JPH028961A - Graphic input device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention utilizes a graphic human ability that recognizes and separates graphic elements such as symbols and lines in a drawing in order to automatically input figures written in a drawing into a computer. It is related to the device.

[Prior Art] Fig. 11 shows, for example, the article ``Expected auto digitizer
It is an overall configuration diagram showing the graphic input device shown in System Application J (p, 84 to P89). In the figure, 1 is an image input means that reads a drawing and creates binary input image information, 2 is an input image storage means that stores the input image information created by the image input means 1, and 3 is this input image storage means. 2 is a thinning means for extracting the center line of the lines constituting a figure from the stored input image information and creating a thinned image; 4 is a thinning image for storing the thinned image created by this thinning means 3; storage means; 11 is a graphic element separation means for extracting graphic elements that are candidates for symbols and lines from the thinned image stored in the thinned image storage means 4; 12 is a symbol or line extracted by the graphic element separation means 11; A graphical element storage means 13 stores graphical elements that are candidates for symbols, and symbol recognition 13 inputs graphical elements that are candidate symbols from among the graphical elements stored in the graphical element storage means 12 and recognizes the type of symbol. means,
14 is a symbol storage means for storing constituent lines of the symbol recognized by the symbol recognition means 13; 15 is a symbol storage means for manually selecting graphic elements as line candidates from among the graphic elements stored in the graphic element storage means 12; Set the line type to H'! !
The line recognition means 16 is a line storage means for storing the lines recognized by the line recognition means 15.

The conventional graphic human input device is constructed as described above, and when a drawing is inputted from the image input means 1, it is stored in the input image storage means 2 as binary digital image data with thickness. The line thinning means 3 inputs this image data, thins the line until the thickness of the line becomes 1 pixel, extracts line information having the coordinate string of the center of the line and the connection relationship between the lines, and stores the image data in the thinned image storage means 4. Store in. The graphic element separation means 11 uses the line information to determine, for example, graphic elements such as a collection of short lines of a certain size or less, a small loop, a collection of small bending points, etc. as symbol candidates, and other graphic elements as line candidates. It is separated as a graphic element and stored in the graphic element storage means 12. The symbol recognition means 13 determines whether or not it is a symbol based on the separately stored information on graphical elements serving as symbol candidates, and stores lines forming the symbol in the symbol storage means 14 for those determined to be symbols. On the other hand, the line recognition means 15 also
It is determined whether or not a line candidate is a line from among the separately stored line candidates, and for those determined to be lines, the line type is determined and stored in the line storage means 16.

[Problem to be solved by the invention] The conventional graphic input device is configured as described above,
For example, if a drawing with symbols and lines touching each other as shown in Figure 2 is input, the symbols and lines will be connected and extracted as graphic elements, making it impossible to separate the symbols and lines correctly. There was a problem.

This invention was made to solve the above-mentioned problems, and its purpose is to provide a graphic input device that can correctly separate symbols and lines even when they are written in contact with each other. do.

[Means for Solving the Problems] A graphical human-powered device according to the present invention extracts a closed region having a predetermined shape as a symbol from input image information stored in an input image storage means to create closed region information. A closed region detection means, a closed region information storage means for storing the created closed region information, a thinned image stored in the thinned image storage means, and a closed region information stored in the closed region information storage means. Deletion information information storage means for creating deletion information for deleting pixels constituting a closed region serving as a symbol; deletion information storage means stored in each of the storage means; input image information stored in each storage means; Pixels constituting a closed region that becomes a symbol are deleted from the input image information based on the thinned image, closed region information, and deletion information, and an image containing graphical elements other than the closed region and a closed region that becomes the deleted symbol are created. The pixel deletion means is provided with a pixel deletion means for creating an image including the pixel deletion means, and an image storage means for storing each image created by the pixel deletion means.

[Operation] In this invention, the thinning image and closed area information created by the thinning means and the closed area detection means are used to -C1
Deletion information information storage means creates deletion information by analyzing the structure of lines constituting graphical elements, and based on the deletion information, pixel deletion means deletes pixels constituting a closed region serving as a symbol from input image information. As a result, an image is created that includes graphical elements other than closed areas, and an image that also includes closed areas that are the deleted symbols.

[Example] An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is an overall configuration diagram showing an embodiment of the present invention.
The same reference numerals are used for the same or equivalent parts as in the conventional example, and the explanation thereof will be omitted. In the figure, reference numeral 5 indicates a closed area detection means for extracting a closed area having a predetermined shape as a symbol from the input image information stored in the input image storage means 2 to create closed area information, and 6 indicates this closed area detection means. 5, a closed area information storage means 7 stores the closed area information created by the thinned image storage means 4; 8 is a deletion information storage means for creating deletion information for deleting pixels constituting a closed region; 8 is a deletion information storage means for storing deletion information in each storage means by the deletion information storage means 7; 9 is an input image storage The input image information stored in the means 2, the thinned image stored in the thinned image storage means 4, the closed region information stored in the closed region information storage means 6, and the deletion stored in the deletion information storage means 8. Based on the input image information, pixels constituting a closed region that becomes a symbol are deleted from the input image information, and an image that contains graphical elements other than the closed region and an image that contains the closed region that becomes the deleted symbol are created. 10 is an image that stores an image containing graphical elements other than closed regions created by the pixel deletion means 9, and an image containing closed regions that become deleted symbols. It is a means of storage. Note that solid line arrows in the figure indicate the flow of processing, and broken line arrows indicate information referenced by each means.

