JPH02285476A - Data generating method for free curve - Google Patents

Abstract

PURPOSE:To compress the data quantity for reproducing faithfully an original curve by setting a tapered sequential decision area which can avoid effectively a data input and executing the data input. CONSTITUTION:A cursor 5 is put at one point on an outline 4C, and by setting its point as a start point P1, tracing of the outline 4C is advanced. When a coordinate value inputted newly is outside of an area whose area is a1, around the point P1, its point is set as P2. A sequential decision area D is set, based on vectors P1, P2 as a reference, symmetrical with regard to a half line L passing through the point P2 by setting the point P1 as an origin, and an area surrounded by a superior arc D1 obtained by cutting a circular arc of a prescribed radius a2 around the point P2 at about + or -45 deg. against the half straight line L, a hyperbola D2 which is connected smoothly at an end point of the superior arc D1 and asymptotic to the half straight line L, and a segment whose length becomes (h) in the direction being orthogonal to the half straight line L. Subsequently, when a point P3 inputted newly is outside of the sequential decision area D, the vectors P2, P3 are set as a new reference. In such a way, since the tapered area is set in the direction running along the half straight line L, it will suffice that the data quantity is small.

Description

[Detailed Description of the Invention] (Industrial Application Field) This invention allows arbitrary curves, such as the outline of a figure, to be traced using an input device such as a pen or cursor, and the curve is expressed as a sequence of coordinate points of Dinotal. This paper relates to a data generation method for free curves.

(Prior Art) Let us take as an example a cutout mask creation device used in a plate-making process. A cutout mask is a film in which only a desired graphic area is transparent and the other parts are opaque, and is used to photographically combine two images.

Creation of the cutout mask is now done electronically, by inputting transparent and smooth extrinsic regions on a digitizer.
The input data is passed to the output device, the beer film is cut, and the mask is completed.

In this case, the graphical area to be made transparent is converted into data by the operator tracing the outline of the graphical area of the original drawing with a pen or cursor on the digitive table. That is, a contour line (a curved line (one line)) is approximated by a plurality of contour points, and the coordinate data of the contour points are sequentially stored in a memory.

By the way, in general, when expressing the outline of a figure as a sequence of points, it is better to take more points at short intervals in order to faithfully reproduce the outline, but there is a problem that the amount of data becomes enormous. Conversely, if points are taken at relatively coarse intervals, the amount of data can be reduced, but the problem arises that the traced original curve cannot be faithfully reproduced. Efforts have been made to solve the conflicting problem between reproduction fidelity and data volume, but one solution is to compress the data volume while maintaining a certain level of fidelity. An example of this is the figure contour data compression method disclosed in Japanese Patent Application Laid-Open No. 59-49071.

To specifically explain this conventional example with reference to FIG. 5, the point P
Coordinate data has already been input for n-x and point Pn, and point P n
et, point Pn◆2 is an input candidate point (sampling point along the contour line), and is a virtual # connecting point Pn and point Pn++.
Let the length of 1 minute be rffiJ, and the angle that this line segment makes with its preceding line segment be ``θ'', find the product value ``r・θ'' of both, and compare this with a preset constant fkJ. 'r・θ
≧k, the coordinate value of point Pn÷1 is determined to be necessary information, and conversely, if l・θ<kJ, it is determined to be unnecessary information and no data input is performed.And 1, αPn+1 If it is determined that the coordinate value is unnecessary information, the same process is subsequently performed for the next sampling point Pn+2 and any other point P++.

(Problems with the conventional technology) However, in the conventional technology, the main focus is basically only on the problem of the amount of data with respect to the fidelity of curve reproduction.
At present, little consideration has been given to the background in which an operator manually traces a curve, which is the object of this invention. That is, some operators trace the curve quickly, while others trace the curve monotonously, slowly, and faithfully.

When tracing a curve quickly, it is certainly more efficient (and has the advantage of reducing camera shake, but it is not possible to trace a long distance monotonously, and the pen or cursor must be moved intermittently, making it difficult to As shown in Figure 6 (a), a backlash-like return portion (B I L (B 2 L...) occurs at the point where the cursor stops moving. Also, in order to move quickly, the trace locus is There is also a high probability that they will leave.

Backlash-like return part (B 1 ), (B 2 )
, ..., according to the conventional example shown in Fig. 5, l is short and θ becomes extremely large (see Fig. 6 (b)), and "r・θ≧k" is satisfied. There is an extremely high possibility that the point Pn+x will be determined to be necessary and input as valid data. Furthermore 1. If αpH+1 is input as data, there is a possibility that the sampling point Px on the way to the next point pH+2 along the curve will also be inputted as valid data because θ will be thick.

If coordinate values are input as data due to such backlash, the reproduced curve will have local jagged edges as shown in FIG. 6 (,), causing a problem in fidelity.

On the other hand, when tracing a curve slowly, although it is true that the trace locus rarely deviates from the curve, the work efficiency is low, and worse still, the influence of camera shake becomes a major problem. In other words, the operator moves his nervous hand rather than following the curve with his eyes, and moves and traces the nervous hand without applying feedback. This means tracing a curve in a zigzag pattern left and right with the original song #! I can't say that I can get it.

