JPS62196772A - Picture clipping system - Google Patents

Abstract

PURPOSE:To shorten time for processing by making previous calculation at each scan on the clipping calculation part applicable in common to run length data in one scan, and calculating using the result in common in the same scan. CONSTITUTION:When making clipping by making graphic conversion calculation on run length vector data, invariable parameter in the same scan is calculated at each scan and applied to run length vector data in the scan utilizing the fact that run length data 42-26 have coordinate values which are the same value in the same scan. Thereby, time for processing is shortened and high speed graphic clipping process is realized. That is, when (y) coordinate of the starting point is below the lower limit for objective vectors of output 43-45, the part below the lower limit is cut off, and when (y) coordinate of the terminal point of the vector is above the upper limit, unnecessary part of vector of parts 43-45 above the upper limit is cut off.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention optically scans a drawn figure using an image scanner to define it as run-length data, and further performs a figure conversion operation to obtain one of the data. The present invention relates to an image clipping method for clipping a rectangular region of a portion.

[Conventional technology]

First, run-length vectors, which are commonly used in image processing technology, will be explained. FIG. 4 is a diagram showing the run length vector of a figure td A IF drawn on the paper. In the figure, 1 is a white drawing to be read, 2 is a figure drawn in black on drawing 1, and 3 is a figure drawn in black on drawing 1. The trajectory of the scan, 4 is a run-length vector indicating from which start point to the end point the black of figure 2 continues within the scan, 5 is the start point of the run-length vector 4, 6 is the end point of the run-length vector, Y is the coordinate axis.

Here, scanning is performed at regular intervals on the X-axis to obtain run length data for the entire drawing.

Next, a calculation algorithm for run-length vector data in the conventional graphic conversion method will be described below.

Equation (1) is a conversion equation applied when correcting graphic distortion, and a=h is a constant.

Here, X and '/ in equation (1) represent the start point or end point of the run-length vector, and are the coordinate values on x and YI,
x' and y' are coordinate values of points on the X and Y axes after transformation corresponding to the coordinate values (x, y).

Parameter a+ br C+ dy e+ f+ g+
As shown in Figure 5, h is the coordinates of the four points in the skin data on the input diagram (X□r ys) + (X2t yz)
+ (Xsr y3) + (X41 y4) and the coordinates on the output diagram corresponding to those four points (X'+' +
'jx' ) + (Xi' r Vz') + (
X3'ly3')l (X4'?y4') and (2
) is obtained by solving the equation.

After calculating equation (1), even if the run length data is within the same scan, the coordinate value x' (or y'
) are generally not equal.

In order to clip a part of the rectangular area from the run-length vector data after calculation of equation (1) as shown in FIG. 6(a) to FIG. 6(b), conventionally, How to use it” (Gijutsu Hyoronsha, August 1983)
The following approach, shown on pages 26-28, was used.

As shown in FIG. 7, the space is divided into nine parts inside and outside the clipping area, and encoded with 4 bits as shown in FIG. These 4 bits are numbered 1st from the left. 2nd゜3rd. Fourth
If the first bit is 1, it means that it is above the upper limit of the edge of the clipping area, and the second bit is 1, it means that it is below the lower limit of the edge of the clipping area, When the third bit is 1, it means that it is to the right of the right edge limit of the clipping area, and when the fourth bit is 1, it means that it is to the left of the left edge limit of the clipping area.

Next, it is classified into which space the start point and end point of the vector fall into, and the AND of each 4-bit code is calculated. If the result is not oooo, the vector is completely outside the clipping area and does not need to be drawn at all.

If the logical product of the 4-bit code is oooo, the vector is cut according to the meaning of the bit that is 1 in the 4bjt code at each of the starting point and ending point of the vector.

For example, as shown in FIG. 8, vector AB is first cut off by AC to become vector CB, then CD is cut out, and finally becomes vector DB, so that a vector within the clipping area is obtained.

Further, in the flowchart of FIG. 9-(1) and (2), 7 is a "correction constant calculation" step, 8 is a "scan line initial setting" step, 9 is a "judgment step indicating scan line end point j, and 10 is a " 11 is the step of "calculating formula (1)",
12 is a "scan line count up" step, 13 is a "start point X' direction classification" step, 14 is a "start point y' direction classification" step, 15 is an "end point X' direction classification" step, and 16 is an "end point y' direction classification" step. 17 is a space code logical product step, 18 is a logical product judgment step,
19 is the "vector equation calculation" step, 2o is r
21 is the step of "determining the sign bit of the starting point", 22 is the step of "calculating the intersection of two straight lines,"
"Cut unnecessary vector" step, 23 is "r bit counter value count up" step, 24 is r 4 b
25 is the r bit counter initial setting J step, and 26 is the "end point sign bit determination"
Step 27 is a "calculate intersection of two straight lines, cut out unnecessary vectors" step, 28 is a "rbit count value count up" step, 29 is a determination step "r 4 bit end", and 30 is a "vector drawing" step.

