JPS59191667A - Picture data converting system - Google Patents

Abstract

PURPOSE:To perform a quick and accurate correction and to reduce the software load of a computer by carrying out both the correction processing and the cut-out line extraction processing after converting a linear graphic data of raster type into the data of vector type. CONSTITUTION:The picture data read by a TV camera are all converted into the linear graphic data with a raster type maintained as it is. Then a computer 5 converts the linear graphic data into the data of vector type via an external memory 7 and writes it to an external memory 6. This data is read out of the memory 6 and displayed to a monitor 9 to perform a checkup. Then the corrected data is fed to the computer 5 by a digitizer 8. Thus a drop-out part or undesired part of a linear pattern is corrected, and the desired cut-out line data is extracted and written to the memory 7. With such cut-line data stored in the memory 7, a cut-out mask is produced by reading out the cut-out line data and supplying it to a plotter 10. As a result, the correction is possible quickly and accurately and the computer load is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for creating cutout line extraction data necessary for creating a cutout mask for printing and a block copy, and in particular, a method for creating cutout line extraction data necessary for creating a cutout mask and block copy for printing, and in particular, a method that facilitates data modification processing necessary for creating such data. Related to image data conversion method.

Conventionally, this type of printing cutout mask or block copy has been created by hand directly from the original image. Therefore, the work required a great deal of skill, and there were also problems in terms of uniformity of quality and production time.

By the way, in recent years, computer-based image processing technology has been widely adopted in the field of printing plate making, and the original image R* is first converted into image data using a television camera or scanner, and then processed. Systems that automate and simplify work have come into use.

Therefore, for the creation of cutout masks and block copies for printing, we apply such computer image processing techniques to first convert the original drawings into image data as described above, then convert this image data into full binarization, extract outlines, A method of creating a cutout mask by converting it into line figure data through arithmetic processing such as line thinning has come to be used, and it has become possible to solve the above-mentioned manual problems.

However, if the line figure data obtained through the above-mentioned image data calculation processing is used to extract the cutout lines as is, it is not possible to obtain accurate results at present, although it depends on the type and condition of the original image. I can hardly do it.

Therefore, in this computer-based printing cutout mask and block production system, correction work is performed on the line figure data obtained through the above-mentioned arithmetic processing, making it possible to extract cutout lines to the extent necessary for practical use. I even tried to correct the data. In other words, the line graphic data obtained through arithmetic processing may be missing some of the necessary cutout line information, or may include unnecessary information other than the cutout lines. Therefore, while comparing this line figure data with the original image data, errors are corrected by repairing disqualified parts or removing unnecessary parts.

This work of correcting line figure data can be done, for example, in Japanese Patent Application No. 57-4.
This is done using a device such as the one proposed in the No. 8734 application, but in the past, this line figure data was in raster format, so handling the data necessary for correction work was troublesome, and it was difficult to perform correction work. This method has disadvantages in that it requires extra time and tends to increase the load on computer software.

An object of the present invention is to eliminate the above-mentioned drawbacks of the prior art, and to provide an image that facilitates handling of data for correction processing, allows correction work to be completed in a short time, and reduces the load on computer software. ('J data conversion method is provided.

In order to achieve this object, the present invention is characterized in that the line figure data 'x4 given in raster format is converted into line figure data in vector format in advance before the actual work.

Hereinafter, all the drawings of the embodiment of the image data conversion method according to the present invention will be described.

FIG. 1 is a block diagram showing an example of a printing cutout mask creation device equipped with an embodiment of the image data conversion method according to the present invention, in which 1 is an original image, 2 is a filter, and 3 is a television (
TV) camera, 4 is a camera controller, 5 is a computer, 6. ．． 7 is an external memory, 8 is a digitizer, 9 is an image monitor, and 10 is a plotter.

This device operates according to the flowchart shown in FIG. 2. First, in step 1, one image of the original image 1 is imaged by the TV camera 3, and the image signal is transmitted vertically and horizontally by the camera controller 4, for example. 5 each
There are 12 pixels, and each pixel is converted into 8-pit digital image data and stored in the external memory 6. The filter 2 consists of four types of filters: red, green, blue, and gray, and is used to sequentially select predetermined filters according to the original image 1 and capture image data of each color.

