JP3537956B2 - Drawing inspection method and drawing inspection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drawing inspection system for a drawing created by a CAD system, and more particularly to a drawing inspection system for detecting an excess or deficiency in dimension entry and reducing the number of inspection steps.

[0002]

2. Description of the Related Art A designer writes the shapes of parts constituting a product, creates a drawing in which dimensions are written, and transfers the drawing to a manufacturing department. If the dimensions are duplicated, the manufacturer does not know which dimension should be processed first. Of course, if there is a shortage in the dimensions, the position and size of the part cannot be known, and the part cannot be manufactured. For this reason, the designer visually inspects the created drawing and checks whether the dimension is written properly or not. However, visual inspection is easy to overlook and takes time.

[0003] In order to solve this problem, there has been proposed an apparatus for inspecting a lack of dimension entry by a computer. For example, Japanese Unexamined Patent Publication No. Hei 6-75620 (Reference 1) discloses a method for detecting shortage of dimension entry for a parallel or vertical straight line written in the drawing.

On the other hand, there is known a CAD system for generating a shape based on a geometric constraint relationship between figures. This C
In the AD system, as described in Japanese Patent Publication No. 7-89382 (Cited Example 2), a dimension is treated as one of the constraint relationships, and a method of detecting a competing geometric constraint relationship has been proposed. I have.

[0005]

In the drawing inspection of the cited reference 1 in the above-mentioned prior art, it is not possible to inspect the overlapping entry of dimensions. Regarding the shortage of the dimensions, it is not possible to inspect a portion where a straight line in an oblique direction or an angular dimension is entered. In the method of detecting conflicting geometric constraints in Reference 2,
If the geometric constraint conditions are insufficient, that is, if there is insufficient dimension entry, the detection process stops halfway, and duplicate dimension entries cannot be correctly detected.

In general, a drawing represents a part shape using a plurality of projections or cross-sections, and a projection in which the length and position defined by dimensions are clear is selected, and there is no duplication or lack of entry throughout the drawing. Dimensions as shown. However, in the methods of the cited examples 1 and 2, there is no consideration of excess or deficiency of dimensions between a plurality of projection views.

Further, when the shortage of dimension entry is detected, it is still unclear which figure in the drawing has a dimension entry omission in the methods of the cited examples 1 and 2.

A first object of the present invention is to provide a drawing inspection system which detects shortage or duplication of dimensions entered in a drawing created by a CAD system and clearly indicates a detection location to a user in view of the above-mentioned problems of the prior art. It is to provide a method and an apparatus.

A second object of the present invention is to provide a drawing inspection method and apparatus for detecting a lack of dimension entry in a drawing and adding the missing dimension to the drawing.

A third object of the present invention is to provide a drawing inspection method and apparatus capable of reliably detecting overlap of dimension entry even when there is a shortage in dimension entry.

[0011] A fourth object of the present invention is to provide a drawing inspection method and apparatus capable of detecting shortage and duplication of dimension entry in the whole drawing when the drawing is composed of a plurality of projection views or cross-sectional views. It is in.

Further objects of the present invention will become clear through the following description.

[0013]

The first object is to provide a CA.
Data generated as CAD graphic data by the D system
In a drawing inspection for detecting a shortage or an excess or a shortage of dimensions of a drawing composed of a number of graphic elements ,
Graphic elements that can determine the relative positional relationship with each other
The graphic elements are grouped, and a different table is displayed for each group.
Achieved by displaying with the display attribute . The figure required
Elementary grouping constrains geometric elements geometrically
Stores the constraint data, and based on this constraint data
It is characterized by being performed .

The purpose of the home 2 is to store graphic data for each graphic element in the drawing and constraint condition data for geometrically constraining the graphic elements, and to store the relative condition data based on the constraint condition data. This is achieved by grouping the graphic data into graphic elements whose positional relationship can be determined, adding predetermined dimensions between predetermined graphic elements of different groups, and eliminating shortage of dimensions.

The predetermined dimensions are determined by searching for a set of graphic elements whose dimensions can be defined between different groups and determining the predetermined graphic elements, and determining the predetermined dimensions from the graphic data of the set of graphic elements. Is calculated and added. Further, the added predetermined dimension is displayed while being distinguished from other dimensions to facilitate user confirmation. Alternatively, the added predetermined dimension is displayed with the same display attribute as the figure defining the dimension to make it easier to see.

The first and second objects are to store graphic data for each graphic element on the drawing and constraint condition data for geometrically constraining the graphic elements. Sequentially, the dependency relationship between other graphic elements can be determined, or a predetermined geometric inference including the constraint data is performed on the premise, and when the geometric inference is interrupted due to lack of the constraint data, This is achieved by assuming insufficient constraint condition data and repeatedly executing the above-mentioned geometric inference for all graphic elements until the above-mentioned geometric inference is established.

[0017] When there is determined constraint condition data between a graphic element whose dependency cannot be determined and a graphic element that has already been determined, the missing constraint condition data is most likely to be assigned to the graphic element that cannot be determined. It is characterized in that a determined point graphic element which is close is determined and is assumed from the distance between the two.

The third object is to store graphic data for each graphic element on the drawing and constraint condition data including dimensional data for geometrically constraining the graphic elements, and sequentially store each of the graphic elements. In the case where a dependent relationship can be determined between other graphic elements or a predetermined geometric inference including the constraint data is performed on a premise thereof, and when the geometric inference is interrupted due to lack of the constraint data, Assuming the lacking constraint data, it is repeatedly executed until the geometric inference is established for all the graphic elements, a flag is set on the dimension data used for executing the geometric inference, and the dimension data without the flag is duplicated. Display as an entry, or
This is achieved by tracing back the execution process of the geometric inference starting from the graphic elements at both ends of the dimension without the flag, and displaying the dimension data in the process as candidates for deletion.

Each time the geometric inference is interrupted, a group number is added to the set of graphic elements for which the inference has been established and / or the assumed constraint data while updating the group number. The element or the assumed constraint data is displayed.

A fourth object of the present invention is to inspect a drawing composed of a plurality of projections created by a CAD system and to check for a lack of dimensioning of the drawing, the figure data for each graphic element of each projection, The constraint condition data for geometrically constraining the space between the projection maps and the projection transformation data for specifying the projection method of each projection diagram are stored, and the constraint condition data of each projection diagram is converted, integrated, and integrated using the projection transformation data. This is achieved by grouping the graphic data into graphic elements that can determine a relative relationship from each other from the restricted condition data, and displaying graphic elements on each projected view with different display attributes for each group. You.

The constraint condition data described above includes dimension data for explicitly constraining the size of the graphic data, the distance or angle between the graphic data, and the positional relationship between the graphic elements such as orthogonal, contact, and parallel. The data includes unrestricted constraint data to be constrained, and the latter is extracted according to a predetermined graphic rule.

It is needless to say that the respective constitutions of the present invention corresponding to the first to fourth objects can be appropriately combined as described in the claims to achieve a plurality of objects at the same time. No.

In the structure of the present invention, the figure grouping means groups together figures whose relative positions can be determined from the constraint conditions. The figures whose relative positions can be determined are a set of figures whose positions can be determined by determining the positions of some figures.

A relative position cannot be determined between figures belonging to different groups. Writing such figures on the drawing is prohibited by drafting rules, and it is necessary to add dimensions to eliminate different groups.

Here, figures belonging to different groups A and B are a and b, respectively. If a dimension representing the positional relationship between the figures a and b can be added, the relative position between the figures a and b can be determined. Since the position of the figure a ′ other than the figure a in the group A can be determined relative to the figure a, the figure b ′ other than the figure b in the group B and the figure a ′ can be determined.
Can be determined, and shortage of dimension entry can be eliminated. However, adding dimensions between shapes belonging to the same group will not only eliminate the shortage of dimensioning, but also duplicate dimensions between shapes whose relative positions are already determined,
This causes inconvenience in the manufacturing process.

Therefore, the grouped figure data is displayed as a figure with different display attributes for each group. According to this, since the difference between the graphic groups can be distinguished at a glance on the screen, the user can add an insufficient dimension without causing duplicate entry.

Further, the dimension adding means for adding dimensions between figures belonging to different groups can automatically add dimensions that can eliminate shortage of dimensions without duplicating entry. It can be greatly reduced.

Further, a figure group from which the relative positional relationship can be determined can be obtained from the dimension data and the implicit constraint data by means for creating dependency data on the position of the graphic data.

If the dependency data cannot be created from existing dimension data or constraint data, the dimension data for the figure is insufficient. Therefore, when the constraint condition is added, the dependent relationship data can be continuously created. However, for a figure that depends on a figure to which a constraint has been added, the dependency cannot be determined without the added constraint, so determine the relative position to the figure for which the dependency was determined before adding the constraint. Can not.

Therefore, the graphic data before the addition and the graphic data for which the dependency data has been created after the addition are classified and grouped, and displayed with different display attributes for each group. By adding the constraint conditions in this manner, all the dependencies can be determined even in a drawing having insufficient dimension entry. Note that the dimension not used for determining the subordination relationship is a duplicate entry, and the dimension of the duplicate entry can be detected based on whether or not it is used.

Further, even if a dimension not used to determine the dependency is used instead of the dimension used to determine the dependency of the figure constrained by the dimension, the dependency of the figure is determined. Can be. Therefore, if one of these dimensions is left and the other is deleted, duplicate entry of dimensions can be eliminated. In addition, by displaying a plurality of dimensions related to the overlapping entry on the screen by default, the user can select only the optimal dimension and eliminate the duplication.

Further, in the case of drawing data composed of a plurality of projection views, dimension conversion means for converting dimension data of one projection view into another projection view converts dimension data on one projection view into another projection view. Can be projected on top. As a result, even when the relative position cannot be determined by data of only one projection view, the relative position can be determined by mutual dimension data between the projection views.

In this way, when dimensions are entered in a plurality of projections, it is possible to indicate to the user where the dimensions are insufficient from the entire projection relationship. In addition, overlapping dimension entries between projection views can be detected by projecting and converting dimension data of another projection view into a projection view.

[0034]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In addition, the same components are denoted by the same reference symbols throughout the drawings.

[Embodiment 1] FIG. 1 shows the configuration of a drawing inspection apparatus according to Embodiment 1. The drawing inspection apparatus includes a drawing data storage means 1 having graphic data 2 and dimension data 3, a graphic data grouping means 4 for grouping graphic data whose relative positions can be determined together, and a graphic data 2 for each group. It comprises graphic data group display means 5 represented by different display attributes, dimension addition means 7 for adding missing dimensions due to grouping, and display means 6 for displaying drawings and dimensions in a predetermined format.

FIG. 2 shows a hardware configuration for realizing the drawing inspection apparatus. This hardware configuration is common to each embodiment. That is, a display device 201 that outputs drawings and inspection results, a central processing unit 202 that controls drawing inspection processing and input / output devices, a keyboard that receives user commands and instructions, an input device 203 such as a pointing device, and drawing data. Main storage device 20 for storing inspection results, inspection processing programs, work data being processed, and the like.
4, an external input / output device 205 for inputting / outputting drawing data and inspection results between an auxiliary storage device and a network,
This is realized by a computer system including:

Next, each component of the drawing inspection apparatus will be described in detail. The drawing data storage means 1 provided in the main storage device 205 receives the graphic data 2 and the dimension data created by the CAD program operating on the same computer as the graphic inspection device according to the instruction from the input device 203 by the user. You. Alternatively, a CA operating on another computer
It may be created by a D program and stored in the drawing data storage unit 1 via the external input / output device 205.

