JP4022710B2 - Drawing processor - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a drawing processing apparatus that performs drawing processing based on drawing data.
[0002]
[Prior art]
In a drawing processing apparatus that processes drawing data such as PostScript (registered trademark) and GDI, a drawing process for developing input drawing data into bitmap data or the like corresponding to an output apparatus is performed. FIG. 17 is a flowchart showing an example of a conventional drawing process. When drawing data is input in S101, an intermediate code corresponding to the input drawing data is generated in S102, and the intermediate code generated in S102 is stored in S103. In S104, it is determined whether or not all the drawing data in the page has been completed, and the processes in S101 to S103 are repeated until the processing for the drawing data for one page is completed. As a result, intermediate codes corresponding to one page of drawing data are accumulated.
[0003]
Thereafter, in S105, an intermediate code is acquired for each unit such as one page unit or a band unit obtained by dividing one page into a band-like region, and in S106, rendering processing is performed for each specific unit. The drawing process at this time is performed using one specific drawing method incorporated in advance. In S107, it is determined whether or not all units of intermediate codes have been processed, and the processes of S105 and S106 are repeated until the processing of all units is completed. In this way, an image corresponding to one page of drawing data can be drawn.
[0004]
As described in S106 above, conventionally, drawing processing is performed using a single drawing processing method in the same apparatus. As one of the drawing processing methods, there is a method using an intermediate code and a band memory for development as disclosed in, for example, JP-A-9-171564. In this method, intermediate codes for one page are accumulated in association with bands obtained by dividing the output page into strips, the intermediate codes are taken out in band units, and the input drawing data is stored in the band memory in order. It will be overwritten. In this drawing processing method, the unit in FIG. 17 is “band unit”, and “overwrite processing using a band memory” corresponds to a specific drawing method. In this method, since the overwriting process is performed on the memory, there is a problem that the performance deteriorates as the writing area increases.
[0005]
As another drawing processing method, for example, as disclosed in Japanese Patent Laid-Open No. 10-79038, etc., vector format data is accumulated, data is acquired for each scan line, clip processing and painting processing are performed. At the same time, the scan line data is output with the overlap removed. In this drawing processing method, the unit in FIG. 17 is “scan line unit”, and “a process for simultaneously performing clip processing and filling processing and simultaneously removing overlap” corresponds to a specific drawing method. However, since this drawing processing method needs to sort the data on the scan line, there is a problem that the performance is worse than the memory overwrite method due to the sort overhead when the input data is small.
[0006]
Thus, each drawing processing method has advantages and disadvantages. In the conventional method, since a single drawing processing method is adopted, there is a problem that the performance may deteriorate due to the characteristics of the drawing processing method.
[0007]
On the other hand, in each drawing processing method, the processing time can be predicted by a certain parameter. Many techniques for predicting the processing time and performing some processing according to the prediction result have been disclosed. For example, Japanese Patent Laid-Open No. 8-279050 discloses a technique for determining whether real-time development is possible from the result of predicting the processing time, and converting to a raster format if real-time development is impossible. Further, for example, Japanese Patent Application Laid-Open No. 10-307689 discloses a technique for changing to an intermediate code format that can shorten the development time from the result of predicting the processing time when real time development is not possible. Similarly, Japanese Patent Application Laid-Open No. 11-198489 discloses a technique for selecting an intermediate code format that can be processed at high speed with respect to a predetermined processing time. Japanese Patent Application Laid-Open No. 10-307924 describes an example in which the processing time is predicted and applied to load equalization in parallel processing.
[0008]
In this way, the processing time is predicted for each drawing processing method, and when real-time development cannot be performed by the drawing processing method, a time-consuming process such as a data format change is performed in advance. For example, even if other drawing processing methods allow real-time deployment, other drawing processing methods have not been used. As a result, the drawing process took time.
[0009]
[Problems to be solved by the invention]
The present invention has been made in view of the above-described circumstances, and an object thereof is to provide a drawing processing apparatus that realizes high-speed drawing processing.
[0010]
[Means for Solving the Problems]
In the present invention, drawing processing is performed using a drawing method selected by the drawing method selection means from a drawing method in which data is overwritten in the memory in the order of drawing and a drawing method in which overlap processing is performed on the intermediate code. It can be performed. For example, based on the processing load or processing time for each drawing method estimated by the processing load estimating means, the drawing method that can be processed at the highest speed can be selected to perform the drawing process. As a result, even when there is a delay in drawing processing when a single drawing method is used, high-speed drawing processing can be performed by taking advantage of other drawing methods. In this case, since the drawing processing time is reduced, a delay in processing time due to conversion of another data format or the like can be suppressed, and the overall processing speed can be increased.
