JP2823925B2 - Vector to raster converter - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vector / raster conversion device that converts vector data output from a CAD or the like into raster data, and in particular, generates output data to a large raster printer. The present invention relates to a vector / raster conversion device suitable for the above.

[Prior Art] Conventionally, a vector / raster conversion device that inputs vector data, converts the data into raster data, and outputs the raster data to a raster output device such as an electrostatic printer, a thermal printer, a laser beam printer, or a CRT display device is known. The input vector data is converted into, for example, DDA (digital differential
After converting the data into bitmap data using a technique such as an analyzer), the data is buffered in a bitmap memory corresponding to all pixels and output to a raster output device.

In a general DDA algorithm, bitmap data is generated from vector data as follows. That is, as shown in FIG. 9, two points P ₁ , P
₂ coordinates _{(x 1, y 1),} (x 2, y 2) when a given, both points P
By comparing the _1, the distance in the X-axis direction of the distance dx and Y-axis direction between the P ₂ dy, who absolute value is large (hereinafter, referred to as dl) while increasing one by one coordinate axis of Add the slope m (= ds / dl) to the coordinates of the smaller (hereinafter called ds) axis,
An increment determination is made as to whether there is a carry in the integer value obtained by rounding the decimal part. And if there is no carry,
If the carry of the ds value is not changed and the carry, the ds side coordinate is set to 1
Increase. In the example of FIG. 9, since dx = 16 and dy = 5, while increasing x by 1, m = 0.3125 is added to y, and the carry becomes an integer value when the decimal part is rounded off. The y coordinate is determined depending on whether or not there is. As an improvement of the above algorithm, Bresenham's algorithm is also known in which speed is increased by performing only integer operation without performing floating point operation at the time of increment determination.

[Problems to be Solved by the Invention] However, in the above-described conventional vector / raster conversion apparatus, when converting from vector data to bitmap data, writing to the bitmap memory is performed one dot at a time. There is a problem that it takes time. Also, there is a problem that a large amount of bitmap data must be provided to buffer the converted bitmap data.

On the other hand, as a memory device with a reduced memory size, a device in which the capacity of a bitmap memory is limited to a certain bandwidth is known. However, even in this case, the conversion still takes time.

The present invention has been made in view of the above problems, and has as its object to provide a vector / raster conversion device capable of performing high-speed processing and reducing the capacity of a buffer memory.

[Means for Solving the Problems] A vector / raster conversion apparatus according to the present invention comprises a vector / run length conversion means for converting vector data into run length data, and sequentially stores the run length data converted by the means. A run-length buffer that converts the run-length data stored in the run-length buffer into raster data. The run-length data is generated such that the number of dots of the run-length data at the time of (1) is smaller than the number of dots of the run-length data at the intermediate portion of the vector.

[Operation] According to the present invention, input vector data is once converted into run-length data. Then, the run length data is written to the run length buffer. At this time, writing to the run length buffer is performed not in dot units but in run length units. For this reason, the number of times of writing to the buffer is reduced, and the processing speed can be increased.

Further, in the present invention, since buffering is performed in units of run-length data, the buffer capacity can be significantly reduced as compared with a conventional device that buffers with bitmap data.

Embodiment Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a block diagram showing a vector / raster conversion device according to an embodiment of the present invention.

Vector data output from a CAD or the like is input to a vector / run-length conversion unit 2 via an interface circuit 1. The vector / run-length conversion means 2 converts the input vector data into run-length data. Run length data output from the vector / run length conversion means 2 is sequentially stored in a run length buffer 3. The run-length data stored in the run-length buffer 3 is supplied to the sort / merge unit 4. The sort / merge unit 4 sorts and merges the run-length data so as to be suitable for raster scanning. The output of the sort / merge means 4 is supplied to a run-length / bitmap conversion means 5.
The run length / bit map conversion means 5 sequentially converts the run length data into raster data according to the reference clock. Then, the output of the run-length / bitmap data converting means 5 is provided to a raster type printer 6 such as an electrostatic plotter, a thermal printer, a laser beam printer, and a CRT display device.

In the above configuration, the vector / run-length conversion means 2, the sort / merge means 4, and the run-length / bitmap conversion means 5 are constituted by a microcomputer and software even if they are constituted only by hardware. But it is good.

Next, the operation of the present apparatus configured as described above will be described.

When the vector data is input to the vector / run-length converting means 2 via the interface circuit 1, the vector / run-length converting means 2 converts the vector data into run-length data according to the following algorithm.

