JPH0616289B2 - Area recognition method - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to a region recognition method suitable for computer graphics, pattern recognition, and shape recognition.

[Background of the Invention]

In recent years, design, manufacturing, design, art, CAD (Computer
Aided Design), CAM (Computer Aided Manufacturin
g), CAE (Computer Aided Engineering), etc.
It has become widespread to embody graphics and concepts with the aid of computers by using workstations or graphics terminals of graphics processing systems. And when embodying figures and concepts,
Generally, a line drawing called a wire frame is input to a graphic processing system, and a stylus pen, a light pen, a keyboard, a function key, etc. are used to input coordinates and input various information. This operation of pointing the coordinates allows the user of the graphics to interact smoothly with the device. In such a graphic processing system, when attribute information (for example, finite element division, surface filling, etc.) is added to the graphic input by the user, the cursor on the display is interlocked with a stylus pen to display the attribute information. The contour lines of the region to be added are designated one by one in order, or the coordinate values of the end points of the contour line are designated to indicate the data of the target region. The computer of the graphic processing system retrieves a series of graphic information from the memory based on the instructed / input data, and performs a region recognition process. The recognition processing of this area includes, for example, the determination of the end points of the contour line group, the extraction of the contour line group having the same end point, the search of the closed loop, and the recognition of the area. Then, it is judged whether or not the area is established by the area recognition processing from the series of information extracted from the memory, and the information that is established as the area is registered in the memory. By this registration, the attribute information is added.

However, the conventional area recognition processing as described above has the following problems.

(1) In general, many graphic shapes handled in the graphic processing system are complicated, and it is not easy to instruct the contour line of the area. Moreover, the number of line drawings displayed to accurately indicate the order of instruction is large, which hinders the instruction of the contour line.

(2) If the contour line is not specified correctly, it will hinder the recognition process of the computer.

(3) Even if the user intends to instruct the contour lines in order, the front and back of the area or the inside and outside are reversed due to the illusion of the user's eyes, etc., and the area requested by the user is not recognized. Things can happen.

(4) When the area is a hexahedron or the like, the method of instructing the contour line of the area itself becomes complicated, which is not practical.

As described above, in the conventional graphic processing system, there are concomitant problems such that the more complicated the graphic is, the more time it takes for the user to interact with the device and the more time the computer occupies.

On the other hand, in Japanese Examined Patent Publication No. 58-16217, a displayed image is sequentially scanned, and a portion (connecting area) where pixels different from the background are continuous in the vertical direction and the horizontal direction is sequentially stored in an external storage means. However, it is disclosed that a number of areas on an image are numbered. However, in the technique disclosed in Japanese Patent Publication No. 58-16217, a part of the area cannot be taken out, and when obtaining a specific area, the same operation as described above must be performed.

Incidentally, as other means or device for area recognition, or as a related technique for shape recognition, for example, Japanese Patent Laid-Open No.
-19882, JP-A-55-91077, JP-A-55-8
3977, JP55-83972, JP55-1
0621 publication and the like.

[Object of the Invention]

An object of the present invention is to provide an area recognition method capable of easily extracting an arbitrary area of a graphic input to a graphic processing device.

[Outline of Invention]

The present invention relates to an area recognition method for inputting and accumulating graphic information to an information processing device and indicating and displaying an arbitrary area of the inputted graphic, in which at least end points of respective line segments forming the graphic are connected. Information including conditions is input to the information processing device in advance, instruction information including at least an end point for designating an area to be recognized of the graphic is given to the information processing device, and the end point is designated by the instruction information. A line segment satisfying the connection condition of the line segment is extracted by the information processing device, and a new endpoint of the extracted line segment is recognized by the information processing device as new instruction information, and the new endpoint is included. Extract line segments that meet the connection conditions,
Repeating the above procedure until the extracted line segment forms the region specified by the instruction information, causing the information processing device itself to search for the line segment that constitutes the region, and easily extract any region of the figure. It is configured to be able to.

Example of Invention

A preferred embodiment of the area recognition method according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a block diagram showing the basic configuration of a graphic processing system for implementing the area recognition method according to the present invention. This graphic processing system includes an input unit 10, a processing unit 20, and an output unit 30. As shown in FIG. 3, the input section 10 constitutes a coordinate information pointing device and an information input device having a stylus pen 12, a tablet 14, a keyboard 16, a function key 18, and the like. The input unit 10 can instruct the already-input component of the graphic displayed on the output unit 30, for example, the data of a line segment, and input parameters.

