JP2990671B2 - Image smoothing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to laser printers and LEs.
The present invention relates to an image smoothing processing method in an image output device that forms a matrix dot image, such as a D printer or other page printer or a CRT display.

[0002]

2. Description of the Related Art Conventionally, a laser beam modulated on the basis of video data serially output from an image controller is repeatedly scanned in a main scanning direction along a generatrix of a photosensitive drum in a main scanning direction. Laser printers for forming are known. Also, like an LED printer, an array of pixel forming means arranged in one line in the main scanning direction is relatively moved in the sub-scanning direction while driving (lighting) one line at a time or in block units based on video data. A dot printer that forms a pixel pattern in a matrix on a recording material (photosensitive drum or the like) to be printed is also known.

[0003] In this type of printer, nx
In order to adopt a method of forming an arbitrary character or figure by arranging pixel patterns in a matrix of m, when forming a curve or oblique line like V or 〇, the printing boundary portion is formed in a step shape. In addition, since a plurality of pixels are arranged close to each other at an intersecting portion like X, the portion becomes thick, and the printing quality inevitably deteriorates. In order to eliminate such a drawback, it is necessary to
Suppose that it should be moved to the right, and suppose that this movement amount is 移動 of the distance between adjacent dots, and calculate the logical sum / product between the center pixel and the adjacent pixel. Is determined by logical sum / product. However, although the maximum value (in the case of an inclination of 45 °) of the movement amount is half of the distance between dots, the distance between the movement amount and the original image increases as the inclination approaches the vertical (horizontal) direction. In other words, in the case of a step portion having a steep or nearly horizontal slope, it is extremely difficult to achieve smooth smoothing processing even if a logical sum is taken.

[0004] For example, Japanese Patent Laid-Open No. 60-251761
As shown in the figure, there is a technique for smoothing by replacing a white dot (in the main scanning direction) adjacent to a reference pixel with a pixel (black dot) having a low energy density. Since only small-diameter dots are simply added to the dot area adjacent to the area, especially in the case of thin line-shaped diagonal lines or curves, the portion becomes thicker, which not only does not necessarily lead to an improvement in image quality, but also the main scanning Since the dot is added in the direction, for example, as the inclination of the oblique line or the curve becomes gentler, the smoother smoothing becomes more difficult.

In order to solve such a drawback, not only a small-diameter dot is simply added to a dot area adjacent to a reference pixel, but the target dot itself is reduced (flattened) in accordance with the dot addition. The dot itself is simulated so as to be shifted to one side, and the addition of the small diameter dot and the flattening of the reference pixel are performed not only in the main scanning (horizontal) direction but also in the sub-scanning (vertical) direction, in other words, right, left, up and down. US Pat. No. 4,847,64.
1 (hereinafter referred to as USP641).

Next, to briefly explain the configuration of the prior art, as shown in FIG. 6, the present technology employs four compensating sub-cells 51a for narrowing the dots in the main scanning direction and dots in the sub-scanning direction. .. Are reduced to four compensating subcells 52a.
First, bit data corresponding to image information is sequentially and serially stored in a FIFO buffer 53 capable of storing a plurality of lines, and a plurality of upper, lower, left, and right bit data adjacent to a target bit is extracted from a sample window 54. By passing through a matching network 55, a large number of templates 56 stored in the network 55 are compared with bitmap data located in the window 54, and when they match, compensation for flattening a reference pixel is performed. The sub-cell generator 51 includes sub-cells 51a, 52a... And compensation sub-cells 51a, 52a... Which add flattened small dots for upper, lower, left and right pixels adjacent to the reference pixel (in general, the pixel is white data). Select from 57, if not coincident, select the reference pixel as it is By outputting these selected video data to the print engine side 58 serial or less, which can perform a predetermined laser printing operation.