The ufjl image data stored in the external memory 6 is so-called raster format image data consisting of 512 pixels x 512 pixels.

What is the data flow in this step 1? This is shown in Figure 3.

Next, in step 2, the image data is sequentially read from the external memory 6, and is subjected to arithmetic processing by the combiator 5 to binarize the data, remove noise, and extract outlines.

Line thinning is performed, and the data converted into line graphic data is written into the external memory 7. Therefore, the data flow at this time is as shown in FIG. Incidentally, the binarization, noise removal, contour line extraction, and line thinning at this time may be performed by, for example, the method proposed in Japanese Patent Application No. 56-2143193.

In this way, once the image data read by the TV camera 3 has been converted into line figure data while remaining in raster format, the process moves to step 3, where the computer 5 sequentially converts the line figure data from the external memory 7 to each pixel in a predetermined order. The vector data is read out, converted into vector format data by vector conversion processing, and written to the external memory 6. Therefore, the data flow at this time is as shown in Figure 5. The details of the conversion process will be described later.

In step 4, vector format line graphic data is read from the external memory 6 and displayed by the computer 5 on the monitor 9 as a line graphic image. Therefore, the line figure image displayed on the monitor 9 is examined, and the corrected data is inputted to the converter 5 using the digitizer 8 to make corrections for missing or unnecessary parts of the line figure, and the necessary cutout line data is made. Extract. Then, the results are written to the external memory 7. Therefore, the data flow at this time is as shown in FIG. Regarding this correction work, please refer to the above! It may also be carried out using the apparatus disclosed in Japanese Patent No. 57-48734.

Once the cutout line data is prepared in the external memory 7 in this way, the process proceeds to step 5, where the necessary cutout line data is read out from the external memory 7 and input to the plotter 10 to create a cutout mask.

Therefore, according to this embodiment, the line figure data is converted from raster format to vector format in step 3, prior to the process of correcting the line figure data in step 4 and the process of extracting cutout line data from the line figure data. Since the above-mentioned correction processing and extraction processing are performed based on the line figure data in vector format, data handling is easy and the load on the conbeater 5 can be reduced. Furthermore, as a result of converting from raster format to vector format, the amount of data is significantly reduced, so you can get the effect of requiring less memory capacity.

In the above embodiment, external memorandums IJ6, 7'e made of magnetic disks or the like are used to transfer data between them, but this is for the purpose of simplifying the explanation and is not actually used. In this case, one external memory may be used and data may be transferred between different memory areas of the external memory.

Next, an embodiment of the conversion process of line graphic data from raster format to vector format, that is, the vector conversion process in step 3 in FIG. 2 will be described.

FIG. 8 is a flowchart showing a simple example of vector conversion processing according to the present invention, and the processing according to this flowchart will be explained below using the conceptual diagram of FIG. 9.

Now, the raster format line figure data prepared in the external memory 7 is being moved in the X direction (horizontal direction) as shown in FIG.
12 pixels, 512 pixels in the Y direction (vertical direction),
Each pixel is represented by an XY coordinate point, and the point is IP
(X, Y), this data is IP (0, 0),
IP(1,O), ・・・・・・.

IP(511,O), IP(1,1),...
IP (X, Y), tp (x+1.'y), ...
..., IP (511, 511), and if the binarized pixel is black, then the point IP (X, Y) = 0. If it is white, take the value of IP (X, Y) = 1 for that point.

Here, we will explain the terms used in the following explanation of peku)/'.

・Continuous points, black pixel points existing within the 3×3 mask shown in FIG. 10 for the point IP(X, Y)=O.

・Connection point, a black pixel point existing within the 5×5 mask shown in FIG. 11 with respect to the point IP(X, Y)=O, which is accompanied by continuous points.

・Starting point, the point at which vector tracing starts for the line figure.

The search is carried out in the order of IP (0, O), IP (1, O), etc., and while these points are white pixels (=1), the search continues until the point becomes a black pixel. Let I P (Xo, Yo)=0 be the starting point. However, only connected points are included, and single points (described later) and isolated points (described later) are not included.

- Branch point, a point where there are two or more branch directions at the point where IP (X, Y) = 0.