Furthermore, an example of the configuration of a general image processing system (for example, an image information file device) is shown in FIG.

That is, (A) is an image input/output device, which is connected to an image processing device (D) via an image bus (C) together with an image memory device (B).

(E) is a registered search processing device and data processor;
It consists of an RT, a keyboard, etc., and is connected to an index management device (G) such as a floppy disk or magnetic disk via a data bus (F), and performs predetermined calculations.

Next, the operations shown in FIG. 9-(1) and (2) will be explained.

First, in step 7 "Calculation of correction constant", the parameter 8. of equation (1) above. bg Q? After calculating d, 8tL g+ h and performing the initial setting of the target scan line in step 8 "Initial scan line setting", there is no target scan line in the judgment of "scan line end" in step 9. If so, the operation is ended, and if there is a scan line to be scanned, the process moves to step 10.

If it is determined in step 10 that the run length within a scan is completed, and there is no run length data that has not been subjected to the conversion operation within the target scan line, then the step 12 is the ``scan line count up''. If it exists, the process moves to step 11, ``calculation of equation (1)''. Then, in step 11, "calculation of equation (1)" is performed on the coordinate values of the start point and end point of the run-length vector to obtain the coordinate values of the corresponding run-length vector.

After step 11 “calculating equation (1)”, step 13
"Start point X' direction classification" in Step 14, "Start point y' direction classification" in Step 15, "End point X' direction classification" in Step 16, and "End point y' direction classification" in Step 16 are performed. For each, the 9-divided space shown in FIG. 7 is classified.

Next, in step 17, ``spatial code logical AND'', the spatial classification code at the starting point of the vector and the spatial classification code at the end point of the vector are logically ANDed. If the logical product is 0000 in the "logical product judgment" in step 18, the process branches to step 19, and if it is not 0000, the vector does not span the clipping area, so the process returns to step 10.

In Step 19, [Calculate vector equation], calculate the equation of the straight line passing through the starting point of the vector. In step 20, "r bit counter initialization", the space code 4 bit
Initialize a counter for counting the bits of t, and count up the rbit counter value in step 23.
and step 24, “r 4 bit end”,
Step 2 for the 4-bit spatial code at the vector starting point
1 and step 22 are performed.

In the sign bit judgment J of the r starting point in step 21, bit
Determine whether the spatial code bit of the starting point indicated by the counter is 1 or not. If it is 1, proceed to step 22 "calculating the intersection of two straight lines and cutting out unnecessary vectors", and calculate the linear equation of the vector and the clipping area. Find the point of intersection of the edge with the linear equation and change the starting point of the vector to that point of intersection to cut out the unnecessary part of the vector.

Steps 20, 21, 22, 23, and 24 are vector cutting with respect to the vector starting point.

In exactly the same way, in steps 25, 26, 27, 28, and 29, the vector is cut at the end point of the vector, and in step 30, "vector drawing", a vector that is only within the clipping area without unnecessary parts is output. Output to device.

For example, if the total number of scans is 1000 lines, there are an average of 10 run-length vectors in one scan.

1/4 of the total run length vector within the clipping area
If there is, the amount of computation for the entire drawing will be approximately as follows.

2 steps of step 7.8 = 1 time II 9, 12 2 steps: total number of scan lines = 1000 times n 10, 1.1, 12, 13, 14, 15. :
Number of vectors in total run length vector 16, 17.18 = 10009 steps
, 23, 24, 26, 28, 29. : 6 steps in the clipping area Number of vectors in the area x number of code bits = 1000 x 10 x l/4
The aim of the programming is that it can be programmed in about two lines, but only step 7, ``Calculation of correction constants'', requires about 100 lines in FORTRAN. Step 7
is executed only once, so it is negligible in terms of the overall amount of calculations.

[Problem that the invention seeks to solve]

Since the conventional clipping method is performed as described above, the coordinate value X (or Y) of the run-length vector data before the graphic conversion operation is not based on the property that it is constant within the same scan. Since the run-length vector after the graphic conversion operation was clipped, the judgment and calculation process associated with the clipping became complicated, and there was a problem in that a large amount of calculation time was required.