[0011]
It should be noted that the selection of the drawing method can be configured such that when a fixed drawing method is set, the set drawing method is selected. The selection unit of the drawing method can be a job unit, a page unit, a band unit, a line unit, or the like. Further, the drawing data can be converted into an intermediate code in a format such as a vector format, a trapezoid format, or a run length format before the drawing process is performed.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram showing an embodiment of a drawing processing apparatus of the present invention. In the figure, 1 is a drawing data input unit, 2 is an intermediate code generation unit, 3 is an intermediate code storage unit, 4-1 to n are drawing load estimation units, 5 is a drawing method selection unit, 6 is a drawing unit, and 7-1. ˜n is a drawing method, and 8 is an output device. The drawing data input unit 1 inputs drawing data. The drawing data is composed of characters, figures, raster images, and the like. The intermediate code generation unit 2 generates an intermediate code from the input drawing data. The intermediate code storage unit 3 stores the generated intermediate code, for example, for one page.
[0013]
One or a plurality of drawing load estimators 4-1 to n are provided, and the processing time corresponding to each drawing method 7-1 to n or the processing load corresponding to the processing time is estimated from the generated intermediate code. Of course, it is not necessary to provide according to the number of drawing methods, the number may be smaller than the drawing method, and the processing time and processing load may be estimated by several estimation methods for one drawing method.
[0014]
The drawing method selection unit 5 selects an optimum drawing method from the plurality of drawing methods 7-1 to n based on the results estimated by one or more drawing load estimation units 4-1 to n. The drawing unit 6 performs drawing processing based on the intermediate code stored in the intermediate code storage unit 3 using the drawing method selected by the drawing method selection unit 5, and outputs it to the output device 8. To do.
[0015]
FIG. 2 is a flowchart showing an example of the operation in the embodiment of the drawing processing apparatus of the present invention. In FIG. 2, as an example, intermediate code data is accumulated for each page, and two drawing methods (drawing methods a and b) are switched within the page.
[0016]
When drawing data is input to the drawing data input unit 1 in S11, the intermediate code generation unit 2 generates intermediate code data corresponding to the input drawing data in S12. Next, in S <b> 13, each drawing load estimation unit 4-1 to 4-n switches the drawing process with respect to the load applied to the drawing process when drawing with each drawing method, using the generated various profile information of the intermediate code. Calculate for each unit. In S <b> 14, the intermediate code generated by the intermediate code generation unit 2 is stored in the intermediate code storage unit 3. Note that the drawing load calculation and intermediate code accumulation may be mixed. In S15, it is determined whether or not drawing data in the page has been completed. If unprocessed drawing data exists, the process returns to S11 and the above processing is repeated. Thus, the processes of S11 to S14 are repeated until the drawing data for one page is completed.
[0017]
When the processing for one page of drawing data is completed, in S16, the drawing method selection unit 5 acquires the load for each drawing method for each processing unit from the drawing load estimation units 4-1 to n, and which processing method is used. It is determined whether processing can be performed at high speed, and a drawing method for performing drawing processing is determined. Here, it is determined in S17 whether or not the drawing load of the drawing method a is larger than the threshold value. If it is within the threshold value, the drawing method a is selected in S18. If the drawing load of the drawing method a is larger than the threshold value, the drawing method b is selected in S19. In S18 or S19, the drawing unit 6 executes the drawing process using the drawing method selected by the drawing method selection unit 5. Then, the drawing result is output to the output device 8.
[0018]
Next, in S20, it is determined whether or not the processing has been completed for all the units. If the processing has not been completed, the process returns to S16 to repeat selection of a drawing method for each predetermined unit and drawing processing by the selected drawing method. When the process is completed for all units, the drawing process for one page is ended.
[0019]
In this way, it is possible to estimate a processing load for each drawing method for each predetermined unit, select a more preferable drawing method, and perform drawing processing using the drawing method. Therefore, even when the processing load increases with a single drawing method, the processing load can be reduced by using another drawing method, and drawing processing can be performed at high speed.
[0020]
FIG. 3 is a conceptual diagram of a drawing method switching process in the embodiment of the drawing processing apparatus of the present invention. FIG. 3 shows the relationship between the processing load estimated by the drawing load estimation units 4-1 and 2 and the processing time when the drawing unit 6 performs the drawing process at the time of the processing load. For example, it is assumed that the drawing method a is a drawing method in which the drawing processing time is short when the processing load is small, but the drawing processing time increases dramatically as the processing load increases. The drawing method b is a drawing method in which the drawing processing time does not become so short even when the processing load is small, but the drawing processing time does not increase so much even if the processing load increases.