[0007]

However, even in the prior art, when determining a reference pixel, it is necessary to always select a sub-cell of two dots together with other adjacent dots. Since four sheets are required in each of the horizontal direction and the vertical direction, at least the number of templates necessary for comparing the bitmap data located in the window is determined in the horizontal direction (2 dots × 2).
A total of 256 basic template types (4 sheets) × vertical direction (2 dots × 4 sheets) are required, and if an application template is prepared to further smooth the smoothing, the number of templates becomes unnecessary. As a result, the memory for preparing such a large number of templates and the circuit configuration for performing the comparison operation are both complicated, and as a result, the circuit configuration must be enlarged.

In view of the drawbacks of the prior art, the present invention provides an image smoothing process for a dot printer which can easily and accurately perform a smoothing process with a simple determination operation without complicating the circuit configuration. The aim is to provide a method . Another object of the present invention is to provide an image smoothing method in a dot printer capable of improving the quality of dots themselves together with image smoothing and obtaining a high quality image that can sufficiently withstand data expansion after the smoothing processing. The purpose is to do.

[0009]

SUMMARY OF THE INVENTION The present invention provides a matrix
M × N or / and N × M (N
≧ M, N and M are odd numbers of 3 or more)
Based on the target dot in the map and the target dot
Forming the dot map with the distribution of dots
Vectors obtained by vectorizing sequentially from a small number of N
In an image output device that performs image smoothing processing based on information
In the image smoothing processing method, a note in the dot map
Eye dot and first surrounding dot group adjacent to the noted dot
Obtained by vectorizing the first dot distribution consisting of
1 vector information, the noted dot, and the first
A second dot adjacent to the surrounding dot group in the X-axis or Y-axis direction
The second-order vector obtained by vectorizing the dot distribution of the
Vector information and before the {(N-1) / 2} order
With obtaining the vector information of the serial target dot, the
Weight vector information up to {(N-1) / 2}
And sum of vector information obtained by weighting
Is the vector information for correcting the position error with respect to the noted dot.
It is characterized by being adopted as information. That is, M × N or / and N × M (N ≧ M, N and M are odd numbers of 3 or more ) coordinates, more specifically, 3 × 3, or 3 × 5 and 5 × 3, or 3 × 7 Based on the 7 × 3 dot map, the dot distribution around the target dot in the map is vectorized, and the vector information and the dot type of the target dot (black or white)
Image smoothing while correcting a position error with respect to the noted dot based on the image
Is the way .

In this invention, in the case of a 3 × 3 dot map, only one vector information is required, but this alone can only correct a position error of ド ッ ト of the dot pitch.
Lack of precision. In such a case, or 3 × 5 and 5 ×
3. Further, by obtaining vector information based on dot maps of 3 × 7 and 7 × 3, precise position error control becomes possible.

For example, in the case of 3 × 7 and 7 × 3 dot maps, first vector information based on the 3 × 3 dot map, the X axis or Y axis
Next-order vector information based on the second dot group adjacent in the axial direction, and further next-order vector information based on the third dot group adjacent in the X-axis or Y-axis direction at 3 × 5 and 5 × 3. Although three pieces of vector information can be obtained, a dot adjacent to the dot of interest is more affected by the position error.
Therefore, the three vector information are sequentially weighted from the one closer to the dot of interest to obtain the sum of the vector information,
By adopting the sum vector information as 3 × 7 and 7 × 3 vector information, it is possible to perform a more precise smoothing process corresponding to the position error of the target pixel.

In the embodiment described later, M × N coordinates and N × N coordinates are used.
Correction vector information is obtained for the directional component of the M coordinate, in other words, for both the ± X component and the ± Y component. For example, first from the 3 × 3 coordinates, the target direction component of the target dot is closer to the vertical direction or to the horizontal direction. After judging whether they are close to each other, if the vertical direction is close, only 3 × 5 and 3 × 7 may be obtained as the sum vector information in the ± Y-axis direction based on the vector information.
If the horizontal direction is close, it is sufficient to obtain only the sum vector information in the ± X-axis direction based on the vector information for only 5 × 3 and 7 × 3.

As a specific measure of the smoothing process using the vector information, a plurality of registers storing the smoothing process data for correcting the position error are prepared, and based on the vector information and the dot type of the dot of interest. It is preferable to select a corresponding register and perform image smoothing while generating a position error correction pulse for the dot of interest based on the correction data in the register.

[0014]

As described above, according to USP641, (3 ×
7) The cell generator 57 selects a bitmap located in the window 54 of a total of 33 dots * (7 × 3) in a total of 33 dots and compares it with a corresponding template selected at random from the cell generator 57. In other words, since the 33-dot bitmap is compared two-dimensionally at random, the number of templates is unnecessarily widened (up to 2 ³³⁾ to perform smooth processing as described above. ), The memory for preparing such a large number of templates and the circuit configuration for performing the comparison operation are both complicated, and as a result, the circuit configuration has to be enlarged and the comparison operation is delayed, As described above, there is a problem that high-speed operation cannot be performed.

On the other hand, the present technical means has a common point that a window as a (3 × 7) * (7 × 3) dot map is used similarly to the above-mentioned prior art, but the present technical means is two-dimensionally random. Instead of performing comparison with the template, any of vector information on the dot distribution for the pixel of interest in the window, in other words, any position error has occurred (-X direction, + X direction, -Y direction, + Y direction) First, the vector information in the one-dimensional direction is obtained, and for example, a register corresponding to the compensation subcell is selected and smoothing processing is performed based on the vector information. The primary processing is performed by hardware, and functional software processing is sufficient. As a result, a large number of templates and a circuit configuration for performing the comparison operation are unnecessary.較動 operation is slowing, may also eliminate the one problem of the prior art that it can not correspond to the speed. Also, register selection is based on vector information, for example, maximum (-X direction, + X direction, -Y direction, + Y direction)
The selection can be made automatically based on the × N integer values 4N and the type of the dot of interest (black or white). In other words, extremely precise control is possible only by preparing 2N registers for each direction component.

[0016]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of the present invention; However, unless otherwise specified, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention, but are merely illustrative examples. Not just. First, the smoothing processing method of the present invention will be described while sequentially explaining the process leading to the present invention. For example, in the case of an original image containing oblique lines and curves, if this is to be developed into matrix coordinates, the Χ axis or Y including a step portion
The combination of vertical or horizontal dots extending in the axial direction is as described above. The image data that has been coordinate-developed in a matrix form is, for example, as shown in FIG. 1A, a dot of coordinates (0, 0) in a step portion is, for example, on a coordinate obtained by matching a diagonal inclined line to the resolution. It is considered that the dot includes an error in the right direction as compared with the position to be actually displayed for placing.

Therefore, when the dot at the coordinates (0, 0) (hereinafter referred to as the center dot) is moved to the right by a predetermined distance, the entire image is approximated by the inclined line of the original image. In a general prior art, it is assumed that the movement amount of the predetermined distance is 1/2 of the distance between adjacent dots, and the presence or absence of the movement is obtained by a logical sum / product. However, although the maximum value (in the case of the inclination of 45 °) of the movement amount is の of the distance between the dots, the inclination is vertical (horizontal).
The smaller the movement amount is, the closer to the direction is, the better the original image matches.

The amount of movement from the center dot is as follows: {coordinate (0, 0) → (1, 0)}: composite vector A1 is : {coordinate (0, 0) → (1, −1)}: vector A2, {coordinates (0,0) → (0,1)} : vector A3 in Table,
In this case, A1 = (A2 + A3), which can be obtained as a composite force of two types of vectors. That is, the movement amount from the center dot can be obtained from the sum of the vectors. However, in the case of a 3 × 3 dot map, as described above, only one-stage position error correction of ド ッ ト of the dot pitch can be performed. ) Ho to be etc. center dot is likely to contain a positional error. For example, when the smoothing process is performed by the logical sum / product, the possibility that the center dot includes many positional errors increases when the dot forms a horizontal diagonal line or a vertical diagonal line. . This is because the ratio of the change in the X direction to the change in the Y direction or the ratio of the change in the Y direction to the change in the X direction is small.

Therefore, among the dots forming the oblique line which is nearly horizontal, it is difficult to obtain a great effect by using the correction vector for the center dot as information from rasterized data of a 3 × 3 matrix. That is, as a matter of course, the larger the amount of information, the more accurate the control. The rasterized data to be referred to in the horizontal direction,
The matrix is extended in the vertical direction to a matrix of 7 × 3 and 3 × 7, and a control is performed by obtaining a correction vector for the center dot. Theoretically, it is also possible to obtain a correction vector using data of an N × 3, 3 × N matrix, and to perform a smoothing process. However, as will be described later in detail, based on the performance of the current print engine, etc. This is because it can be determined that it is appropriate to realize the size of × 3, 3 × 7.

Therefore, as shown in FIG.
A × 7 matrix is defined by coordinates, and the value of the dot at coordinates (X, Y) is indicated by D (X, Y). For example, when the center dot is black, it is represented as D (0,0) = 1. Next, a method of obtaining a correction vector will be described with reference to FIG .
3x3 matrix, 3x5 matrix, 3x7
For each of the above matrices, three types of vectors are obtained as indicated by the thick arrows in FIG. It is to be noted that the shaded broken lines indicate the actual inclination of the image data. And 3
Each of the two vectors is weighted as 4, 2, 1 to facilitate digital processing. Then, a correction amount for a dot to be corrected in a certain input pattern is obtained as a sum of three vectors. Most correction amount of increases, therefore, as in the example of the center pixel (hereinafter referred to as center dot) D (0,0) in FIG. 1 (A), the distance between dots 1
In this case, a correction vector having a magnitude corresponding to / 2 is generated.

The method for obtaining the correction vector will now be described in more detail. 1) First, a first correction vector is obtained from a 3 × 3 matrix. Dots that generate correction vectors for each direction component (-X direction, + X direction, -Y direction, + Y direction) are D (-1, -1) and D (0,1) in the -X direction. Pair, two pairs of D (-1,1) and D (0, -1), in the + X direction, a pair of D (1,1) and D (0, -1), D (1,- 1)
And D (0,1) in the -Y direction, D (-1, -1) and D (1,0)
, Two pairs of D (1, -1) D (-1,0), and in the + Y direction, D
A pair of (1,1) and D (-1,0), a pair of D (-1,1) and D (1,0)
Pair. Further, the above-described condition of the dot including the error is added to the effective condition of the correction vector, and a general 3 × 3
Find the correction vector given from the matrix of
Here, if a vector having a size of 1 and having only an X component or a Y component is defined as a unit vector (X, Y), (X, Y) = ｛(-1,0), (1,0), (0, -1), (0,1)｝. CR ₃ (X, Y) to D (0,0), unit vector
Assuming that the correction vector works in the (X, Y) direction, CR ₃ (X, Y) = {D (0,0) + D (X, Y)} / D (-X, -Y)} ｛D (Y , X) D
(XY, YX) + D (-Y, -X) D (X + Y, Y + X)｝ D (0,0) + D (X, Y) is a term for referring to the condition of boundary line continuity, where D (0,0) is a central component, and D (X, Y) is
This is a correction direction component when the center component is white. D (-Y, -X)
Is a suppression vector having a component opposite to the vector component in the (X, Y) direction. In this case, the logical negation is "/", the logical sum is "+", and the logical multiplication operator is omitted.

2) Next, a correction vector is obtained from a 3 × 5, 5 × 3 matrix. Each direction component (-X direction, + X direction, -Y
(+, Y direction)
In the X direction, a pair of D (-1, -2) and D (0,2), D (-1,2) and D (0,-
2), two pairs of D (1,2) and D (0, -2), two pairs of D (1, -2) and D (0,2) in the + X direction, In the -Y direction, D (-2,
-1) and D (2,0) pair, D (2, -1) and D (-2,0) pair, +
In the Y direction, a pair of D (2,1) and D (-2,0), D (-2,1) and D (2,0)
, Two pairs. In addition, the above-described condition of the dot including the error is added to the effective condition of the correction vector, and the correction vector given from a general 3 × 5, 5 × 3 matrix is centered on C ₃ (X, Y). If the component, S ₃ (X, Y), is the correction direction component, C ₃ (X, Y) = D (0,0) D (Y, X) D (−Y, −X) S ₃ (X, Y) = D (XY, YX) D (X, Y) D (X + Y, Y + X) D (0,0). CR ₅ (X, Y) to D (0,0), unit vector
Assuming that the correction vector acts in the (X, Y) direction, CR ₅ (X, Y) = {CR ₃ (X, Y) + C ₃ (X, Y) S ₃ (X, Y)} / D (−X, Y
-Y) {D (2Y, 2X) D (X-2Y, Y-2X) + D (X + 2Y, Y + 2X)}. CR ₃ (X, Y) indicates that when a correction vector exists in a 3 × 3 matrix, the correction vector in a 3 × 5 or 5 × 3 matrix is valid. CR ₃ (X, Y) +
S ₃ (X, Y) is a term for referring to the condition of boundary line continuity, CR ₃ (X, Y) is a central component, and S ₃ (X, Y) is a central component This is the correction direction component at the time of. D (-X, -Y) is (X,
The suppression vector has a component opposite to the vector component in the Y) direction.

3) A correction vector is obtained from a 3 × 7, 7 × 3 matrix. Dots that generate correction vectors for each direction component (-X direction, + X direction, -Y direction, + Y direction) are -X
In the direction, a pair of D (-1, -3) and D (0,3), D (-1,3) and D (0, -3)
In the + X direction, a pair of D (1,3) and D (0, -3), D
Two pairs of (1, -3) and D (0,3), in the -Y direction, D (-3, -1)
D (3,0) pair, D (3, -1) and D (-3,0) pair, and in the + Y direction, D (3,1) and D (-3,0) Pairs, two pairs of D (-3,1) and D (3,0). In addition, the above-described condition of the dot including the error is added to the effective condition of the correction vector, and the general 3
A correction vector given by a matrix of × 7 and 7 × 3 is represented by C ₅ (X, Y) = C ₃ (C ₅ (X, Y) as a central component and S ₅ (X, Y) as a correction direction component. X, Y) D (2Y, 2X) D (-2Y, -2X) S 5 (X, Y) = S 3 (X, Y) D (X-2Y, Y-2X) D (X + 2Y, Y + 2X). CR7 (X, Y) is converted to D (0,0) by unit vector
(X, Y) When correction vector acting in the direction CR7 (X, Y) = { CR 5 (X, Y) + C 5 (X, Y) + S 5 (X, Y)} / D (-
X, -Y) ｛D (3Y, 3X) D (X-3Y, Y-3X) + D (-3Y, -3X) D (X + 3Y, Y + 3
X)｝. CR ₅ (X, Y) indicates that when a correction vector exists in a 3 × 5, 5 × 3 matrix, a correction vector in a 3 × 7, 7 × 3 matrix is valid. CR
₅ (X, Y) + S ₅ (X, Y) is a term for referring to the boundary line continuity condition, and CR ₅ (X, Y) is a central component, S ₅ (X, Y) )
Is a correction direction component when the center component is white. D (-X,-
(Y) is a suppression vector having a component opposite to the vector component in the (X, Y) direction.

4) Calculation of correction vector acting on center dot Accordingly, CR (X, Y) is calculated from the center of all 3 × 3, 3 × 5, 5 × 3, 3 × 7, 7 × 3 matrix states. Assuming that the dot correction vector is CR ₅ (X, Y), CR (X, Y) = K ₃ CR ₃ (X, Y) + K ₅ CR ₅ (X, Y) + K ₇ CR ₇ (X, Y) ... ... 1) can be expressed as Here, K ₃ , K ₅ , and K ₇ are coefficients in consideration of the weighting. The influence on the center dot decreases as the distance from the center dot increases.
Set as follows. Then, for example, (-1,0)
By substituting the coordinates of (1,0) (0, -1) (0,1), each direction component CR (-1,0) (-X direction), CR (1,0) (+ X direction) , CR (0, -1)
Correction vector information of each of (-Y direction) and CR (0,1) (+ Y direction) can be obtained. When the weights of K ₃ , K ₅ , and K ₇ are set to 4, 2, and 1, CR ₃ (X, Y), CR ₅ (X, Y),
The vector information of CR ₇ (X, Y) is max (1) (absolute value) because of the above-mentioned CR (-1,0), CR (1,0), CR (0, -1), CR (0, -1).
1) is a maximum of 7 each.

5) Selection of correction vector information Of the correction vectors of the respective directional components (-X direction, + X direction, -Y direction, + Y direction), the largest correction vector information is adopted by an arbitrary determination means. Then, based on the correction vector information, the center dot is converted into a correction dot for smoothing processing by a method described later.

6) Conversion of Correction Dot According to the above method, for example, by setting the weighting, CR (-1,0), CR (1,0), CR (0, -1), CR (0, -1) 1)
The maximum vector information of each is 7 each, and 7 types of correction vector information can be obtained, and the center dot is 1 (black) or 0
In the case of (white), the correction information viewed from the black side is inverted, so that 14 types of correction vector information are obtained as a result.
However, in practice, even if the laser beam of the laser printer is turned on / off by, for example, 14 types of precise control, accurate gradation cannot be obtained on the printing side due to variations in output light intensity. Therefore, in this embodiment,
The four types of correction information are sorted into six types, and the dots to be corrected are replaced with six-stage correction dots. In this case, the correction dots are generally generated by pulse width modulation corresponding to the correction information. However, in a laser printer, since the scan line of the laser beam is in the horizontal direction, the dot movement in the horizontal direction is easy, but the vertical movement is easy. Shifting directions is difficult. Therefore, in the vertical direction, the processing is performed by changing the dot size without changing the dot position.

FIG. 2 is a basic configuration diagram showing a hardware configuration according to an embodiment of the present invention for performing the above-mentioned smoothing process.
The image data rasterized to i (300 × 300 per inch ² ) is sequentially written into the video memory while being read from the font RAM 1a in accordance with the image signal from the host computer 2 side, and one page is stored in the video memory 1b. Alternatively, after the video data corresponding to one band width is dot-developed, the video data is converted into serial data synchronized with the video clock and output to the print engine 3 side, for example, while the laser beam is controlled on / off based on the video data. Printing corresponding to video data is performed while performing optical scanning at a pitch interval equivalent to 300 dpi.

In this embodiment, a smoothing process based on the above system is realized by interposing a smoothing processing circuit 10 shown in FIG. 3 between the printer controller 1 and the print engine 3. That is, the smoothing processing circuit 10 performs writing / reading with the 7-line memory 11 composed of an SRAM while forming 3 × 7 and 7 × 3 windows.
A latch circuit 12 for latching × 7 dot data, an arithmetic circuit 13 for calculating the magnitude of the correction vector of the center dot, and a center output from the print controller 1 side by the correction vector output from the arithmetic circuit 13 A conversion circuit 14 for converting dots into correction dots is interposed. Reference numeral 15 denotes a register storage unit for converting a center dot into a correction dot based on correction vector information.

Next, the operation of the above circuit will be briefly described. The 7-line memory 11 is a memory capable of storing video data of 7 scanning lines + 1 scanning line, that is, 8 scanning lines.
A RAM, and the print controller 1
The video data serially output in synchronization with CLK is
After the data is sequentially stored in the first bank of the 7-line memory 11 and VDATA for one scan line is stored in the first bank, the VDATA of the second line is similarly transferred to the second bank, and the VDATA of the third line is similarly stored. Are sequentially stored in the third bank. When VDATA for seven lines is stored in the first to seventh banks, the time until the data of the corresponding address of the eighth line is output from the print controller 1 is used to correspond to the first to seventh banks. The VDATA stored in the address is sequentially read out based on the latch signal, and is sequentially loaded into a seven-line shift register constituting the data latch circuit 12.