This invention was made to solve the above-mentioned problems.Since the coordinate value X (or Y) of run-length vector data before performing a graphic conversion operation is constant within the same scan, A clipping method that improves the efficiency of calculations by pre-calculating parts that can be commonly applied to data within a scan for each scan, and using the results commonly within the same scan. The purpose is to provide

[Means for solving problems]

The image clipping method according to the present invention scans a figure on paper using an image scanner as an input device, converts the run length data of the figure obtained by the operation using an image processing device, etc. In a clipping method that clips a part of a rectangular area and outputs it to an output device, a parameter whose value does not change within one scan is calculated for each scan. Clipping operations are performed using value parameters.

[Effect]

The image clipping method in this invention uses the fact that run-length vector data has coordinate values that are the same within the same scan when clipping is performed by performing graphical transformation operations on run-length vector data. By calculating parameters for each scan and applying them to the run-length vector data within a scan, processing time is reduced and high-speed graphical clipping processing is achieved.

〔Example〕

FIGS. 1-(1) and (2) are flowcharts 1 to 1 showing an embodiment of the present invention, and FIG.

The same functions as in (2) are illustrated with the same symbols. In FIG. 1-(1) and (2), 31 is a "clipping constant calculation" step, 32 is an "X-direction classification" step,
33 is an "outside range" determination step, 34 is a "y direction upper limit calculation" step, 35 is a "y direction lower limit calculation" step,
36 is a determination step for "start point y ≦ upper limit", 37 is a "determination step for end point y ≥ upper and lower limit", 38 is a determination step for "start point y x lower limit", 39 is a "start point y = upper and lower limit step, 40
is the determination step for “end point y>upper limit”, and 41 is the “end point y=upper limit” step.
This is the "upper limit" step.

First, an overview of the operation of the present invention will be explained.

Equation (1) is the conversion equation from (a) to (b) in Figure 2, but conversely, equation (3) is the conversion equation from (b) to (a).
Obtained by solving the equation.

Next, as shown in Figure 2, the clipping rectangular area "vertex (Xexe yet), (x
3□, y ni x) * (XG3# y ot) t
(Xe4+y≦4)" is located before the conversion operation of equation (1). Vertex (xe□v yet) l (
xczt yes) +(X(3t ye3) y
(X(41ye4)J is calculated using equation (3).

Then, the xyX coordinates before the conversion calculation in equation (1) are the vertex (
xe□+Ve□) and vertex (X e2t'/cm)
The straight line connecting the vertices, the vertices (xe2.
+Y e+) straight line connecting the vertex (Xcsv yea)
and the vertex (Xe*yy(:4), the straight line connecting the vertex (xe,
, ye4) and the vertex (xe□+yet) to find four straight line equations.6 The following is an explanation based on the example in FIG. 2.

The scan line of the run-length vector before the conversion calculation in equation (1), that is, the X coordinate is less than the X coordinate of the vertex (Xe4+y≦4), or the vertex (X,:z).

If it exceeds the X coordinate of Ve2), the vector is completely outside the clipping area and does not need to be drawn.

When the X coordinate of the run length vector before conversion in equation (1) is greater than or equal to the X coordinate of the vertex (X(:41 y(:4)) and less than the X coordinate of the vertex (xe□v'jct) has a vertex (X eat ': I cs ) and a vertex (x
e4. The straight line connecting ye4) is the upper limit, and the vertex (x(
-, , ye4) and the vertex (Xei+yet) is the lower limit, so vectors outside this range are cut out and output to the output device.

Similarly, if the X coordinate of the run length vector before conversion in equation (1) is greater than or equal to the X coordinate of vertex (xet+yζ1) and less than or equal to the X coordinate of vertex (xe3.yea), then :3) and the vertex (X e+ .

The straight line connecting yen) is the upper limit, and the vertex (x(1゜ye
t) and the vertex (X(zt yes)) is the lower limit.

Also, X of the run length vector before the conversion operation in equation (1)
If the coordinate exceeds the X coordinate of the vertex (Xe3w yea) and is less than or equal to the X coordinate of the vertex (X e2t'/ex), the coordinate is
) and the vertex (X (131'/ea) is the upper limit, and the line connecting the vertex (x et t V et ) and the vertex (
The straight line connecting XCxv 'jet) is the lower limit.

Checking the upper and lower limits of the run-length vector and cutting out unnecessary parts of the vector are performed as follows.

As shown in Figure 3, when 42, 43, 44, 45 degrees 46 are run length vectors in the target scan, 47 is the upper limit and 48 is the lower limit, first, the X coordinate of the starting point of the run length vector is X that is below the upper limit and the end point
A vector whose coordinates are greater than or equal to the lower limit, that is, 43.
Only 44.45 will be output.