However, the direction of the searched points (penalty points) is except.

・End point, point where the next connection point does not exist. Therefore, the number of branches =
It becomes a point of 0.

・Intermediate points, points between branching points, between endpoints, and points between branching points and endpoints. Therefore, the number of branches = 1°, a single point, a point that does not have continuous points but has connected points r.

- Isolated point, a point that has neither continuous points nor connected points.

・Included in deflection points, points where the direction of travel is changing, and intermediate points.

- Group, a set of points IP (X, Y) = 0 that exist continuously from the starting point IP (Xo, Yo) = O. Therefore, it consists of a loop (described later) and/or an open cluster (described later).

・Line (t) IPs (Xs, Ys) IPm (
Xs.

A straight line represented by Y, ), that is, a straight line between eccentric points.

・Niremen) (H) A collection of straight lines from a branch point (end point) to a branch point (end point).

・Loop (L) A collection of elements included in a closed curve.

・Open cluster (0) A collection of elements that are not included in a closed curve. A closed curve.

・Point (P) I P (X, , Y, )
A single point.

・Number of branches = 0... End point, single point, isolated point.

=1...Middle point (deflection point), starting point.

≧2... Branching point.

Now, returning to FIG. 8, this vector conversion process is roughly divided into vector tracking processing and loop figure detection processing, and for each group, first, seven vector tracking processes are performed,
Next, loop figure detection processing is performed.

Vector tracking process (1) Search for starting point 1 As shown in Figure 9, IP (0, 0), IP (1,
0), ...IP (511,0), IP (0
, 1)...... External memory 7 (Fig. 1, 5
The line figure data in raster format is read from the IP (
Find the point where X, Y) = O.

As mentioned above, black is 20 and white is 1.

If it reaches the point of IP(X, Y)=0, the first
Using the 3x3 mask shown in Figure 0, it is checked whether or not there are continuous points as shown in Figure 12.

If a continuous point exists, that point IP (X, Y) is regarded as the starting point, and all group numbers are counted up.

111 If it is determined that there are no continuous points, use the 5×5 mask in Figure 11 to determine whether it is a single point or not.
Check whether there are other connection points as shown in the figure. Then, in addition to IP (X, Y) in the 5 × 5 mask, there is a black point (-
If there is a ○), use a 3x3 mask to draw with that point as the center, and if there is a continuous point at this point, consider this point to be a connected point and use it as the starting point.

When the connection point is found in this way, the above point IP (
X, Y) is regarded as a single point, its position is used as point data, and it is stored in a prepared database (external memory 6). Note that in reality, the shaded area a in Figure 13
A search for a single point at an intermediate point would require searching the entire 5×5 mask, whereas searching for a single point at an intermediate point would be sufficient.

When it is determined that there are no continuous points or connected points at point IP (X, Y), this point is considered to be an isolated point and is ignored as it is caused by noise etc., and the next point IP (X + 1
．． Start searching again from Y).

(2) END judgment 1 While searching for the starting point, find the point IP (X.

Y) and find that X=511. When Y=511, it is determined that all data conversion processing has been completed, and the processing is terminated.

(3) Starting point processing A branch decision is made at the starting point, and the following processing is performed.

1 When the number of branches = 1 (Fig. 14) ・ Control data tape A/(described later) is fully prepared,
The element number and junction point (end point) number are registered there. Note that this joining point (end point) number is the total point It) number.

・Write the element data to the prepared database and count up the Niremen ID and point ID = i.

・Midpoint: IP (X, Y), set to = 1 (white). If endpoints appear here, process the endpoints.

11 When the number of branches = 2: - Write element data and point data to the database.

- Set the intermediate point = 2, and when a branch point appears, consider it as one of the branches (Figure 15).

Salary If a branch point does not appear, it is regarded as a single loop (closed loop or open loop), writes to the data table, and returns to the starting point when it is completed (Figure 16).

- When an end point appears, use the end point as the starting point and execute the same process as when the number of branches = 1 (Figure 17).

111 When the number of branches > 3: - It is determined that it is a branch point, and '%1' is stored in the database, branch table, and control data table.

・Proceed in the priority direction (described later). Branching point: Set to -1.

