JPH0463435B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] A character string combination device applied to an automatic drawing input device that inputs design information into a CAD system, which combines line segments such as center lines and dimension lines drawn in drawings. By understanding the meaning of the line segments, determining whether to integrate (combining) or not, and combining the divided character strings as necessary, the plotter can This is to obtain the correct string data as intended.

[Industrial Application Field] The present invention is a drawing that reads a drawing drawn by a designer as an image, automatically extracts design information from the image data using pattern recognition technology, and inputs the design information into a CAD system. Regarding a character string integration device used in an automatic input device or a drawing recognition device, in more detail, after classifying character data and line segment data written on a design drawing, the classified individual character data is classified as necessary. The present invention relates to a character string integration device that integrates character string data to obtain character string data intended by a plotter.

In a drawing drawn by a designer, in addition to various line segments, there are various explanatory texts and character strings that indicate the size of a shape, the type of a shape, etc. using character information. and,
Among these character strings, there are character strings that are independent across line segments, and character strings that cross line segments and mean one sentence.

Therefore, for character strings that exist across a line segment, if it is a single sentence, the character strings that exist on both sides of the line segment are combined, and one character string data is created. and 2 are required.

[Prior Art] FIG. 4 is a conceptual diagram showing the principle of a conventional method for extracting character strings from drawings.

As shown in Figure 4 A, the character string ABC is shown in the drawing.
and character string DE, measure the distance l between the two character strings, judge whether this is greater than a predetermined threshold T _l , and if l > T _l , the distance l between the two character strings ABC and DE is determined. are independent character strings, and if l<T _l , they are combined into one character string.
I am trying to recognize it as ABCDE.

On the other hand, as shown in Figure 4B, for example, if the line segment L crosses between the character string ABC and the character string DE, both character strings will be divided and become two independent character strings ABC and DE. recognized as existing.

[Problems to be Solved by the Invention] Such a conventional method for extracting a character string from a drawing is such that if there is a line segment that crosses the character string, the line segment is The two separated character strings were divided to obtain data as two character strings. Therefore, for character strings that need to be interpreted as continuous even if a line segment crosses them, there is a problem in that it is not possible to extract the correct character string data intended by the plotter.

The present invention has been made in view of these points, and it integrates (combines) character strings that are individually divided by line segments by understanding the meaning of the line segments that cross the character strings. ), and to provide a character string combining device that determines whether or not the character string data is correct and obtains the correct character string data intended by the plotter.

[Means for Solving the Problems] FIG. 1 is a block diagram of the principle of the character string combining device of the present invention. In the figure, 1 is an element classification means for classifying line segments and character strings from the vectorized image data applied thereto, and 2 is a storage for storing character string information classified by the element classification means 1. Section 3 is a line segment identification means that inputs the line segment information classified by the element classification means 1 and identifies line segments based on the definition information given therein. A line segment that crosses the character string is identified.

4 is a line segment type determination means that determines the type of line segment that crosses a character string, and it receives character string information from the storage unit 2;
Line segment information that crosses the character string is input from the line segment identification means 3. Reference numeral 5 denotes a string joining means for integrating (combining) two character strings divided by a line segment based on the judgment result of the line segment type judgment means 4, from which integrated string data is obtained. That's what I do.

[Operation] The line segment type determination means 4 determines, from the character string information and the line segment information, what attributes the type of line segment crossing the character string has, for example, does it mean a center line? , or determines whether it means a dimension extension line, and depending on the attributes of each line segment, for example, if the line segment means a center line, convert the two divided character strings into characters. The columns are combined by column combining means 5.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 2 is a block diagram showing an embodiment of the present invention. Components that are the same as in FIG. 1 are designated by the same reference numerals. Here, we will exemplify a case in which the present invention is applied to an automatic drawing input device that inputs design information into a CAD system. In the figure, 6 is a drawing, which includes
Graphic information consisting of line segments, characters, character strings, etc. that is input into the CAD system is drawn. 7 is drawing 6
For example, a CCD that optically reads graphic information is used as a reading means for reading the graphic information drawn on the image. Reference numeral 8 denotes an image file for temporarily storing the binarized image information from the reading means 7. 9
is a vectorization means that thins the image information read out from the image file 8 and creates connections between each pixel,
It is a vector signal that takes into consideration the direction, etc. Conventional signal processing is implemented using known techniques.