Then, for this output target vector, if the X coordinate of the starting point is less than or equal to the lower limit, the portion that is less than the lower limit is cut out, and if the X coordinate of the end point of the vector is greater than or equal to the upper limit, the portion that is more than the upper limit is cut off. That is, unnecessary portions of the vectors 43 and 45 in FIG. 3 are cut off.

Next, the operation will be explained in detail based on the flowchart shown in FIG. First, in step 7 "Calculation of correction constant" (1)
Parameters a, b, QT a of the equation.

ev f + gy h is calculated, and in step 31 "clipping constant calculation, the parameter a/ of equation (3).

b' * O' + d' t '' + f' r g
' * h' and the equations of the straight lines on the four sides of the rectangular area before the calculation of equation (1) that corresponds to the clipping rectangular area after the calculation of equation (1) are calculated, and the Executes the initial setting of the target scan line, and if there is no target scan line in the judgment of [Scan line end] in step 9, the operation is terminated, and if there is a target scan line, the operation is terminated. Step 32
to move to.

Then, in the "X direction classification" of step 32, the target scan line is "Vertex (X c4 * yes)" that does not need to be drawn at all.
or the X coordinate of the vertex (Xez+yex)
``Beyond the coordinates'' or one color of the ring.

The vertex (Xgx
r'/c) or the vertex (Xet+y
is greater than or equal to the X coordinate of e□), and the vertex (X ez + 'I
Is it less than or equal to the X coordinate of ea ), or is it less than or equal to the X coordinate of (
It is classified whether it exceeds the X coordinate of 1) and is below the X coordinate of the vertex (ICx+yc2).

After step 32, if the target scan line is classified as something that does not require drawing at all in step 33, the process moves to step 12; otherwise, step Move to 34. Step 12
In the "scan line count up" step, the scan line count is increased by one and the process moves to step 9.

In step 34 "y direction upper limit calculation", step 32
The upper limit in the y direction is determined based on the classification and the calculated constant in step 31, the y direction is similarly determined in step 35 "y direction lower limit calculation", and the y direction is determined in step 10 "end of scan run length". , if there is no run-length data on which clipping and conversion operations have not been performed within the target scan line, the process described in step 12 above is performed.
If there is a scan line count up, the process moves to step 36.

If it is determined in step 36 that "start point y≦upper limit", the X coordinate of the start point of the run-length vector is less than or equal to the upper limit, the process moves to step 37, and if it exceeds the upper limit, the process moves to step 1o. In the determination of "end point y≧lower limit" in step 37, if the X coordinate of the end point of the run-length vector is greater than or equal to the lower limit, the process moves to step 38, and if it is less than the lower limit, the process moves to step 10.

If the X coordinate of the start point of the run-length vector is less than the lower limit in the determination of "Start point y = lower limit" in step 38, then in Step 39 "Start point y = lower limit" the vector is unnecessary with the X coordinate of the start point set as the lower limit value. Step 40 after cutting out the section
If the X coordinate of the starting point of the run length vector is equal to or greater than the lower limit, the process immediately moves to step 40.

Similarly, if the X coordinate of the end point of the run-length vector exceeds the upper limit in step 40, ``end point y>upper limit'', the unnecessary part of the vector is cut off in step 41, ``end point y=upper limit''.

Then, in Step 11, "Calculation of Equation (1)," Equation (1) is computed for the coordinates of the starting point and end point of the vector that has been cut off, and in Step 30, "Vector drawing," the output device Output to.

That is, in step 31, calculations regarding the entire clipping area are performed, and for each scan, steps 9, 12,
32, 33, 34, and 35 are performed (however, the execution of steps 34 and 35 depends on the determination result of step 33), and the step 10.36,37,38,39,40,41,1
1° (however, steps 38, 39, 40° 4
1, if is performed only for run-length vectors that lie within the clipping region).

For example, if the total number of scans is 1000 lines, there are an average of 10 run-length vectors in one scan, and there are 172 scan lines and 1/4 of the total run-length vectors within the clipping area. For the entire drawing, the amount of calculation is approximately as follows.

Step 7.31.8 3 steps: 1 time #9.
12, 32, 33: Total number of scan lines = 4 steps 1ooo times 34.35: Clipping area 2 steps
Number of internal scan lines = 1000 x 1/2 times II 10, 36, 37: Clipping region 3 steps Total number of run length spectra in internal scan lines = 1000 x 1/Z x 10 times II 38, 39, 40, 41, 11. : 5 steps of clipping area Number of vectors in area x1
000 x 10 x 1/4 times Total of about 13,000 steps It should be noted that each step can be programmed in about two lines in programming using the FORTRAN language, for example. However, steps 7 and 31 require about 100 lines in FORTRAN, but are executed only once, so they can be ignored in terms of the overall processing amount.

