changed to a signal of a subdivided sub-area. Information change means, and output means for outputting a change signal from the information change means to an output unit, and when the feature detection means detects that the white line is formed with a line width of 1 bit width The information changing means does not change the target pixel.

With respect to the fifth problem, it is an object of the present invention to provide an information recording apparatus capable of preventing a harmful effect of kanji characters by smoothing by eliminating a smoothing process by incorporating a prohibition pattern peculiar to kanji. I do. To achieve this object, according to the information recording apparatus of the present invention, an information signal generating means for generating an information signal, a storage means for temporarily storing a part of the bit information signal emitted from the information signal generating means, A first feature detection unit for detecting whether a feature of the stored bit information matches one of a plurality of predetermined features to be subjected to smoothing processing, and a first feature detection unit. When it is detected that one of a plurality of predetermined features to be subjected to the smoothing process is matched, the printing information of the specific pixel of interest in the information group stored in the temporary storage unit is smoothed. The information changing means for changing to the signal of the processed and segmented sub-area, the output means for outputting the change signal from the information changing means to the output unit, and the characteristics of the stored bit information are predetermined. A second feature detecting means for detecting whether or not the pattern matches one of a plurality of specific patterns for prohibiting a plurality of kanji-specific smoothing processes, wherein the first feature detecting means is determined in advance. Even if it is detected that one of the plurality of features matches, the second feature detection unit matches one of the specific patterns that inhibits the plurality of smoothing processes. When it is detected that the target pixel is in agreement, the information changing means does not change the target pixel.

[0029]

Embodiments of the present invention will be described below with reference to the accompanying drawings.

FIGS. 1 and 2 are views showing an engine part of a laser beam printer to which the present invention is applied.

In FIG. 1, reference numeral 1 denotes a sheet as a recording medium;
Reference numeral 2 denotes a paper cassette that holds the paper 1. Reference numeral 3 denotes a feeder for separating only the uppermost sheet of the sheets 1 placed on the sheet cassette 2 and transporting the leading end of the separated sheet to the position of the feed rollers 4 and 4 'by a driving unit (not shown). The paper cam intermittently rotates each time paper is fed, and feeds one sheet of paper corresponding to one rotation.

Reference numeral 18 denotes a reflection-type photo sensor, which detects the absence of paper by detecting the reflected light of the paper 1 through a hole 19 provided at the bottom of the paper cassette 2.

The paper feed rollers 4, 4 '
When the paper 1 is conveyed to the roller section, the paper 1 is rotated while being slightly pressed and conveyed. When the sheet 1 is conveyed and the leading end reaches the position of the registration shutter 5, the conveyance of the sheet 1 is stopped by the registration shutter. Keep spinning. in this case,
By driving the registration solenoid 6 to release the registration shutter 5 in the upward direction, the sheet 1 is fed to the conveyance rollers 7 and 7 '. When the resist shutter 5 is driven, the laser beam 20 is applied to the photosensitive drum 11.
Synchronization is achieved with the transmission timing of the image formed by forming an image thereon. A photo sensor 21 detects whether or not the sheet 1 is present at the position of the registration shutter 5.

Here, reference numeral 52 denotes a rotary polygon mirror, which is driven by a motor 53. The laser driver 50 drives the semiconductor laser 51 in accordance with dot data sent from a character generator (not shown) for generating bit data. The character generator generates kanji character bit data as Japanese characters shown in FIGS. 54 and 57 in addition to the alphabet characters shown in FIG. The laser beam 20 from the semiconductor laser 51 driven by the laser driver 50 is scanned in the main scanning direction by the rotary polygon mirror 52, and is reflected by the f-θ lens 56 disposed between the rotary polygon mirror 52 and the reflection mirror 54. It is guided onto the photosensitive drum 11 via the mirror 54, forms an image on the photosensitive drum 11, scans in the main scanning direction, and forms a latent image on the main scanning line 57. In this case, 300
When a printing speed of 8 pages / minute (: A4 size or letter size) is used at a printing density of dots / inch, the lighting time of the laser for recording one dot is about 540 nanoseconds (one pixel has three small dots). When divided into pixels, the light-on time for one small pixel is 180 nanoseconds), and the laser light-on time for recording one dot at a print density of 300 dots / inch and a print speed of 16 sheets / min. Is about 270 nanoseconds (when one pixel is divided into three small pixels, the lighting time of one small pixel is 90 nanoseconds). Also, when printing at a printing density of 600 dots / inch and a printing speed of 8 sheets / min, the laser lighting time for recording one dot is about 135 nanoseconds (when one pixel is divided into three small pixels, The lighting time of one small pixel is 45 nanoseconds, and the laser lighting time for recording one dot at a printing density of 600 dots / inch and a printing speed of 16 sheets / minute is about 68 nanoseconds ( When one pixel is divided into three small pixels, the lighting time of one small pixel is 2
3 nanoseconds). With the current technology, the driving capability of the laser driver 50 used in this type of laser beam printer is such that the shortest pulse lighting time is about 4 nanoseconds (: lighting rise time about 1 nanosecond, turn off fall time about 1 nanosecond). ) Degree is the limit. Therefore, if the light is turned on for a shorter time, the light cannot be turned on, or even if the light is turned on, the lighting time and the light amount are unstable. Accordingly, the laser lighting pulse width modulated for smoothing is set to a lighting time of at least about 4 nanoseconds or more. The beam detector 55 arranged at the scanning start position of the laser beam 20 detects a BD signal as a synchronization signal for determining an image writing timing in the main scanning direction by detecting the laser beam 20.

Thereafter, the paper 1 is fed to the photosensitive drum 11 by obtaining the transport torque by the transport rollers 7, 7 'instead of the paper feed rollers 4, 4'. A latent image is formed on the surface of the photosensitive drum 11 charged by the charger 13 by exposure to the laser beam 20. The latent image of the portion exposed by the laser beam is visualized as a toner image by a developing device 14, and then the toner image is transferred to the sheet 1 by a transfer charger 15.
Is transferred onto the paper. Reference numeral 12 denotes a cleaner for cleaning the drum surface after being transferred onto the sheet 1.

The sheet 1 onto which the toner image has been transferred is thereafter fixed with the toner image by fixing rollers 8 and 8 ', and is discharged onto a discharge tray 10 by discharge rollers 9 and 9'.

Reference numeral 16 denotes a paper feed table, which enables not only paper feed from the paper cassette 2 but also manual paper feed from the paper feed table 16 one by one. The paper fed manually to the manual paper feed roller 17 on the paper feed table 16 is lightly pressed by the manual paper feed roller 17, and the leading edge of the paper is registered with the registration shutter similarly to the paper feed rollers 4 and 4 '. 5 until it reaches 5, where it slips and turns. The subsequent transport sequence is exactly the same as when feeding paper from a cassette.

The fixing roller 8 houses a fixing heater 24, and controls the surface temperature of the fixing roller 8 to a predetermined temperature based on the temperature detected by the thermistor 23 that makes a slip contact with the roller surface of the fixing roller. The toner image on the sheet 1 is thermally fixed. A photo sensor 22 detects whether or not there is a sheet at the positions of the fixing rollers 8 and 8 '.

Such a printer is connected to a controller by an interface means, and receives a print command and an image signal from the controller to perform a print sequence. The signals transmitted and received by the interface means will be briefly described below.

FIG. 3 is a diagram showing interface signals between the printer engine unit and a controller for generating image data. Each of the interface signals shown in the figure will be described below.

The PPRDY signal is a signal sent from the printer to the controller and indicates that the printer is turned on and the printer is in an operable state.

The CPRDY signal is a signal sent from the controller to the printer and indicates that the controller is turned on and the controller is operable.

An RDY signal is a signal sent from the printer to the controller.
This signal indicates that the printing operation can be started or continued at any time when the RNT signal is received. For example, when the printing operation cannot be performed because the paper cassette 2 runs out of paper or the like, the signal becomes “false”.

PRNT signal-a signal sent from the controller to the printer, which is a signal for instructing the start of the printing operation or the continuation of the printing operation. Upon receiving the signal, the printer starts a printing operation.

The VSREQ signal is a signal sent from the printer to the controller. When the RDY signal sent from the printer is in the "true" state, the PRNT signal is set to "true" from the controller to perform the printing operation. This signal indicates that the printer is ready to receive image data after the start instruction is sent. In this state, it becomes possible to receive a VSYNC signal described later.

The VSYNC signal is a signal transmitted from the controller to the printer, and is a signal for synchronizing the transmission timing of image data in the sub-scanning direction. By this synchronization, the toner image formed on the drum is transferred onto the paper in synchronization with the paper in the sub-scanning direction.

