JPH0546761A - Method for recognizing kind of line for automatic drawing input device - Google Patents

Abstract

PURPOSE:To improve the tracking of a segment and the performance of line kind decision based upon the tracking by accurately continuing the tracking of the segment and efficiently grouping vectors as one line in a line kind recognizing method for an automatic drawing input device to be used for a CAD. CONSTITUTION:In a vector sorting process 1, a threshold is set up in the 'length' information of a core vector to be recognized and a vector shorter than the threshold is set up as a 'point', while sorting a longer vector as a 'line'. When the 'point' is selected as a connection candidate in a segment tracking process 2, a segment is supposed between the 'point' and a tracking reference point to be a reference vector and the segment is tracked as the directional vector of the 'point'. In a line kind deciding process 3, the segment is recognized in accordance with the tracking.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a line type recognition method for a drawing automatic input device used for CAD or the like, and more particularly to a line type recognition method with improved line segment tracking and line type determination performance.

[0002]

2. Description of the Related Art In a drawing automatic input device, a binary image input from an image scanner is thinned or cored to thereby use a set of vectors. According to this method, the data can be compressed, the memory capacity is small, and
There is an advantage that processing can be performed at high speed by software.

In the image data skeletonization process, first, an image scanner or the like is used to read a binary image of white and black as shown in FIG. 6 (a), and a boundary between white and black is extracted from the binary image. To create a contour pixel row. From this contour pixel row,
Further, a contour vector 61 shown in FIG. 9B is generated, two vectors forming a pair are searched from the contour vector data, and a core vector 62 corresponding to the center line thereof is generated. The generation of the skeleton vector 62 is interrupted at the position where the pairing of the two vectors is lost.
Remember as. When all the contour vectors 61 have been searched, a region search for the core line vector 62 to be connected is performed. If there are a plurality of interruption points within a predetermined area, the position of each interruption point is changed to the position shown in FIG. The connection point 64 as shown is calculated, and each interruption point 63 is integrated based on the connection point 64. When a plurality of interruption points are connected, the respective core wire sections are integrated into one core wire section. When the cored data is obtained in this way,
The drawing automatic input device recognizes broken lines and chain lines based on them. Here, the line type is recognized by detecting a vector that may be a broken line candidate from the set of vectors by examining the length of the vector, the direction of the vector, the distance between the vectors, the relationship between angles, etc. , The line segment is traced, and the set of vectors regarded as one line in the traced result is used to calculate the length distribution of each vector.
First, the broken line and the chain line are distinguished, and then the line types of the one-dot chain line and the two-dot chain line are determined. Such a line type recognition method is described in detail in Japanese Patent Application No. 2-291379 by the present applicant.

[0004]

However, in the above-mentioned conventional partial tracking method, first, a vector serving as a tracking reference is taken out, an arbitrary area is set from the end of the reference vector, and a vector serving as a link candidate from that area is set. Has been elected. As the selection method, emphasis is placed on “distance”, “angle”, “positional relationship” between the reference vector and the candidate vector, and “direction” of each vector. Therefore, in the case of a short line segment that appears in a chain line, if the length of the line segment is shorter than a certain length, the direction of the core line vector tends to be unstable when the core line is generated. For example, as shown in FIG. 7A, in the alternate long and short dash line composed of relatively long line segments, a short vector 70 and a long vector 72 are shown as shown in FIG. 7B.
Although the angle formed by and is small and the directionality of each vector is stable, as shown in FIG.
In the alternate long and short dash line composed of 4,76, the directions of the long skeleton vector 75 and the short skeleton vector 76 are too different from each other to be a connection candidate, and a vector that should originally be a connection candidate is not recognized as a part of a line segment. After tracing up to the vectors 73, 74, and 75, the line segment tracing is interrupted at the vector 76, the efficiency of segmenting the line segment is significantly reduced, and the determination of the line type is greatly affected. The present invention has been devised in view of such problems, and it is possible to accurately continue tracing line segments, efficiently perform vector grouping as one line, and perform line segment tracing and line types resulting therefrom. It is intended to provide a line type recognition method that improves the performance of judgment.