[0021]
In this case, when the processing load is small, the drawing method a has a shorter drawing processing time, and when the processing load is large, the drawing method b has a shorter drawing processing time. Therefore, if the estimated processing load is equal to or less than the threshold, the drawing method a is selected when the processing load intersecting the two is set as a threshold, and the drawing method b is selected when the estimated processing load is larger than the threshold. Regardless of the processing load, the drawing processing time is minimized, and the drawing processing can always be performed at high speed.
[0022]
In S17 and FIG. 3 described above, the drawing method is selected using the processing load of the drawing method a. However, this selection condition is merely an example, and various determination conditions can be set. For example, the processing load of the drawing method a and the processing load of the drawing method b are compared, or the drawing processing time is compared. And the drawing method may be selected based on the drawing processing time.
[0023]
Hereinafter, the above-described configuration and operation will be described using specific examples. In the specific example described below, a drawing method a in which vector format data is used as an intermediate code and an overlap process is performed using a band memory, and an overlap process is performed on the intermediate code and then data is written to the band memory. Use scheme b. The processing is switched for each band using the presence frequency of the vector on the scan line as the drawing processing load. As a switching method, for example, as described with reference to FIG. 3 described above, the drawing method a is selected when it is smaller than a predetermined load threshold, and the drawing method b is selected when it is larger, and will be described below. .
[0024]
First, two drawing methods will be described. FIG. 4 is an explanatory diagram of an example of the drawing method a, and FIG. 5 is an explanatory diagram of an example of the drawing method b. Here, as shown in FIGS. 4A and 5A, three drawing data are input. Assume that the drawing order is data that is drawn first and data that is drawn later. That is, the drawing is performed in the order of (1), (2), and (3). In addition, about each drawing data, for convenience of illustration, it distinguishes by attaching | subjecting different hatching.
[0025]
The drawing method a is a method in which data is overwritten on the memory in the order of drawing. That is, in FIG. 4B, the data (1) is first written on the memory. Next, data (2) is written as shown in FIG. 4 (C), and finally data (3) is written as shown in FIG. 4 (D). In this way, data (1) to (3) are drawn.
[0026]
The drawing method b is a method in which the intermediate codes are sorted in the order of coordinate values on the scan line, the upper drawing object is selected in the scan order, and only necessary portions are written in the memory. First, in FIG. 5B, the drawing start point (3) is first obtained, but since no drawing data has been started before this, only the coordinate value is stored as the drawing start point. . Next, in FIG. 5C, the drawing start point {circle around (2)} is acquired, but since drawing {3} already drawn after {2} is the drawing start, nothing is done. Next, in FIG. 5D, the drawing end point of (3) is acquired, and since drawing is the top in the drawing data that has started drawing, (3) is drawn from the drawing start point to this coordinate value. Update the drawing start point. Next, in FIG. 5E, the drawing start point of (1) is acquired. However, since drawing (2) already drawn after (1) is the drawing start, nothing is done. Next, in FIG. 5 (F), the drawing end point (2) is acquired, and since drawing is the top of the drawing data that has started drawing, (2) is drawn from the drawing start point to this coordinate value. Update the drawing start point. Next, in FIG. 5G, the drawing end point of (1) is acquired, and since drawing is the top in the drawing data that has started drawing, (1) is drawn from the drawing start point to this coordinate value. Update the drawing start point. In this way, data (1) to (3) are drawn.
[0027]
FIG. 6 is a flowchart illustrating an example of drawing processing of one band by the drawing method a. First, in S31, the band memory used for output is cleared. Next, in S32, an intermediate code is acquired in the drawing order of the drawing data. Next, in S33, the data in the intermediate code is acquired for one scan line. First, data corresponding to the scan line at the uppermost part (when writing in order from the upper side of the band memory, and conversely when writing from the lower side of the band memory) is acquired.
[0028]
Next, in S34, the coordinate values of the drawing start point and drawing end point held by the intermediate code are acquired, and in S35, the writing start position of the memory is calculated based on these coordinate values. In S36, the width of the drawing data on the scan line is acquired from the intermediate code. In S37, the color data of the drawing data is acquired from the intermediate code. In S38, actual memory writing is performed.
[0029]
Next, in S39, it is determined whether or not data on the scan line has been lost. If there is still data, the process returns to S34 to perform processing on the scan line (from calculation of the memory write position to overwriting of data for the width). repeat. If there is no more data on the scan line, it is determined in S40 whether or not the processing of the acquired intermediate code has been completed. If it still remains, return to S33 to acquire data on one scan line and repeat. When the processing of the acquired intermediate code is completed, it is determined in S41 whether all the drawing data in the band has been processed. If it remains, the process returns to S32, the intermediate code is acquired in the drawing order, and the process for the intermediate code is repeated. If all intermediate codes have been processed, the processing within one band is terminated.