An object of the present invention is to improve the above-mentioned problems caused by manual manipulation by an operator as well as data compression.

(Means for Achieving the Object) A free curve data generation method according to the present invention involves manually tracing a free curve using an input device on a plane giving coordinate values, and calculating coordinate values of sampling points along the free curve. In a method of sequentially automatically inputting data into a memory, a superior arc of a predetermined radius centered on the second point is created between the first point where data is input and the point tpJ2 where data is input next to this fpJl point. from the first point to the second point.
A half-line passing through the point is defined, and a sequential judgment area is defined by the above-mentioned arc and a straight line or curve that asymptotically approaches the half-line from the end point of the above-mentioned arc, and the sampling point is within this sequential judgment area. If it belongs, do not input data; if it does not belong, input the coordinate values of the sampling point, and use the input sampling point as the new second point. The basic feature is that data is input by repeating the process.

(Example) FIG. 1 is a block diagram of an embodiment.

(1) is a digitizer that outputs coordinate values (x+y) at a predetermined period, and is used to create a cutting mask etc.
2) is placed on the table (3) and the outline (4) of the figure (4).
It is equipped with a cursor (5) for tracing C).

(6) is a processing device into which the coordinate values (x, y) are input, and includes a microcomputer, an arithmetic/judgment section (7), a control section (8), and a memory (9) that sequentially stores the coordinate values. Be prepared. The control unit (8) supplies sampling pulses to the digitizer (1), and the calculation/judgment unit (7) performs predetermined calculation processing based on the coordinate values (x+y) input at each sampling period. It is determined whether or not the determined coordinate values satisfy certain conditions, and based on the determination result, it is determined whether or not to transfer the coordinate values to the memory (9).

Specifically, the digitizer (1) has a minimum resolution of 25
The sampling period is set to 30S. The calculation/judgment unit (7) performs processing according to a predetermined program based on the coordinate value data.

An outline of this program 70- is shown in FIG.

There is an initial setting step group (IL) consisting of steps (Sl) to (84') and steps (S5) to (SIO).

First, in step (Sl), the coordinate input device, that is, the cursor (5) is activated, and the processing device (6) inputs the coordinate value (x
, y) is placed in a state where it can be accepted. Next step (S
In 2), place the cursor (5) at one point on the outline (4C), set that point as the starting point (first point) P+, and set the coordinate values by operating the keys of the cursor (5), for example. do. The tracing of the contour (4C) proceeds.

In step (S3), new coordinate values are input every sampling period of 30tes. In step (S4),
It is determined whether the input coordinate values are within or outside a region with a radius a centered on point P1.

If it is within the range, the process returns to step (S3) and the same determination process is repeated for the next input coordinate value.

In this example, a is set to 250 μm.

In step (S4), if it is determined that the point is "outside", the coordinate value of this point is stored in the memory in the next step (S4'), with that point as P2, as shown in FIG. . At this point, the memory (9) contains points P and P.
2 (x=y) coordinate values are stored.

In step (S5), the point P1 and the point P2 form a sequential determination area (D) as shown in FIG. 3(b).
Set. The sequential determination area (D) is set based on P IP 2, and is a half straight line (L
) and is symmetrical with respect to point P2 and has a predetermined radius a2 cut at approximately ±45° with respect to half surface #1l (L). Connected to cow blood! Hyperbolic #X(D 2 ) asymptotic to i(L”)
, and a line segment having a length h in the direction orthogonal to the half-line (L).

Specifically, in the sequential determination area (D) of this example, a2 is 250 μm, similar to al in FIG. 3(a), and the tip width 11 is 5.
The total length is 0 to 100μ and the total length is about IIo11. Note that, as shown by the dashed line in FIG. 3(b), if the total length in the two directions of P 1P is increased and the tip width is set to h, which is smaller than h, the amount of data can be significantly reduced.

In addition, as will be described later, data is not input for points that fall inside the arc of radius a2, so the area surrounded by this arc is
Effectively absorbs minute operator shakes and other noise.

Now, when the sequential determination area (D) is set in step (S5), the process proceeds to step (S6), where new coordinate values are input. In the next step (S7), it is sequentially determined whether the input point is within or outside the determination area (D). If it is within the range, the coordinates of this point are not input into the memory, and the process returns to step (S6), where the coordinate values of the point obtained by the next sampling are input.

In step (S7), if it is determined that the coordinate values of the input point are "outside" the sequential determination area (D), the point P,
In the next step (S8), at the same time as the coordinate values are input into the memory, P2P is set as a new reference.

In step (810), it is determined whether tracing of the curve (contour) has been completed, and if it has been completed, this program is terminated and the process is terminated. If not, repeat. Note that the end of the step (SIO) is determined by detecting an end signal transmitted by forging an input key provided at the cursor (5).