2A and 2B are flowcharts showing the algorithm, and FIG. 3 is a schematic diagram showing a dot row corresponding to run-length data generated by the algorithm. Here, the Y-axis direction is set to the raster direction.

First, as vector data, two points as shown in FIG.
Given the coordinates (x ₁ , y ₁ ) and (x ₂ , y ₂ ) of P ₁ and P ₂ ,
The vector / run length conversion means 2 calculates distances dx and dy between the two coordinates in the X-axis direction and the Y-axis direction (S ₁ ). Third
In the example shown in the figure, dx = 5 and dy = 16 are calculated. If dx is not 0, a slope d = dy / dx is calculated (S ₂ , S ₃ ),
If d has a decimal part, then n ₁ = d, n ₂ = d + 1,
If d has no decimal part, n ₁ = n ₂ = d (S ₃ , S ₄ ,
S _5). Also, if dx is 0, run length data of length dy extending in the Y-axis direction is generated, so that n ₁ = n ₂ =
Let it be dy (S ₂ , S ₆ ). In the example of FIG. 3, dx = 5, dy = 16
Therefore, d = 16/5 = 3.2, n ₁ = 3, and n ₂ = 4.
Therefore, in this case, the run-length data of three dots or four dots is sequentially written except for both ends of the vector.

Therefore, first, run-length data of the start point is generated. At both end points P ₁ and P ₂ of the vector, run-length data having half the number of dots in the middle part of the vector is used in order to reduce the error with respect to the true straight line. That is, n ₁ and n ₂ are each reduced by _one , and the coordinate values x ₁ and y ₁ of the starting point P ₁ are substituted for x and y, respectively (S ₇ ), and (n ₂ +1) / 2 is set as the number of dots K. assignment to (S _8). Then, whether the number of dots K is not too long, determined by the following discriminant (S _9).

2 · dx · K> dy (1) If the number of dots is too long, K is reduced by one to correct an error (S ₁₀ ). In the example of FIG. 3, x = 0, y
= 0 and K = 2 are substituted. Then, K + 1 run-length data is output from the position (x, y) in the y direction (S
₁₁ ). In FIG. 2, such a run-length output command is expressed as “Pixel out (x, y, K + 1)”.
Thus, the starting point P ₁ of Figure 3 is run length of 2 dots are generated, the run-length data (0,0,2) in the run length buffer 3 is written. When run-length data is written, dissipate one increment of x, is incremented the y only K + 1 (S _12).

Subsequently, a run length at the middle of the vector is generated. First, by substituting n ₂ to K (S _22), whether the number of dots K is not too long, determined by the following discriminant (S _23).

2 · dx (y + K) > dy (2 · x + 1) ... (2) Then, only when it is determined that the dot array K is too long, n ₁ is substituted into K (S _24). Then, it outputs the (x, y) run-length data of K + 1 in the y direction from the position of (S _25). In the example of FIG. 4, a run length of 3 dots is generated from the position of (1, 2), and the run length data (1, 2, 3) is written into the run length buffer 3. When run-length data is written, dissipate one increment of x, is incremented the y only K + 1 (S _26). In the middle,
The process x is repeated until the x ₂ (S _21).

Finally, in the end position, by substituting y ₂ -y to dot row K (S _27), and outputs the (x, y) run-length data of K + 1 in the y direction from the position of the (S _28). In the example of FIG. 3, (5,1
A run length of 2 dots is generated from the position of 5), and the run length data (5, 15, 2) is stored in the run length buffer 3.
Is written.

By performing the above processing, run-length data as shown in FIG. 4 is written into the run-length buffer 3. The data shown in Fig. 3 was converted to a conventional DDA
When converted by the algorithm, 16
According to the algorithm of the present embodiment, the conversion process can be completed by writing data to the run-length buffer 3 six times, while writing the data twice is required.
In particular, remarkable effects can be expected for vector data having many horizontal lines, straight lines with small inclinations, and thick lines.

By the way, in the above example, the run-length data is generated in ascending order of the scan line. However, since the vector data is normally generated regardless of the position on the coordinate axis, the run-length data is generated. The positions on the coordinate axes are also different. In addition, for example, as shown in FIG. 5, lines may intersect and run-length data may partially overlap. Therefore, as shown in FIG. 6A, the run-length data stored in the run-length buffer 3 is converted into coordinate values along the scanning line by the sort / merge means 4 as shown in FIG. And the merged portion is applied to the overlapping portion, and the merged portion is sequentially supplied to the run-length / bitmap converting means 5.