[0030]
FIG. 7 is a flowchart illustrating an example of drawing processing of one band by the drawing method b. First, in S51, the band memory used for output is cleared. Next, in S52, all data on the scan line is acquired. Initially, all the data on the scan line at the top (possible at the bottom as well as the drawing method a) is acquired.
[0031]
Next, in S53, the leftmost drawing point (when drawing from left to right; when drawing from right to left, rightmost drawing) is acquired. Next, in S54, it is determined whether or not the drawing point acquired in S53 is a drawing start point.
[0032]
If the drawing point is the drawing start point, it is registered in S55 that the drawing data has started drawing. Next, in S56, it is determined whether or not the data is the uppermost data (drawn later in the drawing order) in the drawing data that has started drawing. It is determined whether or not there is data below (data previously drawn in the drawing order). If there is an area below (or above) the drawing data, in S58, the drawing data color from the drawing start point to the point on the left of the point of interest (the drawing data coordinate point at which drawing has started) is determined. In step S59, the drawing start point is changed to the attention point. If the drawing start data is not the uppermost side or if there is no drawing data starting drawing on the lower side, the process up to the change of the drawing start point in S59 is skipped and the process proceeds to S62.
[0033]
When it is determined in S54 that the drawing point is not the drawing start point (drawing end point), in S60, it is determined whether or not the drawing point is the uppermost in the drawing data that has started drawing. In S61, the drawing data color is used to draw from the paint start point to the point on the left side of the attention point, and in S59, the drawing start point is changed to the attention point. If it is not the uppermost side in the data that has started drawing, the process up to the change of the drawing start point in S59 is skipped and the process proceeds to S62.
[0034]
In S62, it is determined whether or not all the data on the scan line has been processed. If any data still remains, the process returns to S53 for obtaining the leftmost drawing point and is repeated. If the data on all the scan lines has been processed, it is determined in S63 whether or not there is a scan line that has not yet been processed in the band. Return to the data acquisition process and repeat. When all the processes are performed, the process in one band is terminated.
[0035]
Drawing processing is performed by selecting either drawing method a or drawing method b. In this example, the drawing frequency is used to determine which drawing method is used. Use as a load to switch. Here, this vector will be described. FIG. 8 is an explanatory diagram of an example of a vector data format, and FIG. 9 is an explanatory diagram of a specific example of a vector. A vector can take a data representation as shown in FIG. 8, for example. In the vector data format shown in FIG. 8, “order” is the drawing order of drawing commands. “X” indicates the value of the X intercept of the vector crossing the current scan line. “X displacement” indicates the amount of change in the X intercept when the process moves to the next scan line. “Y displacement” indicates the number of remaining scan lines affected by the vector. “Direction” indicates the direction of the vector.
[0036]
The vector shown in FIG. 8 will be further described using a specific example shown in FIG. As shown in FIG. 9A, a triangle represented by three points P1 (0, 300), P2 (600, 0), and P3 (300, 600) is taken as an example. Here, the coordinate system has an origin at the upper left of the drawing, and accordingly, the X coordinate value increases in the horizontal right direction of the drawing and the Y coordinate value increases in the vertical lower direction of the drawing. From these three points P1, P2 and P3, three vectors P1P2, P2P3 and P3P1 are generated in consideration of the direction of the vector. Each vector is as shown in FIG. For example, in P1P2, “X” is 600, “X displacement” is −2, “Y displacement” is 300, and “direction” is −1. The same applies to other vectors.
[0037]
The drawing range to be filled is unknown only with such vector data. Therefore, it is necessary to determine whether the figure is internal or external based on such a vector. For example, it is necessary to determine the drawing start point and the drawing end point in FIG. 6 showing the above-described drawing method a, and the drawing start point in FIG. 7 showing the drawing method b processing. Hereinafter, determination rules used for internal / external determination of such a figure are shown.
[0038]
There are two types of judgment rules: non-zero winding rules and odd-even rules. These are all rules used when determining the inside of one figure formed by a plurality of external shapes (paths). FIG. 10 is an explanatory diagram of an example of a figure internal determination method based on the non-zero winding rule and the odd / even rule. FIG. 10A shows a non-zero winding rule, and FIG. 10B shows an odd / even rule. The non-zero winding rule shown in FIG. 10 (A) takes into account the direction when the outer shape constituting the figure intersects the scan line. In FIG. 10A, the case where the outer shape intersects the scan line upward is positive, and the case where the outer shape intersects downward is negative. If it is a positive intersection in the scanning direction, 1 is added, and if it is a negative intersection, 1 is subtracted. Then, it is determined that the point from the point where the total value is no longer 0 to the point where it returns to 0 is inside. In the example shown in FIG. 10A, the entire inside of the outer path shown by hatching is the drawing area.