The processing unit 20 includes a microcomputer 22, a host computer 24, an auxiliary storage device 26, etc., and the microcomputer 22 is connected to the tablet 14, the keyboard 16, and the function key 18. The processing unit 20 displays the output unit 30 having a hard copy machine, a plotter, etc. on, for example, the color display 32, fetches data from various devices constituting the input unit 10, inspects, searches, recognizes, and assists. It controls the recording in the storage device 26 and the operation control of each device constituting the graphic processing device.

Taking the finite element division of a figure as an example, the area recognition method by the figure processing system described above takes the PAD (Pro) shown in FIG.
blem Analysis Diagram).

A graphic to be divided into finite elements has already been input to the auxiliary storage device 26, and the input graphic is displayed on the color display 32. Further, the figure displayed on the color display 32 is read by the micro computer 22 and the host computer 24 as the end point coordinate position, the vector, the radius, the polygon data and the like as the figure information and accumulated in the auxiliary storage device 26. . And
The figure displayed on the color display 32 by the user
When dividing into finite elements, as shown in step 100 of FIG. 1, the keyboard 16 is used to indicate the topological dimension number of the finite element, and the reference figure corresponding to the topological dimension number is used by the stylus pen 12 and the tablet 14. To instruct (input). As input data, the topological dimension number is 1,
When either 2, or 3 is input, the process is executed based on the topological condition corresponding to each topological dimension number illustrated in FIG. When the topological dimension is 1, the topological condition for recognizing a region is a line segment. Specifically, as shown in FIG. 4 (A), a straight line, an arc, a curve,
It is a line segment such as an elliptic arc. When the topological dimension is 2, the topological condition for recognizing a region is a closed loop, and specifically, as shown in FIG. 4 (B), the surface is a triangle, a quadrangle, or the like. . Further, when the topological dimension is 3, the topological condition for recognizing the region is a solid, and specifically, as shown in FIG. 4 (C), it is a tetrahedron, pentahedron, hexahedron, etc. . The reference figures corresponding to these topological dimensions are as shown in FIG.

That is, when the topological dimension is 1, the reference figure is the end points of V ₁ and V _{2 as} shown in FIG. 5 (A), which is called the reference end point. Similarly, the reference figure in the case where the topological dimension is 2 is the line segment e ₁ ~ shown in FIG. 5 (B).
It refers to one of e ₄ and this is called the reference edge. Taking the pentahedron as an example, the reference figure when the topological dimension is 3 is
Line segment e ₁ −e ₂ − forming the loop shown in FIG. 5 (C)
e ₃ , e ₄ −e ₇ −e ₅ −e ₁ etc., and these are called reference loops.

The topological dimension number and the reference figure instructed by the user are used as instruction information for specifying the area to be recognized, and are processed by the processing unit 20.
Are read by the micro computer 22 constituting the above, and transferred to the host computer 24. As shown in step 101, the host computer 24 accesses the graphic information stored in the auxiliary storage device 26 on the basis of the transferred instruction information, and includes the instructed reference figure, and has a predetermined geometry. All regions satisfying the dynamic condition and the topological condition are searched. The geometrical condition in this embodiment is that when an endpoint of a line segment is shared by two line segments, the two line segments are considered to be connected. The line segment satisfying the target condition is a line segment in which the end point of the line segment and the certain end point are common. That is, when an example is shown in FIG. 6, assuming that the display on the color display 32 is as shown in FIG. 6 (A), the internal structure model of the graphic data is the sixth model.
As shown in FIG. 6 (B), when the endpoints indicated by a circle are shared, it is assumed that the geometric conditions are satisfied, and as shown in FIG. 6 (C), the endpoints are not shared. When there is a part, it is assumed that the geometric condition is not satisfied. The topological condition is as shown in FIG. 4 above. For example, the topological condition input to the triangular prism shown in FIG. That the closed loop is formed of a triangle or a quadrangle.

The geometrical condition and the topological condition are input as information indicating the connection condition at the end points of the line segment and the topological condition of each figure as described above.