In the above embodiment, as shown in FIG.
Xe) + yes), vertex (XCxr yes), vertex (X(Xls ye3) - vertex (XCxr yes)
However, in general, the vertices of the clipping area before formula (1) are sorted in descending order of the X coordinate value, and the vertices (X C2 + 'j) and the vertices (X C2 + '/
e2), vertex (x(:Zl ycz) and vertex (Xe3+
ye3) - vertex (XCxr yei) and vertex (Xe4
9 y(:4)-vertex(Xe4t ye4) and vertex(x
The upper limit value and lower limit value for a certain scan are determined from the relationship in which e□+ye 工) are connected.

Further, the graphic conversion calculation in equation (1) may be performed using a high-speed calculation method as shown in Japanese Patent Application No. 7213/1983.

In the above embodiment, a part of the rectangular area of the input image after the conversion calculation of equation (1) is clipped, but the rectangular area may generally be any square. Furthermore, although an example has been described in which only vectors within the clipping area are drawn, it is also possible to similarly draw only vectors within the clipping area.

〔Effect of the invention〕

As described above, according to the present invention, the clipping calculation part that can be commonly applied to run length data within one scan is calculated in advance for each scan, and the results are commonly applied within the same scan. This has the effect of shortening the processing time.

[Brief explanation of drawings]

Figures 1-(1) and (2) are a flowchart of a clipping method showing an embodiment of the present invention, Figure 2 is an explanatory diagram of the clipping area before and after performing graphic conversion operations, and Figure 3 is an embodiment of the present invention. Figure 4 is a general explanation of run-length vectors; Figure 5 is an illustration of the entire image area before and after the graphic conversion operation; Figure 6 is a general explanation of clipping; The figure is an explanatory diagram of space codes in the conventional clipping method, FIG. 8 is an explanatory diagram of clipping of one vector in the conventional clipping method, FIG. The figure is a diagram of an image processing system. 11 is one type of calculation, and (A) is an image input/output device. (B) is an image memory device, and (D) is an image processing device. In addition, in the figures, the same reference numerals indicate the same or corresponding parts. Patent Applicant Mitsubishi Electric Corporation No. 1 Leap-(1) (a) (bl fig. 4 fig. 5(a) (b) fig. 7 fig. 8 fig. 9-(1) (D), Image Pro't=, Shin7゛den〈ist continuation 1 mountain positive "+'F (spontaneous) 61.6.-2 Showa year month H 't='f, i'l'L'I'L<Palace 1 , 'j Usou's indication 1, 'j Gansho No. 61-38540 2, Name of the invention Image clipping method 3, Representative Moriya Shiki 5, Subject of amendment 6, Contents of amendment (1) Details (2) Amend page 15 of the attached specification. (3) Amend attached page 1-(1). (4) Amend attached page 1 of the attached specification. Amend Figure 1 - (2). (5) Amend Attached Sheet, Figure 9 - (1). (6) Amend Attached Sheet, Figure 9 - (2). 7. Attached documents Catalog (1) One document stating page 15 of the amended specification (2) One document stating Figure 1-(1) after the amendment (
3) Document stating the amended Figure 1-(2) 1 copy (4
) 1 document containing amended Figure 9-(1) (5)
One or more documents describing Figure 9-(2) after correction 21)
, (X'c 21'+2), (n, + 133),
Where is (xQ4+1o4)J located before the conversion operation in equation (1)? Vertex (X + y), (xo8.yc,
) 1el C1 (x + y) and (xc4+yc4)J are calculated using equation (3). Then, at the xy coordinates before the conversion calculation in equation (1), the vertex (
Direct el cl line connecting the vertex (x = y) and the vertex (X c2I yc, ), the vertex (x, y) and the vertex (x, 3r'
A straight line connecting IC ri with c2 C2, apex (x, y) and apex (z C
4mas cs y ), the vertex (X c 4* yc 4) and the vertex Figure 1-(1) Figure 9-(1)

Claims (1)

[Claims] A figure is scanned by an image scanner, the run length data of the figure obtained by the scanning is imported into an image memory device, a conversion calculation formula is executed by an image processing device, and a rectangular area of a part of the result of the calculation is In the image clipping method of clipping and outputting to an output device, a parameter whose value does not change within one scan of the image scanner is calculated for each scan, and the constant value is calculated for run length data within the same scan. An image clipping method characterized by performing a clipping operation using parameters.