[0039]
On the other hand, according to the odd / even rule shown in FIG. 10B, the number of paths crossing the scan line is counted, and an odd point to an even point are determined to be inside. According to this determination method, as shown by hatching in FIG. 10B, the drawing area is an annular area sandwiched between the outer path and the inner path.
[0040]
FIG. 11 is a flowchart illustrating an example of a vector internal determination process. It is assumed that an attribute area corresponding to the graphic is separately provided in the drawing graphic, and an internal determination rule, an internal determination value, and the like are stored therein. Further, it is assumed that the data of these attribute areas can be acquired from the value indicating the “order” in the vector data shown in FIG. 8, for example.
[0041]
First, in S71, an internal determination rule for a graphic composed of this vector is acquired from the “order” of the vector, and the internal determination rule is examined in S72. If the type of determination rule is a non-zero winding rule, an internal determination value is acquired in S73, and whether or not the value is 0 is checked in S74. If the internal determination value is 0, the vector direction (+1 or -1) is added in S75, and it is determined in S76 that "region start". If the internal determination value is not 0 in S74, the vector direction (+1 or -1) is added in S77, and then it is checked again in S78 whether the internal determination value is 0 or not. If the internal determination value is 0, it is determined in S79 that the region is “end”, and if the value is other than 0, it is determined that the region continues in S80.
[0042]
If the type of the internal determination rule is an odd / even rule in S72, an internal determination value is acquired in S81, and the internal determination value is increased by 1, and then a remainder of 2 is obtained. In S82, it is determined whether the internal determination value is an odd number or an even number depending on whether the remainder of 2 of the internal determination value is 1 or 0. If the internal determination value is an odd number, the figure is inside. . If the internal determination value is an even number, the figure is outside the figure, so it is determined in S84 that “region end”.
[0043]
Note that “region start” corresponds to the drawing start point, “region end” corresponds to the drawing end point, and if this processing is applied to each acquired vector, the drawing start point and drawing end point are set. Can be determined.
[0044]
Next, a method for calculating the drawing processing load will be described. FIG. 12 is a flowchart illustrating an example of a drawing load calculation process. As described above, in this example, the vector presence frequency on the scan line is used as the drawing processing load. First, in S91, it is determined which band the generated vector belongs to, and the maximum value and minimum value on the Y axis of the vector in each band are obtained. In S92, the number obtained by subtracting the minimum value from the maximum value for each band, that is, the length on the Y-axis is obtained. In S93, the calculated value is added to the value corresponding to the load for each band. In this way, the vector presence frequency can be calculated and used as the load of the drawing process.
[0045]
FIG. 13 is an explanatory diagram of a specific example of a drawing load calculation method. In FIG. 13A, the height of each band is 128. It is assumed that a square shown in black is added for the first time. This rectangle belongs to each band with lengths of 64, 128, and 64. Usually, since a vector parallel to the X axis is excluded, this quadrangle is expressed by two vectors. At this time, 64, 128, and 64 are calculated in the respective bands as the load on the left vector. As a load for the right vector, 64, 128, and 64 are calculated in each band. Therefore, the load of each band is 128, 256, and 128, respectively, as shown in FIG.
[0046]
The drawing method is switched using the drawing load thus calculated. FIG. 14 is an explanatory diagram of an example of a drawing method switching method. In FIG. 14, it is assumed that the four figures {circle around (1)} to {circle around (4)} shown in FIG. 14A are overlaid as shown in FIG. Note that the height of the band is 128. Each figure is drawn in contact with the band boundary, figure (1) is in bands B1-B8, figure (2) is in bands B2-B7, figure (3) is in bands B3-B6, figure (4) Belong to bands B4 and B5, respectively. In FIG. 14 (B), the figure (2) is drawn on the figure (1), the figure (3) is further drawn thereon, and the figure (4) is further drawn on the figure (1). The part underneath the figure is not visible.
[0047]
With respect to such figures {circle around (1)} to {circle around (4)}, when the drawing load in each band is calculated as described with reference to FIGS. 12 and 13, it is as shown in the “load” column of FIG. For example, since there is one figure for band B1, the drawing load is 256, and for band B4, there are four figures including a figure that cannot be seen due to overwriting, so the drawing load is 1024.