(4) Branch Point Judgment After proceeding by one pixel from the starting point, a branch point judgment consisting of a single point search, branch point detection, and line judgment is performed.

1. Single point search: Using the 5×5 mask shown in FIG. 11, points included in the mask are picked up as shown in FIG. 18.

- Using the 3×3 mask shown in FIG. 10 for each point picked up, check the continuity by checking the presence or absence of continuous points as shown in FIG. 18.

・ If there is no continuity and it is determined that it is a single point, it is convoluted into the database as point data, and single point = 1 (
white).

Also, if it is not a single point, leave it as is.

11 Branch point detection (using 3X3 mask @)a
Number of 0 branches = 0 - Considered as an end point.

- Refer to the branch table, return to the latest branch point, and proceed in the priority direction. Note that the determination of the priority direction will be described later.

b0 branch number = 1 ・Regarded as the midpoint.

- Perform a direction search, and if the direction is straight, proceed to the next one, and if it is curved, write the coordinates in the element coordinate play.

・Set IP=1.

C1 number of branches ≧2 - Regarded as a branch candidate point (Fig. 19), a branch point is detected from this point. Detection of branch points will be described later.

[Determination of priority direction]

The preferred direction at a branch point is determined as follows.

1. When you reach a branch point, proceed in the direction that makes the most acute angle to the input direction, as shown in Figure 1.

When returning from the end point to the latest branch point, the process proceeds in the direction 1@, which is at an acute angle to the input direction, as shown in FIG.

2. If there are two directions that make the most acute angle with respect to the direction of human power, as shown in Figure n, give priority to the right direction.

3. If the starting point is a branch point, determine the priority order as shown in Figure 5. For example, in the case of Figure U, the order is as shown.

[Detection of branching point]

A branch point is detected from branch candidate points using the following method.

1. When a branch candidate point is detected, IP it.

, then proceed to the next point IP in the direction of travel.

2. If the next point IP in the direction of travel is a white point, as shown in Figure 5, if there is no black point next in the direction of travel, then
This point IP0 is defined as a branch point.

This branch point becomes the END coordinate of the previous element, and also becomes the 8TART coordinate of the next element. Also, at this time, the element ID number is counted up and the point data is written into the database, and the process proceeds in the priority direction.

3. When the point IP, =o in the traveling direction, that is, the point IP, is black, it is further searched to see whether the next point IP, =O in the traveling direction.

3-1. If the point IP in the traveling direction is not Q, and there are branch candidate points in the traveling direction at two points Po and IP, that is, in the case of diagram n (al, Φ), the point IP
, is the branching point. However, as shown in FIGS. These points IP. or IP, with the next point as the inflection point 0 3-2. Three points in the direction of travel: IPo, IP,
When IP is a branch candidate point.

(2) At point IP1, using the 5×5 mask shown in FIG. 11, determine the branching direction relative to the traveling direction as shown in FIGS. A (a) and (b).

■ If the branch direction spans two directions ■ -1,5X
When a point exists in the direction of 1 or 7 of the 5 mask, for example in the case of Figure 4, the point IP! There are 6 branching points.

(2) When a point exists in the 2 or 6 direction of the -2,5×5 mask, for example in the case of FIG. 30, the point IP is set as the branch point.

(2) When a point exists in the 3 or 5 direction of the -3,5×5 mask, for example in the case of FIG. 31, (2) Po is set as the branch point.

■ If there are three or more branching directions, process in the same way.

The data of the li+ line determination element contains coordinates of branch points, end points, and deflection points. Therefore, the following process is performed as a line judgment.

- The directionality is checked every time one pixel point advances from the starting point consisting of a branch point or an end point, and as shown in Fig. 32 (a),
Φ), if the direction is straight as shown in (C), it is considered as an arrangement point [see section 3-1 of branch point detection]), write the coordinate data of that point to the element coordinate play, and count the coordinate count. rise. Furthermore, if a branch point or end point is detected, it is E 1'J D of the element, so it is written in the entire element data database.

- If the pixels are connected one step at a time, as shown in FIGS. 33(a) and 33(b), they are not regarded as eccentric points, but are treated as mere intermediate points. Note that since the line graphic data to be processed has already been thinned, it will not result in 1ffll connections as shown in FIG. 33(C).