As a result, in the 7-line shift register constituting the latch circuit 12, while sequentially updating the contents by transferring data from the 7-line memory 11, a dot map (3 lines before and after the center dot and 3 dots left and right) × 7)
Are stored and arranged, and the vector arithmetic processing in the arithmetic circuit 13 can be performed using the dot map. In the arithmetic circuit 13, as shown in FIG.
3 from dot map (7 × 7) from line shift register
Extract window data of × 7 (X, Y) and 7 × 3 (X, Y),
± X based on the correction vector calculation formula shown in 1) above, respectively.
Direction vector information, ± Y direction vector information or ± 4
5 ° vector information is obtained, and the vector information corresponding to the center dot and the type of the center dot (black or white) are obtained from the information.
After converting the video data corresponding to the central dot into corrected video data corresponding to the correction vector information by the conversion circuit 14 based on the information, the video data is output to the print engine 3 side.

That is, the position error of the dot to be corrected is at most 1/2 of the distance between adjacent dots. Therefore, even if the information of the correction vector is the same, if the center dot is (black) or (white), the position error from the black side The viewed correction information is inverted. (E.g. white 70
The% position error is achieved by hitting a black dot corrected to 30%. Therefore, for example, when seven types of correction vector information are obtained, when the center dot is (black) or (white), the correction information viewed from the black side is inverted, so that fourteen types of information are obtained.

However, in practice, even if the 14 types of fine control are performed, the configuration becomes complicated, and even if the laser beam is ON / OFF controlled by the fine pulse control, the output luminous intensity varies on the side where printing is performed. Thus, accurate gradation cannot be obtained. Therefore, in the present embodiment, the 14 types of correction information are sorted into 6 types, and the dots to be corrected are replaced with 6-stage correction dots. In this case, in this embodiment, the type of the conversion register is set to CR (-1,0) (-X direction) and CR (1,0) (+ X direction) for each direction component.
6 types for each of CR (0, -1 0,1) (± Y direction), total 1
Eight types of conversion registers are prepared. The correction dots are generally generated by pulse width modulation of the center dot. However, in a laser printer, since the scan line of the laser beam is in the horizontal direction, the dot movement in the horizontal direction ± X direction is easy. It is difficult to shift in the vertical direction ± Y direction. Therefore, in the present embodiment, processing is performed by performing pulse control such that the dot size is changed without changing the dot position in the present embodiment.

For example, as shown in FIG. 4 (I), in the conversion circuit 14, a curve including the center dot (including a straight line) is a curve close to a vertical line of 45 ° or more based on the ± 45 ° vector information. Or a horizontal curve,
For a nearly horizontal curve, the center dot ｛black: (D (0,0 =
1)), white: (D (0,0) = 0) information and correction vector information (CR
The correction data selected from the following registers A to F based on (χ, y) = 1 to 7) is converted in parallel to a 16-dot conversion circuit 1.
4, the video data corrected as shown in FIG. 4 is output to the print engine 3 by outputting the serial data from the shift register based on a clock having a cycle of 16 times 300 dpi. Can be done. As a result, as shown in FIG. 5 (I), the print sequence as shown in FIG. 5A can be replaced with a print sequence as shown in FIG. (Data contents of conversion registers A to F) A: 1111111001111111 B: 1111110000111111 C: 1111100000011111 D: 1111000000000111 E: 1110000000000111 F: 1100000000000011