The BD signal is a signal transmitted from the printer to the controller, and is a signal for synchronizing the transmission timing of image data with respect to the main scanning direction. By this synchronization, the toner image formed on the drum is transferred onto the paper in synchronization with the paper in the sub-scanning direction.
This signal indicates that the scanning laser beam is at the start of the main scan.

VDO signal-a signal sent from the controller to the printer, and a signal for transmitting image data to be printed. The signal is VCLK which will be described later.
It is sent out in synchronization with the signal. The controller receives code data such as a PCL code transmitted from the host device, generates a character bit signal generated from a character generator corresponding to the code data, or generates a vector such as a PostScript code transmitted from the host device. Upon receiving the code, it generates graphic bit data corresponding to the code, or generates bit image data read from the image scanner, and transmits the data to the printer as a VDO signal. The printer prints a black image when the signal is “true” and a white image when the signal is “false”.

The VCLK signal is a signal transmitted from the controller to the printer, and is a synchronization signal for transmitting and receiving the VDO signal.

The SC signal is a bidirectional serial signal for bidirectionally transmitting and receiving the "command" which is a signal transmitted from the controller to the printer and the "status" which is a signal transmitted from the printer to the controller. . An SCLK signal described later is used as a synchronization signal for transmitting or receiving the signal. In addition, an SBSY signal and a C
BSY signal is used. Here, the “command” forms an 8-bit serial signal. For example, the controller determines whether the paper feeding mode is a mode for feeding paper from a cassette or a mode for feeding paper from a manual feed slot. Is instruction information for instructing. "Status" is a serial signal consisting of 8 bits.
For example, the printer informs the controller of various states of the printer, such as a wait state in which the temperature of the fixing device of the printer has not reached a printable temperature, a paper jam state, or a paper cassette being out of paper. It is information to do.

SCLK signal-for the printer to take in a "command" or for the controller to be "status"
Is a synchronizing pulse signal for capturing

The CBSY signal is a signal for occupying the SC signal and the SCLK signal before the controller transmits a "command".

SBSY signal-Printer is "status"
Is a signal for occupying the SC signal and the SCLK signal prior to transmitting the signal.

After the VDO signal is input to the printer together with the VCLK signal, it is input to the VDO signal processing unit 101 disposed in the printer engine and performing the signal processing of the present embodiment. The VDO signal processing section converts the input VDO signal into a signal by a signal processing described later and inputs the converted signal to a laser driver (not shown) as a converted signal VDOM to drive the semiconductor laser 51 to blink.

The operation of such an interface will be described below.

When the power switch of the printer is turned on and the power switch of the controller is turned on, the printer initializes the internal state of the printer, and then sets the PPRDY signal to "true" to the controller. On the other hand, the controller similarly initializes the internal state of the controller and sets the CPRDY signal to "true" for the printer. As a result, the printer and the controller confirm that the power has been turned on.

Thereafter, the printer energizes the fixing heater 24 housed inside the fixing rollers 8 and 8 ', and when the surface temperature of the fixing roller reaches a temperature at which fixing can be performed, an RDY signal is output.
After the controller confirms that the RDY signal is "true", if there is data to be printed, the controller sets the PRNT signal to "true" for the printer. Is confirmed, the photosensitive drum 11 is rotated to uniformly initialize the potential of the photosensitive drum surface, and at the same time, the paper feed cam 3 is driven in the cassette paper feed mode to move the leading end of the paper to the position of the registration shutter 5. In the manual paper feed mode, the paper manually fed from the paper feed table 16 by the manual paper feed roller 17 is transported to the position of the registration shutter 15. Thereafter, when the printer is ready to receive the VDO signal, VSREQ is output. The signal is set to “true.” After the controller confirms that the VSREQ signal is “true,”
The SYNC signal is set to “true” and the VDO signal is sequentially transmitted in synchronization with the BD signal. The printer is VSYNC
When it is confirmed that the signal has become "true", the registration solenoid 6 is driven in synchronization with this to release the registration shutter 5. Thereby, the sheet 1 is conveyed to the photosensitive drum 11. In response to the VDO signal, the printer turns on the laser beam when the image is to be printed in black, and turns off the laser beam when the image is to be printed in white, thereby forming a latent image on the photosensitive drum 11. Next, the developing device 14 attaches toner to the latent image and develops the latent image to form a toner image. Next, the toner image on the drum is transferred onto the sheet 1 by the transfer charger 15, and is fixed on the fixing rollers 8 and 8 ′.

FIG. 8 shows a VDO signal processing for performing a smoothing processing installed at an input portion of the printer engine unit according to the first embodiment of the present invention, which is adapted to a laser beam printer having a printing density of 300 dots / inch. Part 1
It is a figure which shows the circuit block of No. 01.

In the first embodiment, as shown in FIG. 5, a pixel A to be printed (hereinafter, this pixel is called a pixel of interest) is surrounded by a peripheral region (11 pixels in the main scanning direction) surrounding the pixel of interest. The characteristic of the pixel data (× 9 pixels in the sub-scan) is examined, and the target pixel is changed according to the result. More specifically, for example, the resolution 300 shown in FIG.
When printing the target pixel A in the dot data group of the dot character "a" of dot / inch, the area S surrounding the target pixel A (11 pixels in main scan × 9 pixels in sub-scan = 9)
(9 pixels) is stored in the temporary storage means. As a result, dot information as shown in FIG. 7 is stored. Thereafter, the characteristics of the dot data group in the area S are checked, and the data of the target pixel A to be printed is changed and printed according to the characteristics. In this case, the data is changed so that the outline of the figure composed of the dot group is printed smoothly. In the first embodiment, as shown in FIG.
Is a sub-division (x1, x2, x3,
x4). Therefore, at the printing stage, printing is performed at a print density of 1200 dots / inch main scanning × 300 dots / inch sub-scanning equivalently.

In FIG. 8, reference numerals 25 to 33 denote line memories which convert an input image signal VDO into a clock signal VC.
The data is stored while being sequentially shifted in synchronization with the LK, and each line memory stores dot information of the main scanning length for a page to be printed. Each line memory is connected in the order of line memory 1 → line memory 2 → line memory 3 →... → line memory 9 and stores dot information having a main scanning length of 9 lines in the sub-scanning direction. I do. Reference numerals 34 to 42 denote shift registers. Each of the shift registers 1 to 9 receives an output from each line memory corresponding to each of the line memories 1 to 9. Each shift register is composed of 11 bits, and as shown, 1a to 1k, 2a to 2k, 3
.., 9a to 9k constitute a dot matrix memory of 11 dots in the main scanning direction × 9 lines in the sub-scanning direction. In the matrix memory, a central dot 5f is defined as a target dot. Reference numeral 43 denotes a processing circuit which detects the characteristics of data stored in the dot matrix memory for smoothing and changes the pixel of interest 5f as necessary. 9k), and the changed parallel signal MDT is output. The parallel signal MDT is input to the parallel-serial conversion circuit 44.
After converting the input parallel signal MDT into a serial signal VDOM, the parallel-serial conversion circuit 44 drives the laser 55 by a laser driver (not shown).
In the first embodiment, the parallel signal has four bits (X1, X
2, X3, X4). Similarly, processing for one line in the main scanning is sequentially performed. 45 is a clock generation circuit,
A BD signal which is a main scanning synchronization signal is input, and a clock signal VCK is generated as a clock signal synchronized with the signal.
The clock signal VCK has a clock frequency f0 required for printing at 300 dots / inch in the main scanning direction.
Has four times the frequency of The serial signal VDOM is sequentially transmitted in synchronization with the clock VCK. A frequency dividing circuit 46 receives the clock signal VCK and divides the frequency by 4 to generate a clock signal VCKN having a frequency f0.
The clock signal VCKN is used as a synchronous clock when fetching dot data from the dot matrix memory into the processing circuit 43.

In FIG. 8, when an image signal VDO of 300 dots / inch is transmitted from the controller to the printer in synchronization with the image clock signal VCLK, the image dot data is sequentially stored in the line memories 1 to 9. At the same time, dot matrix information of 11 dots in the main scan × 9 dots in the sub-scan is extracted from the dot data of the line memories 1 to 9 in the shift registers 1 to 9. Thereafter, the processing circuit 43 detects the feature of the dot matrix information, and generates and prints modified data including four data X1 to X4 obtained by dividing the target pixel into four in the main scanning direction in accordance with the detected feature. It works like it does.