(5) Detection of group end The 0 branch table is checked, and if there is no data there, it is determined that the group is at the end and all vector ruins processing is performed.

Here, the tables created and used in the above vector tracking process will be explained.

1. Branch table When a new branch point is detected during vector tracking, write the point (point) ID number of the branch point, the coordinates of the branch point (9 branch directions (absolute direction)) all in the branch table. The branching direction (absolute direction) at this time is determined as shown in FIG.

・　Total pressure in the direction away from the branch point, negative in the opposite direction.

Therefore, the direction toward the branch point is marked with -.

・Set the searched direction data to O.

Therefore, in the example of FIG. 35, the table becomes as shown in FIG. 36. In addition, IP is not only the coordinates of the branch point, but also the branch point key (-10) and point ID number, so the actual format of the branch table is the third one.
It will look like Figure 7.

- When all branching directions become zero, return to the previous branching point.

- When reaching the end point, refer to the latest branch information and trace the vector in the direction that makes the most acute angle with respect to the negative direction.

・　Used to judge the end of the group.

・The branch table is replaced sequentially.

- In the case of the branch point -8 TART point, there is no negative direction, so in this case, the first direction traveled is regarded as the negative direction, and a - mark is added to the data.

2. Control table - This is a table in which element IIJ numbers representing the extracted elements and point numbers representing their junctions are written in the order of extraction. Here, a junction point refers to a starting point and an ending point of an element, and is a branching point or an end point.

- It is used to create a junction matrix in loop figure detection processing.

- It has not only element information but also group information.

Therefore, in vector tracking as shown in Fig. 38, the third
The table is as shown in Figure 9.

Loop figure detection (1) Creation of a junction matrix A junction matrix represents the relationship between elements in a group and their junction points (branch points and end points). ) number, and point from top to bottom in the column direction) Take the L and D numbers as shown in Figure 40, and mark the intersections of elements and points that are related to each other with 1, and the intersections that are not with 0. Each was given.

Therefore, this junction matrix B can be created from the control data table, and if this table is like the first rule on the left in Figure 41, then the junction matrix B generated from it will be as shown on the right in Figure 41. It will look like the figure.

(2) Horizontal line system Here, all elements not included in the loop are detected.

- Examine the junction matrix fJi, and check whether there is a row in which %fl of one row in the row direction is 1, as shown in FIG. 42. In the example of FIG. 42, the sum of the first row is tc to l.

- If there is a row in which O is set to 1, the elements in the column that contain the summation set to 1 in that row are judged to be elements that are not included in the loop, that is, elements with endpoints, Data representing this is registered in the database as open cluster data.

・ Mask the column containing element 1 of the row whose sum is 1, and as a result, if a row appears where the oro in the row direction is 1,
Repeat the same thing.

(3) Column search Here, both single loop and multiple loop detection are performed. Note that the single loop here refers to a loop consisting of one element (Hn) and therefore one junction Pm as shown in Fig. 43, and the multiple loop refers to a loop consisting of multiple elements as shown in Fig. 44. E5. E6 or E6. E7 and therefore multiple junctions P4, P5'! This refers to the loop that exists in .

1 Among the columns of the junction matrix B, the sum in the column direction is the 45th
If there is a column that is 1 as shown in the figure, the elements in that column are considered to form a single loop, and are written into the loop table as IJm = (En). In the example of Figure 45, I
Jm = (E2).

・Mask that column.

11 Elements in columns with the same matrix element configuration are the fourth
As shown in FIG. 5, Dk = (Ei, Ej), which are regarded as elements that mutually form multiple loops, are stored in the loop table.

- If there is more than one, they are considered to form a loop in order. Therefore, in the example of FIG. 45, 1), = (E5.
E6), D, = (E6. E7).

(4) Detection of a generative tree A generative tree is a graph that connects all vertices in a graph consisting of L
It is a collection of minimum elements that do not include ooP, and is used here for minimum loop detection.

1 Regarding the junction matrix that has been checked in (3) above,
All elements in each row are deleted so that the sum of each row in the row direction is 1.

11 At this time, among the elements that are 1 in each column, the one with the most element number a is deleted first, and the number of elements that are 1 in each column is 2 or O. Erase it in such a way that it looks like this.