On the other hand, in the case of a curve that is nearly vertical, the vector information CR (χ, y) or the positive (+ X direction) vector information CR is used.
It is determined whether (1,0) or negative (-X direction) vector information CR (-1,0), and the center dot ｛black: (D (0,0 = 1)), white: (D (0,0)
= 0) and the correction data selected from the following G to L and M to R corresponding registers based on the correction vector information (CR (χ, y) = 1 to 7).
Write to the shift register inside, and from the shift register,
By performing the serial output based on a clock having a cycle of 16 times 300 dpi, the corrected video data can be output to the print engine 3 as shown in FIGS. 4 (II) and 4 (III). As a result, as shown in FIG. 5 (II), the print sequence as shown in FIG. 5A can be replaced by the print sequence as shown in FIG. (Data contents of conversion registers G to R) G: 1111111111111000 H: 1111111111100000 I: 1111111110000000 J: 1111110000000000 K: 1111100000000000 L: 1110000000000000 M: 0001111111111111 N: 0000011111111111 O: 0000000111111111 P: 0000000001111111 Q: 0000000000011111 R: 0000000000000111

[0035]

According to the described as the present invention EFFECT above, without complicating the circuit configuration, Thus also easily and accurately smoothing process that can be performed by a simple determination operation. The present invention aims to quality of dot itself with image smoothing, it is possible to obtain a high-quality image which can withstand sufficiently the expansion of the data after the smoothing process. And so on.

[Brief description of the drawings]

FIGS. 1A and 1B are operation diagrams showing the principle of the present invention, wherein FIG. 1A is a basic configuration diagram, FIG. 1B is coordinate data required for the calculation of FIG.
(C) shows a calculation procedure for obtaining correction vector information for the center dot.

FIG. 2 is a basic configuration diagram of a printer incorporating a smoothing processing circuit according to an embodiment of the present invention.

FIG. 3 is a block diagram showing an entire configuration diagram of a smoothing processing circuit of the present invention.

FIG. 4 is a pulse waveform chart showing types of video data for smoothing generated by the smoothing processing circuit, (I) is a pulse of a nearly horizontal distribution line, and (II) and (III) are near vertical distribution lines. 2 shows a pulse waveform of the first embodiment.

5A and 5B show print dot distributions generated based on the video data of FIG. 4, wherein FIG. 5A shows a state before smoothing processing and FIG. 5B shows a state after smoothing processing.

FIG. 6 is a block diagram showing an entire configuration diagram of a smoothing processing circuit according to a conventional technique.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Print controller 10 Smoothing processing circuit 11 7-line memory 12 Latch circuit 13 Operation circuit 14 Conversion circuit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ identification symbol FI H04N 1/409 G06F 15/62 325A (58) Investigated field (Int.Cl. ⁶ , DB name) B41J 2/485 G06F 3 / 12 G06T 11/60 G09G 5/28 610 H04N 1/387 H04N 1/409

Claims (2)

(57) [Claims] 1. A method for forming a matrix dot image.
,M × N or / and N × M (N ≧ M, N and M are 3 or less
Based on the dot map of the (odd) coordinates above
Distribution of the target dot and the dots around the target dot
From the small number of N forming the dot map
Image smoothing based on vector information obtained by vectorization
Do the processingImage smoothing process in image output deviceMethodTo
AndThe dot In the mapAttention dot andThe attention dotNext to
A first dot consisting of a first group of surrounding dots in contactDistribution
Into a turtleFirst vector information obtained by The target dot and the first surrounding dot group have an X-axis
Or the dot distribution of the second dot group adjacent in the Y-axis direction
Second-order vector information obtained by vectorization, In the following, the dot of interest up to {(N-1) / 2}
And obtain the vector information of The vector information up to the {(N-1) / 2} order is further overlapped.
And the total of vector information obtained by this weighting
The sum is a vector for correcting the position error with respect to the noted dot.
Adopt as information Image smoothing processingOne
Law. 2. A plurality of registers storing data for a smoothing process for correcting a position error of the target pixel are prepared, and a corresponding register is selected based on the vector information and a dot type of the target dot. Image smoothing is performed while generating a pulse for correcting a position error with respect to the noted dot by using correction data in a register.
The image smoothing method according to claim 1 .