JPH0650903B2 - Image recording method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an image recording method for recording an image by a pixel matrix, and more particularly to an image recording method capable of smoothly recording and outputting oblique lines such as characters and line drawings.

[Prior Art] Conventionally, outputting both high-quality image information, particularly image information having a halftone, and font output of characters and the like is different from each other, and the cost and performance are poor, so that most of them are practical. It hasn't reached the end. For example, with regard to characters, if high-resolution characters can be output if bit-mapped character fonts of 32 × 32, 64 × 64, etc. are provided, the jagged edges of the edge portion will still appear in the enlarged image of the characters. Is conspicuous and the quality is inevitable. Further, if it is attempted to provide a character font having a high degree of resolution for a large number of characters, a large-capacity font memory is required and it becomes extremely expensive.

[Object] The present invention eliminates the above-mentioned drawbacks of the prior art, and an object of the present invention is to provide an image recording method that enables the output of high-quality characters and line drawings.

[Embodiment] An embodiment according to the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of the multi-valued output of this embodiment, in which one pixel 6 to be recorded is subdivided into a plurality of fine pixels 1.
It is divided into zero. In the figure, an output method is shown in which one pixel 6 is divided into five fine pixels 10 into five, that is, one pixel 6 is output in six values. This output method can be obtained, for example, in a laser beam printer by setting the main scanning direction of beam scanning to be the x direction and dividing the emission pulse width of light within one pixel into five. In a thermal transfer type printer using a full-multi thermal head, the thermal heads are arranged in the y direction, and a 6-value output is similarly obtained by changing the pulse width applied to each head in the feeding direction (x direction). be able to.

FIG. 2 (A) shows the output state of the conventional multi-valued output. In the case of 6-value output, (1) → (1) as shown in FIG. 2 (A) →
By thickening the fine pixels 10 in the direction of (5),
An example is shown in which gradation output of 5 steps (6 steps including all white) is obtained per pixel.

In this embodiment, the state corresponding to 1/5 pixel width as shown in FIG. 2 (B) can be set to five states as shown in the figure. Therefore, as shown in FIG. 2C, in the state where two fine pixels 10 each having a 1/5 pixel width are made to emit light, as shown in FIG.
There are only ₅ CS ₂ = 10 states. Similarly, when there are three fine pixels, the number of states becomes very large, ₅ C ₃ = 10, ...
The degree of freedom increases. Therefore, the total number of states at this time is Becomes

In this embodiment, each fine pixel arbitrarily emits light (or prints) in this way.
It is configured so that it can be multivalued. However, in general, as the number of states increases, the amount of information increases, the amount of data for instructing the states increases, and control tends to be complicated. However, in this embodiment, as described below,
The above method can be realized with the same data capacity as that used for handling ordinary binary data.

FIG. 3 is a diagram showing a case where a character is output. In the figure, a case where a part of the character "A" is output is illustrated. Here, the coordinates (x ₀ , y ₀ ) of each vertex 30 to 34
Given that (x ₄ , y ₄ ) is given, among the pixels of the dots to be printed, the pixels around the vertex 31 are displayed in association with the 3 × 3 pixel matrix 35.

FIGS. 4 (A) and 4 (B) show a state in which the data is fleshed around the data of the straight line passing through the pixel matrix 35 and connecting the vertices 30 and 32. FIG.
(B) shows the case of thickening. With this amount of padding, it is possible to deal with various fonts and to prevent the occurrence of jaggedness on diagonal lines.

Further, as shown in FIG. 4 (C), the fine pixels passing through the straight line 36 are blackened, and the fine pixels 37 slightly apart from the black line are also blackened, whereby the straight line 36 can be changed to a continuous and easy-to-see output. The applied signal in the AA ′ cross section of (C) is shown in FIG. 4 (D). In the case of a laser beam printer, since scanning is performed with a finite laser spot diameter (the laser spot diameter is often widened to about one pixel size), the light energy distribution due to scanning is shown in FIG. As shown in E), the blackening density is high in the center and low in the periphery. For this reason, it appears as continuous smoother diagonal lines.

FIG. 5 shows a method for obtaining the coordinates of the center position of the flesh. The two coordinates of the end point of the straight line 36 are A point (x ₀ , y ₀ , l ₀ ) and B point (x ₂ , y ₂ , l).
₂ ). (However, here, l is the distance in the x direction measured from the left end of the pixel within one pixel.) At this time, the coordinates (xi, yi,
li) is obtained as follows.

Converting the coordinates of the point A _{(X A,} Y _A) = equation of _{(x 0, l 0, y} 0) converting the coordinates of the point B _{(X B, Y B) =} (x 2, l 2, y 2) linear 36 The value of Xi is sequentially obtained by giving yi from the following equation.

However, the value of yi is an integer value between y ₀ and y ₂ .

Once Xi is obtained, xi and li are separated as follows. Now, assuming that xi is an integer value and li is, for example, 6 values, five values of 0/5, 1/5, 2/5, 3/5, 4/5 can be taken, and xi = [Xi] ... (2) However, [] is a Gauss symbol and indicates only the integer part. li
Takes the closest value from 0/5 to 4/5 of the decimal part. With this, it is possible to obtain how many fine pixels of one pixel the straight line passes through.