111 When calculating the above-mentioned row sums, perform the calculations sequentially from the top row to the bottom row, excluding elements that form pairs in each column.

'This will be explained with reference to FIG.

Through the processing up to (3) above, the number of elements that are 1 in each column of the junction matrix is always 0 or 2.

So now, A = Element to be deleted ■: Element pile to be retained: If we compare the elements that are left because they form a pair with the element of ■ and are not included in the calculation of the sum of rows, we get the following in Figure 46. In the first row, In in the first and seventh columns is 1, so the upper me l! According to the rule, the 7th column's support meeting is eliminated (=
O)L, which makes the sum of the first row 1.

In the second row, the elements in columns 1, 3, and 5 are 1
It has become. Therefore, element 1 in the first column is paired with the element in the first row, so according to rule 111 above, this element is excluded when calculating the sum, and according to rule 11 above, it is not in the third column. to delete the element in the 5th column and make the row sum 1.

This is repeated until the bottom row.

IV Let it be the Nilementra-produced water in the column with two elements of 1 remaining. Therefore, in the example of FIG. 46, El.

E3. hi4. E6 becomes produced water.

■ Also record the number of the deleted element.

(5) Searching for the minimum loop For line drawings that include multiple loops, it is not necessary to register all the data for all the loops.If you import only the data for the minimum unit of loop, the data for all the loops will be played back. I can do something.

Therefore, the minimum loop is searched and only that data is stored, thereby making it possible to obtain all the loop data with the least amount of data.

Regarding the junction matrix B that has completed the generated water detection process in step 1(4), the process is performed in the order of the deleted elements.

11 Find the shortest route from one (Pm) to the other (Pn) of the elements that were 1 among the deleted elements.

- At this time, you can freely move between elements that are set to 1 in the same row or tag.

In reality, the junction matrix B'tl - elements Pm to Pn
Pick up all the routes to (elements connected by 1),
The loop with the least number of elements is the minimum loop.

This will be explained below using the example shown in FIG.

Now, if the line figure is as shown in FIG. 47(a), its junction matrix B will be as shown in the figure.

Therefore, element E3. [4, column E6 is deleted,
As shown in Figure 47 (C), the produced water is composed of El, E2, E
It is 5.

Then, the detection of the minimum loop becomes as follows.

■ The 1-pupil path from the junction point P1 of E3 (Erased) to P3 is 1) P1 → P2 → P3 2) P1 → P2 → P4 → P3 3) P1 → P4 → P3 4) P1 → P4 → P2 → There are four types: P3. Among these, the minimum loop is 1) or 3). Then, when there are multiple minimum loops, the minimum loop is determined based on the following priority order.

*Priority order when there are multiple minimum loops <a>
Priority is given to those containing more generated water.

Φ) When the number of produced waters becomes the same, check to see if there is any overlap in other loops and give priority to the one that overlaps.

(C)　Kilometers away from the one with the closest generated water number.

Therefore, in this case, according to the rule (a) above, P1
→P2→P3 is assumed to be the minimum loop.

■ In the case of E4 (Erased Niremen) 1) P1 → P3 → P4 2) P1 → P2 → P4 Therefore, according to rule (b) or (C), P1 → P3 → P
4 is the minimum loop.

■ Case of deleted element E6 1) P2→P1→P4 2) P2→P3→P4 Therefore, according to rule (a), P2→P3→P4 is the minimum loop.

111 The minimum loop is Dm = (Di l,
El, Ek --). Therefore, in the case of ■, ■, ■ above, D,'' (El, E2. E3) ・・・・・・・・・
...■D, -(E3. E4. E5) ...
...... ■D, = (E2. E5. E6) ・
・・・・・・・・・・・・■.
(,”・(also) Loop table creation・Detected minimum loop data Dn = (Et, E
lEk,...) is written to a loop table such as a meeting number chart.

- Similarly, oven clusters are also written in the open cluster table as shown in Figure 49.

Next, writing of a data file in one embodiment of this vector conversion processing will be described.

In this embodiment, the following six types of data are written to the file.

(1) Element l) When an element is detected by performing data vector tracking processing, perform the following processing.