As can be seen from the sequential determination area (D) shown in Figure 3(b), the setting is tapered in the direction along the bovine blood acL), so the straight line part of the traced curve is the frequency of data input. is small, with a maximum of 25 only in areas with extremely large curvature.
Data is only input at intervals of 0. Therefore, the amount of data stored in the memory (9) can be relatively small.

Regarding the determination of whether or not it belongs to the area (D), in the above embodiment, the coordinate values of the points inputted at every 30a+s sampling time are used, but this is done using the coordinates of multiple points. This may be done by taking the average of the values. For example, if you sequentially memorize 7 coordinates of points that are input every 30 m5, and when the 8th coordinate is input, calculate the average value of these 8 coordinate values, and then calculate the point given by this average value. Determine whether it belongs to area (D). When the ninth data is input, moving average processing is performed in which the oldest seven data are discarded and the average of eight new data points is calculated together with the eighth data. When this moving average processing is performed, the left and right knob-zab components of the zigzag trace trajectory caused by small camera shakes by the operator are canceled out, and the input data means that the original curve is faithfully traced, which improves the curve reproduction. As the fidelity improves, the smoothness of the linear approximation also increases. Note that this process is performed in step S3 in FIG.
), (S 6 ) is effective, and may be performed in step (S2) instead of specifying with a cursor.

Furthermore, in the above embodiment, the coordinate values of points that have gone out of the determination area (D) are stored in the memory one after another, but it is also possible to perform processing on these coordinate values and store the processed coordinate data. You may also do so. For example, point P in FIG. 3(b),
In, among the coordinate values of point P, a coefficient a smaller than 1 is applied to the component in the direction orthogonal to the half line (L), for example, 0.9.
Multiply by 0.8 to obtain data. Qualitatively, point P3 is connected to a half-line (L) parallel to a direction orthogonal to the half-line (L).
). It has been confirmed that if this process is performed, the degree of matching with the original curve is greatly increased compared to the case where it is not performed. The reason for this is that the coordinate values are taken in by sampling, so the judgment area (D)
When the previous point is located on the periphery inside the , there is a possibility that it will be far away from this judgment area (D) in the next sampling, and this will be multiplied by the coefficient a to make the judgment area (D
) is considered to be able to effectively absorb the inconvenience caused by sampling, the influence of camera shake in a direction intersecting the direction in which the trace is advanced, or noise K.

In the above embodiment, the sequential determination area (D) is shown in FIG. 3(b), but may be modified to be shown in broken lines in FIG. Instead of a circular arc, use a polygon (M),
A straight line (S) is used instead of a hyperbola. Alternatively, lengthen this straight line (S) by W to intersect the straight line M(L) with straight line #i(S'
). A combination of these may also be applied.

(Effects of the Invention) As described above, according to the present invention, data input is performed by setting a tapered sequential judgment area that can effectively avoid data input. In addition to compressing the amount of data, data input is avoided by setting a circular area around the already input point, which prevents inconvenient data input based on manual input by the operator, and faithfully reproduces the edge of the original piece. It is possible to improve the degree of

[Brief explanation of drawings]

Fig. 1 is a block explanatory diagram of an embodiment of the present invention, Fig. 2 is a data input 70-chart, and tj43 figures (a) and (b).
) is an illustrative diagram of the processing of the embodiment, FIG. 4 is a diagram showing a modification of the sequential judgment area, FIG. 5 is a diagram explaining the principle of the prior art, and FIG. 6 (a) = (b), (c) is FIG. 2 is an explanatory diagram of the drawbacks of the conventional example. DESCRIPTION OF SYMBOLS 1... Di7izer, 4C... Contour, 5... Cursor, 7... Calculation/judgment section, 9... Memory, D...
- Sequential determination area, S7...Step of determining whether the input point belongs to the sequential determination area.

Claims (4)

[Claims] (1) In a method in which a free curve is manually traced using an input device on a plane giving coordinate values, and the coordinate values of sampling points are sequentially and automatically input into memory along the free curve, the first and a second point whose data is input next to the first point, define a superior arc of a predetermined radius centered on the second point, and from the first point to the second point.
A half-line passing through the point is defined, and a sequential judgment area is defined by the above-mentioned arc and a straight line or curve that asymptotically approaches the half-line from the end point of the above-mentioned arc, and the sampling point is within this sequential judgment area. Determine whether it belongs to the point, and if it does, do not input data; if it does not, input the coordinate values of that sampling point, and use the inputted sampling point as the new second point as described above. A free curve data generation method characterized by inputting data by repeating processes. (2) The free curve according to claim (1), wherein the sampling period is set such that at least one of the sampling points subsequent to the new second data-input point is always within an arc of the predetermined radius. Data generation method. (3) The point according to claim (1) or (2), wherein the point to be determined whether or not it belongs to the sequential determination area is a virtual sampling point given by an average value of coordinate values of a plurality of sampling points. Free curve data generation method. (4) For the coordinate values of the sampling points determined not to belong to the sequential determination area, give a value obtained by multiplying the coordinate component in the direction perpendicular to the half line by a predetermined coefficient smaller than 1, and use this value. The free curve data generation method according to any one of claims (1) to (3), wherein the coordinate values to be input are used as coordinate values.