The drawing data storage means 1 stores line segments, circles, arcs, and the like in the graphic data 2, and stores the size of the graphic data 2 and the dimensions indicating the distance and angle between the graphic data in the dimension data 3. .

FIG. 3 shows an example of the drawing. The graphic data 2 on this drawing has a data structure as shown in FIG. 4, and the dimension data 3 has a data structure as shown in FIG. That is, the graphic data 2 stores the type of graphic (for example, a straight line) and geometric information (coordinates) for each graphic ID. Also,
The dimension data 3 stores a dimension type (for example, distance between straight lines) and a target graphic element (graphic ID) for each dimension ID.

The graphic data grouping means 4 groups the graphic data into graphic data whose relative positional relationship can be determined from the positional relationship between the dimension data and the graphic data. A graphic data group whose relative positions can be determined is a set of graphic data from which the absolute positions of some of the graphic data can be determined if the absolute positions of some of the graphic data are determined.

For example, the straight line 16 and the straight line 17 in the drawing of FIG. If the position of the straight line 16 on the drawing is given, since the distance between two parallel straight lines is determined, the straight line 17
Can be determined on the drawing. Conversely, given the position of the straight line 17 on the drawing, the position of the straight line 16 on the drawing can be determined. Such a straight line 16 and a straight line 17 can be said to be graphic data from which a relative position can be determined.

The graphic data grouping means 4 groups the straight line 16 and the straight line 17 into one group. Also, straight line 16 and straight line 18
Is also specified by distance 24 to be 80. The graphic data grouping means 4 groups the straight line 16 and the straight line 18 into one group. In addition, straight line 16 and straight line 17, straight line 16 and straight line
Since the relative position of the line 18 is determined, the relative position of the line 17 and the line 18 can be determined. Figure data grouping means 4
Makes the straight line 16, the straight line 17, and the straight line 18 into one group from these relationships. Similarly, straight line 16, straight line 17, straight line 18,
The straight line 19 and the straight line 20 constitute one group.

On the other hand, for the vertical straight line, since the dimension 21 is given, the straight line 11 and the straight line 12 are put into one group. Also, given the dimensions 22, the straight line 13 and the straight line
Make 14 into one group. However, no dimension is drawn between the straight line 11 or the straight line 12 and the straight line 13. Therefore, even if the absolute position of the straight line 11 on the drawing is determined, the position of the straight line 13 is not determined. That is, the relative position between the straight line 11 and the straight line 13 cannot be determined.

Therefore, the graphic data grouping means 4
Classifies a figure whose relative position cannot be determined into another group. That is, the straight line 13 is a different group from the group of the straight lines 11 and 12 and the group of the straight lines 13 and 14. Similarly, the straight line 15 is another group.

Finally, the graphic data grouping means 4 selects only one reference point from the drawing, and if an angle between straight lines passing through the reference point is given, the group to which each straight line belongs is combined into one group. Group. The reference points are selected to determine the relative positions of the lines that are independent of each other. If the reference point is not determined, the relative positions of the vertical straight line and the horizontal straight line cannot be determined in the drawing of FIG.

For example, if the intersection of the straight line 11 and the straight line 16 is set as a reference point, the straight line 11 and the straight line 16 are orthogonal. In the drafting method, the entry of the angle dimension between the orthogonal straight lines is omitted. Therefore, the processing is performed assuming that the angle is instructed even in the case of orthogonality. If an angle with a straight line passing through one point is given by the geometric drawing theorem, the position of the other straight line can be determined, so that these two straight lines can determine the relative positions to each other. Therefore, the straight lines 11, 12, 16, 17, 18, 19, 20
Into one group.

FIG. 6 shows the result of grouping the drawing data of FIG. 3 by the graphic data grouping means 4. As the grouping result, the ID of the graphic element is stored for each group number.

The graphic data group display means 5 sets different display attributes for each of the grouped graphic data groups and displays them on the display means 6 (display 201). FIG. 7 shows a display example of grouped graphic data. Here, different line widths and line types are set for each of the three groups in FIG. In this way, by changing the display attribute for each group of graphic data, the graphics whose relative positions are determined are displayed with the same display attribute, and can be easily distinguished from other graphic groups.

The dimension adding means 7 writes dimensions between figures having different display attributes on the screen. For example, in accordance with an instruction from a user with a mouse or the like, a dimension line to be additionally entered is drawn on the screen, and stored in the dimension data storage unit 3 together with a dimension value input from the keyboard. An example of automatic entry of a dimension to be added will be described later.

In the screen display as shown in FIG. 7, it can be seen at a glance that the straight lines 12 and 13 having different line types belong to other groups. Therefore, if the user adds a distance dimension between the straight line 12 and the straight line 13, the straight line 12 and the straight line 13 are parallel to each other, so that their relative positions are determined. As a result, the group 2 of the straight lines 13 and 14 is absorbed by the group 1 and one of the dimensional shortages can be eliminated.

By the way, when dimensions are written between figures of the same line type, the added dimensions become excessive. For example, if a dimension is added between the straight line 17 and the straight line 18, the relative positions of the straight line 17 and the straight line 18 are already determined by the dimension 23 and the dimension 24, and therefore, the added dimensions overlap. But,
According to the present embodiment, the difference of the group to which each figure belongs can be seen at a glance, so that the user can easily grasp the entry point where the dimension is truly insufficient, and can prevent duplicate entry.

The graphic data group display means 5 may set different display colors for each group. Also, GKS
The graphic data grouped by the graphic data grouping means 4 is assigned to different display data groups for each group by using a display data grouping function of a graphic library such as PHIGS and OPEN-GL. May be. In the display data group, it is possible to change the display color of the graphic elements in the group all at once, or to highlight or blink the graphic elements collectively when picked.

The display means 6 is preferably a display device such as a CRT having the above display data group function. However, the graphic data group display means 5 of this embodiment
By changing the display attribute for each group, a hard copy output device such as a printer or a plotter can be used instead.

Further, when a figure belonging to the group containing the largest number of figure elements is displayed with the same display attributes as the figure in the original drawing, if the number of missing elements is smaller than the number of elements in the drawing as shown in FIG. Since the number of graphics whose display attributes change is smaller than that of the original drawing, graphic elements whose display attributes have changed stand out. Therefore, the user can easily find a portion with insufficient dimensions.

[Embodiment 2] Next, Embodiment 2 of the present invention.
Will be described. This embodiment is characterized by a dimension adding unit for automatically adding a missing dimension.

FIG. 8 shows a processing procedure of the dimension adding means for automatically filling in missing dimensions. The dimension adding means 7 automatically adds the minimum required dimension data between the different figures of the figure data group and displays them on the display means 6.

First, a group having the largest number of graphic elements is set as a variable G1, and graphic elements belonging to the group G1 are stored in a list L1 (S31). In the example of FIG. 6, group 1 has seven graphic elements, group 2 has two graphic elements, and group 3 has one graphic element. Therefore, the group 1 is set as the variable G1, and the graphic elements 11, 12, 16, 17, 18, 19, and 20 are set as the list variable L1.

Next, for a group other than the group G1, a variable G2 is set each time the loop is repeated, and the following loop processing is executed (S32). Group 2 is set to the variable G2 in the first iteration and group 3 in the next iteration.

Next, figures belonging to the group G2 are set in the list L2. For example, the graphic element 13 and the graphic element 14 are set in the list L2. Then, each graphic element of the list L1 is sequentially set to the variable E1, and the following processing is repeated (S34).

In the example of FIG. 6, the graphic element 11 is first displayed, and the graphic element 12. . . And the graphic elements stored in the list L1 are sequentially set to the variable E1 every time the loop is performed. Similarly, each of the graphic elements in the list L2 is set as a variable E2 for each loop, and the following processing is repeated (S3).
5).

Next, it is determined whether a dimension can be set between graphic elements stored in the variables E1 and E2 (S
36). If it can be set, a dimension D according to the geometric information of the graphic elements of the variables E1 and E2 is generated, and a list L3
(S37).

When the dimension D is generated, the group G is added to the dimension D.
The group number 2 is set (S38). For example, when the graphic element 11 is stored in the variable E1 and the graphic element 13 is stored in the variable E2, since these are parallel straight lines, a distance dimension 41 between the parallel straight lines is generated. The distance dimension 41 is calculated from the graphic geometric data 303 of the graphic elements 11 and 13.

FIG. 9 shows the data structure of the list L3.
When the dimension 41 is generated, since the group 2 is set in the variable G2, the group number 2 is set to the group number of the dimension 41 as shown in the list L3.

When the relative positions of all the graphic elements of group 2 are determined, the flow returns to S32. Here, the group G2
When one dimension is generated, the relative position with respect to the graphic element of the group G1 is determined, so that additional dimension addition is not necessary for the graphic element of G2. Returning to S32, the next group number 3 is set as the variable G2, and the process is repeated.

Similarly, in the group number 3, S36
It can be seen that the distance dimension can be set between the graphic element 11 and the graphic element 15 in FIG.
In step 8, the group number 3 is set to the dimension 42. After S32 has been repeated for all the groups, the additionally generated dimensions stored in the list L3 are displayed on the display means 6 with display attributes according to the group numbers.

If the graphic elements stored in the variables E1 and E2 are non-parallel straight lines, if the two straight lines are not orthogonal from the drawing and the angular dimension is not defined,
Define an angular dimension between two straight lines. The angle dimension is calculated from the graphic geometric data 303 of both graphic elements.

For example, the graphic element 16 and the variable E
When the graphic element 13 is stored in 2, both straight lines are orthogonal to each other on the drawing, so that it is not necessary to define the angular dimension. If any of the graphic elements stored in the variables E1 and E2 is a circle or an arc, the process branches to S37 to define a distance dimension such as a center or a radius.

FIG. 10 shows a display example of a drawing including dimensions added to the drawing of FIG. The added dimension is displayed in accordance with the display attribute of each group displayed by the graphic data group display means 5. For example, since the group number is 2 for the dimension 41 of the list L3, the graphic element 1 of the group number 2
Display with the same display attributes as 3 and 14. Also, since the dimension 42 is the group number 3, it is displayed with the same display attribute as the graphic element 15.

As described above, by adding one dimension between the graphic elements of different groups, the shortage of the dimensions can be automatically eliminated, and the man-hour for correcting the drawing can be greatly reduced. In addition, by changing the display attribute for each group number of the figure element to which the dimension has been added and superimposing it on the figure group before the addition, it is easy to determine which part of the added dimension has been added to eliminate the lack of dimensions. I understand. This allows the user to easily confirm the automatically added dimensions.

[Embodiment 3] Next, Embodiment 3 of the present invention.
Will be described. When the constraint relationship is insufficient, the drawing inspection apparatus according to the present embodiment continues the grouping on the assumption that the constraint relationship is insufficient, and the dimension is insufficient (assumed value) on the drawing.
And duplicates are detected.

FIG. 11 is a functional block diagram of the drawing inspection apparatus according to the present embodiment. Drawing storage means 1 equivalent to FIG.
In addition to the above, there are a constraint relationship extraction unit 101, a constraint network storage unit 102, a dependency relationship determination unit 103, a dimension shortage / assuming detection unit 104, and an overlap dimension detection unit 105. Although not shown, it has graphic data group display means 5 and display means 6 equivalent to those in FIG. Further, the hardware configuration of the present embodiment is realized by the computer system of FIG.