[0048]
Here, if the processing switching threshold is 1000, the load of the band B4 and the band B5 is 1024, so the drawing method b is selected and used during the drawing process. For other bands, the drawing method a is selected.
[0049]
As described above, in the drawing method “a”, since the figures {circle around (1)} to {circle around (4)} shown in FIG. Do. Therefore, the more overlapping the figures, the longer the drawing processing time. On the other hand, in the drawing method b, since the drawing process is performed while performing the overlap removal process, no unnecessary drawing process is generated, but the process of determining the drawing area is performed, and the overhead of the process is increased. Therefore, when there is little overlap, the determination processing overhead affects the drawing time. In the example shown in FIG. 14, the bands B1 to B3 and the bands B6 to B8 are assumed to have little overlap, and the drawing method a that can perform drawing processing at high speed when there is little overlap is used. In a portion where there are many overlaps, a drawing method b that can perform high-speed drawing processing even if there are many overlaps is used. In this way, the overall drawing processing time can be shortened and high-speed drawing processing can be performed.
[0050]
Next, another specific example will be described. In the specific example described above, an example is shown in which vector format data is used as the intermediate code, and the rendering method is selected using the vector presence frequency on the scan line as the rendering processing load. In this example, trapezoidal data is used as an intermediate code. The drawing methods that can be used are the drawing method a that performs overlapping processing using a band memory, and the drawing method b that performs overlapping processing on an intermediate code and then writes data to the band memory, as in the above example. And As for the drawing processing load in this example, a trapezoidal area (number of pixels) is used for the drawing method a, and the existence frequency of the side on the scan line is used for the drawing method b. As a switching method, the predicted time is calculated from the load, the processing method with the smaller predicted time is selected, and the drawing method is switched for each band.
[0051]
FIG. 15 is an explanatory diagram of an example of trapezoid data. When trapezoidal data is used as the intermediate code, the figure represented by the drawing data is first divided into trapezoids having two sides parallel to the scanning line direction using various existing methods. Then, the information about the position and size of the trapezoid on the drawing area is expressed in the XY coordinate system with the sub-scan line direction as the Y coordinate axis and the scan line direction as the X coordinate axis.
[0052]
In the trapezoidal data structure shown in FIG. 15A, the maximum value YMAX and minimum value YMIN of the Y coordinate value, and the X coordinate value X1 and Y coordinate value at the position of the minimum value of the Y coordinate for the left side are unit quantities. The X coordinate value increment DX1 when increased, and the right side is trapezoidal by the X coordinate value X2 at the position of the minimum value of the Y coordinate value, and the X coordinate value increment DX2 when the Y coordinate value is increased by a unit amount. At least the area is defined. Each coordinate value is shown in FIG. 15B using an example of a trapezoid.
[0053]
Besides the values (YMIN, YMAX, X1, X2, DX1, DX2) related to these drawing areas, there are a fill attribute value (“attribute”) such as a color, a value indicating the drawing order (“order”), and the like. Stored. Note that there are various representations of the filled area by the trapezoid data, and an example is shown here, and other representation methods may be used. Of course, other information may be included.
[0054]
Since the trapezoid represents the internal figure to be filled, the internal determination processing shown in FIG. 11 that is performed when the above intermediate code is expressed by a vector can be omitted. Instead, the left oblique side may be treated as the start point of the region, and the right oblique side may be treated as the end point of the region. However, when the trapezoids expressing the same figure may overlap each other, and it is necessary to fill the overlapping area, the non-zero winding rule is applied as the determination rule.
[0055]
A method for calculating the drawing processing load when the intermediate code is expressed as a trapezoid will be described. As described above, the drawing processing load in this specific example uses a trapezoidal area (the number of pixels) for the drawing method a, and uses the frequency of edges on the scan line for the drawing method b. To do. The trapezoidal area used as a drawing processing load corresponding to the drawing method a can be calculated as follows with reference to FIG.
((XMIN2-XMIN1) × 2 + (DX1 + DX2) × (YMAX−YMIN)) × (YMAX−YMIN) / 2
Here, XMIN1 is the smaller of the maximum Y coordinate value or the minimum X coordinate value of the left side, and XMIN2 is the maximum Y coordinate value or the minimum X coordinate value of the right side. Whichever is smaller.
[0056]
Note that the drawing processing load corresponding to the drawing method b may be calculated in the same manner as in the case where a vector is used as the intermediate code by regarding the hypotenuse of the trapezoid as a vector.