In the search for this region, for example, when the input figure is a quadrangle in FIG. 5 (B), the topological dimension number is given as 2,
Assuming that e ₁ is input as a reference figure (reference edge), the host computer 24 determines that the line segment e ₂ and the line segment that share one of the two end points V ₁ and V ₂ of the reference figure (reference edge) e _1. Search for e ₄ . After that, the host computer 24 determines the reference figure (reference edge) e ₁ of the line segments e ₂ and e _4.
If a line segment that shares the end points V ₃ and V ₄ that are not shared with is searched for and it is confirmed that e ₃ exists as such a line segment, it is determined that the given topological condition is satisfied. e ₁
Temporarily stores the loop information −e ₂ −e ₃ −e ₄ . However, when the area that satisfies the geometrical condition and the topological condition is not detected by the above search, it is determined that "the area cannot exist", and the processing ends.

Next, the host computer 24 checks, as shown in step 102, whether or not there is a contradiction in the detected area. This inspection is performed from the following two points.

(1) Whether the center of gravity of the currently detected region matches the center of gravity of the previously detected region.

(2) Whether or not there are two or more detection areas including the designated reference graphic.

Regarding (1) of this inspection, attention is paid to the fact that the object has a center of gravity, and the center of gravity differs depending on the shape of the object. The center of gravity is obtained from the coordinate values of the detected area, and the center of gravity is different. If it is recognized as a new area, the data is discarded as an already recognized area if the centers of gravity are the same. Regarding the inspection (2), the area including the reference figure is
As a general rule, by setting the assumption that there is only one, when there are two or more detected regions,
Leave one and discard the others. In this case, in this embodiment, as shown in step 104, an interactive check method is adopted in which it is left to the user's recognition which area is necessary. However, it is also possible to allow the host computer 24 to select in the automatic mode.

If the detected area does not include the above contradiction, the area is registered with the graphic data input to the auxiliary storage device 26 as shown in step 103. continue,
The host computer 24 moves to step 105, sets the region that has just been registered as the reference region, and selects from among the line segments forming the reference region, excluding those designated as the reference graphic, for example, FIG. Any one of e ₂ , e ₃ , and e _{4 in} B) is reset as a new reference figure (reference edge), the process returns to step 101, and the area search processing is continued. By setting the newly set reference graphic so that it is always processed in the counterclockwise direction, for example, when searching for a loop, the reference graphic with the earliest search order becomes the next reference graphic. That is, in the above example, e ₂ becomes a new reference figure (reference edge).

On the other hand, in the host computer 24, all of the line segments that make up the area used as the reference area in step 105 are
It is inspected whether or not it has been selected as a reference figure (reference edge). The step 105 updates the reference area by the same means as in the case of the reference figure. That is, after all the line segments forming the reference area are selected as the reference figure, the area accumulated next to the area previously used as the reference area among the areas accumulated in the auxiliary storage device 26 is selected. It is updated to a new reference area, and it is checked whether all the line segments forming the area have been selected as the reference figure.

In this way, it is possible to recursively repeat the operations of steps 101 to 105 and recognize the area of the input graphic. Then, the processing ends, all the searched areas become the reference areas, and all the line segments in each of these reference areas are selected as the reference figures, the areas including these reference figures are searched, and the recognition processing ends. It is time (step 106). Then, steps 107 and 108
Then, all the recognized areas are automatically divided into finite elements. If a hard copy of the result is desired, the instruction is given by the function key 18.

Next, the area recognition method will be described in more detail with reference to FIG.

In the case of FIG. 7 (A), first, when the user designates the line segment A using the stylus pen 12 and the tablet 14, and the phase dimension number 2 is input using the keyboard 16, the micro computer 22 It interprets that the line segment A is instructed and transfers it together with the phase dimension number to the host computer 24, and the host computer 24 interprets the line segment A as a reference figure (reference edge) (step 10).
0). Subsequently, the host computer 24 searches for line segments B and D that start from both end points of the reference figure (reference edge) A, and the line segments B and D do not share the reference figure (reference edge) A. A line segment C sharing the other end point is searched for, and a region ABCD whose end point is indicated by a black circle is extracted (step 101). Next, the host computer 24 determines whether or not there is a contradiction in the extracted area ABCD.
The center of gravity of the CD is obtained, and it is determined whether or not there is another area sharing the reference figure (reference edge) (step 102). In this embodiment, since there is no contradiction in the area ABCD, this area is registered in the database secured in the auxiliary storage device 26 (step 103).