The constraint relation extracting means 101 reads the graphic data 2 and the dimension data 3 from the drawing data storage means 1, creates a constraint network, and creates a constraint network.
Write to 2. The constraint network uses graphic data as nodes, and converts explicit constraint conditions based on distance and angle dimension data and implicit constraint conditions based on the positional relationship of graphic elements such as contact and intersection into undirected arcs with no directionality. .

The dependency determining means 103 executes the inference based on the geometric inference rule to determine the dependency of the graphic element nodes. For the principle of determining the position of a shape element by the geometric inference rule, see, for example, Aldefeld B., “Variation of geometries based on the geometric inference method.
based on a geometric-reasoning method); Computer-
Aided Design; Volume 20, Number 3, 1988, PP.117
~ 126 "are well known.

The geometric inference rule is divided into a "premise part" and a "consequence part", like the inference rule of the expert system or the like. If a graphic element and a constraint relationship satisfying the condition of the premise part exist in the database, the position of the graphic element of the consequent part can be determined. In other words, it indicates that the graphic element included in the consequent part depends on the graphic element included in the premise part of the established geometric inference rule.

For example, <premise> geometric inference that there are two fixed-point positions, there is a straight line whose position has a distance constraint between these points and the fixed position, and <consequence> that straight line can be determined. If the rule holds, we know that the line is dependent on two points. In other words, the relative position can be determined between the graphic elements having the subordinate relation, indicating that they belong to the same group in the first and second embodiments. In this case, dependency data indicating that the straight line element is dependent on the two point elements is added to the constraint network storage means 102 as described later.

The geometric inference of the dependency relationship determining means 103 is performed by selecting some of the graphic element nodes of the constraint network stored in the constraint network storage means 102 as reference positions, and performing geometric inference with the starting point as a starting point. Create the dependency data of the constraint network storage means 10
Store in 2. This is repeated until the geometric inference rule cannot be applied.

In the case of a constraint network using graphic element data as nodes and constraint relation data as arcs, the dependency relation data is expressed by the constraint target nodes and the direction of the constraint arc that are in a master-slave relationship. In addition, a used flag is set for the constraint arc used for creating the dependency relationship data. This flag is used for detecting an overlap dimension described later.

The dimension shortage detection / assuming means 104 is activated when the geometric inference rule cannot be applied before the dependency of all the graphic element nodes stored in the constraint network storage means 102 is determined. Remaining graphic element nodes for which the dependency relationship cannot be determined indicates that the constraint relationship is insufficient. Thus, a graphic element having an insufficient constraint relationship is specified, the constraint relationship is assumed, and the geometric inference by the dependency determining means 103 can be continued.

In this case, one graphic element for which the dependent relationship cannot be determined is selected, and it is checked what constraint relationship can be satisfied to continue the geometric inference rule. The missing relation is calculated and assumed from the geometric data of the graphic, and the dependency data of the graphic element node is determined.

For example, when there are two fixed-point points and there is a straight line whose position is not fixed and has a distance constraint between the two points, a distance constraint is imposed between another point and the straight line. If so, the above-described geometric inference rules can be applied.
Therefore, such a missing distance constraint is assumed from the geometric data 303 of the graphic, and dependency data indicating that the straight line depends on two points is created.

When new dependency data is added to the constraint network storage means 102, a new geometric inference rule can be applied to the graphic element node.
Perform step 103 again.

When determining the dependent relationship by assuming the constraint relationship, there is no lack of dimension entry for the group of graphic elements for which the dependent relationship could be determined before the assumption. However, since the graphic elements determined after the assumption assume a constraint relationship, that is, a dimension, there is a dimension for which entry is insufficient. Moreover, this assumption may be repeated. It is not appropriate to treat them as one group.

Accordingly, a dependency relationship group number is prepared as data for distinguishing a graphic element for which the dependency is determined before assuming the constraint relationship and a graphic element for which the dependency is determined after the assumption, and the degree of the constraint is assumed. Is added to and stored, and when the geometric inference rule is applied to determine the dependency, the value is added to the graphic data.

As a result, it can be understood that the graphic data having the dependency relationship group number of 0 has no insufficient dimension entry, and the graphic data having at least one has the insufficient dimension. Further, the maximum value of the dependency relationship group number indicates the number of missing dimension entries in all the drawing data stored in the drawing data storage unit 1.

The overlap size detecting means 105 searches the constraint network stored in the constraint network storage means 102 and refers to the constraint arc for which the used flag is not set. The constraints that were not used to determine the dependencies are the dependencies of the figure, ie, the dimensions of the duplicate entries that are not needed to determine the position of the figure. However, unused constraint condition data is not always subject to deletion. From the drafting rule, it may be appropriate to delete the used constraint conditions and leave unused data.

Therefore, the graphic element nodes at both ends of the unused constraint arc are referred to, and the nodes are determined by the dependency determination means 10.
While tracing the process determined in 3 in the reverse direction, the constraint condition data related to the duplicate entry is collected. Then, an overlapping group number is assigned to the constraint condition for each set, that is, the dimension data, as data indicating an overlapping entry.

Hereinafter, each component in the present embodiment will be described in detail. FIG. 12 shows a processing flow of the constraint relationship extracting means, and FIGS.

FIG. 13 shows an example of a drawing used for describing the present embodiment. In the figure, reference numerals 601 to 608 denote linear graphic elements, and reference numerals 609 and 610 denote point graphic elements. Also,
Reference numerals 611 to 617 are dimension data.

FIG. 14 shows the data structure of the drawing of FIG. The graphic data 2 shown in FIG. 3A includes a graphic identifier 301 for distinguishing individual graphic elements, a graphic type 302 representing the type of a graphic such as a point, a line segment, a circle, an arc, and the like. Coordinate values, coordinate values of start and end points for straight lines, coordinate values and radii of center points for circles, coordinate values of center points, radii for arcs, and geometric data 303 for storing coordinates of start and end points, displayed on display device 201 The display color 304 includes a dependent relationship group number 305 representing a group of graphic elements whose relative positions are determined by the dependent relationship. The dependent relationship group number 305 is written by the dependent relationship determining means 103 and the dimension shortage detecting / assuming means 104.

The dimension data 3 shown in FIG. 9B includes a dimension identifier 306 for identifying each dimension, a distance, an angle, a radius,
Dimension type 307 representing the type of dimension such as diameter, constrained graphic element data 308 for storing graphic identifiers 301 of one or two graphic elements whose dimensions are constrained, and duplication when it is determined in the drawing inspection that they are duplicates The overlap flag 309 is set by the dimension detection means 105, and includes an overlap group number 310 representing a group of dimensions having an overlapping relationship with each other.

FIGS. 15 and 16 show a graphic element node table and a constraint arc table of the constraint network generated by the constraint relationship extracting means and the like.

The graphic element node table 400 includes a node identifier 401, a node type 402 representing the type of graphic element such as a point, a straight line, a circle, and the like. Corresponding graphic identifier 403 for storing 301, node geometric data 404 for storing geometric information of the graphic represented by the node, determined flag 405 indicating that the relative position from the reference position can be determined by executing geometric inference, Dependency relationship determination order number 406 indicating whether the dependency was determined by the second inference rule application,
0 when determined from the reference element without lack of constraint
When determined by the addition of the constraint relation, it is composed of a dependent relation group number 407 indicating the number of additions.

The node identifier 401, the node type 402, the corresponding graphic identifier 403, and the node geometric data 404 are determined by the constraint extracting unit 101, and the determined flag 405, the dependent relationship determining order number 406, and the dependent group number 407 are determined by the dependent relationship determining unit.
By 103, they are respectively written into the constrained network storage means 102.

The constraint arc table 410 contains an arc identifier 4
11, arc type indicating the type of constraint such as distance, angle, radius, etc.
412, a constraint target node 413 indicating the first and second graphic element nodes in a constraint relationship, if the dimension corresponding to the constraint arc is in the drawing data storage unit 1, corresponding dimension data 414 storing the dimension identifier 306 thereof; A used flag 415 indicating whether or not the constrained arc has been used for geometric inference. In the geometric inference, 0 is set when a second node is determined from the first node of the node 413 to be constrained, and 1 is set in the opposite case. It is composed of dependent direction data 416 indicating the direction of the dependent relationship.

The arc identifier 411, the arc type 412, the constraint target node 413, and the corresponding dimension data 414 are the constraint relation extracting means.
According to 101, the used flag 415 and the dependent direction data 416 are written into the restricted network storage means 102 by the dependent relationship determining means 103.

The processing procedure of the constraint relation extracting means 101 will be described below with reference to the processing flow of FIG. First, data of a constraint network is generated based on a graphic element. For this reason, as shown in FIG. 14A, for each of the graphic elements stored as the graphic data 2,
301 (601 to 610) is set as the variable E1, and the following processing is repeated (S501).

In S502, referring to the graphic data 2,
A graphic element node table 400 corresponding to the graphic identifier 301 set in the variable E1 is created and set in the variable N1. For example, when the identifier “601” is set in the variable E1, the graphic type 302 and the graphic geometric data 303 of the identifier 601 of the row indicated by the reference numeral 320 are referenced from the graphic data, and the identifier L1 of the node table 400 is referred to. Create a row,
L1 is substituted for the variable N1, and the graphic type 302, graphic identifier 301, and graphic geometric data 303 of the graphic data are copied in the columns of the node type 402, the corresponding graphic identifier 403, and the node geometric data 404.

In S503, the graphic element node N1 created in S502 and the graphic element node table 40
It is checked whether there is an intersection or a contact point with the graphic element node other than N1 registered in 0. If there is an intersection or a contact point, a graphic element node having an intersection or a contact point with N1 is set to N2, and the coordinate value of the intersection or the contact point is set to a variable C1,
The process branches to S504.

For example, when the identifier "601" is set in the variable E1, there is no graphic element node other than N1 in the graphic element node table 400, so that the flow does not branch to S504. In the next iteration, the identifier “60” is assigned to the variable E1.
When “2” is set, the node table 400 has graphic element nodes with identifiers “L1” and “L2”, and the variable N1
Is set with the identifier “L2”. L1 and L2 are
Since the intersection occurs at the coordinates (0, 0), L1 is set to the variable N2, (0, 0) is set to C1, and the process branches to S504.

In S504, it is determined whether or not a point element node having the same coordinates as C1 has already been registered in the graphic element node table 400 for each of the plurality of intersections C1 obtained. If not registered, the flow branches to S505. If registered, the identifier is substituted for the variable N3, and the process proceeds to S507. In the graphic element node table 400, the identifier "L1",
If there is only the graphic element of “L2”, there is no point element node at (0, 0), so the flow branches to S505.

In S505, the graphic element node table 40
0, a new graphic element node is added with an identifier “P1”, “point” is set as the node type 402, “no corresponding element” is set as the corresponding figure 403, and the coordinate value (0, 0) is set. The identifier “P1” of the created graphic element node is set in the variable N3, and the process proceeds to S506.

In S506, the constraint arc table 410
Is added to the constraint arc with the identifier “701” indicated by reference numeral 423, and “distance” is added to the arc type 412, and the constraint target node 4
The L1 of the line graphic element N2 is set to the first node (# 1) 13 and the P1 of the point graphic element N3 is set to the second node (# 2). This indicates that the point graphic element N3 is constrained on the line graphic element N2. Note that “no corresponding dimension” is set in the corresponding dimension data 414 of the constraint arc “701” newly created in S506.

In S507, for example, a constraint arc “702” indicated by reference numeral 424 is newly added to the constraint arc table 410, “Distance” is entered in the field of the arc type 412, and the node N1 created in S502 is added to the constraint target node 413. The identifier "L
2 "and the identifier" L3 "of the point graphic node N3. The corresponding dimension data 414 sets “no corresponding dimension”.

In S508, the node N1 in S502
It is determined whether or not there is a contact relationship (tangent line or tangent circle) between the graphic element node and the existing graphic element node in the graphic element node table 400. If there is a contact relationship, the identifier of the graphic element node is set in the variable N4, and the flow branches to S509.

Although the contact relationship exists between the circle or arc figure and the surrounding graphic elements, in the example of the drawing of FIG. 13, there is no circle or arc, so that the flow does not branch to S509. However, by adding this processing, it is possible to cope with graphic processing including rounding of corners.

In step S509, a constrained arc is newly created in the constrained arc table 410, “distance” is set in the arc type 412, and the node identifier 4 of the variable N1 is set in the constrained node 413.
01 and the node identifier 401 of the variable N4 are set. "No corresponding dimension" is set in the corresponding dimension data 414. For example, when registering a line segment, if the line segment touches a circle already registered in the constraint network, a constraint arc of arc type “distance” is generated between the line segment and the circle.

In this manner, the processing of S501 to S509 is repeated, intermediate generation of the constraint network based on the graphic data 2 is performed, and then S5 based on the dimension data 3 is performed.
Then, the process proceeds to the generation of the constraint network from S11.

In S511, each dimension stored as the dimension data 3 is sequentially substituted into the variable D1 (S51).
1), the processing of S512 to S515 is repeated. FIG.
In the example of the dimension data 3 in (b), the identifiers 611 to 617 are sequentially assigned to the variable D1.

In S512, the column of the dimension type 307 of the dimension data indicated by the variable D1 is referred to.
The process branches to S514 for an angular dimension and to S515 for a diameter dimension added to a circle or an arc.

In S513, a new constraint arc is created in the constraint arc table 410, and "distance" is set in the field of the arc type 412. Further, the value of the node identifier 401 that matches with the constraint target graphic element 308 of the dimension data corresponding to the dimension identifier of the variable D1 and the value of the corresponding graphic identifier 403 of the graphic element node table 400 is set in the constraint target node 413. . Finally, the value of the variable D1 is set in the corresponding dimension data 414 of the constraint arc table 410.

For example, when the variable D1 is the dimension identifier "611"
In the case of, first, a cell of the constraint arc "425" is created, and "distance" is set to the arc type 412. Next, the dimension identifier “6
Referring to the value of the constraint target graphic element data 308 of “11”,
Its values are "602" and "608". On the other hand, the graphic identifier 40
The values of the node identifier 401 in the graphic element node table 400 in which 3 is “602” and “608” are “L2” and “L8”, respectively. Therefore, the “L2” and “L”
8 "is set to the constrained arc" 425 ". And the variable D1
Is set to the corresponding dimension data 414.

In S514, “angle” is set to the arc type 412 of the new constrained arc, and the same processing as in S513 is performed. Also in S515, the arc type 41 of the new constraint arc
“Diameter” is set to 2 and the same processing as in S513 is performed. As a result, 611 to 617 of the dimension data in FIG.
After the processes of 513 to S515, constraint arcs of 719 to 725 are created on the constraint arc table 410.

Finally, in the loop of S521, usually, a right-angled relationship (referred to as an implicit right-angled relationship) between graphic elements not explicitly shown in the drawing is extracted and added to the constraint network. In S521, the identifier 401 is sequentially set to the variable Q for all the graphic element nodes whose node type 402 is “point” in the graphic element node table 400.

In S522, the data passes through the point Q and the node type 40
Among the graphic element nodes in which “2” is a “straight line”, all combinations of nodes that are orthogonal to each other are obtained, and the combination
To be stored. For example, when the identifier P1 is set in the variable Q, straight lines passing through P1 are L1 and L2.
A combination of {(L1, L2)} is set in the list R1.

In S523, each combination of the list R1 is set to the variable NP1 in order, and S524 and the subsequent steps are repeated. For example, when the list R1 is {(L1, L2)},
First, (L1, L2) is set to the variable NP1 and S524
Execute Then, the loop of S523 is repeated until there is no unprocessed combination in the list R1.

In S524, for a set of straight-line nodes of the variable NP1, a node group connected by dimension to each node in the list, that is, a parallel straight-line group is obtained and set to the list variables R2 and R3, respectively. For example, when (L1, L2) is set in the list NP1, L1
Is L3. Therefore, (L1, L3) is set in the list variable R2. A straight line parallel to L2 is L8, and (L2, L8) is set in the list variable R3.

In S525, it is checked with reference to the constrained arc table 410 whether there is a constrained arc whose arc type 412 is “angle” between the graphic node of the variable R2 and the graphic node of the variable R3. If not, the flow branches to S526. For example,
List R2 is (L1, L3) and list R3 is (L2, L
In the example of 8), there is no angle-restricted arc between the straight lines L1 and L2, the straight lines L1 and L8, the straight lines L3 and L2, and the straight lines L3 and L8.

In S526, a new constraint arc is created in the constraint arc table 410, "angle" is set as the arc type 412, and a combination of the linear graphic element nodes set in the variable NP1 is set as the constraint target node 413. . For example, when (L1, L2) is set in the list NP1, a new constraint arc 426 with the identifier 726 is created in the constraint arc table 410, “angle” is set as the arc type 412, and the variable NP1 is set as the constraint target node 413. (L1,
L2) is set.

Similarly, when P2 to P9 are sequentially set to the variable Q and the process is repeated, an angle constraint arc with the identifier 727 is created in the constraint arc table 410.

Next, the operation of the dependency relationship determining means 103 and the dimension shortage detecting / assuming means 104 will be described. The dependency determining means 103 selects a plurality of graphic element nodes from the constraint network stored in the constraint network storage means 102 based on the selected graphic element nodes, and then traces a constraint arc based on geometric inference. Check whether the relative position of the node from the reference element node can be determined. The result of the check is stored in the graphic element node table 400 and the constrained arc table 410 of the constrained network storage means 102 as dependency data between graphic nodes.

FIG. 17 shows a subordinate relationship determination flow including a dimension shortage detection process. With reference to the drawing of FIG. 13 as an example, a procedure of the dependency determination processing including the dimension shortage detection using the above-described constraint network will be described in detail.

First, in S801, the reference element of the dependency is determined. Therefore, in the graphic element node table 400,
One pair of straight nodes that have angular constraints on each other and have the largest number of constraint arcs due to dimensions are selected.

In the example shown in FIG.
Arc type 412 searches for an arc with an angle, and the identifier "72
6 "and" 727 "constrained arcs are obtained. From each of the constraint target nodes 413, the arc of the identifier 726 is a combination of the graphic element node L1 and the graphic element node L2, and the arc of the identifier 727 is the combination of the graphic element node L4 and the graphic element node L5.

Next, for each combination of graphic element nodes,
The number of constrained arcs according to the dimension added to the graphic element node is searched. In the case of a combination of the node L1 and the node L2, the column of the constraint target node 413 is L1 or L2.
In addition, there are four identifiers 719, 720, 721, and 723 for which the dimension identifier is set in the column of the corresponding dimension data 414. On the other hand, in the case of the combination of the node L4 and the node L5, the two are the constraint arc identifiers 724 and 725. As a result, a straight line of the set (L1, L2) is selected.

Then, the intersection of the selected set of straight lines and the straight line with the larger number of constrained arcs depending on the dimensions are used as reference elements.
In the above example, the point element node at the intersection of the node L1 and the node L2 is P1 from the constraint arcs “701” and “702”. Further, there are two restraining arcs depending on the dimensions added to the node L1 and the node L2, and either one may be used as the reference element. Here, L1 and P1 are set as reference element nodes.

FIGS. 18 to 20 show intermediate results of the dependency determination processing. 18 is FIG. 15, FIG. 19 is FIG. 16, FIG.
0 respectively correspond to FIG.

As shown by reference numerals 901 and 902, the determined flag 405 of the reference element nodes L1 and P1 nodes is set to 1, the dependency determination order number 406 is set to 0, and the group number 407 is set to 0. Further, as indicated by reference numeral 930, the group number 305 of the graphic element “601” corresponding to the reference element node L1 is set to 0.
Set to. Since there is no figure corresponding to the reference element node P1, the group number 305 is not set. In addition, code 911
As shown by, the used flag 415 of the constrained arc "701" that constrains the reference element nodes L1 and P1 is set to 1 to indicate that it has been used.

In S802, variables used for the dependency determination processing are initialized. A group number variable representing the number of times assuming an insufficient dimension is initialized to 0, and a dependency determining order variable (propagation order variable) representing the order of the graphic element nodes for which the dependency has been determined is initialized to 1. The values of these variables are
Group number 40 of the graphic element node determined in 05
7 and the dependency determination order number 406.

In S803, the graphic element node table 40
Until the determined flags 405 of all 0 nodes become 1,
Loop from S804. In S804, it is checked whether there is a graphic element node to which the geometric inference rule can be applied. If there is an applicable node, the flow branches to S805.

The geometric inference rule is a rule in which a graphic element whose position can be newly determined based on a constraint relationship with a graphic element whose position has been determined is classified according to the type of element and constraint. Representative rules are shown below.

[0131] If there is a straight line element that has a distance constraint between two fixed points, the straight line can be determined, and. If there is a linear element having a distance constraint between one fixed linear element and the linear element, the linear element can be determined, and. If you have a fixed point and a distance constraint,
If there is a linear element having an angle constraint between one linear element whose position is determined, and the position of the linear element can be determined,. If there is a point element having two fixed lines and a distance constraint, the position of the point element can be determined.

The above rules are based on the following geometric theorems (a) to (d).

(A) Draw one straight line passing through two points, (b) draw one straight line if given a distance parallel to the straight line, and (c) give a distance from point and an angle to the straight line. If one straight line is drawn and (d) the distance from the two straight lines is given,
The point is fixed.

The above-mentioned geometric inference rules are rules for straight lines and points, but by adding geometric inference rules corresponding to the geometric theorem, more complicated figures such as circles and three-dimensional shapes can be obtained. It can also be applied to the inspection for excess or deficiency of the dimensions of drawings with restrictions and restrictions.

By the way, in the first iteration of S804, only the graphic element nodes L1 and P1 have been determined. In order to check whether there is a graphic element node to which the geometric inference rule can be applied, the determined graphic element node is set as a restriction target node in the restriction arc table 410, and a restriction arc for which a used flag is not set is searched.

As shown in FIG. 18, when the determined graphic element nodes are only L1 and P1, as shown in FIG. 19, unused constrained arcs for which L1 is to be constrained are 717, 720, and 72.
3, 726, and 702 is an unused constrained arc whose constraining target is P1. The graphic element nodes connected to these constraint arcs are P4, L3, P5, and L2. Among them, the geometric inference rules can be applied to L2 and L3.

L2 is a constraint arc "702" of type "distance" between the determined graphic element node P1 and a constraint arc "726" of type "angle" between the determined graphic element node L1. There is. Therefore, the geometric inference rule can be applied to the graphic element node L2. Further, since the constraint arc “720” of the type “distance” exists between L3 and the determined graphic element node L1, the above-described geometric inference rule can be applied.

On the other hand, between the point element node P4 and the determined graphic element node L1, there is a constraint arc "717" of type "distance". The other restraining arc “718” that restrains P4 restrains P4 and L8, but L8 has not been determined. Therefore, the geometric inference rule cannot be applied to P4, and the position cannot be determined. Similarly, the geometric inference rule cannot be applied to the graphic element node P5.

In step S805, one of the graphic element nodes for which the position can be determined by applying the geometric inference rule is determined.
One is set to the determined flag 405 of the graphic element node table 400, the value of the dependent relationship determining order variable is set to the dependent relationship determining order number 406, and the value of the group number variable is set to the group number 407. The value of the group number variable is also set to the subordinate group number 305 of the graphic data.

Further, 1 is set to the used flag 415 of the constraint arc used for position determination, and the first (#
0 is set when the second (# 2) constraint target node is determined from the constraint target node of 1), and 1 is set in the opposite case. Finally, 1 is added to the dependency determination order variable.

For example, if the positions of the graphic element nodes L2 and L3 can be determined by the geometric inference rule, it is assumed that the graphic element node L2 is selected first. The determined flag 405 of the graphic element node L2 indicated by reference numeral 904 is set to 1, the dependent relationship determining order number 406 is set to the dependent relationship determining order variable value 1, and the group number 407 is set to the group number variable value 0. Also, the corresponding graphic 40 of the graphic element node L3
Since 3 is “602”, the value 0 of the group number variable is set to the subordinate group number 305 of the identifier 603 of the graphic data indicated by reference numeral 931.

Next, the constrained arcs used to determine the position of L2 are "702" and "726". First, the used flag 415 of the constrained arc 702 indicated by reference numeral 912 is set to 1 and P1 of the constrained node # 2 is determined from L1 of # 1. Therefore, 1 is set to the dependent direction data 416. In addition, code 9
Since the used flag 415 of the constraint arc 726 indicated by 13 is set to 1 and the L2 of the constraint target node # 1 is determined from the L1 of the constraint target node # 1, 0 is set to the dependent direction data 416. Finally,
Since the dependent relationship determination order variable is currently 1, it is set to 2 by adding 1.

The above processing is repeated until the geometric inference rule cannot be applied to the drawing data of FIG. 13 as it is. The processing results are shown in FIGS. Each figure corresponds to FIGS. In the example of the drawing in FIG. 13, the dependency is determined by the following procedure.

(1) Since the passing point P1 is determined at right angles to the straight line L1, the straight line L2 is determined by the geometric inference rule. The determination order number is 1, (2) Since the straight line L1 is determined and the distance is given by the dimension D2, the straight line L3 is determined by the geometric inference rule.
The determination order number is 2, (3) Since the straight line L2 is determined and the distance is given by the dimension D1, the straight line L8 is determined by the geometric inference rule.
The determination order number is 3. (4) Since the straight line L1 and the straight line L8 are determined, the point P4, which is the intersection thereof, is determined by the geometric inference rule. The determination order number is 4, (5) Since the straight line L8 is determined and the distance from L1 is given by the dimension D5, the straight line L8 is determined by the geometric inference rule.
The upper point P5 is determined. The determination order number is 5, (6) Since the straight line L3 is determined and the distance from L2 is given by the dimension D3, the straight line L3 is determined by the geometric inference rule.
The upper point P3 is determined. The determination order number is 6, (7) Since the straight line L2 and the straight line L3 are determined, the point P2, which is the intersection thereof, is determined by the geometric inference rule. The determination order number is 7, (8) Since the points P3 and P4 are determined, the straight line L4 passing through the two points is determined by the geometric inference rule. The determination order number is 8, (9) Since the straight line L4 is determined and the distance is given by the dimension D6, the straight line L7 is determined by the geometric inference rule.
The decision order number is 9.

Here, graphic elements whose positions can be determined by directly applying the geometric inference rules are eliminated from the drawing data of FIG. However, as shown in FIG. 21, the graphic element nodes L5, L6, L7, P6, P7, P
8, P9 remains. Then, the process exits the loop of S804 and detects the missing dimension and makes its assumption.
Proceed to 807. Note that if there is another constraint arc between the determined graphic element nodes, the geometric inference rule can be applied. If there are a plurality of graphic element nodes whose positions can be determined, the straight line is prioritized.

In S806, 1 is added to the group number variable to update from 0 to 1. That is, the figure element node group number determined so far is 0, and the group numbers determined thereafter are 1 or more.

Next, in S807, one more space is left between the part where the dimension is insufficient, that is, the determined graphic element node.
If there is one constrained arc, one graphic element node to which the geometric inference rule can be applied is searched. That is, the determined arc 405 of the graphic element node table 400 is 0, the constrained arc in which the arc used flag 415 of the constrained arc table 410 is not set to 1 and the determined flag 40
A graphic element node whose column 5 is connected to one graphic element node is searched.

Further, S804 is repeated again on the assumption that the constraint arc is insufficient for the graphic element. When all the graphic element nodes have been determined, the process ends.

FIG. 17B shows the processing flow of S807 assuming a lacking constraint relationship. First, S807
1, the straight line node for which the dependency relationship of the graphic element node table 400 cannot be determined, that is, the node type 402 is “straight line”
Then, for each node whose determined flag 405 is not 1,
The following processes are sequentially looped by setting the straight line node to N.

In S8072, it is checked whether there is an angle constraint between the undetermined node N and the determined graphic element node. Specifically, the constrained arc table 410 is searched, the arc type 411 is “angle”, one of the constrained nodes 413 includes the node N, and the determined flag of the other constrained node is “1”. Check for the presence of a constrained arc. If there is such a restraining arc, the process proceeds to S8073. If not, the process proceeds to S8074.

In S8073, a point node P closest to the node N is searched, and a constrained arc with an arc type of "distance" is assumed between the nodes N and P, and the processing is terminated.

In S8074, it is checked whether there is a distance constraint between the node N and the determined graphic element. Specifically, the constrained arc table 410 is searched, the arc type is “distance”, and one of the constrained nodes includes the node N,
In addition, the determined flag of another constraint target node is 1
Investigate the presence or absence of the constrained arc. If there is such a constrained arc, the process proceeds to S8075.

In S8075, the geometric data 40 of the node N
Referring to FIG. 4, a search is made for a node P that is the closest determined point node and has no constraint arc between the node N and the node N, and the constraint of the arc type “distance” is set between the node N and the node P. The arc is assumed, and the process of S807 ends.

Next, in the case of point nodes for which a dependency cannot be determined even by the above processing, in step S8076, the following loop processing is performed by sequentially setting those point nodes as nodes Q.

In S8077, it is checked whether there is a distance constraint between the undetermined node Q and the determined graphic element node. Specifically, the constrained arc table 410 is searched, and the arc type is “distance”, the constrained target node includes the node Q, and the determined flag of another constrained target node is “1”. Check for the presence of

If there is such a restraining arc, S8078
To the node P, which refers to the geometric data 404 of the node Q and searches for a node P that is the closest determined point node and has no constraint arc between itself and the node Q. Assuming a constraint arc of type "distance",
The process of S807 ends.

The processing of S807 will be specifically described through an example in which the geometric inference is interrupted due to the lack of the constraint relation as shown in FIGS. When a straight line node for which a dependency relationship cannot be determined is searched, a graphic element node table shown in FIG.
From L400, L5 and L6 are obtained.

When L5 is a node N, when a constraint arc table with L5 as a constraint target node is examined in the constraint arc table 410 of FIG. 22, a constraint arc with the identifier "727" is obtained. Since another constraint target node of the constraint arc “727” is L4, the determined flag 405 of L4 in the graphic node table 400 in FIG. 21 is 1 and L5 is an angle between the determined graphic element L4 and the determined graphic element L4. Captive arc "72
7 "is detected.

Then, the flow advances to S8073, where the node L5
Find the determined point node closest to. In FIG. 21, there are five determined point nodes P1 to P5, and the coordinate values are (0,0), (0,60), (60,60), (100,
0) and (100,20). L5 coordinates start point (70,50),
Since the end point is (50, 30), when the shortest distance between L5 and each point is calculated, the closest distance is P3. Therefore, P3 and L
4, a distance constraint can be assumed.

FIGS. 24 to 26 show that the group number variable = 1.
The processing result in the case of is shown. Each figure corresponds to FIGS. 21 to 23, respectively. The determined flag 405 of the node L5 denoted by reference numeral 971 is set to 1, the dependent relationship determining order number 406 is set to the dependent relationship determining order variable value 10, and the group number 407 is set to the group number variable value 1. Corresponding figure of node L5
Since 403 is “605”, the graphic data “605” indicated by reference numeral 981
The value 1 of the group number variable is set to the group number 305. Further, the used flag 415 of the constraint arc “727” indicated by reference numeral 991 is set to 1 and L5 is determined from L4, so 0 is set to the dependent direction data 416.

By the processing in S807, the number of determined graphic elements increases, and a graphic element to which the geometric inference rule can be applied appears again. Therefore, S804 is repeated. The following describes the process of determining the dependency of the remaining graphic elements through the dimension shortage / assuming process.

(10) Assuming that there is a distance constraint arc between the points P3 and L5, the position of L5 is determined by a geometric inference rule.

(11) Since the straight line L5 and the straight line L4 are determined, the point P6 which is the intersection thereof is determined by the geometric inference rule.

(12) The straight line L5 is determined, and the distance is the dimension D7.
, The line L
6 is decided.

(13) Since the straight line L5 and the straight line L7 are determined, the point P8, which is the intersection thereof, is determined by the geometric inference rule.

(14) Since the straight line L6 and the straight line L7 are determined, the point P9 which is the intersection thereof is determined by the geometric inference rule.

(15) Since the straight line L6 and the straight line L4 are determined, the point P7 which is the intersection thereof is determined by the geometric inference rule.

Next, a description will be given of a procedure for detecting a candidate for a dimension to be deleted by overlapping entry by the overlapping dimension detecting means 105.

FIG. 27 shows a flow of the overlap size detection processing. First, in S1001, a duplicate group number variable is initialized to 1. In S1002, the processing from S1003 on is performed for each unused constraint arc in the constraint arc table 410, that is, for each constraint arc whose used flag 415 is not 1.

In S1003, the constraint arc used when applying the geometric inference rule for the first node (# 1) in the column 413 of the constraint target node of the unused constraint arc is set in the direction opposite to the dependent direction data 416. Tracing until the reference element node is reached, and all the constrained arcs in the middle are stored in a list. This is List A. The list A is a set of constraint conditions necessary for determining the first node of the constraint target node 413 of the unused constraint arc.

In S1004, the constraint arc used when applying the geometric inference rule is traced in the same manner as described above for the second node (# 2) in the column 413 of the constraint target node of the unused constraint arc. List all constraint arcs
To save.

In S1005, the union of the list A and the list B is obtained, the intersection of the list A and the list B is removed from the union, and this is stored in the list C. When this is represented by an equation, it is described as in Equation 1.

[0173]

(Equation 1)

The intersection of the list A and the list B is a constraint condition necessary for both the first node and the second node. Therefore, if the necessary constrained arcs that are to be a product are removed, both nodes cannot be determined and cannot be deleted.
On the other hand, the constrained arc included in list C is one of them.
Even if one is deleted, it can be determined by either the first node or the second node, and the node that cannot be determined by using an unused constraint arc can also be determined. Therefore, list C
Is a candidate for a constraint arc that can be deleted to solve overlapping dimension entry.