[0057]
In this example, the processing time is further predicted from the load of the drawing process. The prediction of the processing time based on the drawing processing load differs depending on the platform on which the drawing processing is performed. Therefore, it is necessary to determine a function from the load of the drawing process to the processing time in accordance with the platform that actually performs the drawing process. As an example,
Drawing method a processing time = Drawing method a drawing processing load ÷ 16384
Processing time of drawing method b = drawing processing load of drawing method b ÷ 128
Can be obtained.
[0058]
FIG. 16 is an explanatory diagram of an example of a drawing method switching method when the trapezoidal format is used as the intermediate code. Here, the case where the four figures {circle around (1)} to {circle around (4)} shown in FIG. The height of the band is 128 as in FIG. 14, and the widths of the figures (1) to (4) are 128, 256, 384, and 512, respectively.
[0059]
For example, the load of the drawing process in the drawing method a of the band B1 is
(128 × 2 + 0) × 128/2 = 128 × 128 = 16384
So the processing time is
16384 ÷ 16384 = 1
It is. The processing time of the drawing method “a” of the band B2 is 32768, which is 16384 for the drawing process for the graphic (1), and twice for the drawing (2) for the graphic (2).
(16384 + 32768) ÷ 16384 = 3
It becomes. The same applies to the bands B3 to B8, and the processing time of the drawing method a in each band can be predicted as shown in the column of prediction 1 in FIG.
[0060]
Further, for example, the processing time of the drawing method b of the band B1 is determined using the load value shown in FIG.
256 ÷ 128 = 2
It is. In addition, the processing time of the drawing method b of the band B2 is
512/128 = 4
It becomes. The same applies to the bands B3 to B8, and the processing time of the drawing method b in each band can be predicted as shown in the column of prediction 2 in FIG.
[0061]
When the predicted values of the processing times of the drawing method a and the drawing method b shown in FIG. 16B are compared, the processing time of the drawing method b (prediction 2) is the processing time of the drawing method a (prediction 1) in the bands B4 and B5. ) Smaller. Accordingly, the drawing method b is selected in the bands B4 and B5, and the drawing method a is selected in other cases. Drawing processing may be performed using the drawing method selected in this way. Thus, since the drawing process is performed by selecting the drawing method that shortens the drawing process time, a high-speed drawing process can be realized.
[0062]
Furthermore, run-length data may be used as the intermediate code. Even in this case, the drawing method can be switched in the same manner. Similar to the specific example described above, the drawing method a for performing the overlap processing using the band memory and the drawing method b for performing the overlapping processing on the intermediate code and then writing the data to the band memory are used. The trapezoidal area (number of pixels) is used as the load of the drawing method, and the existence frequency of the sides on the scan line is used as the load of the drawing method b. In addition, as a switching method, the predicted time is calculated from the load, and the processing method with the smaller predicted time is selected for each band.
[0063]
The run length can be considered as a special trapezoid. That is, in the above equation for obtaining the trapezoidal area, YMAX-YMIN is always 1 and the values of DX1 and DX2 are 0. That is,
((XMIN2-XMIN1) × 2) / 2 = XMIN2-XMIN1
It becomes. This expression is a value obtained by subtracting the coordinate value of the drawing start point from the coordinates of the drawing end point in the scan line.
[0064]
Since the run length also represents the internal figure to be filled, like the trapezoid, the internal determination process as shown in FIG. 11, for example, can be omitted. Instead, the start point of the run may be treated as the start point of the region, and the end point of the run may be treated as the end point of the region. However, when run lengths representing the same figure may overlap each other and it is necessary to fill the overlapped area, a non-zero winding rule may be applied as a determination rule. The following processing may be performed in the same manner as in the case where the intermediate code has a trapezoidal shape, and thus description thereof is omitted.
[0065]
In the above description of the embodiment and specific example of the drawing processing apparatus of the present invention, the drawing processing time is determined based on the drawing processing load or load inside the drawing processing apparatus and the drawing method is switched. For example, a setting unit that receives an instruction from the user may be provided, and the drawing method may be determined in accordance with the instruction from the user set in the setting unit. For example, a method for specifying a processing method is considered as one of print settings of a printer driver on a client. Further, the drawing method may be set by software that requests drawing. In addition, in the control panel on the drawing processing apparatus side, a fixed processing method may be set and automatic selection and manual setting may be configured. In such a configuration, since the user can select a processing method, the degree of freedom is high and processing as desired by the user can be performed.
[0066]
In the above description, the drawing method has been described using an example in which the drawing method is switched in band units. However, the processing may be switched in page units, job units, or scan line units. Of course, the drawing methods that can be switched are the drawing method a in which the overlapping process is performed using the band memory as in the above-described specific example, and the drawing method b in which the overlapping process is performed on the intermediate code and then the data is written to the band memory. The drawing method is not limited to the above, and any drawing method can be used. The number of drawing methods to be provided is not limited to two, and may be three or more.