FIG. 11 and FIG. 12 show the features of the dot pattern extracted from the matrix area of 11 dots in the main scanning direction × 9 dots in the sub-scanning direction over the entire area of the matrix area.
FIG. 9 is a diagram illustrating an algorithm for checking whether or not a dot pattern is to be smoothed; Hereinafter, this figure will be described. FIG. 11A shows the main scanning direction 11.
FIG. 5 is a diagram showing a reference area of 9 dots in the sub-scanning direction and a, b, c, d, e, f, g, h, i,
j, k, 1, 2, 3, 4, 5, 6, with respect to the sub-scanning direction
A matrix of 7, 8, and 9 represents 99 pixels. For example, the center pixel is represented by 5f. The center pixel is selected as a pixel to be changed for smoothing. FIG. 11 (b)
The reference areas in FIG. 11A are represented by X1 to X8, Y1 to Y
It is divided into 17 areas of 8.5 f. Here, the region X1 is 3d, 3e, 3f, 4d, 4e, 4f,
The area X2 is 3f, 3g, 3h, 4f, 4g, 4h, the area X3 is 6d, 6e, 6f, 7d, 7e, 7f, and the area X4
Are 6f, 6g, 6h, 7f, 7g, 7h, and the area X5 is 3
d, 3e, 4d, 4e, 5d, 5e, area X6 is 5d,
5e, 6d, 6e, 7d, 7e, area X7 is 3g, 3
h, 4g, 4h, 5g, 5h, and the area X8 is 5g, 5h,
It is composed of 6 dots of 6g, 6h, 7g and 7h.
The area Y1 is 1a, 1b, 1c, 2a, 2b, 2c,
3a, 3b, 3c, area Y3 is 1i, 1j, 1k, 2
i, 2j, 2k, 3i, 3j, 3k, area Y4 is 4i,
4j, 4k, 5i, 5j, 5k, 6i, 6j, 6k, the region Y5 is 7i, 7j, 7k, 8i, 8j, 8k, 9i,
9j, 9k, area Y7 is 7a, 7b, 7c, 8a, 8
b, 8c, 9a, 9b, 9c, the area Y8 is 4a, 4b,
It is composed of 9 dots each of 4c, 5a, 5b, 5c, 6a, 6b, 6c. The area Y2 is 1d, 1e, 1f,
1g, 1h, 2d, 2e, 2f, 2g, 2h, area Y6
Are 8d, 8e, 8f, 8g, 8h, 9d, 9e, 9f,
It consists of 10 dots each of 9g and 9h. As described above, the above-mentioned area is composed of eight areas composed of 6 dots (: X1 to X8).
And six areas consisting of 9 dots (: Y1, Y3, Y4,
Y5, Y7, Y8) and two areas (: Y2, Y6) each composed of 10 dots and a central pixel 5f.

Here, the characteristics of each area are represented as Xn and Yn. If all the dots in each area are the same (: all pixels are ＜< white pixels> or all pixels are ＜< black pixels>), the feature (Xn, Yn) of each area is set to “0”.
If the dots in each area are different from each other (: ○ <white pixels> and ● <black pixels> are mixed), the feature (Xn, Yn) of each area is set to “1”. For example, if all the dots in the area X1 are ○ dots, the characteristic of the area X1 is X
When 1 = “0” and all the dots in the area X1 are ● dots, the feature of the area X1 is X1 = “0”, and when the dots in the area X1 are composed of ○ dots and ● dots, The feature of X1 is X1 = "1". The characteristics of each area described above are detected by the circuit shown in FIG. In the figure, A1 to A16 are exclusive logic circuits, and each of the exclusive logic circuits A1 to A16 is an area (: X1 to X8, Y1 to Y8).
Exclusive logic is applied to all the pixel signals in (i.e., if the input signals are all the same, the output is "0", and if the input signals are different from each other, the output is "1"). In this way, X1 to X8 and Y1 to Y8 are obtained as characteristics of each area.
In addition, the circuit shown in FIG.
One or more of the features Yn for the regions
This is a circuit for detecting that the value is "0". In the figure, B
1 to B8 are inverter circuits, and C1 is an OR circuit. That is, the characteristic signals Y1 to Y8 of each area are logically inverted by the inverter circuits B1 to B8, respectively,
1 is input. Thereby, the output Z1 of the OR circuit C1 is obtained.
Outputs "1" when any one of Y1 to Y8 is "0".

An embodiment of the present invention in which the first problem described in the prior art is improved will be described below.

FIGS. 23 (a) to 23 (d) are diagrams for explaining a smoothing process for a figure whose boundary portion has a slope of 1/2 (a horizontal line of 45 degrees or less). In each figure, when the bit pattern shown in the left figure is detected, the target pixel (:
The processing for changing the center pixel) is performed as shown in the right figure. The details of the algorithms shown in FIGS.
FIG. FIG. 25 is a diagram illustrating a specific algorithm corresponding to FIG. As shown in FIG. 25B, the areas X5 = 0 and X2 = 0 and the areas Y1 to Y1
At least one of the regions Y8, X3, and X4 is 0, and the bit pattern is 7a as shown in FIG.
= 7b = 6c = 6d = 5e = 5f = 4g = 4h = 3i =
3j = 2k = 0, 8a = 8b = 7c = 7d = 6e = 6f
= 5g = 5h = 4i = 4j = 3k = 1, the target pixel (: central pixel) 5f is set to x1 = 0, x2 =
0, x3 = 1, x4 = 1 and output. FIG. 29 shows a circuit for realizing the algorithm. In the figure, B1 to B15 are inverter circuits, E1 to E3 are AND circuits, and C1 is an OR circuit. Information of the areas X2 and X5 is input to the input of the AND circuit E2, and the OR circuit C
In the case where the information of the regions X3, X4, Y1 to Y8 (: Z1) is input to the input of 1, the information of the bit pattern is input to the input of the AND circuit E1, and the output of the AND circuit E3 satisfies the same condition. Outputs "1" as the PTN1 output, and outputs "0" if they do not match. This PTN1 output is connected to an OR circuit Q4 of the circuit of FIG.
Connected to the input of FIG. 26 is a diagram illustrating a specific algorithm corresponding to FIG. FIG.
As shown in (b), the area X1 = 0 and the area Y1
26, at least one region among Y8, X7, X3 and X4 is 0, and the bit pattern is 7a = 6b = 6c = 5d = 5e = 4f = 4g = 3h as shown in FIG.
= 3i = 2j = 2k = 0,8a = 7b = 7c = 6d = 6
If e = 5f = 5g = 4h = 4i = 3j = 3k = 1, the target pixel (: center pixel) 5f is set to x1 = 0,
The output is changed to x2 = 0, x3 = 1, and x4 = 1. FIG. 30 shows a circuit for realizing the algorithm. In the figure, B1 to B15 are inverter circuits, E1 and E3 are AND circuits, and C1 is an OR circuit. Inverter circuit B
The information of the area X1 is input to the input of the area 12, and the areas X3, X4, X7, Y1 to Y are input to the input of the OR circuit C1.
8 (: Z1), the bit pattern information is input to the input of the AND circuit E1, and the AND circuit E3 is input.
Is output as PTN2 output when the same condition is satisfied, and "0" is output when the same is not satisfied. This PTN2 output is connected to an input of an OR circuit Q13 of the circuit of FIG. 15 described later. FIG.
It is a figure explaining the concrete algorithm corresponding to 3 (c). As shown in FIG. 27B, the areas X8 = 0, X3 =
0, and at least one of the regions Y1 to Y8, X1, and X2 is 0, and the bit pattern is 7a = 6b = 6c = 5d = 5e as shown in FIG.
= 4f = 4g = 3h = 3i = 2j = 2k = 1,8a = 7
b = 7c = 6d = 6e = 5f = 5g = 4h = 4i = 3j
= 3k = 0, the pixel of interest (: central pixel)
5f is changed to x1 = 1, x2 = 1, x3 = 0, x4 = 0 and output. FIG. 31 shows a circuit for realizing the algorithm. In the figure, B1 to B15 are inverter circuits, E1 to E3 are AND circuits, and C1 is an OR circuit. Information of the areas X3 and X8 is input to the input of the AND circuit E2, and the areas X1 and X8 are input to the input of the OR circuit C1.
2, the information of Y1 to Y8 (: Z1) is input, the information of the bit pattern is input to the input of the AND circuit E1, and if the same condition is satisfied as the output of the AND circuit E3, P
"1" is output as the TN3 output, and "0" is output if they do not match. This PTN3 output is connected to an input of an OR circuit Q4 of the circuit shown in FIG. FIG.
FIG. 8 is a diagram for explaining a specific algorithm corresponding to FIG. As shown in FIG. 28B, the area X4 =
0, X8 = 0 and the regions Y1 to Y8, X1, X
2. At least one area of X6 is 0, and the bit pattern is 7a = 7b = 6 as shown in FIG.
c = 6d = 5e = 5f = 4g = 4h = 3i = 3j = 2k
= 1,8a = 8b = 7c = 7d = 6e = 6f = 5g = 5
If h = 4i = 4j = 3k = 0, the target pixel (: central pixel) 5f is set to x1 = 1, x2 = 1, x3 =
0, x4 = 0 and output. FIG. 32 shows a circuit for realizing the algorithm. In FIG.
16 is an inverter circuit, E1 to E3 are AND circuits, C1
Is an OR circuit. Area X is input to the input of AND circuit E2.
4 and X8, information of the areas X1, X2, X6, Y1 to Y8 (: Z1) is input to the input of the OR circuit C1, and bit pattern information is input to the input of the AND circuit E1. Then, if the same condition is satisfied as the output of the AND circuit E3, "1" is output as the PTN4 output, and if not, "0" is output. This PT
The N4 output is connected to an input of an OR circuit Q13 of the circuit shown in FIG.