In S1006, the corresponding dimension data 414 of the constraint arc and the unused constraint arc included in the list C are checked. If there is the corresponding dimension data, the overlap flag 309 of the dimension data is set, and the overlap group is set. Number 3
Set the value of the duplicate group number variable to 10. S1007
Then, 1 is added to the overlapping group number variable, and the next unused constraint arc is processed.

The operation of the overlap size detecting means 105 will be described in detail with reference to FIG. 25 which is a result of processing the dependency data as an example. In this figure, a constrained arc that is not used for determining the dependency and whose arc used flag 415 is not 1 is a constrained arc “722” indicated by reference numeral 992. Regarding the constraint target node 413 of the constraint arc “722”, the first node is P3 and the second node is P5.

Next, the constraint arc used to determine the position of P3, that is, the column # 1 of the constraint target node 413 is P3, the column of the dependent direction data 416 is 1, or the column # 2 of the constraint node 413 is P3. Then, when the constraint arc in which the column of the dependent direction data 416 is 0 is searched from the constraint arc table 410 in FIG. 25, constraint arcs “705” and “721” are obtained. These arcs are added to list A. The constraint target node of the dependent element of P3 of the constraint arc "705" is L3.
The constraint arc used to determine the position of is searched, and the constraint arc “720” is added to the list A. Captive arc "72
The dependent node of the dependent source of "0" is L1 and is a reference element node, so the search is stopped here.

Similarly, since the constraint target node of the dependent element of the constraint arc "721" is L2, the constraint arcs "702" and "726" used to determine L2 are obtained, and these are added to the list A. . The constraint target nodes of the dependent elements of the constraint arcs "702" and "726" are P1 and L1 and both are reference element nodes, so the search is stopped here. The list A obtained as described above is represented by Expression 2.

[0179]

A = {arc 705, arc 721, arc 720, arc 702, arc 726} Similarly, if the subordinate order of P5 of the second node is reversed,
List B is obtained and is represented by Equation 3.

[0180]

## EQU3 ## B = {arc 716, arc 723, arc 719, arc 702, arc 726} Then, list C is expressed by equation 4.

[0181]

C = {arc 705, arc 721, arc 720, arc 716, arc 723, arc 719} Here, the arc 726 common to A and B is excluded.

In the list C, the constraint arc table shown in FIG.
Arc 721 and Arc 72 have dimensions corresponding to 410
0, Arc 723 and Arc 719, dimensions 613 and 61 respectively
2, dimensions 615 and 611. Unused constraint arc 7
22 corresponds to dimension 614.

These dimensions are candidates for constraint arcs that can be deleted to resolve duplicate entries. 1 is added to the column of the overlap flag 309 of the dimension data, and the overlap group number 310
Is set to the value 1 of the duplicate group number variable. The result is shown as dimensional data 3 in FIG.

By displaying the same size with the same overlap group number 310 of the size data on the display device 201, for example, by changing the color, it is possible to show the user the location of the overlap size and the candidate of the size to be deleted. In addition, if the maximum value of the overlap group number is displayed on the display device 201, it is possible to present to the user how many overlap dimensions exist.

FIG. 29 shows a display example of the inspection result of the missing dimension and the overlapping dimension of the drawing in the present embodiment. The missing dimension is based on the graphic data of FIG. 26, and the overlapping dimension is based on the dimension data of FIG. A message box 1101 indicates the number of duplicate entry dimensions and the number of dimension entry shortages. The number of overlapping entry dimensions can be found by the maximum value of the overlapping group number 310 of the dimension data, and the number of dimension entry shortages can be found by the maximum value of the group number 305 in FIG.

Dimension 1102 highlighted by changing line type
Indicates candidates to be deleted in order to resolve duplicate entries.
Also, the graphic element 1103 highlighted by a bold line is
A figure element whose position cannot be determined due to insufficient dimension entry is shown.

As described above, it is possible to present to the user all the lack of dimensions and duplicate entries on the drawing in an easy-to-understand manner, so that it is easy for the user to deal with and incorrect work can be prevented.

[Embodiment 4] Next, a fourth embodiment for inspecting a dimension including a plurality of projections for excess or deficiency will be described. In the present embodiment, a front view, a side view, an auxiliary projection view,
Inspect the drawings composed of a plurality of projection views, such as partial enlarged views, for insufficient dimensions and overlapping dimensions.

FIG. 30 is a functional block diagram of the drawing inspection apparatus according to the present embodiment. This drawing inspection apparatus can be realized, for example, on the computer system shown in FIG. The drawing data storage means 1201 stores a plurality of projection drawing data 1202. Each projection map data 1202 is represented by a figure arranged on a three-dimensional space, and graphic data 1203 representing a projected shape of an object, a center line, etc., dimension data 1204 representing a distance, a radius, and an angle of the graphic data 1203. It comprises a projection transformation matrix 1205 for obtaining a projection shape from a part. The data structure of the graphic data 1203 and the dimension data 1204 is the graphic data 2 shown in FIG.
And the data structure of the dimension data 3 may be the same.

The constraint relation extracting means 1206 generates the drawing data 1201
The figure data 1203 and the dimension data 1204 are read out for each of the projection view data 1202, and a constraint network 1208 is created in which the figure data is a node, and the phase relation and angle relation between the dimension data and the figure elements are arcs. Restraint network
The data structure of 1208 may be the same as the data structure shown in FIGS.

The constraint relation adding means 1207 includes projection view data 1202 other than the projection view data 1202 to be processed by the constraint relation extracting means 1206 (this is called target projection view data).
For each of these (referred to as transformed projection map data), the dimension data 1204 and the projection transformation matrix 1205 are read, and the projection transformation matrix 1205 of the target projection diagram and the projection transformation matrix 1205 of the transformed projection diagram are read.
From, a transformation matrix for transforming the coordinate values on the transformed projection view into the coordinate system on the subject projection view is obtained, and using this transformation matrix, the dimension data 1204 is transformed into the coordinate system of the subject projection view, Add to constraint network 1208.

The dependency determining means 1209 checks whether the position of the graphic element node of the constraint network 1208 can be determined by geometric inference in the same manner as the procedure shown in FIG. 17, and determines the graphic element node based on the inspection result. The dependency data between them is added to the constraint network 1208 as the direction of the arc representing the constraint relationship.

The overlapping dimension detecting means 1210 obtains a dimension candidate to be deleted in order to eliminate the duplicate entry of the dimension from the constraint network 1208 by the processing procedure shown in FIG.
Set the dimension data 1204.

The undersize detecting / assuming means 1211 performs the same processing as the undersize detecting / assuming means 104 in FIG. That is, when the geometric inference rule cannot be applied before the dependency determining means 1209 determines the dependency of all the graphic element nodes, the dependency data necessary for determining the dependency from the constraint network 1208 is activated when the geometric inference rule cannot be applied. Assume. This enables geometric inference of an undetermined graphic element node, and writes the number of constraint relations assumed up to the time of determining the dependent relation into the graphic data 1203.

[0195] Restriction relation addition means 1207 in this embodiment.
Can be said to be a function substantially added to the drawing inspection means 4 of FIG. FIG. 31 shows a processing procedure of the constraint relation adding means. First, in S1300, the processing from S1301 onward is repeated for the transformed projection view data in the projection view data 1202.

In step S1301, the projection transformation matrix 12
From 05 and the projection transformation matrix 1205 of the transformation projection, a matrix for transforming the coordinate values on the transformation projection to the coordinate system on the target projection is obtained. In S1302, S303 and subsequent steps are repeated for each of the dimension data 1204 included in the converted projection view data. S1303
Then, if the dimension type 307 of the dimension data 1204 is a distance dimension, the procedure goes to S1304.
If it is an angular dimension, the process branches to S1310.

In S1304, using the transformation matrix calculated in S1301, for example, the dimension line 1401 of the distance dimension shown in FIG.
Are converted into coordinate values on the coordinate system of the target projection view. However, since the coordinate value of the end point of the dimension line is two-dimensional, it is treated as a three-dimensional coordinate value where the value of the Z coordinate value is 0. If the Z coordinate values at both ends of the converted dimension line match, that is, if the converted dimension line is parallel to the target projection view, the flow branches to S1305. Dimension data whose converted dimension line is not parallel to the target projected view is ignored because it is not correctly projected on the target projected view.

At S1305, the feet 1404 of the two dimension auxiliary lines,
Convert 1405 to the coordinate system of the target projection, create a straight line passing through that point and perpendicular to the converted dimension line,
1208 is added as a graphic element node. Then, a distance constraint arc is added between the straight nodes. If a straight line overlapping on a straight line perpendicular to the converted dimension line already exists in the constraint network 1208, no node is added, and a distance constraint arc is added between the existing node and the existing node. if,
When there is an existing point node on the added straight line node and when the added straight line node contacts an existing circular element arc, a distance constraint arc is added between them.

At S1306, a process is performed for a case where the dimension is a radius dimension or a diameter dimension. FIG. 33 shows an example of the diameter dimension.
First, a center point and a radius of a circle or an arc whose dimensions constrain the diameter are obtained on the transformed projection view. Next, a straight line passing through the center point and perpendicular to the transformed projection is obtained, and this is set as the central axis on the transformed projection.

For example, when the coordinate value of the center point is (5, 10), a straight line passing through points (5, 10, 0) and (5, 10, 1) is the center axis. This center axis is converted to the coordinate system of the target projection using the conversion matrix calculated in S1301. If the converted center axis is perpendicular to the target projection view, that is, if both the X component and the Y component of the direction vector of the center axis are 0, S13
Branch to 07. If the converted center axis is not perpendicular to the target projection view, that is, if both the X component and the Y component of the direction vector of the center axis are not 0, the process branches to S1308.

In S1307, since the center axis after the conversion is perpendicular to the target projection, the circle on the conversion projection is directly converted into a circle on the target projection. Therefore, in the constraint network 1208 of the target projection data, a circle graphic element node having a radius equal to the center of the converted circle is searched for, and a radius constraint is added to the node.

In S1308, it is checked whether or not the converted center axis is parallel to the target projection, that is, whether or not the Z component of the converted center axis direction vector is zero. If parallel, the flow branches to S1309. If the central axis after the conversion is parallel to the target projection, the target projection is viewed from a direction perpendicular to the central axis of the cylindrical surface described by a circle on the conversion projection. Therefore, the circle on the transformed projection is drawn as a parallel line parallel to the central axis and separated by a radius on the target projection.

At S1309, it is determined that there is a distance dimension between the parallel lines, and a distance constraint arc is added to the constraint network 1208.

In the above description, the case of the diameter and the radius dimension given to the circle has been described. However, when the diameter dimension is attached to the arc, the processing is performed in the same manner as the circle.

In the case of an angular dimension, in S1310, it is determined whether to add to the constraint network 1208 as follows. FIG. 34 shows an example of the angle dimension. An axis that passes through the intersection 1603 of the two straight lines 1601 and 1602 whose angle dimensions are constrained and that is perpendicular to the transformed projection is transformed into the coordinate system of the target projection. If the converted axis is perpendicular to the target projection view, the flow branches to S1311 to add an angular dimension to the constraint network 1208.

In step S1311, the two straight lines 1601 and 1602 whose angular dimensions are constrained are converted to the coordinate system of the converted projection. A straight line node matching these two straight lines is added to the constraint network 1208.
And add an angle constraint arc between them.