[0067]
【The invention's effect】
As is apparent from the above description, according to the present invention, the load on the drawing process, the drawing processing time calculated based on the load, etc. are used, and the load is reduced or drawn out of a plurality of drawing methods. A drawing method that is predicted to have a short processing time is selected and used. As a result, there is an effect that it is possible to perform graphic drawing processing at high speed.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an embodiment of a drawing processing apparatus of the present invention.
FIG. 2 is a flowchart showing an example of operation in the embodiment of the drawing processing apparatus of the present invention.
FIG. 3 is a conceptual diagram of a drawing method switching process in the embodiment of the drawing processing apparatus of the present invention;
FIG. 4 is an explanatory diagram of an example of a drawing method a.
FIG. 5 is an explanatory diagram of an example of a drawing method b.
FIG. 6 is a flowchart illustrating an example of a drawing process of one band by a drawing method a.
FIG. 7 is a flowchart illustrating an example of a drawing process of one band by a drawing method b.
FIG. 8 is an explanatory diagram of an example of a vector data format;
FIG. 9 is an explanatory diagram of a specific example of a vector.
FIG. 10 is an explanatory diagram of an example of a method for determining the inside of a graphic based on a non-zero winding rule and an odd / even rule.
FIG. 11 is a flowchart illustrating an example of a vector internal determination process;
FIG. 12 is a flowchart illustrating an example of a drawing load calculation process.
FIG. 13 is an explanatory diagram of a specific example of a drawing load calculation method.
FIG. 14 is an explanatory diagram illustrating an example of a drawing method switching method;
FIG. 15 is an explanatory diagram of an example of trapezoid data.
FIG. 16 is an explanatory diagram illustrating an example of a drawing method switching method when a trapezoidal format is used as an intermediate code;
FIG. 17 is a flowchart illustrating an example of a conventional drawing process.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Drawing data input part, 2 ... Intermediate code production | generation part, 3 ... Intermediate code storage part, 4-1 to n ... Drawing load estimation part, 5 ... Drawing method selection part, 6 ... Drawing part, 7-1 to n ... Drawing method, 8... Output device.

Claims (11)

  From the drawing data, drawing data input means for inputting drawing data to be drawn, first drawing processing means for performing drawing processing using a drawing method in which data is overwritten on the memory in the order of drawing, A second drawing processing unit for performing a drawing process using a drawing method for performing a process of removing an overlap of the drawing figures in the state of the generated intermediate code, and a processing load or a process in the first and second drawing processing units A processing load estimation means for estimating time and a drawing for selecting a drawing processing means capable of processing at the highest speed based on the processing load or processing time in the first and second drawing processing means estimated by the processing load estimation means A drawing processing apparatus comprising a method selection unit.   The drawing processing apparatus according to claim 1, wherein the processing load estimation unit uses, as a processing load, an existing frequency on a scan line of a vector for drawing a graphic.   The processing load estimation means uses a drawing area as a processing load of the first drawing processing means, and uses a presence frequency on a scan line of a graphic side to be drawn as a processing load of the second drawing processing means. The drawing processing apparatus according to claim 1.   A setting unit configured to set a fixed drawing method, and the drawing method selection unit selects the set drawing method when the fixed drawing method is set by the setting unit; The drawing processing apparatus according to claim 1.   5. The drawing processing apparatus according to claim 1, wherein the drawing method selection unit selects the first or second drawing processing unit in units of jobs. 6.   5. The drawing processing apparatus according to claim 1, wherein the drawing method selection unit selects the first or second drawing processing unit in units of pages.   The drawing processing apparatus according to claim 1, wherein the drawing method selecting unit selects the first or second drawing processing unit in a band unit.   The drawing processing apparatus according to claim 1, wherein the drawing method selecting unit selects the first or second drawing processing unit in units of lines.   It has intermediate code generation means for generating an intermediate code format in vector format from the drawing data, and the first and second drawing processing means perform drawing processing from the intermediate code format. The drawing processing apparatus according to claim 1.   It has intermediate code generation means for generating a trapezoidal intermediate code format from the drawing data, and the first and second drawing processing means perform drawing processing from the intermediate code format. The drawing processing apparatus according to claim 1.   Characterized in that it has intermediate code generation means for generating an intermediate code format in a run-length format from the drawing data, wherein the first and second drawing processing means perform drawing processing from the intermediate code format. The drawing processing apparatus according to any one of claims 1 to 8.

Images