Next, the host computer 24 uses the area ABCD as the reference area and the line segment B as a new reference figure (reference edge) (step 105), and returns to step 101 to search for a new area. However, the area BCDA searched for the line segment B as the reference figure (reference edge) is
Since it is the same area as ABCD) and the centers of gravity are the same, it is not recognized as a new area (step 102). Therefore, the host computer 24 updates the line segment C as a new reference figure (reference edge) (step 105), searches for an area for the reference figure (reference edge) C, and outputs the area CDAB.
And the area CEFG are extracted (step 101). But,
Since the area CDAB has the same center of gravity as the area ABCD, it is not recognized as a new area and only the area CEFG is recognized as a new area and registered (step 102, step 103). After that, the host computer 24 displays the line segment D.
Is operated as a reference figure (reference edge), and the same processing as in the case of the line segment B is performed.

In this way, when the area search is completed by using all the line segments forming the reference area ABCD as reference figures (reference edges), the host computer 24 updates the reference area and sets the area CEFG as a new reference area. , This new reference area CE
The line segment E of FG is set as a new reference figure (reference edge) (step 105). However, a search for an area using the line segments E, F, and G as reference graphics (reference edges) has the same result as in the case of the line segments B and D, and a new area is not recognized. When the area search for the line segments E, F, and G is completed, the reference area to be updated and the reference figure (reference edge) are eliminated, and the area search is completed (step 106). Finally, the host computer 24 sets the number of divisions for each line segment according to the given instruction, and automatically performs finite element division for all recognition regions.

Even when the phase dimension number is 1 or 3, it can be easily understood based on the case where the phase dimension number is 2 as shown in FIG. 7 (B).
Shows an example in which the number of phase dimensions is three. In FIG. 7 (B), the reference figure (reference loop) that is first designated is a line segment that constitutes a hatched surface. Then, the first obtained reference area is a hexahedron Q composed of line segments whose end points are indicated by black circles.

A specific example of this embodiment is shown in FIG. What is shown in FIG. 8 (A) is a graphic input to the graphic processing system,
The line segment with * and # is the reference figure (reference edge). The topological dimension number at this time is two. 8 (B) to (M) show the searched areas on the display, and the numbers attached to these indicate the orientation of the areas (counterclockwise), and the number 1 is It is a reference figure (reference edge) in each area. Further, FIG. 8 (N) is the result of finite element division of the recognized area.

As described above, according to the present embodiment, the user of the graphic processing apparatus only first specifies and inputs the reference graphic and the topological dimension number, and the computer specifies and searches the area (step 101), and the area is searched. To perform the inspection (step 102),
The required area can be retrieved quickly and very efficiently. In addition, various modifications can be made to the geometrical and topological conditions of the area to be searched, and more efficient and general-purpose effects can be obtained. For example, it is possible to add a condition that if the endpoints of all the line segments of the input figure are on the same plane, the three-dimensional recognition area cannot be obtained. Furthermore, when the two-dimensional recognition area is a polygon, by adding a condition that no line element other than the outermost line should be applied in the area, performance and function improvement suitable for the purpose can be easily performed. .

Further, in the present embodiment, the user only needs to first instruct and input the reference figure and the topological dimension number, so that the erroneous operation can be extremely reduced and the correction is easy.

Although the finite element division has been described in the above embodiment, as shown in FIG. 9, it can also be applied to surface drawing or painting of the drawing. That is, it is also possible to divide the figure input into the figure processing system of FIG. 9 (A) into finite elements as shown in FIG. 9 (B), and to draw the figure as shown by the diagonal lines in FIG. 9 (C). Filling is also possible. Further, in the above-mentioned embodiment, the case where the region as shown in FIG. 4 is given as the topological condition has been described, but it is also possible to easily give a polygon or a polyhedron as the region. 10 to 12 show the results of finite element division performed by an actual graphic processing system, and FIG. 10 (A) to FIG. 12 (A) are input to the graphic processing system first. Similarly, the figure (B) is a figure obtained by performing finite element division. Comparing the time required for more detailed finite element division of the figure corresponding to FIGS. 10 to 12 with the conventional example, the following Table 1 is obtained. However, the description of FIGS. 10 to 12 in Table 1 is not the same as FIGS.
It means the "similar figure" corresponding to these.