FIGS. 34 (a) to 34 (d) are diagrams for explaining the smoothing process for a graphic whose boundary portion has an inclination of 2/1 (: a vertical line of 45 degrees or more). In each figure, when the bit pattern shown in the left figure is detected, the target pixel (:
The processing for changing the center pixel) is performed as shown in the right figure. FIG.
Details of the algorithms (a) to (d) are shown in FIGS.
FIG. FIG. 35 is a diagram illustrating a specific algorithm corresponding to FIG. As shown in FIG. 35 (b), the regions X1 = 0 and X6 = 0, and the regions Y1 to Y1
At least one of Y8, X4 and X7 is 0, and the bit pattern is 1h as shown in FIG.
= 2g = 3g = 4f = 5f = 6e = 7e = 8d = 9d =
0,1i = 2h = 3h = 4g = 5g = 6f = 7f = 8e
= 9e = 1, the pixel of interest (: central pixel)
5f is changed to x1 = 0, x2 = 0, x3 = 0, x4 = 1 and output. FIG. 36 is a diagram illustrating a specific algorithm corresponding to FIG. FIG. 36 (b)
As shown in the figure, the area X5 = 0, and the areas Y1 to Y
8, at least one of X3, X8 and X7 is 0
And the bit pattern is 1h = 2h = 3g = 4g = 5f = 6f = 7e = 8e = as shown in FIG.
9d = 1, 1g = 2g = 3f = 4f = 5e = 6e = 7d
= 8d = 9c = 0, the target pixel (: central pixel) 5f is set to x1 = 0, x2 = 1, x3 = 1, x4 =
Change to 1 and output. FIG. 37 is a diagram illustrating a specific algorithm corresponding to FIG. FIG.
As shown in FIG. 37B, the areas X4 = 0 and X7 = 0, and at least one of the areas Y1 to Y8, X1 and X6 is 0, and the bits are set as shown in FIG. The pattern is 1h = 2h = 3g = 4g = 5f = 6f = 7e =
8e = 9d = 0, 1g = 2g = 3f = 4f = 5e = 6e
= 7d = 8d = 9c = 1, the target pixel (: central pixel) 5f is set to x1 = 1, x2 = 0, x3 =
0, x4 = 0 and output. FIG.
It is a figure explaining the concrete algorithm corresponding to (d). As shown in FIG. 38B, the areas X4 = 0 and X8 = 0
38, and at least one of the regions Y1 to Y8, X1, and X8 is 0, and the bit pattern is 1h = 2g = 3g = 4f = 5f =
6e = 7e = 8d = 9d = 1, 1i = 2h = 3h = 4g
= 5g = 6f = 7f = 8e = 9e = 0, the target pixel (: center pixel) 5f is set to x1 = 1, x2 =
1, x3 = 1, x4 = 0 and output.

Actually, each pattern shown in FIGS. 23A to 23D is a set (eight patterns in total) of the feature extraction of a pattern in which the left and right are interchanged around the target pixel (center pixel).
Having. Each of the patterns in FIGS. 34A to 34D has a set (eight patterns in total) of feature extraction of a pattern in which the left and right are switched around the target pixel (center pixel). For example, FIG. 24 shows a result obtained by exchanging the left and right with respect to the feature extraction shown in FIG. The smoothing algorithm in this case is 2a = 3b = 3c =
4d = 4e = 5f = 5g = 6h = 6i = 7j = 7k = 0
(: Dot), 3a = 4b = 4c = 5d = 5e = 6f
= 6g = 7h = 7i = 8j = 8k = 1 (: ● dot),
If X7 = X1 = 0 and at least one of Y1 to Y8, X3, and X4 is “1”, the pixel of interest 5
f is changed to x1 = 1, x2 = 1, x3 = 0, x4 = 0. Similarly, a left-right symmetric algorithm is set for FIGS. 23 (b), (c), and (d). By making the feature extraction algorithm symmetrical in this manner, for example, smoothing of characters such as "O", "U", "V", and "W" is performed by a symmetrical algorithm, and the appearance of the characters is naturally increased. I do.

FIGS. 15 and 16 show a data generation circuit which inputs each output signal of a plurality of feature detection circuits including the above-described feature detection circuits and generates data of the target pixel 5f in accordance with the input signals. . FIG. 16 shows a part of the circuit shown in FIG. 15 in detail. In FIGS.
Q16 is an OR circuit, R1 to R64 and U1 to U2 are 2-input AND circuits, S1 to S4 are 16-input OR circuits, E4 is an inverter circuit, and T1 is a NOR circuit. Each of the output signals of the feature detection circuit extracted as the plurality of features,
It is connected to one of the OR circuits Q1 to Q16. As for the outputs of the above-described feature detection circuits, PTN1 and PTN3 are output to an OR circuit Q4, and PTN2 and PTN4 are output to an OR circuit Q4.
13, etc. are connected. In addition, the above PTN1
Signals from all the feature detection circuits including .about.PTN4 are connected to the NOR circuit T1. Corresponding to the output “1” of the OR circuits Q1 to Q16, the AND circuit groups of R1 to R64 are each 4
20 (: R
4 outputs), 21 (: R3 output), 22 (: R2 output),
23 (: R1 output) as a 4-bit code from "0" to
Generate a code with "F" as it is. The 20 digit of these code outputs is ORed by the OR circuit S1 and output as the output of the OR circuit S1: x1. Also, the digit 21 of the code output is ORed by the OR circuit S2 and the output of the OR circuit S2: x
Output as 2. The 22 digit of the code output is off.
The output is ORed by the OR circuit S3 and output as the output of the OR circuit S3: x3. The 23 digit of the code output is ORed by the OR circuit S4 and output as the output of the OR circuit S4: x4. In this manner, one code "0" corresponding to the outputs of Q1 to Q16 that is not selected at least two at the same time
To “F” are obtained as outputs x1 to x4 of the OR circuits S1 to S4. For example, if the code is “3”, x1
= 1, x2 = 1, x3 = 0, x4 = 0, and when the code is “9”, x1 = 1, x2 = 0, x3 =
0, x4 = 1. Since all the feature matching signals are connected to the input of the NOR circuit T1, the output of T1 is "1" when one of the feature matching signals does not become "1" (when no feature matches). become. At this time, when the target pixel 5f is a dot, the output of the two-input AND circuit U1 becomes "1", the output of the OR circuit Q1 becomes "1", and x1
Output code “0” to x4 (: x1 = 0, x2 = 0, x
3 = 0, x4 = 0), and when the target pixel 5f is a dot, the output of the two-input AND circuit U2 becomes "1", the output of the OR circuit Q16 becomes "1", and the codes "x1" to "x4" are written. F "(: x1 = 1, x2 = 1, x3 = 1, x4
= 1). When the feature does not match the predetermined feature, the data of the target pixel 5f is stored and printed as it is.

The outputs x1 to x4 of the data generation circuit are converted into x1, x
A signal VDOM output in synchronization with the clock signal VCK is generated in the order of 2, x3, x4, and the VDOM signal drives a laser through a laser driver.

Using the above algorithm, FIG.
FIG. 22B shows an output image signal when the horizontal line having a slope of 45 degrees or less shown in FIG. As can be seen from the figure, the pulse width signal added to the left boundary of the line image and the width signal removed from the center of the line image at the left boundary are processed so as to have the same time. . In this case, the same processing is performed on the right boundary portion of the line image. Further, the processing is performed so that the pulse width added to the left boundary of the line image is equal to the pulse width added to the right boundary.

FIG. 33B shows an output image signal when a vertical line having an inclination of 45 degrees or more shown in FIG. 33A is subjected to smoothing processing. As can be seen from the figure, the same signal width as the signal width added (or deleted) to the left end boundary of the line image is deleted (or added) from the right end boundary.
As a result, it is possible to make the thickness of the line after the smoothing process equal to the case where no smoothing is performed.

By such processing, it is possible to prevent the line after the smoothing processing from becoming thicker or thinner. Further, the appearance of the left end and the right end of the line can be made uniform, which is effective for improving the image quality of the line.

An embodiment of the present invention in which the second problem described in the above prior art is improved will be described below.

In the conventional smoothing technique, an algorithm for smoothing a line image is shown in FIGS.
An algorithm as shown in (d) is used to perform a smoothing process on the line image shown in (a) of FIG. As a result, lines were printed smoothly. FIG.
The line (a) is a line in which the dot configuration at the boundary of the line is 2 dots in the vertical direction and 2 dots in the horizontal direction. Such a line having a dot configuration of two or more dots in the vertical direction and two or more dots in the horizontal direction should not be subjected to smoothing processing. This is because such a line created by a computer should have been basically developed into dot data for the purpose of printing in a jagged manner. This is because it should be easy for the computer to develop dot data of one vertical dot and one horizontal dot if necessary in a memory area that can be managed by the computer. Therefore, performing the smoothing process on such a boundary line changes image data.
Therefore, in the embodiment of the present invention, such smoothing processing for lines having two or more dots in the horizontal direction and two or more dots in the horizontal direction is prohibited, and the original dots are not changed as shown in FIG. Print.

An embodiment of the present invention in which the third disadvantage of the conventional technique is improved will be described below.

FIGS. 43 (a) to 43 (h) are diagrams showing a smoothing algorithm proposed by the present applicant for a horizontal line whose boundary is nearly horizontal. Since the detailed algorithm is the same as that described above, the description is omitted here. When smoothing is performed using the above algorithm, the signal resulting from smoothing for one dot line as shown in FIG. 42A is shown in FIG. 42B, and as a result, as shown in FIG. Printed on paper. A signal obtained by smoothing one dot line as shown in FIG. 42D is shown in FIG.
As a result, printing is performed on paper as shown in FIG. As can be seen from FIG. 42 (c) and FIG. 42 (f), even if the same one dot line is
In the case of (a) and the case of FIG. 42 (d), printing is performed by one dot less in the case of FIG. 42 (d) where the black dot area added to the smoothing processing unit is. As a result, smoothing printing as shown in FIG.
In the case of (f), the line becomes thin, and a line break occurs. Conversely, when the dot area added to the smoothing processing unit is increased by an algorithm that eliminates thinning in FIG. 42 (f), thinning due to smoothing processing in FIG. 42 (d) can be prevented. As a result, fattening occurs in the smoothing processing unit for a).

FIGS. 45 (a) to 45 (p) are diagrams showing a smoothing processing algorithm for a boundary of a horizontal line which is nearly horizontal according to the embodiment of the present invention. Since the detailed algorithm is the same as that described above, the description is omitted here. The algorithm in FIG. 45A has a set of a total of four patterns that are symmetrical left and right and vertically symmetrical with respect to the target pixel (center pixel) as shown in FIG. Similarly, each of the algorithms shown in FIGS. 45 (b) to (p) has a set of a total of four unillustrated patterns which are symmetrical left and right and vertically symmetric with respect to the target pixel (: central pixel). . In the algorithm of FIG. 45, it is possible to determine which one of the two types of one-dot line is the one-dot line. Therefore, an optimal process that prevents thickening or thinning of the line according to each type. Can be implemented. FIG. 44 is an explanatory diagram in the case where the smoothing processing is performed on the two types of one-dot lines using the algorithm according to the embodiment of the present invention. As can be seen from FIG. 44, the algorithm for discriminating the two types of one-dot lines is incorporated, and printing is performed so that the black area of the pixel in the smoothing processing unit is equal for both types of one-dot lines. And
As a result, thinning and thickening of lines in the smoothing processing unit can be prevented from occurring, and high-quality printing can be performed.

FIGS. 48A to 48F are diagrams showing a smoothing algorithm proposed by the present applicant for a vertical line whose boundary is nearly vertical. Since the detailed algorithm is the same as that described above, the description is omitted here. When smoothing is performed using the above algorithm, a signal obtained by smoothing one dot line as shown in FIG. 47A is as shown in FIG. 47B. FIG. 47D shows a signal obtained by smoothing one dot line as shown in FIG. FIG.
As can be seen from FIGS. 7 (b) and 47 (d), black dots added to the smoothing processing unit in the case of FIG. 47 (a) and the case of FIG. In the case where the area is as shown in FIG. 47 (a), one dot is printed more. As a result, the line in the case of FIG. 47B as the smoothing printing for FIG. Conversely, when the dot area added to the smoothing processing unit is reduced by an algorithm that eliminates the fatness in FIG. 47A, the fattening due to the smoothing process in FIG. 47A can be prevented. As a result, thinning in the smoothing processing unit occurs for c).

FIGS. 50 (a) to 50 (m) are diagrams showing a smoothing processing algorithm for a boundary portion of a nearly vertical vertical line according to the embodiment of the present invention. Since the detailed algorithm is the same as that described above, the description is omitted here. The algorithm in FIG. 50A has a set of a total of four unillustrated patterns that are symmetrical left and right and vertically symmetrical with respect to the target pixel (: central pixel), as described above. In the algorithm of FIG.
Since it is possible to determine which of the two types of 1 dot line, that is, a line where pixels overlap or a line where pixels do not overlap, it is possible to prevent thickening or thinning of the line according to each type. Optimal processing can be performed. FIG. 49 is an explanatory diagram in the case where the smoothing process is performed on the two types of one-dot lines using the algorithm according to the embodiment of the present invention. As can be seen from the figure, the algorithm for identifying the two types of one-dot lines is incorporated, and the black area of the pixel in the smoothing processing unit is one of the two types.
Printing is performed equally on the dot lines. As a result, thinning and thickening of the lines in the smoothing processing unit can be prevented, and high-quality printing can be performed.

An embodiment of the present invention in which the fourth problem of the prior art is improved will be described below.

FIG. 51 shows the smoothing process for a horizontal line having a width of one dot with white and having an inclination of 45 degrees or less. As can be seen from FIG. 51 (c), as a result of the smoothing processing, the white line is printed as a broken line, and the image quality is degraded. In this case, the effect of the smoothing is also easily affected by the process conditions and the environment such as temperature and humidity.

FIGS. 53 (a) to 53 (h) are diagrams for explaining a processing algorithm according to the embodiment of the present invention for coping with the above inconvenience. The detailed algorithm is the same as that described above, and thus the description is omitted here. In addition,
Each of the algorithms shown in FIGS. 7A to 7H has a set of a total of four patterns (not shown) that are symmetrical left and right and vertically symmetric with respect to the target pixel (: central pixel), as described above. . When the pattern matches the pattern, the data of the target pixel is output as it is without changing the target pixel (: center pixel). FIG. 52 shows printing in the case where such processing is performed. As can be seen from the figure, although the printed image after the processing is not smoothed, the image is printed in such a manner that the line is not thinned (same as the original image).

An embodiment of the present invention in which the fifth problem of the prior art is improved will be described below.