Hereinafter, the operation of the drawing inspection apparatus according to the present embodiment will be described in detail with reference to the example of FIG. 35 including a plurality of projection views.

FIG. 35 is composed of three projections: a front view 1701, a left side view 1702, and a right side view 1703. In this example,
If the front view 1701 alone is inspected for over / shortage of dimensions, there is no duplicate entry, and the graphic elements 1710, 1711, 1712
Is determined to be undersized. However, looking at the whole drawing,
The graphic element 1711 and the graphic element 1712 correspond to the dimension 17 in the right side view 1703.
It can be seen that it is restricted by 26. Furthermore, the front view
The dimension 1723 of the 1701 and the dimension 1725 of the right side view 1703 are overlapped, and it can be seen that they are excessive dimensions.

The drawing of FIG. 35 is the same as the stepped shaft 18 shown in FIG.
01, front view 1701 from the line of sight 1802, left side view 1702
From the line of sight 1803, the right side
, The shape projected by trigonometry.

FIG. 37 shows a three-dimensional space diagram of FIG.
That is, the front view 1701 is arranged on the YZ plane in the three-dimensional space, and the origin of the three-dimensional coordinate system is projected on the coordinate system origins 1704, 1705, and 1706 of the respective projection views in consideration of the line of sight, A left side view 1702 and a right side view 1703 are arranged in a three-dimensional space. The plane 1901 is the projection plane on which the front view 1701 is placed, the plane 1902 is the projection plane of the left side view 1702, and the plane 1903 is the right side view 17
A projection plane 03, reference numeral 1904 indicates an origin and a coordinate system of the three-dimensional space where the projection view is arranged.

By arranging the projections in this manner, a matrix for converting points on each projection into the three-dimensional coordinate system shown in FIG. 37 can be determined from a three-dimensional projection conversion algorithm of computer graphics. This algorithm is, for example, the "Fundamentalsof Interactive Compute
r Graphics (JD Floey, A. Van Dam, Addison-We
sley Publishing Companay; published in 1982; PP268-31
6)).

This transformation matrix is defined as a projection transformation matrix 1205 of the projection drawing data 1202. The projection transformation matrix of the front view 1701 is M
1. The projection transformation matrix of the left side view 1702 is M2, and the right side view 1703
Let M3 be the projection transformation matrix of, each transformation matrix is represented by Equation 5.

[0213]

(Equation 5)

In the drawings, when representing the shape of an object,
In some cases, not only a projection view but also a cross-sectional view is used. Since the line-of-sight direction is determined in the sectional view as in the case of the projected view, the projection transformation matrix 1205 of the sectional view can be obtained in the same manner as in the case of the illustrated projected view.

FIG. 38 shows the contents of the drawing data 1201 of the drawing shown in FIG. The line with the code number 2001 is the graphic data 120
3. The row denoted by reference numeral 2002 is the dimension data 1204, and the row denoted by reference numeral 2003 is the projection transformation matrix 1205. Also, each row of reference numeral 2004 is a front view.
1701 projection view data 1202, each row of reference numeral 2005 is a right side view 17
Projection data 1203 of 03, each row of reference numeral 2006 is a left side view 1702
Is projection view data 1202 of FIG.

FIG. 39 shows that the relation extracting unit 1206
6 is a graphic element node table of a constraint network 1208 of a front view 1701 extracted from graphic data 1201. FIG.
In the constraint arc table of the constraint network 1208 of the front view 1701 extracted from the graphic data 1201 by the constraint relationship extracting means 1206 and the constraint relationship adding means 1207, the constraint arcs with identifiers A1 to A20 are identified by the constraint relationship extracting means 1206. It is a constrained arc extracted from the projection view data 2004 of FIG. The constraint arcs with identifiers A21 to A24 are:
The constraint arc is extracted from the projection view data 2005 in the right side view 1703 and the projection view data 2006 in the left side view 1702 by the constraint relation adding means 1207. Below, identifiers A21 to A2
The process of adding the fourth constraint arc will be described.

A front view 1701 is a target projection view, and a right side view 1703.
Is a conversion projection view, in S1301, the right side view 17
A matrix for converting the coordinate value of 03 into the coordinate system of the front view 1701 is obtained by Expression 6.

[0218]

(Equation 6)

In step S1302, each dimension in the right side view 1703 is added to the constraint network created from the graphic data and dimension data of the front view 1701 using the conversion matrix _M3-1 . Since the dimension 1725 is a diameter dimension, the process branches to S1306.

In step S1306, since the dimension 1725 is added to the circle of the graphic element 1718, the center point is referred to (0,
A straight line passing through two points (0, 0) and (0, 0, 1) is the central axis of the dimension 1725. Using the transformation matrix _M3-1 , the front view 170
When converted to a coordinate system of 1, (100, 0, 0) and (101, 0,
0). Since this straight line is parallel to the front view 1701, the process branches to S1309 via S1308, and the distance between two straight lines parallel to the central axis and separated by a radius of 50 of the circle of the graphic element 1718 is defined as the constraint network of the front view 1701. Add to

When the central axis of the dimension 1725 is converted into the front view 1701, the coordinates are (100, 0, 0) and (101, 0, 0). In the projection view, the coordinate value in the Z-axis direction is ignored because it is a two-dimensional plane coordinate system. if,
(100, 0)-(101, 0), which is on the same straight line as the graphic element node L8 in FIG. Straight line (100, 0)-(101,
Two straight lines parallel to 0) and separated by a radius of 50 of the circle of the graphic element 1718 are (100, 50)-(101, 50), (100, -50)
− (101, −50), and are on the same straight line as the graphic element nodes L2 and L3 in FIG. 39, respectively.

Therefore, reference numerals 2021 and 2021 in FIG.
The straight lines L2 and L8 and the straight lines L3 and L shown by 2022
The constraint arcs A21 and A22 having the constraint node 8 as the constraint target node are added to the constraint arc table in the front view 1701. Similarly, when the dimension 1726 is converted into the front view 1701, the constrained arcs A23 and A24 indicated by reference numerals 2023 and 2024 are obtained. In addition,
Since there is no dimension in the left side view 1702, the dimension addition processing ends here.

FIG. 41 shows a graphic element node table obtained by applying the dependency relation determining means 1209 and the dimension shortage detecting means 1211 to the constraint network after the processing of the constraint relation adding means 1207 is completed. 42.
As indicated by reference numerals 2031 and 2032, the dependent relationship group number of the graphic element nodes L6, P3, P4, P5, and P6 is 2, which is different from other graphic element nodes. This corresponds to graphic element nodes L6, P3, P4, P5,
P6 indicates that the relative position cannot be determined from other graphic elements due to insufficient dimension entry. Of these, the corresponding graphic is set only in L4, and its graphic data identifier is a straight line 1710. This indicates that the straight line 1710 is undersized.

On the other hand, the constraint arc in the constraint arc table in FIG.
Arcs A21 and A22 indicated by 2041 and 2042
It is. When the candidate for the size to be deleted is obtained by the duplicated size detecting means 1210, the arc A
17 and arc A18 are obtained from arc A22.
As shown in FIG. 42, dimensions 1723 are set for A17 and A18,
Since dimension 1725 corresponds to A21 and A22, dimension 1723
And the dimension 1725 are candidates for the dimension to be deleted.

Next, the case where the left side view 1702 is set as the target projection view will be described. FIG. 43 shows the constraint relationship extracting means 12.
06 and the constraint addition means 1207
FIG. 70 is a graphic element node table of the constraint network 1208 of the left side view 1702 extracted from FIG. L1, L2, C1, C
2. P1 is a left side view 1702
This is a graphic element node extracted from FIG. L3 and L4 are graphic element nodes extracted from the dimension data of the front view 1701 by the constraint addition unit 1207.

FIG. 44 is a constraint arc table of the constraint network 1208 in the left side view 1702 extracted from the graphic data 1201 by the constraint relationship extracting means 1206 and the constraint relationship adding means 1207. The constrained arcs with identifiers A1 to A5 are the constrained arcs extracted from the projection data 2006 of the left side view 1702 by the constrained relationship extracting means 1206. Identifier A
The constraint arcs 6 to A7 are constraint arcs extracted from the projection view data 2005 of the right side view 1703 by the constraint relationship adding means 1207. The constraint arcs with identifiers A8 to A10 are:
This is a constraint arc extracted from the projection view data 2004 of the front view 1701 by the constraint relationship adding means 1207.

The graphic element nodes L3 and L3 shown in FIG.
4 and the process of adding the constraint arcs of identifiers A6 to A10 in FIG. 44 will be described. First, the right side view 1703 is a transformed projection view. In step S1301, a matrix _M3-2 for converting the coordinate values in the right side view 1703 into the coordinate system in the left side view 1702 is obtained by Expression 7.

[0228]

(Equation 7)

[0229] In S1302, the dimensions of the right side view 1703 using a transformation matrix M _3-1, adding to the constraint network 1208 created from graphic data and dimension data of left side view 1702. First, since the dimension 1725 is a diameter dimension, the process branches to S1306.

In S1306, since the dimension 1725 is added to the circle of the graphic element 1718, a straight line passing through two points (0, 0, 0) and (0, 0, 1) is referred to by referring to the center point. It is the central axis of 1725. This is converted to a left side view 170 by using a transformation matrix M _3-2 .
When converted to the coordinate system of 2, (0, 0, -100) and (0, 0,-
101).

Since this straight line is perpendicular to the left side view 1702,
The process branches to step S1307, and a circle on the left side view 1702 that is the same as the radius 50 of the circle of the graphic element 1718 is searched around the point (0, 0) on the left side view 1702 through which the central axis passes. In this case, graphic element 17
Since the circle of 21 corresponds to this, a constraint arc A6 indicated by reference numeral 2121 in FIG. Set to. Further, a dimension data identifier 1725 before conversion is set in the corresponding dimension data column.

Similarly, when the dimension 1726 is converted into the left side view 1702, the central axis becomes vertical. Therefore, as the diameter dimension A7 of the circular figure node C2 on the left side view 1702 indicated by reference numeral 2122,
Add to the constraint network in left side view 1702.

Next, the front view 1701 is a converted projection view. A matrix _M1-2 for converting the coordinate values of the front view 1701 into the coordinate system of the left side view 1702 is obtained by Expression 8.

[0234]

(Equation 8)

[0235] In S1302, using the transformation matrix M _1-2 the dimensions in the front view 1701, adds to the constraint network created from graphic data and dimension data of left side view 1702. First, since the dimension 1723 is a distance dimension, the process branches to S1304.

In S1304, the coordinate values on the front view 1701 of the both ends of the dimension line of the dimension 1723 are represented by (−10, 50) and (−10, −50).
Then, on the left side view 1702, the coordinate values of both end points after the conversion are (0, 50, 10) and (0, -50, 10), and the Z coordinate values match. It is parallel to the left side view 1702. Therefore, the flow branches to S1305.

In S1305, the coordinate values (0, 50) and (0, -50) of the foot of the dimension auxiliary line of the dimension 1723 are converted to the left side view 1702, and the coordinate values (0, 50, 0), (0, -50, 0). When two straight lines perpendicular to the dimension line after conversion are obtained and projected on the left side view 1702, the points (0, 50)
And a straight line passing through points (1, 50), and points (0, -50) and (1,
-50) to obtain two straight lines. Since the graphic element nodes that are on the same line as these linear elements are not in the graphic element node table of FIG. 43, they are added to the constraint network of left side view 1702 as L3 and L4 indicated by reference numerals 2113 and 2114.

Since the added straight lines L3 and L4 are in contact with C1, as shown by reference numeral 2123 in FIG.
1. Restricted arc A whose type is distance between straight line L4 and circle C1
8. Add A9. Further, as indicated by reference numeral 2124,
A constraint arc A10 for restricting the distance between L3 and L4 is added, and the identifier of the dimension data before conversion is added to the corresponding dimension data.
Set 1723.

Next, the dimension 1724 is processed. Since the dimension 1724 is a distance dimension, the process branches to S1304, and the coordinate values (0, -60) and (100, -60) of both ends of the dimension line are converted into a left side view 1702. The coordinate values after conversion are (0, -60, 0), (0, -60, -1
00), and the converted dimension line is not parallel to the left side view 1702. Therefore, dimension 1724 is not added to left side view 1702.

FIG. 45 shows a graphic element node table, and FIG.
Indicates a constrained arc table, and shows the result of applying the dependency relationship determining means 1209, the dimension shortage detecting means 1211 and the overlapping dimension detecting means 1210 to the constrained networks shown in FIGS. 43 and 44.

Since the dependent relationship group numbers in the graphic element node table are all the same, there is no lack of dimension entry in the left side view 1702. The constraint arc for which the used flag is not set is A10 indicated by reference numeral 2125. When this is traced back in the dependent direction by the overlapping dimension detecting means 1210, arcs A8, A9, A6, and A3 are obtained. A6 has the corresponding dimension data among these constrained arcs, and the corresponding dimension is 1725. Since the corresponding dimension of A10 is 1723, the dimensions 1723 and 1725 are obtained as candidates for the dimension that can be deleted.

As described above, when the inspection results for all the projection views are summarized, it can be detected that the dimension 1723 and the dimension 1725 overlap in the drawing shown in FIG. 35 and the graphic element 1710 in the front view 1701 is insufficient in dimension.

FIG. 47 shows the inspection result of the drawing of FIG.
By highlighting the underfilled graphic element 2301 and displaying the overlapping dimensions 2302 in different display colors to present the user with errors in dimensioning, the three-dimensional In the drawings, the parts that need to be corrected can be shown at a glance.

[0244]

According to the present invention, the graphic elements whose positions can be relatively determined are grouped and the drawing is displayed with different display attributes for each group, so that the user can easily add the missing dimensions. Also, between different groups,
Since a missing dimension is automatically added by selecting between graphic elements to which dimensions can be added, the number of drawing inspections can be greatly reduced.

According to the present invention, the dependency between graphic elements in the drawing, that is, graphic elements whose positions can be relatively determined is determined by a geometric inference rule including a constraint condition in the premise, and the constraint condition is determined. If the inference is interrupted due to lack of data, the inferred constraints are assumed for the undetermined graphic elements, and the inference is performed repeatedly until the subordination of all the graphic elements is established. Can be determined.

Further, when the inference is completed, by tracing the inference assumption of the unused constraint condition in the reverse direction, all the candidate dimensions of the duplicate entry can be extracted. In other words, even when the drawing has insufficient dimensions, all the dimension candidates for duplicate entry are listed up, so that the user can leave the truly required dimensions and correctly delete the others.

According to the present invention, for a drawing including a plurality of projection views or cross-sectional views, coordinate conversion between the projection views is performed to detect an excess or deficiency of the dimensions of the whole drawing. Accurate inspection results are obtained, and the time and effort for correction can be greatly reduced.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a drawing inspection apparatus according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of a computer system for realizing the drawing inspection apparatus according to the present invention.

FIG. 3 is an example of a drawing having a lack of dimension entry referred to in the description of the first embodiment.

FIG. 4 is a table showing graphic data in the drawing of FIG. 3;

FIG. 5 is a table showing dimension data in the drawing of FIG. 3;

FIG. 6 is a table showing a result of grouping according to the first embodiment with respect to the drawing of FIG. 3;

7 is a screen displaying the graphic elements of the drawing of FIG. 3 with different display attributes for each group.

FIG. 8 is a flowchart showing a process of adding a missing dimension according to the second embodiment of the present invention.

FIG. 9 is a table showing a list of additional dimensions.

FIG. 10 is a screen in which additional dimensions are displayed with different display attributes from the drawing of FIG. 7;

FIG. 11 is a functional block diagram of a drawing inspection apparatus according to a third embodiment of the present invention.

FIG. 12 is a processing flowchart of a constraint network creating unit.

FIG. 13 is an example of a drawing with insufficient or insufficient dimension entry referred to in the description of the third embodiment.

FIG. 14 is a table diagram showing a storage format of drawing data and dimension data.

FIG. 15 is a format diagram of a graphic element node table in a constraint network storage unit.

FIG. 16 is a format diagram of a constraint arc table in a constraint network storage unit.

FIG. 17 is a processing flowchart of a dependency determination unit.

FIG. 18 is a transition diagram showing the contents of a graphic element node table in the process of being processed by the dependency determination unit.

FIG. 19 is a transition diagram showing the contents of a constrained arc table being processed by the dependency determination means.

FIG. 20 is a transition diagram showing the contents of graphic data being processed by the dependency determining means.

FIG. 21 is a transition diagram showing contents of a graphic element node table when inference is interrupted.

FIG. 22 is a transition diagram showing contents of a constrained arc table when inference is interrupted.

FIG. 23 is a transition diagram showing the contents of graphic data when inference is interrupted.

FIG. 24 is a transition diagram showing the contents of a graphic element node table as a result of inference based on a constraint condition assumption.

FIG. 25 is a transition diagram showing contents of a constrained arc table as a result of inference based on a constrained condition assumption.

FIG. 26 is a transition diagram showing contents of graphic data as a result of inference based on a constraint condition assumption.

FIG. 27 is a processing flowchart of an overlap size detection unit.

FIG. 28 is a table diagram of dimension data showing dimensions of overlapping entries in the drawing of FIG. 13;

FIG. 29 is a display screen showing the inspection result of FIG. 13 according to the third embodiment.

FIG. 30 is a functional block diagram of a drawing inspection apparatus according to a fourth embodiment of the present invention.

FIG. 31 is a flowchart showing a process of adding a constraint condition by projection transformation.

FIG. 32 is an explanatory diagram showing an example of a distance dimension.

FIG. 33 is an explanatory view showing an example of a diameter dimension.

FIG. 34 is an explanatory diagram showing an example of an angle dimension.

FIG. 35 is an example of a drawing having a projection view referred to in the description of the fourth embodiment;

FIG. 36 is a perspective view showing a direction of a line of sight of each projection view of FIG. 35;

FIG. 37 is a perspective view in which the projections of the drawing of FIG. 35 are arranged in a three-dimensional space.

FIG. 38 is a transition diagram showing the contents of the drawing data of FIG. 35;

FIG. 39 is a transition diagram showing the contents of the graphic element node table in the front view of FIG. 35 created by the constraint relationship extracting means and the constraint relationship adding means.

FIG. 40 is a transition diagram showing contents of a constrained arc table in the front view of FIG. 35;

FIG. 41 shows a processing result obtained by the dependency determination unit, and FIG.
Transition diagram corresponding to FIG.

FIG. 42 shows a processing result obtained by the dependency relationship determining means.
Transition diagram corresponding to FIG.

FIG. 43 is a transition diagram showing a part of the contents of the graphic element node table created by the constraint relationship extracting means and the constraint relationship adding means.

FIG. 44 is a transition diagram showing a part of the contents of the constraint arc table created by the constraint relationship extracting means and the constraint relationship adding means.

FIG. 45 shows a processing result obtained by the dependency relationship determining means;
The transition diagram showing a part of the contents of the graphic element node table in the left side view of FIG.

FIG. 46 shows a processing result obtained by the dependency determining means,
FIG. 6 is a transition diagram showing a part of the contents of a constrained arc table in the left side view of FIG.

FIG. 47 is a display screen showing the test results of FIG. 35 according to the fourth embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Drawing data storage means, 2 ... Graphic data, 3 ... Dimension data, 4 ... Graphic data grouping means, 5 ... Graphic data group display means, 6 ... Display means, 7 ... Dimension addition means,
101: Constraint relation extraction means, 102: Constraint network storage means, 103: Dependency relation determination means, 104: Dimension shortage detection / assuming means, 105: Overlapping dimension detection means, 400: Graphic element node table, 410: Constraint arc table, 1201 ... Drawing data storage means, 1202 ... Projection drawing data, 1203 ... Graphic data, 12
04: Dimension data, 1205: Projection conversion data, 1206: Restriction relation extraction means, 1207: Restriction relation addition means, 1208: Restriction network storage means, 1209: Dependency relation determination means, 1210: Overlapping dimension detection means, 1211: Dimension shortage Means of detection and assumption.

──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Katsushi Machii 5030 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Pref. Software Development Division, Hitachi, Ltd. (56) References JP-A-6-119424 (JP, A) JP-A-6-75620 (JP, A) JP-A-7-200667 (JP, A) JP-A-63-315772 (JP, A) JP-A-6-119446 (JP, A) (58) Fields investigated (Int) .Cl. ⁷ , DB name) G06F 17/50

Claims (4)

(57) [Claims] 1. A drawing inspection method for detecting, by a computer processing, a lack of dimension entry of a drawing generated as CAD graphic data, the drawing data for each graphic element of the CAD graphic data, Storing constraint condition data defining a relative positional relationship between the graphic elements; dividing the graphic data into graphic elements capable of determining a relative positional relationship to each other based on the constraint condition data; A drawing inspection method, wherein the graphic element is displayed with different display attributes in units. 2. A drawing inspection method for detecting, by computer processing, a lack of dimension entry of a drawing composed of a plurality of projections created by a CAD system, comprising: a graphic data for each graphic element of each projection; Phase between elements
Constraint condition data for defining a relative positional relationship and projection transformation data for specifying a projection method of each projection view are stored. For a target projection view, constraint condition data other than the projection view is stored in the projection transformation data. Is used to convert to constraint data on the projection view, and is added to the constraint data attached to the graphic element of the projection view. Based on the added constraint data, the relative positional relationship between each other is calculated. Wherein the graphic data is grouped into graphic elements each of which can determine a graphic element, and graphic elements on the projection view are displayed with different display attributes for each group. 3. A drawing inspection apparatus for detecting an excess or deficiency of dimension entry for a drawing represented by CAD graphic data created by a CAD system, wherein the geometric data for each graphic element of the CAD graphic data is Drawing data storage means for storing constraint condition data for defining the relative positional relationship of the CAD data; and grouping the CAD graphic data into graphic elements whose relative positions can be determined based on the constraint data. A drawing inspection apparatus, comprising: a graphic grouping unit that performs graphic grouping; and a display unit that displays graphic elements of the CAD graphic data with different display attributes for each group. 4. A drawing inspection apparatus for detecting lack of dimension entry of a drawing composed of a plurality of projections created by a CAD system, comprising:
Drawing data storage means for storing constraint condition data for defining a relative positional relationship, projection transformation data for specifying a projection method of each projection view, and for the constraint condition data of the target projection view, A constraint relation adding means for converting and adding constraint condition data of a projection view other than the projection conversion data by the projection conversion data, and graphic elements capable of determining a relative positional relationship with each other based on the added constraint condition data. And a graphic data grouping means for grouping the graphic data.