[0005]

[Means for Solving the Problems] Means for solving the above-mentioned problems in the present invention are as follows: a relation of "direction" and "length" of a vector from a set of vectors, "distance" and "angle" between the vectors. The line segment is traced by and the result is 1
In the line type recognition method of the drawing automatic input device that distinguishes the line type from the set of vectors regarded as the line segments of the book, a threshold is set in the "length" information of the core line vector to be recognized, and the threshold value is set. Vectors shorter than this are classified as "points", vectors longer than the threshold are classified as "lines", and a line type recognition method that traces line segments from those core line vectors is used.
When a "point" is selected as a connection candidate, it is preferable to trace the line segment by using the portion assumed between the reference point and the "point" as the direction vector of the "point". is there.

[0006]

According to the present invention, a threshold is set for "length" of the information of "length" and "direction" of a core vector to be recognized, and a "point" is set for a vector shorter than the threshold.
The line segment is traced by setting the flag. In tracing the line segment, a line segment is assumed between the "point" and the trace reference point of the reference vector, and the line segment is traced with this line segment as the direction vector of the "point".

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a process chart of an embodiment of the present invention. In the figure, 1 is a vector classification process, 2 is a line segment tracking process, and 3 is a line type determination process. The vector classifying step 1 is a step particularly added in the present invention, and the line segment tracking step 2 is modified based on it.

FIG. 2 is a flow chart showing an example of the above vector classification step 1. As shown in the figure, in this process, all the vectors to be traced by the line segment are taken out one by one, their lengths are calculated, and compared with a preset threshold value, the shorter one is As a "point" vector, a long one is given as a "line" vector with "length" information,
Take over to the next line segment tracking step 2. FIG. 3 is a flowchart showing an example of the line segment tracking step 2. In the figure, when the flow is started, a "line" vector serving as a reference for tracking is taken out from the vector data classified in the step 1, and a vector existing in the search area is searched. As shown in FIG. 4A, this reference vector V
If the "line" vectors v ₁ and v ₂ are found in the search area S1 of ₁ , the candidate vectors are set as candidate vectors, and the candidate vector v ₁ has the distance L1 from the reference vector V1 and the angle θ1 as information. The candidate vector v ₂ is selected with the distance L2 and the angle θ2 as information. When the "point" vector is found, as shown in FIG. 4 (b), the end of the reference vector V1 as the track reference point Vp _1, a candidate from the tracking reference point Vp ₁ "point" P1 the distance L1 to the length Satoshi, a line segment having an angle θ1 of the reference vector V1 assumed between Vp ₁ -P1 as the direction vector,
Continue to follow on. For example, even if the information of "length" and "point" is added to the conventional example of FIG. 7, if it is only a short line segment and the direction is greatly different, it will not be a connection candidate and tracking will be interrupted. However, in the present embodiment, as shown in FIG.
The line type can be recognized by continuing the tracking with 4-vector 75-point 76-vector 77.

[0009]

As described above, according to the present invention, it is possible to accurately continue tracing line segments, efficiently perform vector grouping as one line, and trace line segments and the resulting lines. A line type identification method that improves the performance of type determination can be provided.

[Brief description of drawings]

FIG. 1 is a process drawing of an embodiment of the present invention.

FIG. 2 is a flowchart of vector classification.

FIG. 3 is a flowchart for tracing a line segment.

FIG. 4 is an explanatory diagram of line segment tracking.

FIG. 5 is an explanatory diagram of line segment tracking.

FIG. 6 is an explanatory diagram of tracing a line segment.

FIG. 7 is an explanatory diagram of a conventional example.

[Explanation of symbols]

1 ... Vector classification process, 2 ... Line segment tracking process, 3 ... Line type determination process, 61, 71 ... Contour vector, 62, 7
0, 72, 73, 74, 75, 76 ... Core line vector, 6
3 ... break point, 64 ... connection point, V ₁ ... reference vectors, v _1,
v ₂ ... Candidate vector, points 74, 76 ... Point vector.

Claims (2)

[Claims] 1. A "direction" of a vector from a set of vectors
And the line length is traced according to the relationship between the "length", the "distance" and the "angle" between the vectors, and the line type is discriminated from the set of vectors regarded as one line segment based on the result. In the line type recognition method of the device, a threshold is set for the "length" information of the core vector to be recognized, a vector shorter than the threshold is a "point", and a vector longer than the threshold is a "line". A line type recognition method for an automatic drawing input device, characterized in that the line segment is traced from those core line vectors. 2. The line type recognition method according to claim 1, wherein when a "point" is selected as a connection candidate, a portion assumed between the interruption point of the reference vector and the "point" is a "point". A line type recognition method for an automatic drawing input device, characterized by tracing a line segment as a direction vector.