By repeating the above operation, the value of (xi, yi, li) can be obtained for the value of the integer value yi between y ₀ and y ₂ .

FIG. 6 shows a flow of the above method. First, in step S1, (x ₀ , y ₀ , l ₀ ), (x ₂ , y ₂ ,
l ₂ ) is input. Here, y ₀ <y ₂ . Next, in step S2, xi and li are obtained for the integer value yi satisfying y ₀ <yi <y ₂ . At step S3, Δl is determined based on each font information. That is, as shown in FIG. 4 (C), this means a process for arranging the fine pixels 37 to reduce the jagged feeling. Next, in step S4, yi is incremented by one. In step S5, the value of yi is y
_See if it equals _2, that is, if the processing for the straight line is complete. When the process is not completed, the process returns to step S2 again to execute the above operation.

By repeating the above steps, it is possible to output characters for any font style. Such a method is performed by performing real-time calculation with a CPU or the like and transferring it to an image or a memory (not shown). Therefore, it is possible to calculate the light emission part of the multi-valued fine pixel only by holding the vector data, and the data capacity is much larger than that of the high resolution character pattern in the bit map font (bit image). It is extremely small and can flexibly respond to changes in font.

FIG. 7 shows an output method for a halftone image. The image data 14 is assumed to be, for example, digitalized 8-bit data. RO storing the dither threshold
A dither threshold by the systematic dither method is periodically output from M13, and is compared with the image data 14 by the comparator 11, and as a result of comparison, a binary output signal 1 of 0 or 1 is output.
It is output as 2.

The multivalued method in this embodiment is shown in FIGS. 8 (A) and 8 (B).
The threshold matrix shown in FIG. FIG. 8 (A) shows an example of a dot concentrated type, and FIG. 8 (B) shows an example of an oblique 45 ° screen, each of which is composed of a 3 × 3 matrix. It can be said that the resolution is increased in the x direction by adopting a configuration in which one pixel is divided into five fine pixels as described above. Further, in the case of a matrix of 3 × 3, 5 × 9 + 1 = 46 levels, and the gradation reproducibility is increased.

Such multi-valued data is directly output to a printer or stored in a memory (not shown). When storing in memory, as described in the character output above, per pixel In the case of the halftone image, when the halftone image is output, as shown in the threshold matrix of FIG.
Since five fine pixels in the pixel sequentially emit light from the left side, only six states are possible, and a bit number of l _0gd 6 is sufficient. Therefore, in the case of a halftone image, 3 bits per pixel is sufficient.

FIG. 9 (A) prevents the blackening center position from moving in FIG. 8 (B), and shows an example in which the line in FIG. FIG. 10 shows the light emission order (blackening order) of the fine pixels in one pixel.
(A) is a centralized type, and (B) is a distributed type. 11 (A) and 11 (B) show the light energy distribution generated by laser beam scanning based on FIGS. 10 (A) and 10 (B), and the numbers in each figure indicate the light emission of the fine pixels 1 to 5. It corresponds to. Compared to the concentrated type of FIG. 11 (A), the dispersed type of FIG. 11 (B) shows more change in the intensity of light energy (in the vertical axis direction of the figure).
It can be said that it is suitable for recording methods that allow analog drive. In addition, in this embodiment, one pixel is divided into five, but the present invention is not limited to this.

[Effect] As described above, according to the present invention, when recording and outputting an oblique line such as a character or a line drawing, black pixels having a width narrower than the recording width of one pixel in the pixels on both sides adjacent to the pixels forming the line. The diagonal lines can be smoothly recorded and output by recording and outputting the pixels at intervals from the pixels forming the line.

[Brief description of drawings]

FIG. 1 is a diagram showing an example of a multi-valued output of an embodiment according to the present invention, FIGS. 2 (A) to (C) are diagrams showing an arrangement example of fine pixels, and FIG. 3 is an example of character output. 4 (A) to 4 (E) are diagrams showing examples of weighting of pixels, FIG. 5 is a diagram for obtaining center coordinates of weighting of diagonal lines, and FIG. 6 is a process of the present embodiment based on FIG. FIG. 7 is a flow chart, FIG. 7 is a block diagram showing image signal processing by the systematic dither method, FIGS. 8 (A) and 8 (B) are diagrams showing an example of dither matrix, and FIG. 9 (A). Is a diagram showing an example of the dimatrix, FIG. 9 (B) is a diagram showing the thickening of the line in FIG. 9 (A), and FIGS. 10 (A) and (B) are the increasing order of the fine pixels. 11A and 11B are diagrams showing the relationship between fine pixels and light energy. In the figure, 10, 37 ... fine pixels, 11 ... comparator,
12 ... Output signal, 13 ... ROM, 14 ... Image signal, 30-34 ... Vertex, 35 ... Pixel matrix.

Claims (1)

[Claims] 1. An image recording method for recording an image using a pixel matrix, wherein the recording width of a diagonal line is narrower than the recording width of one pixel in the pixels on both sides adjacent to the pixels forming the line. An image recording method characterized in that diagonal lines are smoothly recorded and output by recording and outputting black pixels at intervals from the pixels forming the line.