Next, the operation of the above embodiment will be explained with reference to FIGS. 2 to 10. For example, the image input means 1 reads an isolated symbol or line having a closed area of a predetermined shape, or a drawing as shown in FIG. 2 in which the aforementioned symbols and lines are in contact with each other, and 2. FIG. 3 shows a digitized and enlarged view of the touching symbols and lines. When a drawing as shown in FIG. 3 is input, the thinning means 3 extracts the center line as a dot sequence, creates a thinned image as shown in FIG. 4, and stores it in the thinned image storage means 4.

The closed region detecting means 5 extracts a closed region having a predetermined shape formed by the symbols from the input image information, and stores a mass region of white pixels inside the closed region in the closed region information storage means 6 as closed region information. That is, the area shown in FIG. 5 is extracted from FIG. 3 as closed area information. Note that a closed region having a predetermined shape is determined to be a closed region having a predetermined shape if the area and vertical and horizontal sizes of a cluster region of white pixels are within a certain range.

Next, the deletion information storage means 7 extracts the center line of the common portion caused by the contact between the symbol and the line as deletion information and stores it in the deletion information storage means 8. The point sequence shown in FIG. 6 is extracted as deletion information from the thin line image shown in FIG. 4 and the closed area information shown in FIG. 5. The common part between the symbol and the line is extracted as follows (see Figure 7). First of all,
Among the point sequence of the thinned image surrounding the closed region represented by the closed region information, we pay attention to the three branch points (point α in Fig. 7).
2 points β). Next, the angle formed near the three-junction point by the three point sequences gathered at the three-junction point is determined. This is done by calculating the angle formed by the vector connecting the 3-branch point and the 01st point from the 3-branch point to each point sequence (Oa to Of in FIG. 7).

Then, select two dot sequences in which the angle formed by the adjacent backs 1 to 1 is closest to 180°, and define the one that surrounds the closed area as the common part of the symbol and the line.

Next, the one-pixel deletion means 9 expands the block region of white pixels (A in FIG. 8) represented by the closed region information from the input image information as white pixels, thereby eliminating black pixels constituting the closed region that becomes the symbol. (B in Figure 8) is deleted. However, the number of dilations depends on the point sequence of the thinned image (see Figure 8!□
) is set at an upper limit of 12 times the number of times required to be deleted, and expansion is not performed beyond the point sequence (β2 in FIG. 8) represented by the deletion information. From the input image information shown in FIG. 3, the thinned image shown in FIG. 4, the closed area information shown in FIG. 5, and the deletion information shown in FIG. An image of a line from which the black pixels constituting the area have been deleted and an image of a symbol consisting of the deleted black pixels as shown in FIG. 10 are obtained and stored in the image storage means 10, respectively. Therefore, even if a drawing in which symbols and lines are in contact is input, the symbols and lines can be correctly separated.

[Effects of the Invention] As described above, according to the present invention, the structure of lines constituting graphic elements is analyzed, and based on this, pixels constituting symbols are deleted from input image information, and symbols, lines, etc. Since separation is performed, even if a drawing in which symbols and lines are in contact with each other is input, graphical elements can be separated correctly.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram showing a graphic input device according to an embodiment of the present invention, FIG. 2 is a diagram showing an example of a drawing subject to human input, and FIG. 3 is an example of binarized input image information. figure,
FIG. 4 is a diagram showing an example of a thinned image, FIG. 5 is a diagram showing an example of closed area information, FIG. 6 is a diagram showing common parts of symbols and lines extracted as deletion information, and FIG. 7 is a diagram showing an example of closed area information. A diagram for explaining the process of extracting a common part between a symbol and a line, FIG. 8 is a diagram for explaining the process of deleting pixels that constitute a symbol, and FIG. 9 shows an example of a line extracted by the deletion process. 10 is a diagram showing an example of a symbol deleted by the deletion process, and FIG. 1.1 is an overall configuration diagram showing a conventional graphic input device. 1 is an image input means, 2 is an input image storage means, 3 is a thinning means, 4 is a thinning image storage means, 5 is a closed area detection means,
6 is a closed area information storage means; 7 is a deletion information information storage means;
8 is a deletion information storage means, 9 is a pixel deletion means, and 10 is an image storage means. Agent: Masuo Oiwa (2 idiots) No. 1B Engraving of drawings No. 11 Lustful 10th section

Claims (1)

[Claims] A graphic input device that recognizes graphical elements such as symbols and lines in a drawing, separates them, and inputs them into a computer, which includes an image input means that reads the drawing and creates binary input image information, and the created input image information. an input image storage means for storing, and a thinning means for extracting center lines of lines constituting a figure from the stored input image information to create a thinned image;
A thinned image storage means for storing the created thinned image; and a closed region for creating closed region information by extracting a closed region having a predetermined shape as a symbol from the input image information stored in the input image storage means. A detection means, a closed region information storage means for storing the created closed region information, and a symbol based on the thinned image stored in the thinned image storage means and the closed region information stored in the closed region information storage means. Deletion information information storage means for creating deletion information for deleting pixels constituting a closed region, and input image information based on the input image information, thinned image, closed region information, and deletion information stored in each of the above storage means. Delete the pixels that make up the closed region that becomes the symbol from the image and create an image that contains graphical elements other than the closed region.
A graphic input device comprising: a pixel deletion means for creating an image including a closed region serving as a deleted symbol; and an image storage means for storing each image created by the pixel deletion means. .