FIGS. 54 and 57 show patterns of bit data of kanji characters as Japanese characters. The bit patterns are generated by the character generator and transmitted to the printer. When the smoothing process is performed on the kanji character pattern as shown in FIG. 43 using the conventional algorithm shown in FIGS. 43 and 48, the pattern of FIG. Is smoothed as shown in FIG.
That is, mustache-like pulses and bumps as shown in (A) to (D) are printed, respectively, and the image quality deteriorates with respect to the original kanji character. The above problems often occur in kanji characters in which the font pattern of a character includes many straight and right-angled portions, but some of the alphabetic characters, for example, "E", "F", etc. The image quality is similarly degraded for a specific font of characters. An algorithm according to an embodiment of the present invention will be described with reference to FIGS. FIG. 60 shows an embodiment of the present invention for the pattern of FIG. 45 (d). In FIG. 60, (a) is a verification pattern, and (b) is a prohibition pattern for (a).
That is, even if the pattern matches the pattern of (a), if it matches the pattern of (b), it is regarded as non-corresponding, and the target pixel (: central pixel) is not changed. . FIG. 61 shows a detailed description of the algorithm in this case.
This will be described with reference to FIG. In the figure, the area is X1 = X2 =
0, and at least one of Y1 to Y8, X3, and X4 is “0”, and 5a = 5b = 5c =
5d = 5e = 5f = 5g = 5h = 5i = 4j = 4k = 0
(: Pixels) and 6a = 6b = 6c = 6d =
6e = 6f = 6g = 6h = 6i = 6j = 5j = 5k = 1
(: Pixels) and “9i = 0 (: pixels),
6k = 7j = 7k = 8j = 8k = 9j = 9k = 1 (: ●
Target pixel (: central pixel) 5f only when it is not
Is changed to x1 = 0, x2 = 0, x3 = 1, x4 = 0 and printing is performed. This algorithm can be realized by the circuit shown in FIG. In the figure, B1 to B17 are inverter circuits, E1 to E5 are AND circuits, and C1 is an OR circuit. One input of the AND circuit E5 is connected to the output of the AND circuit E1 as a result of the comparison of the pattern (a), and the other input is connected to the other input of the pattern A as a result of the comparison of the prohibited pattern (a '). The output of inverter B17 is connected. The collation output PTN10 thus obtained
Are connected to the OR circuit Q5 of FIG. FIG.
A similar prohibited pattern is set for each of the patterns (a) to (p). Therefore, according to this algorithm, even if the pattern of FIG.
When the pattern matches (a '), the processing can be eliminated and the target pixel can be changed. FIG.
FIG. 3 shows an embodiment of the present invention for the pattern of FIG. In FIG. 63, (a) is a verification pattern, and (b) is a prohibition pattern for (a). That is, even if the pattern matches the pattern of (a), if it matches the pattern of (b), it is regarded as non-corresponding, and the target pixel (: central pixel) is not changed. . FIG. 6 is a detailed description of the algorithm in this case.
4 will be described. In the figure, the area is X5 = X6
= 0, and at least one of Y1 to Y8, X7, X8 is "0", and 1e = 2e = 3f
= 4f = 5f = 6f = 7f = 8f = 9f = 0 (: o pixel) and 1f = 2f = 2g = 3g = 4g = 5
g = 6g = 7g = 8g = 9g = 1 (: pixels) and “3j = 0 (: pixels), 1g = 1h = 1i =
Only when “1j = 2h = 2i = 2j = 1 (: ● pixel)” is not satisfied, the target pixel (: center pixel) 5f is set to x1 = 0, x2 =
0, x3 = 0 and x4 = 1 are printed. This algorithm can be realized by the circuit shown in FIG. In FIG. 65, B1 to B17 are inverter circuits, and E1 to E17.
E5 is an AND circuit, and C1 is an OR circuit. The output of the AND circuit E1 is connected to one input of the AND circuit E5 as a result of comparison of the pattern of FIG. 64A, and the other input of the AND circuit E5 is the inhibition pattern of FIG. The output of the inverter B17 is connected as a result of the comparison. The collation output PTN20 thus obtained
Are connected to the OR circuit Q9 of FIG. FIG.
A similar prohibited pattern is set for each of the patterns (a) to (m). Therefore, according to this algorithm, even if the pattern of FIG.
If the pattern matches (a '), the processing is eliminated and the pixel of interest is not changed.

According to the present embodiment, the kanji character "day" of FIG. 54 is printed as shown in FIG. 56, and the kanji character "plate" of FIG. 57 is printed as shown in FIG. As can be seen from the figure, it is understood that the image quality does not deteriorate.

[Other Embodiments] In the embodiments described above, the controller performs main scanning, with respect to the printer engine having a printing function of 300 dots / inch in the sub-scanning direction.
When the image data of 300 dots / inch is transmitted in both the sub-scanning, the resolution in the printer engine is equivalent to four times the resolution in the sub-scanning direction (1200 dots / inch) in the printer engine.
Inch) and printing at a printing density of 300 dots / inch in the sub-scanning direction. However, the equivalent printing density in the main scanning direction need not be limited to four times that in the sub-scanning direction, , 5 times, 6 times, 7 times, 8 times,... For example, in the case of performing the smoothing process at 8 times (: 2400 dots / inch) in the main scanning direction, the pattern generation unit of the change pattern generation circuit described with reference to FIG. ), It is sufficient to constitute one pixel by an 8-bit small signal (: x1 to x8).

Next, when the controller transmits image data of 300 dots / inch in both the main scanning and sub-scanning to the printer engine having a printing function of 600 dots / inch in the sub-scanning direction, Another embodiment in which printing is performed at a printing density of 1200 dots / inch equivalently in the main scanning direction and 600 dots / inch equivalently in the sub-scanning direction will be described.

FIG. 17 shows how to set a small area for printing a pixel of interest in this embodiment. In this embodiment, the target pixel (: 5f) at the center of a dot matrix memory consisting of 11 dots in the main scanning direction of 300 dots / inch × 9 dots in the sub-scanning direction is quadrupled in the main scanning direction × sub-scanning direction. Is a set of sub-areas with double printing density (: x1, x2, x
3, x4, x1, y2, y3, y4).

In this embodiment, the characteristics of the image data in the peripheral area (11 pixels in the main scanning direction × 9 pixels in the sub-scanning) surrounding the pixel of interest are checked with respect to the data transmitted from the controller, and the pixel of interest is determined in accordance with the result. To change. More specifically, for example, the resolution 3 shown in FIG.
When printing a pixel of interest in the dot data group of the alphabetical character "a" of 00 dots / inch, an area surrounding the pixel of interest (11 pixels in main scanning × 9 pixels in sub-scanning = 99)
The dot data of the pixel is stored in the temporary storage means. Thereafter, the characteristics of the dot data group in the area are examined, and the data of the target pixel to be printed is changed according to the characteristics and printed.
In this case, the data is changed so that the outline of the figure composed of the dot group is printed smoothly.
In the present embodiment, as shown in FIG. 17, the pixel of interest is divided into four sub-divisions (x1, x2,
x3, x4, y1, y2, y3, y4). Therefore, in the printing stage, the main scanning 1200 is equivalently performed.
Printing is performed at a print density of dot / inch × 600 dots / inch in sub-scanning.

FIG. 18 is a diagram showing a circuit block of the VDO signal processing unit 101 for performing a smoothing process, which is provided at the input unit of the printer engine of 600 dots / inch, and is a diagram described in the first embodiment. FIG. 8 is a diagram corresponding to FIG. In FIG. 18, devices having the same reference numerals as those in FIG. 8 indicate devices having the same function.

In FIG. 18, SW1 to SW9 are switch means for switching the positions of ".alpha." And ".beta." In FIG. 18 to switch the signals input to the line memories 25 to 33. The switching position of the switch means is controlled by a control signal SWC issued by a control circuit 47 described later. The control circuit 47 inputs a synchronizing signal BD 'corresponding to 600 dots / inch in sub-scanning for printing at 600 dots / inch, and every time the synchronizing signal BD' is input, BD '
A control signal SWC which is inverted in synchronization with the signal is generated. The synchronization signal BD for interfacing with the controller
Is generated as a signal corresponding to the sub-scanning 300 dots / inch, which is obtained by extracting the synchronizing signal BD ′ for each main scanning line. First, the switch means SW1 to SW9 are set to the “α” position. The controller transmits the image data VDO of 300 dots / inch in synchronization with the BD signal.
The line memories 1 to 9 store the 300 dot / inch image signal VDO while sequentially shifting the image signal VDO in synchronization with the clock signal VCLK, and each line memory stores dot information of the main scanning length for a page to be printed. . Each line memory is connected in the order of line memory 1 → line memory 2 → line memory 3 →... → line memory 9 and stores dot information having a main scanning length of 9 lines in the sub-scanning direction. I do. Thereafter, the switch means SW1 to SW9
Is switched to the position “β” by a control signal SWC issued from the control circuit 47. Reference numerals 34 to 42 denote shift registers. Each shift register 1 to 9 receives an output from each line memory in synchronization with the clock signal VCKN corresponding to each of the line memories 1 to 9. At this time, the data output from each line memory is re-input to the line memories 1 to 9 via the switch means SW1 to SW9. Each shift register is composed of 11 bits, and as shown, 1a-1k, 2a-2k, 3a-3k,
... A dot matrix memory of 11 dots in the main scanning direction 9a to 9k × 9 lines in the sub-scanning direction is configured.
In the matrix memory, a central dot 5f is defined as a target dot. Reference numeral 43 denotes a processing circuit which detects the characteristics of data stored in the dot matrix memory for smoothing and changes the pixel of interest 5f as necessary. Each bit of the shift registers 1 to 9 (: 1a
-9k), and the changed parallel signal MDT (: x1, x2, x3, x4) is output. The parallel signal MDT (x1, x2, x3, x4)
Is input to the parallel-serial conversion circuit 44. The parallel / serial conversion circuit 44 receives the input parallel signal M
DT (x1, x2, x3, x4) is converted to a serial signal VDO.
After the conversion into M, the semiconductor laser 55 is driven by the laser driver 50. Similarly, processing for one line in the main scanning is sequentially performed.

Thereafter, the switch means SW1 to SW9 are switched to the position "α". The data is similarly transferred from the line memories 1 to 9 to the next line memory by reading from the line memories 1 to 9 and output to the shift registers 1 to 9 in synchronization with the synchronization signal BD ′ input at the next timing. The processing circuit 43 detects characteristics of data stored in a dot matrix memory of 11 dots in the main scanning direction × 9 dots in the sub-scanning direction output from the shift register, and changes the target pixel 5f as necessary. The parallel signal MDT (y1, y2, y3, y4) is output. The parallel-serial conversion circuit 44 converts the input parallel signal MDT (y1, y2, y3, y4) into a serial signal VDOM, and then drives the semiconductor laser 55 by the laser driver 50. Similarly, processing for one line in the main scanning is sequentially performed.

Thereafter, the switch means SW1 to SW9 are switched to the position "α". Then, the image signal VDO of the next sub-scanning line of 300 dots / inch transmitted from the controller is input.

In this embodiment, the parallel signal is 4 bits as described above, but the first MDT signal (x1, x2, x3, x4) and the second MDT signal are changed according to the synchronization signal BD '.
The signals (y1, y2, y3, y4) are output alternately. Reference numeral 45 denotes a clock generation circuit which receives a BD 'signal as a main scanning synchronization signal and generates a clock signal VCK as a clock signal synchronized with the signal. The clock signal VCK has a frequency twice as high as the clock frequency f0 required for printing at 600 dots / inch in the main scanning direction. In synchronization with the clock VCK, the serial signal VDOM (: x1, x2, x3, x4 or y1, y
2, y3, y4) are sequentially transmitted. A frequency dividing circuit 46 receives the clock signal VCK and divides the frequency by 2 to generate a clock signal VCKN having a frequency f0. The clock signal VCKN is used as a synchronous clock when fetching dot data from the dot matrix memory into the processing circuit 43.

In the processing circuit 43, the feature extracting circuit section is the same as the circuits shown in FIGS. 13, 14, 29 to 32, 62, and 65 described in the first embodiment. FIGS. 19 to 21 show a data generation circuit of the processing circuit 43 used in this embodiment. In the figure, devices having the same reference numerals as those in FIGS. 15 and 16 indicate devices having the same functions. FIG. 20 shows the data generation unit 1 of FIG.
Shows the data generator 2 in FIG. 19 in detail.

FIGS. 19, 20 and 21 show data generating circuits for generating data of the target pixel 5f in accordance with the characteristics of the detected data string. Q1 to Q1 in FIGS.
6 and Q1 'to Q16' are OR circuits, R1 to R61 and R1 'to R64' and U1 to U2 are 2-input AND circuits,
S1 to S4 and S1 'to S4' and S5 to S8 are 16-input OR circuits, E4 and E18 are inverter circuits, and T1 is a NOR circuit. When the first MDT signal described with reference to FIG. 18 is generated, the control signal SWC output from the control circuit 47 outputs the “1” level. In this state, the data generator 1 is selected by the two-input AND circuits U3 to U6 and U3 'to U6' and the two-input OR circuits S5 to S8, and x1, x2, x3, and x4 are output as parallel signals. When the second MDT signal described with reference to FIG. 18 is generated, the control signal SWC output from the control circuit 47 outputs the “0” level. In this state, the two-input AND circuits U3 to U6 and U3 'to U6' and 2
The data generation unit 2 is selected by the input OR circuits S5 to S8, and y1, y2, y3, and y4 are output as parallel signals.

Each of the output signals of each feature detection circuit corresponding to a plurality of patterns is connected to one of OR circuits Q1 to Q16 to select output data of x1 to x4, and at the same time y1 to y4 OR circuit Q to select data
1 'to Q16'.

The outputs x1 to x4 of the data generation circuit are converted into x1, x
2, x3, x4 in order to generate a signal VDOM which is output in synchronization with the clock signal VCK.
4, a signal VDOM output in synchronization with the clock signal VCK is generated in the order of y1, y2, y3, y4,
The VDOM signal drives a semiconductor laser via a laser driver. The above algorithm is equally applicable to the 600 dot / inch printer engine described above.

In the above embodiment, 6
For a printer engine having a printing function of 00 dots / inch, the controller sends 30 dots for both main scanning and sub-scanning.
When the image data of 0 dot / inch is transmitted, the resolution is equivalent to four times the resolution in the sub-scanning direction (1200 dots / inch) in the printer engine and 600 in the sub-scanning direction in the printer engine. Although the case of printing at a dot / inch print density has been described, the equivalent print density in the main scanning direction need not be limited to four times that in the sub-scanning direction, but may be two times, three times.
, 5 times, 6 times, 7 times, 8 times,...
For example, in the case where the smoothing process is performed eight times (: 2400 dots / inch) in the main scanning direction, the processing shown in FIG.
The pattern generation section of the change pattern generation circuit described in 9 to 21 is a small signal of 4 bits (: x1 to x4, y1 to y4).
But 8-bit small signals (: x1 to x8, y1 to y8)
May constitute one pixel.

As described above, according to the embodiment of the present invention, the vertical line inclined at 45 degrees or more is added (or reduced) at the left end of the vertical line.
By reducing (or adding) the signal width (area) equal to the calculated signal width (area) at the right end of the line, it is possible to keep the line thickness unchanged. In addition, for a horizontal line having a slope of 45 degrees or less, a signal having a width equal to the pulse width added to the edge of the line is deleted from the original signal, so that the line thickness can be kept unchanged. .

Further, according to the present embodiment, even if the lines have the same inclination, it is unnecessary to determine whether smoothing should be performed or not by examining the dot configuration at the boundary. It is possible to prevent smoothing (having an adverse effect).

Further, according to the present embodiment, it is possible to determine which one of two types of one dot line a one dot line is, and to change the smoothing process for each type of one dot line. Regardless of the type of one-dot line, it is possible to provide a smoothing technique in which the image quality of the one-dot line does not deteriorate due to thickening or thinning of the line, broken lines, or the like.

Further, according to the present embodiment, it is determined whether or not the line is a one-dot white line, and if the line is a one-dot white line, the smoothing process is not performed. By this, the white line of 1 dot is prevented from thinning,
It is possible to maintain the stability of the image quality against the reverberation.

Further, according to this embodiment, the harmful effect of kanji characters can be eliminated by incorporating a prohibition pattern peculiar to kanji characters to eliminate the smoothing process for kanji characters.

[0105]

As described above, according to the present invention, with respect to a line formed with a line width of 1 bit width, whether the line width at the variation portion of the line is 1 bit width or more, By discriminating whether the 1-bit width is not secured and changing the algorithm of changing the pixel of interest accordingly, the image quality of any type of line can be reduced by making the line thicker or dashed. Smoothing without deterioration can be performed. Further, according to the present invention, when a white line formed with a line width of one bit width is detected, the pixel of interest is not changed by smoothing, thereby preventing the thinning of the white line of one bit width. it can. Further, according to the present invention, it is possible to eliminate the smoothing process for kanji by incorporating a kanji-specific prohibition pattern, and to prevent the harmful effect of kanji characters due to smoothing.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an engine part of a laser beam printer according to an embodiment of the present invention.

FIG. 2 is a diagram showing interface signals between an engine unit and a controller in FIG. 1;

FIG. 3 is a diagram showing interface signals between a printer engine unit and a controller.

FIG. 4 is a diagram illustrating an example of a pattern represented by dot data.

FIG. 5 is a diagram showing a matrix memory.

FIG. 6 is a schematic diagram of storing image data from a dot pattern of FIG. 4 in a matrix memory.

FIG. 7 is a schematic diagram of storing image data from a dot pattern of FIG. 4 in a matrix memory.

FIG. 8 is a diagram showing blocks of a first embodiment of the present invention.

FIG. 9 is a diagram for explaining the inconvenience of the related art.

FIG. 10 is a diagram showing a change area of a pixel of interest applied in the first embodiment of the present invention.

FIG. 11 is a diagram illustrating a data feature extraction algorithm applied to the present invention.

FIG. 12 is a diagram illustrating a data feature extraction algorithm applied to the present invention.

FIG. 13 is a diagram illustrating an example of a circuit of a feature extraction unit according to the present invention.

FIG. 14 is a diagram illustrating an example of a circuit of a feature extraction unit according to the present embodiment.

FIG. 15 is a diagram illustrating a circuit example of a feature extraction unit according to the present embodiment.

FIG. 16 is a diagram showing a part of FIG. 15 in detail.

FIG. 17 is a diagram showing another embodiment in which a target pixel is divided into four in the main scanning direction and into two in the sub-scanning direction.

FIG. 18 is a block diagram in which a pixel of interest is divided in a main scanning direction and a sub-scanning direction, and a smoothing process is performed.

FIG. 19 is a diagram illustrating a circuit example of a feature extraction unit according to another embodiment.

FIG. 20 is a diagram illustrating details of a data generation unit 1 of FIG. 19;

FIG. 21 is a diagram illustrating details of a data generation unit 2 in FIG. 19;

FIG. 22 is a diagram for describing the effect of smoothing of the present embodiment on a horizontal line of 45 degrees or less.

FIG. 23 is a diagram illustrating a smoothing process on a horizontal line of 45 degrees or less.

FIG. 24 is a diagram in which the pattern of FIG.

FIG. 25 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 26 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 27 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 28 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 29 is a diagram illustrating a feature extraction circuit corresponding to FIG. 16;

FIG. 30 is a diagram showing a feature extraction circuit corresponding to FIG. 17;

FIG. 31 is a diagram showing a feature extraction circuit corresponding to FIG. 18;

FIG. 32 is a diagram showing a feature extraction circuit corresponding to FIG. 19;

FIG. 33 is a diagram for explaining the effect of smoothing on a vertical line of 45 degrees or less.

FIG. 34 is a diagram for describing smoothing processing for 45 or more vertical lines.

FIG. 35 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 36 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 37 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 38 is a diagram illustrating an example of a feature extraction algorithm according to the present embodiment.

FIG. 39 is a diagram showing a conventional smoothing process for a line having a step of two dots in the vertical direction and two dots in the horizontal direction.

FIG. 40 is a diagram illustrating the processing of this embodiment for a line having a step of two dots in the vertical direction and two dots in the horizontal direction.

FIG. 41 is a diagram showing an algorithm of the smoothing process shown in FIG. 39.

FIG. 42 is a diagram showing a smoothing process for two types of one-dot lines.

FIG. 43 is a diagram showing a smoothing algorithm for a horizontal line whose boundary is nearly horizontal.

FIG. 44 is a diagram illustrating the effect of smoothing according to the present embodiment on a horizontal line that is nearly horizontal.

FIG. 45 is a diagram showing the smoothing algorithm of FIG. 44;

FIG. 46 is a view showing a pattern obtained by modifying FIG. 45 (a).

FIG. 47 is a diagram showing an example of a conventional smoothing process.

FIG. 48 is a diagram showing a smoothing algorithm for a vertical line whose boundary is nearly vertical.

FIG. 49 is a diagram illustrating the effect of smoothing according to the present embodiment on a vertical line that is nearly vertical.

FIG. 50 is a diagram showing an algorithm of the smoothing of FIG. 49.

FIG. 51 is a diagram showing an example of a conventional smoothing process for a white line.

FIG. 52 is a diagram showing the effect of the processing in this embodiment on a white line.

FIG. 53 is a diagram showing an algorithm for smoothing a white line.

FIG. 54 is a diagram showing a dot pattern of a kanji character.

FIG. 55 is a diagram showing an example of a conventional smoothing process for the kanji characters of FIG. 54;

FIG. 56 is a diagram showing the effect of the processing in this embodiment on the kanji characters of FIG. 54;

FIG. 57 is a diagram showing another example of a dot pattern of a kanji character.

FIG. 58 is a diagram showing an example of a conventional smoothing process for the kanji character of FIG. 57;

FIG. 59 is a diagram showing the effect of the processing in this embodiment on the kanji characters of FIG. 57;

FIG. 60 is a diagram for describing smoothing prohibition processing in this embodiment for kanji characters.

FIG. 61 is a diagram showing an example of a feature extraction algorithm for Kanji characters.

FIG. 62 is a diagram showing a smoothing prohibition circuit for Kanji characters.

FIG. 63 is a diagram for explaining another example of the smoothing prohibition process for the Chinese character.

FIG. 64 is a diagram showing another example of a feature extraction algorithm for Kanji characters.

FIG. 65 is a diagram showing another example of a smoothing prohibition circuit for Kanji characters.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Photosensitive drum 14 Developing device 51 Semiconductor laser 52 Polygon mirror 8-8 'Fixing device 200 Controller 100 Printer 101 VDO signal processing part 25-33 Line memory 34-42 Shift register 43 Processing circuit 44 Parallel-serial conversion circuit 45 Clock generation circuit 46 Frequency divider

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Tetsuo Saito 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Inventor Kiyoshi Kanaiwa 3-30-2 Shimomaruko, Ota-ku, Tokyo Within Non Corporation (72) Inventor Hiroshi Mano 3-30-2 Shimomaruko, Ota-ku, Tokyo Within Canon Corporation (72) Inventor Hiromichi Yamada 3-30-2 Shimomaruko, Ota-ku, Tokyo Within Canon Corporation (72) Inventor Michio Ito 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (56) References JP-A-59-133665 (JP, A) JP-A-63-80669 (JP, U) (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1/387 101 G06T 3/40

Claims (3)

(57) [Claims] And 1. A data signal generating means for generating an information signal, storage means for temporarily storing a part of bit information signals emitted from the information signal generating means, determine characteristics of the stored bit information is Me pre a feature detection means for detecting whether or not matches one of a plurality of smoothing to be characteristic which is, one of a plurality of smoothing to be characteristic of the feature detection unit is a predetermined When it is detected that the target pixel matches, the printing information of the specific pixel of interest in the information group stored in the temporary storage means is smoothed and finely processed.
Information changing means for changing the differentiated small image gu signal, and an output means for outputting a change signal from the information changing means to the output unit, the feature detection unit, a line width of 1 bit wide Formed
Against that line, or line width at inflection portions of said line is secured over one bit wide, 1 bit width to identify and not ensured, the information changing means the corresponding if each An information recording apparatus characterized in that an algorithm for changing a pixel of interest is changed. 2. A data signal generating means for generating an information signal, storage means for temporarily storing a part of bit information signals emitted from the information signal generating means, determine characteristics of the stored bit information is Me pre Multiple
Whether meets the one of the features to be smoothed, and pre-determined characteristic of said stored bit information
1-bit white lane that inhibits the smoothing process
Characteristic detecting means for detecting whether or not the pattern matches the pattern indicating the in, and a plurality of predetermined smoothing wherein the characteristic detecting means
When it is detected that one of the features to be processed is matched, smoothing processing is performed on the print information of a specific target pixel in the information group stored in the temporary storage means to perform fine processing.
Information changing means for changing the differentiated small image gu signal, and an output means for outputting a change signal from the information changing means to the output unit, the feature detection unit, a line width of 1 bit wide Formed
The information recording device according to claim 1, wherein the information changing unit does not change the pixel of interest when the white line is detected. 3. A data signal generating means for generating an information signal, storage means for temporarily storing a part of bit information signals emitted from the information signal generating means, determine characteristics of the stored bit information is Me pre a first feature detecting means for detecting whether or not conform to one of a plurality of smoothing to be characteristic which is the first feature detection means to be more smoothing that is determined in advance When it is detected that one of the features is matched, the printing information of the specific pixel of interest in the information group stored in the temporary storage unit is smoothed, and information changing means for changing the signal, and output means for outputting a change signal from the information changing means to the output unit, a plurality of Chinese characters specific smoothing processing characteristics has been determined Me pre of the stored bit information And a second feature detecting means for detecting whether or not matches one of specific pattern to be stopped, one of the first feature detection means a plurality of features previously determined one even when it is detected that meets the and is detected and meets the one of the specific pattern to prohibit a plurality of smoothing processing by said second characteristic detecting means An information recording apparatus, wherein the information changing unit does not change the target pixel.