1. Fill in the element data ID number.

2. Starting point) Enter the ID number.

3. Write starting point data.

4. Count up the coordinate count.

Filling in the coordinate array.

5.Enter the end point ID number.

6. End point data included.

7. Write element data.

(2) Point (p) data Point data is the start point, end point, and single point of an element. Among these, the start point and end point of an element are written to a file when the element is detected, and the single point is written to a file at any time.

(3) Junction matrix data (When the END of jitOUP is detected, a junction matrix is created from the control data table, and the matrix data is written to a file.

(4) Loop (L) Data Write loop data to a file by referring to the loop table created by loop detection.

(5) Open cluster (0) Refer to the open cluster table created by data loop detection and write open cluster data to a file.

(6) Count data When all line figures have been searched, write the total number of points, total number of elements, total number of groups, total number of loops, and total number of open clusters to a file.

Therefore, if this data is written into a file in a flowchart, it will be as shown in FIG.

According to the vector conversion process according to this embodiment, line figure data is converted into a vector format and stored as minimum loop data, so the amount of data can be sufficiently reduced without losing necessary information. can.

In addition, according to this embodiment of vector conversion processing, data in vector format is hierarchically divided into points, elements, oven clusters, single loops, multiple loops, groups, etc., so correction processing (see Fig. 2) is performed. The handling of data in step 4) is simplified, and correction processing can be performed easily and accurately.

As explained above, according to the present invention, after converting raster format line figure data to vector format data, correction processing and cutout line extraction processing are performed.
What are the drawbacks of the prior art, which makes it easier to handle data for corrective processing? However, it is possible to provide an entire image data conversion system that allows correction work to be performed in a short time and accurately, and that requires less load on the converter.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of a printing cutout mask creation device to which an embodiment of the image data conversion method according to the present invention is applied, and FIG. 2 is a flowchart for explaining its operation.
Figures 3, 4, and 5. 6 and 7 are explanatory diagrams of the operation shown in terms of data flow, FIG. 8 is a flowchart illustrating the operation of an embodiment of vector conversion processing in the present invention, FIG. 9 is a conceptual diagram of the operation, and FIG. Figures 11 and 11 are explanatory diagrams of search masks, Figures 12 to 25, and Figures 26 (a) and (b).
, Figure n (a), (1)), Figure 4 (a), (b)
, Figures 29 to 31, Figures 32 (a), (b)
, (C), Figure 33 (a), (b). (C), Figures 34 to 36 are explanatory diagrams of vector tracking operations, Figure 37 is a conceptual diagram of a branch table, and Figure North is an example of vector tracking operations? 39 is an explanatory diagram showing an example of a branch table, FIG. 40 is a conceptual diagram of a junction matrix, FIG. 41 is an explanatory diagram showing the relationship between a junction matrix and a control table, and FIG. 42 is an explanatory diagram showing an example of a branch table. Figures 43, 44, and 45 are illustrations of row search, and Figures 4 and 4 are illustrations of column search.
Figure 6 is an explanatory diagram of detection of produced water, Figure 47 (a), (b
), (C) is an explanatory diagram of the minimum loop search, FIG. 48 is an explanatory diagram of the loop table, FIG. 49 is an explanatory diagram of the open cluster table, and FIG. 50 is a flowchart showing data writing. ■・・・Original picture, 2...Filter, 3...
Music television camera, 4...Camera controller, 5...Music/computer, 6, 7...External memory, 8...Music digitizer, 9...Picture 1 monitor , 10... Plotter. Representative Patent Attorney Junjibu Take (and 1 other person) Figure 31 O Geunehana, Figure 34 Figure 35 Figure 36 Figure 38 Figure 39 Figure 40 Figure 41 Figure 42 - Kanshin 7th Scroll 〔Figure 48Figure 49; Figure ■

Claims (1)

[Claims] (1) In an image data conversion method in which raster format drawing data read from an original image is converted into line figure data, and this line figure data is modified to obtain data for extracting cutout lines, the line figure data is converted into line figure data. % that the data for extracting the cutout line is obtained by converting it to line figure data in vector format and modifying the converted data.
Image data conversion method. (2. The image data conversion method according to claim 1, characterized in that the line graphic data converted to the vector format data is generated in a hierarchical manner.