The measurement time is defined by the area from the time of parameter input,
Or until recognition is completed. Further, the measurement time is somewhat affected by the computer used, the response time, and the environment of the computer at the time of measurement, but there is no great difference in this comparison value.

In each of these embodiments, the finite element division is performed for all the regions of the input graphic, but it is needless to say that the finite element division, the surface covering, and the filling can be performed for a part of the input graphic. In this case, not only the interactive method but also the automatic mode can be used.

〔The invention's effect〕

As described above, according to the present invention, the area required by the input graphic can be easily and quickly obtained in order to cause the graphic processing system itself to search and inspect the area.

[Brief description of drawings]

FIG. 1 is a PAD display diagram showing a procedure of an embodiment of an area recognition method according to the present invention, and FIG. 2 is a block diagram of a basic configuration of a graphic processing system for implementing the area recognition method according to the present invention.
FIG. 4A is a block diagram showing a concrete embodiment of a graphic processing system for carrying out the area recognition method according to the present invention.
(C) is explanatory drawing which shows the example of topological conditions, (A)
If the topological dimension is 1, then (B) the topological dimension is 2
, (C) is an explanatory view of the reference figure when the topological dimension is 3, and (A) is an explanatory view of the reference figure. (A) is (B) when the topological dimension is 1. ) Is a case where the topological dimension is 2, (C) is a case where the topological dimension is 3, FIG.
FIG. 7 (C) is an explanatory view showing an example of geometric conditions, FIGS. 7 (A) and 7 (B) are explanatory views showing a specific procedure of the area recognition method according to the embodiment of the present invention, and FIG. 8 (A). ) To (N) are explanatory diagrams of finite element division, which is a specific embodiment of the area recognition method according to the present invention, and FIGS. 9A to 9C are application of the area recognition method according to the embodiment of the present invention. FIG. 10 to FIG. 12 are explanatory diagrams of the finite element division and the filling, and FIGS. 10 to 12 are diagrams showing the result of the finite element division by the actual graphic processing system. 12 ... Stylus pen, 14 ... Tablet, 16 ...
… Keyboard, 18 …… function keys, 22 ……
Microcomputer, 24 ... Host computer,
26 ... Auxiliary storage device, 32 ... Color display.

Claims (4)

[Claims] 1. An area recognition method for inputting and accumulating graphic information to an information processing device and indicating and displaying an arbitrary area of the input graphic, comprising: at least an end point of each line segment constituting the graphic. Information including a connection condition is input to the information processing device in advance, instruction information including at least an end point for designating an area to be recognized of the graphic is given to the information processing device, and the end point designated by the instruction information is included. , Causing the information processing device to extract a line segment satisfying the connection condition of the line segment, causing the information processing device to recognize the new endpoint of the extracted line segment as new instruction information, and including the new endpoint. An area recognition method, wherein a line segment satisfying the connection condition is extracted, and the above procedure is repeated until the extracted line segment forms an area specified by the instruction information. . 2. The area recognition method according to claim 1, wherein the information including the connection condition at least at the end points of each line segment forming the figure is two end points of the line segment forming the figure. When shared by line segments, the geometric condition is that the two line segments are connected, and the instruction information including at least the end point that specifies the area to be recognized is at least the end point of the figure. A region recognition method, characterized in that it includes a part including "," and a topological dimension number of a region to be recognized. 3. The area recognition method according to claim 1, wherein the information including the connection condition at least at the end points of each line segment forming the figure is two end points of the line segment forming the figure. The geometric condition that the two line segments are connected when shared by line segments and the topological condition of the figure are included, and the topological condition of the figure is that the figure is It defines whether each is an independent line segment having end points, a closed figure formed by line segments sharing an end point, or a closed solid formed by a combination of the closed figures. An area recognition method characterized by, and. 4. The area recognition method according to claim 1, wherein the information including the connection condition at least at the end points of each line segment forming the graphic is two end points of the line segment forming the graphic. The geometric condition that the two line segments are connected when shared by line segments and the topological condition of the figure are included, and the topological condition of the figure is that the figure is It represents whether it is an independent line segment having end points, a closed figure formed by line segments sharing an end point, or a closed solid formed by a combination of the closed figures. An area in which the instruction information specifying at least an end point that specifies the area to be recognized includes a part of the figure including at least the end point and the topological dimension number of the area to be recognized. Recognition method .