The run-length / bitmap conversion means 5 can be constituted by, for example, a circuit as shown in FIG. That is, the run-length data supplied in the order of coordinates along the scanning line is stored in a first-in-first-out (FIFO) memory 11 and then input to a dot row generation circuit 12. The output of the pixel counter 14 that counts up with the reference clock signal from the clock generation circuit 13 is supplied to the dot row generation circuit 12. The dot row generation circuit 12 compares the output of the pixel counter 14 with the run occurrence position data in the run length data to determine the black and white, and outputs a dot row signal synchronized with the reference clock. This dot row signal is converted by the serial / parallel conversion circuit 15 to, for example, 32
The data is converted into bit parallel data and output to the raster printer 6 or the like via the FIFO memory 16.

As described above, according to the buffering system of the present embodiment, the number of times of writing to the run-length buffer 3 is reduced as compared with the related art, so that the processing speed can be improved and the bitmap memory is required. And not
The capacity of the buffer memory can be significantly reduced.

The present invention is not limited to the above embodiment. For example, a small-capacity bitmap memory that can hold several lines of data may be provided at the subsequent stage of the run-length / bitmap conversion means 5.

It is also possible to convert the data in the run-length buffer 3 into raster data while sorting and merging the data in the run-length buffer 3 using an SRAM (static RAM) having a capacity for one scanning line. This process is shown in FIG. That is, for each scan line indicated by x, the run length data (x, *, *) (where * is an arbitrary value) of the corresponding line is extracted from the run length buffer 3 ( _S34 ). If the extracted data is (x, y, m), SR
To address y or address y + m-1 of the AM will write the dot data (S _35). When all of the run-length data of the corresponding line is written to the SRAM in the form of a bit map (S _33), and outputs to convert the contents of the SRAM into raster data (S _36). By executing this for all scan lines (S ₃₁ , S ₃₂ ), run-length bitmap conversion can be performed with one to several lines of memory.

The format of the run-length data is not limited to the illustrated one, but may be a format that specifies the start point position and the end point position of the dot row. In any case, in the case of the practical drawings of A4 to A0, the run-length data can be expressed by 32 bits = 1 word, and the processing can be performed by a general-purpose CPU.

[Effects of the Invention] As described above, according to the present invention, vector data is temporarily converted into run-length data and buffering is performed. Thus, the buffer capacity can be greatly reduced.

Moreover, according to the present invention, writing to the buffer can be performed in units of run-length data instead of in units of dots, so that the number of times of writing to the buffer can be greatly reduced and the conversion processing speed can be improved. Can be planned.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a vector / raster conversion apparatus according to an embodiment of the present invention, FIG. 2 is a flowchart showing a vector / run-length conversion procedure in the apparatus, and FIG. 3 is generated by the conversion procedure. FIG. 4 is a schematic diagram showing run-length data generated by the conversion procedure and stored in a run-length buffer. FIG. 5 is a schematic diagram showing an example in which a plurality of straight lines intersect. FIG. 6 is a diagram showing run-length data before and after being processed by the sort / merge means in the apparatus, and FIG. 7 is a view showing run-length data in the apparatus.
FIG. 8 is a detailed block diagram of the bitmap conversion means, FIG. 8 is a flowchart showing the run-length / raster conversion procedure in the vector / raster conversion apparatus according to the second embodiment of the present invention, and FIG. 9 is generated by the conventional DDA method. FIG. 3 is a schematic diagram showing a dot row that is displayed. 1; interface circuit, 2; vector / run-length conversion means, 3; run-length buffer, 4; sort / merge means,
5; run-length bitmap conversion means, 6; raster printer, 11, 16; FIFO memory, 12; dot string generation circuit, 13;
Clock generation circuit, 14; pixel counter, 15; serial / parallel conversion circuit

Claims (1)

(57) [Claims] 1. Vector run-length conversion means for converting vector data into run-length data, a run-length buffer for sequentially storing the run-length data converted by the means, and a run-length buffer stored in the run-length buffer. Run-length / raster conversion means for converting length data into raster data, wherein the vector / run-length conversion means is configured such that the number of dots of the run-length data at the start point and the end point of the vector is in the middle of the vector. A vector / raster conversion device for generating run-length data such that the number of dots is smaller than the number of dots in the run-length data.