The element classification means 1 for line segments and character strings includes an element classification means 11 for line segments and characters, and a character string extraction means 12 for extracting character strings from among those classified as characters. .

13 is a character recognition processing means, and a character string combination means 5
The character string data that has been combined and the character string data that is deemed not to be combined are read from the character string storage section 2, and each character data is used as data of an actual character pattern. Reference numeral 14 denotes a synthesizing means, which inputs the character string pattern data from the character recognition processing means 13 and the line segment data from the line segment identification processing means 3, synthesizes both data, and outputs the synthesized data to the CAD system. ing.

The operation of the device configured as described above will be explained as follows. FIG. 3 is an explanatory diagram showing an example of the operation. Here, two character strings 20, 2 divided by a line segment L as shown in FIG.
1 exists.

The line segment and character string element classification means 1 classifies line segments and characters from the image data vectorized by the vectorization means 9, and extracts character strings from the classified characters. Then, this character string data is stored in the storage section 2. Here, when extracting character strings, if the distance between each character is longer than a certain constant, the character strings are different, and if it is shorter, the character strings are the same.
If there is a line segment that crosses between characters, it is determined whether or not to combine the character strings on both sides of the line segment, according to the determination result by the line segment type determining means 4.

That is, the line segment identification means 3 and the line segment type determination means 4
refers to the information defined in advance for each attribute of the line segment that will cross the character string, determines what type of line segment that crosses the character string has, and determines the type of line segment, for example. If it is determined that it means the center line, the string combining means sends an instruction to combine the two character strings 20 and 21 on both sides of the line segment into one character string as shown in FIG. 3A. Output to 5. For example, if it is determined that the line segment type means a dimension extension line, the two character strings 20 and 21 on both sides of the line segment are left as they are (no combination), as shown in Figure 3 (b). do. The character string combining means 5 receives an instruction from the line segment type determining means 4 and combines two character strings into one character string.

The synthesizing means 14 combines character string data that has been combined into one character string, character string data that has not been combined, character string data that is not crossed by line segments, and line segment data from the line segment identification means 3. Combine and output data in a format that can be read by a CAD system.

[Effects of the Invention] As explained in detail above, according to the present invention,
For character strings divided by line segments, it understands the meaning of the line segments, determines whether to combine the divided character strings according to the attributes of the line segments, and obtains character string data. This allows the correct character string data intended by the drafter to be output to the CAD system.

[Brief explanation of the drawing]

FIG. 1 is a principle block diagram of the character string combination device of the present invention, FIG. 2 is a configuration block diagram showing an embodiment of the present invention, FIG. 3 is an explanatory diagram showing an example of operation, and FIG.
The figure is a conceptual diagram of the principle of a conventional method for extracting character strings from drawings. In FIGS. 1 and 2, 1 is an element classification means;
2 is a storage section, 3 is a line segment identification means, 4 is a line segment type determination means, and 5 is a character string combination means.

Claims (1)

[Scope of Claims] 1. A character string combination device applied to a recognition device for drawings containing character information, comprising: element classification means 1 for classifying elements of line segments and character strings from vectorized image data; The storage unit 2 stores the character string information classified by the element classification means, and the line segment information classified by the element classification means 1 is input, and a line segment that crosses two character strings having the same direction is extracted. A line segment identification means 3 identifies the line segment based on predetermined definition information, inputs the character string information from the storage section 2 and the line segment information from the line segment identification means 3, and identifies the type of line segment that crosses the character string. and a character string combining means 5 that combines two character strings divided by the line segment according to the determination result of the line segment type determining means 4. A string joining device characterized by: