JPH1166112A - Method for extracting shape feature from cad data and recording medium where shape feature extracting program is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately and speedily extract shape feature parameters by reading dimension values of shape features out of two-dimensional CAD data by pattern matching. SOLUTION: In this method for detecting shape features of a design object from the two-dimensional data according to a previously given model, CAD data as two-dimensional data are input (Step ST1) and dimension lines and character information from the CAD data are deleted to extract only shape data (step ST2). Then a model which prescribes shape feature parameters including shape features, geometric restriction conditions, and a dimension permissible range is input (step ST5), and on the basis of the values of the shape feature parameters prescribed by the model, various shape features are detected from the shape data in the CAD data (step ST6). Further, dimension values of the respective shape features are detected to detect the shape feature parameters (step ST7).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for extracting shape features from CAD data in a computer-aided design (hereinafter abbreviated as CAD) system. The present invention relates to a method for automatically extracting a shape feature (feature) relating only to a shape to be designed other than non-shape data including character information and the like, and a recording medium storing a shape feature extraction program used in this method.

[0002]

2. Description of the Related Art Computer-Aided Design / Computer-Aided Design / Computer Aided Design / Computer Aided Design / Computer Aided Design / Computer Aided Design (CAE)
Aided Manufacturing- ) Is no longer a new concept, and in recent years concurrent engineering and CALS (Continuous Acquisition a)
nd Life-cycle Support-Integrated information system for production, procurement and operation support-), PDM (Product Data Management-
Such as product data management) by sharing information in all aspects of manufacturing, such as product manufacturing, and allowing information to flow smoothly where and when it is needed.
The need to be more competitive is being called out. At that time, it is often premised that a three-dimensional CAD is used as a design tool, and a design as an upstream process and a solid model as a design result in the design process are flowed to a downstream process. Using the solid model created in the design process,
Numerous examples of performing various analysis operations and using them for mold manufacturing have been reported so far.

As a method of integrating CAD / CAM and a method of manufacturing a solid model in a manufacturing process, mold manufacturing is often cited as an example. However, the efficiency of cutter path generation of NC (Numerical Control) has been improved. Examples of automation have also been reported. In general, when CAD data is used in a manufacturing process, it is of course necessary to have shape data on a shape to be designed, but it is also necessary to specify non-shape (additional) data on dimensional tolerances and materials. Come. Further, in order to determine the processing conditions, not only mere shape data but also performances and functions required for the work and the component are required. Each work, each part, and each part have a required function. For example, holes, pockets,
From general-purpose parts such as chamfered parts to parts such as bolts, nuts, bearings, gears, etc., various different requirements are required for strength, processing accuracy, finish and the like. Therefore, in order to process a predetermined workpiece or a component of a predetermined material, processing conditions are required such as what kind of tool is used, and what kind of rotation speed, feed speed, and the like are suitable for processing. That is, information on what kind of functional component is required.

Many of three-dimensional CADs have a function called feature-based modeling. Shape features in three-dimensional CAD are often described as follows. The shape of a product to be designed / produced is not simply a collection of points, lines and surfaces, but a collection of shapes each having a certain function. In the present specification, such a partial shape unit of a product that has some meaning in relation to a product function or a manufacturing method is called a shape feature.

[0005] Feature-based modeling is a method of constructing a solid model using shape features (features) such as holes, bosses, shells, drafts, and fillets. It is said that a solid model can be created. Further, as examples of the shape features, features such as "the radius of the hole can be changed" and "the property as a through hole is maintained even when the height of the base shape changes" are often emphasized. This is realized by a parametric function in which when a dimension value is changed, another dimension value is automatically changed so as to satisfy a predetermined constraint condition and the shape is reconstructed.

The above-mentioned constraint conditions must be specified by the user. In general, the conditions are insufficient and a solution cannot be uniquely determined, or conversely, the conditions are excessive and there is no solution satisfying all conditions. It is said that the problem of the completeness of the condition often burdens the user. However, once a model has been created, changing the dimensions of some of the model does not require rebuilding the model. The system reconfigures the geometry by changing only the dimensions that you want to change. Willing to. Further, by transmitting the design intention through the shape feature, the interface between subsystems can be improved, such as automation of machining planning. There is an advantage that the design can be easily changed or a similar product can be easily designed.

It has long been pointed out that the introduction of three-dimensional CAD has various advantages. There have been many reports of actual applications, but many have been applied to mold design. While expectations for the introduction of three-dimensional CAD are high in this way, many have pointed out that the spread is actually less advanced than predicted for the following reasons.

(1) Three-dimensional CAD design requires a sense of three-dimensional shape modeling, which is different from two-dimensional design. Therefore, it takes time to become proficient.

(2) Also, a three-dimensional CAD that can be operated by a so-called personal computer has appeared, and although the price has been reduced, there has been a problem that the price is higher and the cost is higher than the two-dimensional CAD.

(3) Although it is possible to leave an ambiguous part in a two-dimensional design drawing, a three-dimensional CA
In D, since it is necessary to create a complete shape,
There was also a problem that the ambiguous part could not be left and the flexibility of the design process was lost.

On the other hand, there has been reported an example in which NC data is automatically generated in two-dimensional CAD data currently in widespread use, such as in sheet metal processing or punching. In these examples,
A limited shape feature such as a hole is automatically extracted from drawing data for a limited use such as a sheet metal, and processing conditions are associated with each other.

A problem in coordinating such two-dimensional CAD and CAM is that figures drawn by CAD are not always accurate. The dimension values in the drawing are accurate,
Figures are not always drawn exactly. Also, communication between humans does not need to be accurate. However, when NC data is created from the CAD data, it must be accurate. In addition, it is necessary to consider a tolerance when obtaining the accuracy of the shape feature, and the use of the two-dimensional CAD data in the CAM field has not been advanced at present.

[0013]

As described above, in order to increase the efficiency by utilizing the data of the design process in the manufacturing process, not only mere shape data but also a group of shape groups each having a certain function. There is a need to capture the product and obtain from the design process a shape feature that is a partial shape unit of such a product. Although feature-based modeling in three-dimensional CAD is one of the methods, it has not been widely spread due to the difficulty of three-dimensional shape modeling and the price of three-dimensional CAD. On the other hand, attempts have been made to automatically detect shape features such as holes for limited applications such as sheet metal from two-dimensional CAD data, which is becoming more widespread. If it is not accurate, it may not work well and there is no sign of its spread.

An object of the present invention is to provide a method for extracting shape features from CAD data which can automatically extract shape feature parameters from widely used two-dimensional CAD data using a general-purpose shape feature detection device. It is in. Also,
By reading dimension values specified in the drawing instead of using graphic data, it is possible to accurately detect shape characteristic parameters even when a graphic is not drawn accurately.
It is another object of the present invention to provide a method for extracting shape features from AD data.

[0015]

According to a first aspect of the present invention, there is provided a method for extracting shape features from CAD data, comprising the steps of: A step of inputting CAD data as the two-dimensional data in detecting a shape feature that is a partial unit in the shape of the design object and has some meaning for the function of the design object; Extracting non-shape data including dimension lines and character information to extract shape data relating only to the shape of the object, including the shape feature, geometric constraint, and allowable dimension from the shape data Inputting a model defining a shape feature parameter; and determining a shape feature parameter based on a value of the shape feature parameter defined in the model. Detecting various shape features from the shape data in the CAD data, and detecting each shape feature parameter by detecting a dimension value defining a dimension of each shape feature. It is characterized by:

According to a second aspect of the present invention, there is provided a method for extracting shape features from CAD data, comprising the steps of:
After extracting the shape data, the method further includes a step of dividing each drawing data from various drawing data including a front view, a top view, and a side view included in the extracted shape data. I have.

According to a third aspect of the present invention, in the method of the first aspect, after the drawing data is divided, a center of a symmetrical graphic is selected from the divided drawing data. The method further includes a step of detecting a simplified expression on a design drawing including an expression in which a half is omitted from a line and automatically correcting the detected simplified expression.

According to a fourth aspect of the present invention, there is provided a recording medium on which a shape characteristic extracting program is recorded, based on a model given in advance, from a two-dimensional data for computer-aided design, as a partial unit in the shape of an object. CAD that detects shape features that have some meaning to the function of an object
A procedure for inputting CAD data as the two-dimensional data, which is used when extracting shape features from data, and removing non-shape data including dimension lines and character information in the CAD data. A step of extracting shape data relating only to the shape of the object, a step of inputting a model defining shape characteristics parameters including the shape characteristics, geometric constraint conditions, and allowable dimensions from the shape data; A procedure for detecting various shape features from the shape data in the CAD data based on the values of the shape feature parameters specified in the above, and a shape value by detecting a dimension value defining a dimension in each shape feature. And a procedure for detecting each of the characteristic parameters.

According to a fifth aspect of the present invention, in the recording medium according to the fourth aspect, after the step of dividing the drawing data, a front view is selected from the divided drawing data. 1 including top view, side view
It is characterized in that a procedure for dividing each drawing data is further recorded.

According to a sixth aspect of the present invention, there is provided a recording medium on which the shape characteristic extracting program according to the sixth aspect is recorded. The method further includes a step of detecting a simplified expression on a design drawing including an expression in which a half is omitted from the center line of the shape figure, and a step of automatically correcting the detected simplified expression. .

As described above, according to the present invention, the shape and geometric constraint conditions, the range of possible dimensions, etc., and the shape feature parameters to be detected which are defined and given by the user in advance are defined. Based on the model, two-dimensional CA
Each shape feature is obtained from the D data by pattern matching, and the shape feature is extracted by reading the value of the dimension value defining the size of each shape feature.

According to the first aspect of the present invention,
A model that prescribes the shape of each shape feature, geometric constraint conditions, a range of possible dimensions, and shape feature parameters to be detected is input in advance. First, CAD data is input, and only portions defining shape data such as straight lines, arcs, and ellipses are extracted from the data. In the shape feature detecting step, each shape defined in the model is extracted from the shape data. A shape feature is detected based on values such as a feature shape, a geometric constraint condition, and a possible size range. Finally, in the shape feature parameter detecting step, a dimension value defining the size of each shape feature is detected.

According to the second aspect of the present invention,
In the drawing data dividing step, a group of shape data is found from each of a plurality of drawing data such as a front view, a top view, and a side view included in the shape data extracted in the shape data extracting step. Are output as shape data of each drawing, and a shape feature is extracted in a shape feature detecting step based on the shape data.

According to the third aspect of the present invention,
From the shape data extracted in the shape data extraction step,
In the simplified expression correction step, a simplified expression in design drafting including an expression omitting a half from the center line of the symmetrical figure is detected, the simplified expression part is automatically corrected, and then the shape is detected in the shape feature detecting step. Extract features.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of a method for extracting shape features from CAD data according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows a CA according to a first embodiment of the present invention.
5 is a flowchart showing a processing procedure of a method for extracting a shape feature from D data. In FIG. 1, a step ST1 of inputting CAD data from a CAD data file created by two-dimensional CAD, and deleting non-shape (additional) data such as dimension lines and character information in the CAD data to obtain only shape data ST2 for extracting the shape and geometric constraint conditions of each shape feature, a range of possible dimensions, and the like, and a model ST3 for inputting a model defining the shape feature parameter to be detected. A step ST4 for detecting individual shape features from the shape data in the CAD data based on values such as the shape, geometrical constraint conditions, and possible dimensions of each defined shape feature; Detecting individual shape feature parameters by detecting dimension values defining the dimensions of the shape features;
ST5.

Step ST1 for inputting the CAD data
In the above, the CAD data of the CAD data file created by using the two-dimensional CAD is input, and the contents included in the CAD data file are converted into an internal data format and stored. Each CAD data file corresponds to a design drawing. FIG.
An example of the AD data file is shown.

In the example of the CAD data shown in FIG.
“LIN” represents a straight line, and the numbers following the three-dimensional coordinates represent the two end points of the straight line. Although three-dimensional coordinate values are used, Z coordinate values and 0 are set in data created by two-dimensional CAD. “ARC” represents an arc, and “ELL” represents an ellipse. The following numbers respectively represent the coordinate value of the center of the arc, the radius, the start angle, the end angle, the coordinate value of the center of the ellipse, the X-direction radius, the Y-direction radius, the start angle, the end angle, and the rotation angle. On the other hand, "D
“MP” specifies a dimension line parameter specifying a method of displaying a dimension value, and “LDV” specifies coordinates of a dimension line.

In step ST2 for extracting the shape data, only the portion representing the shape in the CAD data as described above is extracted. “LIN”, “ARC”, “EL
Only data records representing a shape such as L "are extracted.
As a result, non-shape (additional) data such as dimension lines, center lines, tolerance designations, and various types of character information are excluded from extraction targets. However, at this stage, strictly, data other than the shape data is still included. For example, a frame or the like surrounding a character may be created using a straight line, in which case the frame portion is represented by a “LIN” command.

In the step ST2 for extracting the shape data, after extracting only the data records representing the shapes such as "LIN", "ARC", "ELL", etc. As processing, each straight line
The connection relation between each shape element such as an arc and an ellipse is obtained. The connection relation means which shape element is connected to which shape element and where, and specifically, to obtain the following data. Connection relationship of shape element i Shape element connected at end point A and angle: straight line no.k (90 degrees) Shape element connected at end point B and angle: straight line no.l (90 degrees), arc no.c (0 degree) ) Shape element and angle connected in the middle: Straight line no.m (90 degrees) Shape element connected at two end points and angle at the time of connection at those end points are determined, and shape element connected at the middle and connected Find the angle of

For example, in the example shown in FIG.
The connection relation for no.l is as follows. FIG. 4 shows only the straight line no.m on the coordinates. End point A: straight line no.k (90 degrees) End point B: straight line no.m (45 degrees) Middle: none The above connection relationship is obtained, for example, as follows.

At each position (x, y) in the design drawing, an array through [y] [x] indicating whether or not a shape element passes.
Prepare through [y] [x]: No. of the shape element. (0- nothing passes) The value of each element of the array through [y] [x] is obtained as follows. First, all elements are initialized to zero. After that, for each shape element, the passing coordinate value is calculated from one end point to the other end point, and each time, the value of the corresponding element of the array through [y] [x] is calculated as
No. of the shape element And By performing the above operation for all the shape elements, the array through [y]
All values of [x] will be set.

The connection relation of each shape element can be obtained as follows using the value of the array through [y] [x] obtained as described above.

First, for each shape element, the value of the array through [y] [x] near the coordinate value (x, y) of each end point is examined. The neighborhood is, for example, a 5 × 5 region around the coordinate value (x, y) (that is, [x−2] [y−2] to [x−y]).
+2] [y + 2]). Find all non-zero values in the neighborhood. Those values are a set of connected shape elements at the end points. By examining not only the coordinates of the end point but also the vicinity thereof, two slightly separated straight lines are regarded as being connected.
Therefore, the problem that the CAD data cannot be used for the CAM or the like due to the slight deviation of the straight line from the other straight lines is avoided, and the slight deviation between the straight lines does not pose a problem.

When the connected shape elements are obtained, the angle between them and the shape elements is calculated. After performing the above processing for both end points, in order to obtain a shape element connected between the end points, the same processing as described above is performed on each coordinate value between the end points to obtain a connected shape element. Then, by registering them as shape elements to be connected in the middle, the connection relationship data for each shape element as described above is completed. After the connection-related data is completed,
Deletes shape elements other than shape data in shape elements such as straight lines and arcs such as frames surrounding characters. The framework surrounding the characters is deleted as follows:

The frame surrounding the character is a rectangle containing the character inside. Therefore, first, a rectangle is detected using the connection relation data described above. For a rectangle, four straight lines connected at an angle of 90 degrees at an end point may be searched for one that is continuous and returns to the original straight line. For each shape element,
Since the shape element to be connected at the end point and the angle formed by the end point are already in the connection relation data, for all the straight line elements, the connection form element at the end point is examined. A rectangle can be detected by repeating the search for examining the connection data of the connected straight line and checking whether there is a straight line connected at 90 degrees at the other end point. Further, in order to prevent a rectangle as a shape from being erroneously deleted, the area of each rectangle is calculated and stored each time a rectangle is detected in order to add a determination based on the area.

On the other hand, the CAD data file also contains data on characters. This includes not only the character code but also the position of the character. A record designating a character position is searched in advance from the CAD data file, and the position of all character strings is stored as an array.

When a rectangle is detected by the above method,
It checks all the values in the array specifying the position of the character and determines whether the character is inside the rectangle. If a character exists and its area is smaller than a certain area, the rectangle is removed from the shape data as a frame surrounding the character. By performing the above processing for all the detected rectangles, the frame surrounding the character is excluded from the shape data.

In step ST3 for inputting the model, a model given in advance is input and converted into internal data. As an example of this model conversion, FIG. 6 shows a model of a part having a hole in a plate having a thickness of 10 mm as shown in FIG. Further, the definition in the model file is performed by the program data shown in FIG. 7 using a dedicated description language.

The step ST3 of inputting a model has a procedure (function) for obtaining parameters of a basic geometric figure such as a circle, a line segment (straight line), an ellipse, and a rectangle. For example, Diameter () is the procedure to get the diameter of a circle, Center
() Is a procedure for obtaining the center coordinates of the circle, mun () is a procedure for obtaining the number of argument array elements, Length () is a procedure for obtaining the distance between two coordinate values serving as arguments, and GetRec
Para () is a procedure for obtaining the lengths of the two sides of the rectangle, and Thickness () is a procedure for obtaining the thickness of the plate-like work by description such as “t10” in the drawing. Also, I
n () is a function for determining whether the shape element of the first argument is included in the shape element of the second argument. In addition, Diameter
Procedures such as (), Center (), and GetRecPara () refer to the shape data but read and output the dimension values from the CAD data. The execution of each of the above procedures is performed in step ST4 for detecting a shape feature and step ST5 for detecting a shape feature parameter. The program is also included in step ST4 for detecting a shape feature and step ST5 for detecting a shape feature parameter. In step ST4 for detecting shape features, basic shape elements such as rectangles and circles are first detected and registered as system variables as preprocessing from the shape data obtained in step ST2 for extracting shape data. , _Rectangle, _circle
It can be used with such a name.

A feature item is defined for a shape feature, and a parameter item is defined for a shape feature parameter. The above-mentioned model constants are obtained from the shape data such as circles and rectangles included in the CAD data file, which have been converted into the internal data format in step ST1 of inputting CAD data and step ST2 of extracting shape data. This defines the procedure for obtaining the value of the shape feature parameter using the above procedure used in step ST3 of inputting the shape feature. Most of the steps are performed in step ST4 for detecting the shape feature as it is and in step ST5 for detecting the shape feature parameter. Processing algorithm. In the model input step ST3, the definition input from the model file is processed into a form usable in the shape feature detection step ST4 and the shape feature parameter detection step ST5. The above-described model definition uses the procedure (function) used in the model input step ST3, but has almost the same format as the C language, and can be easily used in the shape feature detection step ST4 and the shape feature parameter detection step ST5. It is possible to take any other form.

The operation of the shape feature detecting means 4 and the shape feature parameter detecting means 5 based on the above-described model definition will be described below. The shape feature detection means 4 detects a plate and a hole as shape features from the shape data output from the shape data extraction means 2. After detecting a rectangle, a circle, or the like by the preprocessing, first, in order to detect a plate, it is determined whether or not each arc is included in the inside of the rectangle.
As a result, for each rectangle, the number of arcs included therein is determined. A rectangular “plate” having the largest number of arcs included therein is assumed.

In the detection of the hole, since the arc included in the shape data has already been obtained at the time of detecting the plate, it is determined whether the hole is included in the rectangle recognized as the “plate”. Is determined, and all arcs included in the “plate” are defined as “holes”.

It is defined in the model that the plate and the hole, which are the shape features, are a rectangle and an arc, respectively. In addition to the fact that each shape feature is a rectangle and an arc, the number of the element in the rectangle, the number of the arc included in the shape data, and the number of holes are the shapes. It is obtained in the feature detection step ST4 and used in the shape feature parameter detection step ST5.

The shape characteristic parameter detecting means 5 detects the value of the shape characteristic parameter shown in FIG. 6 as defined in the model.

Thickness () searches the CAD data for character string data that starts with “t” and has a number immediately following it, and is located outside the shape element specified by the argument. Read the number in the string in and answer. Therefore, the character string "t10" is detected from the CAD data of the drawing shown in FIG. 5, a numeral 10 is answered, and the result is substituted for the plate thickness value "plate_depth".

GetRecPara () returns the lengths of the two sides of the rectangle. However, the dimension values described in the drawings are read and output instead of the graphic information. The dimension value is defined as a pair with the dimension line in the CAD data. Therefore, first, a dimension line is searched, and a dimension value defined as a pair with the dimension line is read. The coordinate values of both ends of the dimension line are defined as parameters. For example, C of two-dimensional CAD
In S-21, the dimensional line as shown in FIG.
The coordinate values of B, C, D, and E are defined as parameters.

Then, the coordinate values, which are the parameters of each dimension line, are checked, and the coordinate values (x0,
y0) (x1, y1) is searched for. If two of the five points, which are the parameters of the dimension line, are found within a predetermined distance from the coordinate values (x0, y0) (x1, y1) at both ends of the long side of the rectangle, the dimension line is determined. Set the dimension line as the dimension point that specifies the length of the long side of the rectangle, read the paired dimension value, and read the plate length pla
te_length. Although there are two long sides of the rectangle, the above processing is performed on both sides. Since there is usually only one dimension line specifying the length of the plate (the length of the long side of the rectangle), only one dimension line is detected by the above processing. The same processing is performed for the length of the short side of the rectangle, and the width of the plate is set to plate_width.

Diameter () in FIG. 7 returns the diameter of the arc specified in the argument. Again, a value specified by the dimension value, not the value of the shape data, is detected and answered. Among the dimension lines near the center position of the arc specified by the argument, a character string that includes a value such as φ that specifies the diameter is detected in the character string paired with that dimension line, The diameter of the arc is determined by reading the number.

Center () in FIG. 7 answers the relative position from the dimension reference position of the shape element specified in the argument. Again, a value specified by the dimension value is detected instead of the value of the shape data, and an answer is given. For the dimension reference position, search for the character string “dimension reference position”, search for a dimension line extending from the vicinity,
It can be obtained by finding the point where the line is inserted on the dimension line. When the shape element specified by the argument is a circular arc, Center () searches for a center line passing through the center position in the shape data. When two orthogonal center lines that intersect near the center position are determined, for each of them, a pair of dimension values or a dimension line where one end point of the dimension line exists on an extension of the center line is determined, and the pair is determined. Find the dimensional value of.

In general, since the distance from the dimension reference position is not always specified, it is necessary to add a plurality of dimension values or subtract from the entire dimension. Therefore, we sequentially search for the dimension value toward the dimension reference position,
The distance from the dimension reference position is obtained by adding all the values.
For example, regarding the coordinate values in the left-right direction of the second hole from the left in FIG. 5, first, a vertical center line passing through the center of the second hole from the left is found, and one end of the dimension line extends on the extension. Find a dimension line. Although there are two in FIG. 5, a dimension line extending in the direction of the dimension reference position is selected.
The dimension value 80 is stored. Since this dimension line does not specify the distance from the dimension reference position, searching for a dimension line that specifies the length in the dimension line direction from the end of this dimension line, the dimension with a paired dimension value of 100 A line is found. Since one end of this dimension line has reached the dimension reference position, 1
By adding 00 and 80, 180 becomes the horizontal distance (coordinate value) from the dimension reference position.

As for the vertical coordinate values, first, a distance 70 from the opposite end of the dimension reference position is found. Since the distance to the dimensional reference position is not found, the dimensional value is not searched for to the dimensional reference position, and 70 is subtracted from 250 by using the already determined width 250 of the plate.
You get a value of 80. The above processing is performed for the leftmost hole and the second hole from the left to determine the center position, and then the distance between the center positions of the two arcs is calculated.
ace. In this way, individual shape feature parameters can be obtained.

Next, the CAD according to the second embodiment of the present invention will be described.
The method for extracting the shape feature from the tank will be described with reference to the flowchart of FIG. 9 and the drawing of the tank of FIG. The shape feature extraction method from CAD according to the second embodiment shown in FIG. 9 divides each drawing data from a plurality of drawing data between steps ST2 and ST5 constituting the first embodiment. Step ST3 is added. In FIG. 9, the presented feature extraction method according to the second embodiment uses a two-dimensional C
CAD data input step ST1 for inputting a CAD data file created by AD, shape data extraction step ST2 for deleting dimension lines, character information, etc. in CAD data and extracting only shape data, front view / top view Drawing data dividing step ST3 for dividing each drawing data from a plurality of drawing data such as figures and side views, the shape and geometric constraint conditions of each shape feature, the range of possible dimensions, and the shapes to be detected A model input step ST5 for inputting a model defining feature parameters; and CAD data in the CAD data based on values such as the shape and geometric constraint conditions of each shape feature defined in the model and a range of possible dimensions. A shape feature detection step ST6 for detecting each shape feature from the shape data, and each shape feature is detected by detecting a dimension value defining the size of each shape feature. And a shape feature parameter detecting step ST7 for detecting parameters.

In the CAD data input step ST1, a CAD data file created using two-dimensional CAD is input, and the contents included in the CAD data file are converted into an internal data format and stored. Each CAD data file corresponds to a design drawing.

In the shape data extraction step ST2,
Only the part representing the shape in the CAD data as described above is extracted. Only data records representing shapes such as “LIN”, “ARC”, and “ELL” are extracted. as a result,
Dimension lines, center lines, tolerance designations and various character information are excluded. However, at this stage, data other than the shape data is still included. For example, a frame or the like that surrounds a character may be created using a straight line. In this case, the frame portion is represented by a “LIN” command.
The shape data extraction step ST2 includes “LIN”, “AR
After extracting only data records representing shapes such as "C", "ELL", etc., as a pre-process for use in subsequent detection of shape features, etc., a connection relationship between shape elements such as straight lines, arcs, ellipses, etc. Ask for.

When the connected shape elements are obtained, the angle formed between these shape elements is calculated. After performing the above processing for both end points, in order to obtain a shape element connected between the end points, the same processing as described above is performed on each coordinate value between the end points to obtain a connected shape element. Then, by registering as a shape element to be connected in the middle, connection relation data for each shape element as described above is completed.

After the connection-related data is completed, those other than the shape data in the shape elements such as straight lines and arcs such as a frame surrounding the character are deleted. The framework surrounding the characters is deleted as follows:

The frame surrounding the character is a rectangle containing the character inside. Therefore, first, a rectangle is detected using the connection relation data described above. For a rectangle, four straight lines connected at an angle of 90 degrees at an end point may be searched for one that is continuous and returns to the original straight line. For each shape element,
Since the shape elements to be connected at the end points and the angles are already in the connection relation data, for all the linear elements,
The connected shape element at the end point is examined, and if there is a straight line connected at 90 degrees, the connection data of the connected straight line is checked, and it is determined whether there is another straight line connected at 90 degrees at the other end point. A rectangle can be detected by repeating the search for checking. Further, in order to prevent a rectangle as a shape from being erroneously deleted, the area of each rectangle is calculated and stored each time a rectangle is detected in order to add a determination based on the area.

On the other hand, the CAD data file also contains data on characters. This includes not only the character code but also the position of the character. A record designating a character position is searched in advance from the CAD data file, and the position of all character strings is stored as an array.

When a rectangle is detected by the above-described method, all values in the array specifying the position of the character are checked to determine whether or not a character exists inside the rectangle. If a character exists and its area is smaller than a certain area, the rectangle is removed from the shape data as a frame surrounding the character. By performing the above processing for all the detected rectangles, the frame surrounding the character is excluded from the shape data.

In the step ST3 for dividing the drawing data, the x, y directions (horizontal direction and vertical direction) of each linear element, arc element and elliptic element included in the shape data extracted in the step ST2 for extracting the shape data And calculate the sum of all the projections. For the design drawing shown in FIG. 10, the sum of the projections of the shape data in the x and y directions is as shown in FIG. If the projection in the x direction and the projection in the y direction are each divided into two or more, the drawing is divided at the center of the division. In FIG. 11, since the projection in the y direction is divided into two, the shape data is divided into two groups at the divided center. After the division, the projection in the x and y directions is again performed for each group, and if the projection is divided into two or more, the shape data is grouped again. The above procedure is repeated until division becomes impossible. In FIG. 11, after the upper and lower parts are divided into two parts, the processing ends without further division. As a result of the above procedure, if a plurality of drawings such as a front view, a top view, a side view, and a cross-section are included in one CAD data, division into each drawing is performed. Complete.

In the model input step ST5, a given model is input and converted into internal data. FIG. 10 is an assembly drawing of a tank including a pipe, a flange, and a post flange. Although it is possible to read the shape characteristics of the components described in the assembly drawings together with the assembly drawings, the detailed shapes and dimensions of the components are generally described in the component drawings. Also in the second embodiment,
An example in which the following shape characteristic parameters such as the mounting position of each component as the shape characteristic of the tank will be described with reference to FIG.

The shape feature parameters and shape features in this example are as shown in FIG. The definitions used in the model file are made using a dedicated description language as shown in FIGS. 13 and 14 show a program for defining a pipe, FIG. 15 shows a program for defining a flange, FIGS. 16 and 17 show a program for defining a post flange, and FIG. 18 shows a tank length and a post flange. Is a program showing parameters related to the angle and the like.

The definition of the pipe is made for each of the side view and the sectional view. In the side view, areas are first obtained for all the rectangles in the side view, and the rectangles are sorted in descending order of the area. After that, it is determined whether there is no rectangle in contact with the rectangle in order from the rectangle with the largest area, and if there is, the array touch_l
Record in ist []. For a rectangle that has two or more touching rectangles, the function CheckPipe () determines whether or not there is a touching rectangle on two opposite sides of the rectangle. There is a definition. Area () finds the area of the shape element of the argument. Sort () sorts the array specified by the first argument using the value of the array specified by the second argument as a key. Touch () is a function that determines whether the shape element specified by the first argument and the shape element specified by the second argument are in contact, and Align () is a function that determines whether the shape element specified by the first argument is This is a function for finding a straight line tangent to the shape element specified by two arguments. The straight line can be answered by any of the numbers 0, 1, 2, and 3 indicating which side of the rectangle is, so that a difference of 2 corresponds to two opposite sides of the rectangle. I do. Each function is one of the basic shape operation functions used in the shape feature detection step ST6 and the shape feature parameter detection step ST7.

The definition in the sectional view of the pipe is as follows. The circle having the largest area in the sectional view is obtained, the circle having the second largest area is obtained, and the circle having the largest area is sandwiched between the circle having the largest area and the second largest circle. Area.

The definition of the flange is only a side view, and is substantially the same as that of a pipe. The definition of the post flange is made for the side view and the sectional view. The definition of the post flange is a rectangle on the pipe in the side view, and an ellipse (including the circle) on the side of the pipe
That is. The rectangle on the pipe is
For all rectangles, a determination is made as to whether or not they are in contact with the pipe. If they are, they are checked for two flanges. If the rectangle is neither flange, it is considered a post flange. Also, for the ellipses on the side of the pipe, for all ellipses, it is determined whether or not the ellipse is included in the pipe, and after obtaining the ellipses included in the pipe, those ellipses that further include the ellipse Judgment is made as to whether or not there is a double ellipse, and the post flange is found. The definition of the post flange in the cross-sectional view is a region surrounded by three straight lines and an arc, and satisfies the following constraints. A straight line a and a straight line b are perpendicular; a straight line b and a straight line c are perpendicular; a straight line c and an arc d are in contact; a straight line a and an arc d are in contact; And there are two parallel straight lines between the straight line a and the straight line c.

Solve () is a solver that returns a list of shape elements satisfying the constraint specified in the first argument.
It is the same kind of solver that solves the geometric constraint problem used in parametric modeling of three-dimensional CAD. ve
rtical indicates vertical and touch indicates touch. Parallel2 () indicates that the straight lines specified in the arguments are arranged in the order specified in the arguments and are parallel.

In the shape feature detecting step ST6, first, a parts table is checked. The BOM is a character string surrounded by rectangles.
By searching for a character string including a predetermined character string such as a figure number, a name, a number, etc., the CAD data can be easily detected. If the entire parts table can be detected, the drawing number, name, and number of each part in the parts list are the character string below the character string "drawing number" and the character string below the character string "name". , "Numerals" can be detected by detecting a character string below the character string. Through the above processing, the number of components and each component can be determined.

Next, according to the above-described model definition input by the model input means 9, the shape feature, pipe, flange, and post flange corresponding to each part are detected. Those that are defined in both the side view and the cross-sectional view are detected in both drawings.

The pipes are first sorted and rearranged in the side view, starting from the one with the largest area, and then searching for the rectangle in which two or more rectangles are in contact on both sides in order from the one with the largest area. Therefore, the processing is ended to make a pipe. In the sectional view, a circle having the largest area and a circle having the second largest area are found, and a region surrounded by the two circles is defined as a pipe.

After sorting and rearranging the flanges in the side view starting from the one having the largest area in the side view, searching for the rectangle in which two or more rectangles are in contact on both sides in order from the one having the largest area, and processing it if found. Is completed, and the rectangles contacting on both sides are defined as flanges.

In the side view, if there is a rectangle which is located above or below the pipe and which is not a flange in the side view, the post flange is used as a post flange, and among the ellipses located inside the pipe, an ellipse is further added. If any exists, it is also regarded as a post flange. In the cross-sectional view, the straight line a and the straight line b are perpendicular, the straight line b and the straight line c are perpendicular, the straight line c is in contact with the arc d, the straight line a is in contact with the arc d, Parallel to the straight lines a and c, and the straight lines a and c
A straight line a, a straight line b, a straight line c, and a circular arc d satisfying the constraint that there are two parallel straight lines between them are obtained, and a region surrounded by them is defined as a post flange.

In the shape feature parameter detection step ST7, the following shape feature parameters such as the mounting position of each component are detected. The shape characteristic parameters are: the total length of the tank, the number of post flanges, and the mounting position and mounting angle of each post flange.

The total length of the tank is the distance between the two longest sides between the rectangles corresponding to the two flanges in the side view. After all, instead of the shape data, the two furthest sides of the rectangle corresponding to the two flanges are obtained, a dimension line that specifies the extension thereof is searched, and a pair of dimension values is obtained. The number of post flanges is determined from the bill of materials. For the mounting position of each post flange, the center of a rectangle or an ellipse in the side view is obtained, and a vertical centerline passing through the center is searched for. A dimension line defining the length from the extension line to the dimension reference position is found, and a pair of dimension values is obtained. For the mounting angle of each post flange, a character string "angle reference position" is first searched for in the cross-sectional view, and a point to which a dimension line paired with the character string is found. Search for an arc dimension line that specifies the angle from that point. Since it is not found in the case of FIG. 10,
In that case, the mounting angle is calculated by searching for a center line passing through the center of each post flange and obtaining the angle formed with the center line passing through the angle reference position. Angle ()
This function calculates the angle of the shape element specified in the argument from the angle reference position, and calculates the angle formed by the center line passing through the center of the shape element specified in the argument and the center line passing through the angle reference position. Further, it searches for an arc dimension line, and if an angle is specified, reads the dimension value.

The drawing includes a balloon element () for designating each part. In the case where a plurality of candidates remain in the procedure defined in the above-described model, it is effective to add a process for eliminating ambiguity by using information indicated by the balloon element.

Next, a method for extracting shape features from CAD according to a third embodiment of the present invention will be described with reference to the drawings. FIG. 19 is a flowchart showing a schematic configuration of the third embodiment, and FIG. 20 is an example of a drawing of a tank including a drawing in which the side opposite to the center line is omitted.

In FIG. 19, the CA according to the third embodiment
The shape feature extraction method from D includes a CAD data input step ST1 for inputting a CAD data file created by two-dimensional CAD, and extracting only shape data by deleting dimension lines, character information, and the like in the CAD data. A shape data extraction step ST2, a drawing data division step ST3 for dividing each drawing data from a plurality of drawing data such as a front view, a top view, and a side view, and a simplified expression in each drawing is detected and used. A simplified expression correcting step ST4 for correcting to missing shape data, and a model input step ST5 for inputting a model defining a shape, a geometric constraint condition, a range of possible dimensions, etc. of each shape feature, and a shape feature parameter to be detected. And the values in the CAD data based on values such as the shape of each shape feature specified in the model, geometrical constraints, and the range of possible dimensions. A shape feature detection step ST6 for detecting each shape feature from the shape data, and a shape feature parameter detection step ST for detecting each shape feature parameter by detecting a dimension value defining the size of each shape feature
7 is comprised.

In the CAD data input step ST1, a CAD data file created using two-dimensional CAD is input, and the contents included in the CAD data file are converted into an internal data format and stored. 1
Each CAD data file corresponds to a design drawing.

In the shape data extraction step ST2,
Only the part representing the shape in the CAD data as described above is extracted. Only data records representing shapes such as “LIN”, “ARC”, and “ELL” are extracted. as a result,
Dimension lines, center lines, tolerance designations and various character information are excluded. However, at this stage, data other than the shape data is still included. For example, a frame or the like surrounding a character may be created using a straight line, in which case the frame portion is represented by a “LIN” command. In this shape data extraction means step ST2,
After extracting only data records representing shapes such as "LIN", "ARC", and "ELL", each shape such as a straight line, an arc, and an ellipse is used as a pre-process for use in subsequent detection of shape features. Find connection relationships between elements.

When the connected shape elements are obtained, the angle between them and the shape elements is calculated. After performing the above processing for both end points, in order to obtain a shape element connected between the end points, the same processing as described above is performed on each coordinate value between the end points to obtain a connected shape element. Then, by registering as a shape element to be connected in the middle, connection relation data for each shape element as described above is completed.

After the connection-related data is completed, the shape data such as a frame or a straight line or an arc, such as a frame surrounding the character, is deleted. The deletion method is the same as in the first and second embodiments.

In the drawing data dividing step ST3,
Each of the linear elements, arc elements, and ellipse elements x, x included in the shape data obtained in the shape data detecting step ST2.
Projections in the y direction (left and right directions and up and down directions) are taken, and the sum of all projections is obtained. If the projection in the x direction and the projection in the y direction are each divided into two or more, the drawing is divided at the center of the division. In FIG. 20, since the projection in the y direction is divided into two, the shape data is divided into two groups at the center of the division. After the division, the projection in the x and y directions is again performed for each group, and if the projection is divided into two or more, the shape data is grouped again. The above procedure is repeated until division becomes impossible. In FIG. 20, after being divided into two parts vertically, the lower half of the figure is further divided into two parts left and right. As a result of the above procedure, if a plurality of drawings such as a front view, a top view, a side view, and a cross-section are included in one CAD data, division into each drawing is performed. Complete. 20 (b) is a front view of the tank, and FIG. 20 (a) is the WW of FIG. 20 (b).
FIG. 20C is a cross-sectional view taken along a line, and FIG.

In the simplified expression correcting step ST4, for each center line, it is determined whether or not the shape data is symmetric with respect to the center line. As shown in FIG. When an object that is not symmetrical is detected, the omitted shape is obtained from the difference in the shape on both sides of the center line, and the shape data is added to the omitted side, so that the omitted shape is omitted. The existing shape data is corrected to complete shape data.

In a model input step ST5, a given model is input and converted into internal data. Shape feature detection step ST6 and shape feature parameter detection step
The processing operation in ST7 is the same as the processing in the shape feature extraction method according to the second embodiment, and a duplicate description will be omitted. As described above, the shape feature can be accurately extracted using the CAD data also by the shape feature extraction method according to the third embodiment.

Finally, FIG. 2 shows a recording medium on which a shape feature extraction program according to the fourth embodiment of the present invention is recorded.
This will be described with reference to FIG. The recording medium 1 or 2 on which the shape characteristic program according to the fourth embodiment is recorded,
The readout is performed by the recording medium driving device 16 or 17 mounted on the computer system 10 as shown in FIG. 21 to extract the shape feature.

As shown in FIG. 21, the computer system 10 includes a computer main body 11 housed in a housing such as a Minnie tower, a display device 12 such as a CRT (Cathode Ray Tube), and a recording output device. Printer 13, a keyboard 14a and a mouse 14b as input devices, a floppy disk drive 16 for driving a floppy disk to read information in the floppy disk 1 as the recording medium, and a CD-ROM as the recording medium. ROM (Compact Disk-Read Only)
Memory) 2 is provided with a CD-ROM drive 17 for reading information from the memory 2.

FIG. 2 is a block diagram showing these components.
As shown in FIG. 2, an internal memory 15 such as a RAM (Random Access Memory) and an external memory such as a hard disk unit 18 are further provided in a housing in which the computer main body 11 is stored. The floppy disk 1 on which the shape characteristic extraction program according to the fourth embodiment is recorded is inserted into a slot of the floppy disk drive 16 and installed in the computer main body 11 by a predetermined procedure, as shown in FIG. The recording medium on which the program is recorded is not limited to the floppy disk 1 but may be a CD-ROM 2 or an MO (Magneto Optical) disk (not shown).
An optical disk, a DVD (Digital Versatile Disk) or the like may be used.

The computer-readable shape extraction program recorded on the recording medium according to the fourth embodiment is the same as the shape extraction method according to the first to third embodiments shown in FIGS. Steps such as 19 are recorded in the form of a computer program as shown in FIGS. 7 and 13 to 18, and only the procedures thereof will be described below, without redundant description.

That is, the recording medium on which the shape feature extraction program according to the fourth embodiment is recorded is obtained from computer-assisted design two-dimensional data based on a given model in partial units of the shape of an object. The purpose is to extract a shape feature from a design object used when extracting a shape feature from CAD data for detecting a shape feature having some meaning to the function of the object, and to input the CAD data as the two-dimensional data. A step of extracting non-shape data including dimension lines and character information in the CAD data to extract shape data relating only to the shape of the object; A step of inputting a model defining a shape characteristic parameter including a constraint condition and a dimension allowable range; A procedure for detecting various shape features from the shape data in the CAD data based on the values of the shape feature parameters, and detecting each of the shape feature parameters by detecting a dimension value that defines a dimension in each shape feature. The detection procedure is recorded, and the program is executed by a computer to extract a shape feature to be designed.

The recording medium on which the shape characteristic extraction program according to the fifth embodiment is recorded is included in the extracted shape data after the procedure for extracting the shape data in the recording medium on which the above-described procedures are recorded. A procedure for dividing each drawing data from various drawing data including a front view, a top view, and a side view is recorded.

Further, the recording medium on which the shape characteristic extraction program according to the sixth embodiment is recorded is a recording medium on which the procedure of the fifth embodiment is recorded, after the procedure of dividing the drawing data, A procedure for detecting a simplified expression on a design drawing including an expression omitting a half from a center line of a symmetrical figure from data and automatically correcting the detected simplified expression is recorded. Things.

[0092]

As described above, the CAD according to the present invention is
According to the shape feature extraction method from the data, the shape feature parameter is automatically determined from the widely spread two-dimensional CAD data by reading the dimension value defining the size of each shape feature from the CAD data. The shape feature parameter can be extracted by a general-purpose shape feature detection device to be extracted.

Instead of using graphic data,
By reading the dimension values specified in the drawing, even if the figure is not accurately drawn, the shape characteristic parameters can be accurately detected, whereby the CAD / CAM integration and the CAD data It is expected that the efficiency of work will be improved in various applications such as automatic generation of NC programs, automation of production planning, and arrangement of jigs and materials.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a method for extracting shape features from CAD data according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram illustrating an example of a CAD data file used in the first embodiment.

FIG. 3 is an explanatory diagram illustrating a connection relationship between shape elements according to the first embodiment.

FIG. 4 is an explanatory diagram showing a part of the connection relationship in FIG.

FIG. 5 is a plan view showing specific shape features to be extracted in the first embodiment.

FIG. 6 is an explanatory diagram showing internal data of a specific model shown in FIG. 5;

FIG. 7 is an explanatory diagram showing definitions in a model in the first embodiment in a dedicated description language.

FIG. 8 is an explanatory diagram illustrating a coordinate system of dimension lines according to the first embodiment.

FIG. 9 is a flowchart illustrating a shape feature extraction method according to a second embodiment of the present invention.

FIGS. 10A and 10B are a cross-sectional view and a front view showing, in a design drawing, specific shape features to be extracted in a second embodiment.

11A and 11B are a sectional view and a front view specifically showing a drawing dividing step in the second embodiment.

FIG. 12 is an explanatory diagram showing internal data of the specific model shown in FIGS. 10 and 11;

FIG. 13 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 14 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 15 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 16 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 17 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 18 is an explanatory diagram showing definitions in a model in a second embodiment in a dedicated description language.

FIG. 19 is a flowchart illustrating a shape feature extraction method according to a third embodiment of the present invention.

20A is a sectional view, FIG. 20B is a front view, and FIG. 20C is a side view showing specific shape features to be extracted in the third embodiment.

FIG. 21 is a perspective view showing a computer system using a recording medium recording a shape extraction program according to the fourth to sixth embodiments of the present invention.

FIG. 22 is a block diagram showing a computer system using a recording medium recording a shape extraction program according to the fourth to sixth embodiments of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Recording medium (floppy disk) 2 Recording medium (CD-ROM) ST1 CAD data input step ST2 Shape data extraction step ST3 Drawing data division step ST4 Simplified expression correction step ST5 Model input step ST6 Shape feature extraction step ST7 Shape feature parameter detection step

Claims (6)

[Claims] 1. A shape feature that is a partial unit in a shape of a design object and has some meaning to a function of the design object is detected from two-dimensional data for computer-aided design based on a given model. CAD
In the method for extracting a shape feature from data, a step of inputting CAD data as the two-dimensional data, and deleting non-shape data including dimension lines and character information in the CAD data, and removing only the shape of the design target Extracting shape data relating to the shape characteristics, geometric constraints, and dimension allowable ranges from the shape data. Detecting various shape features from the shape data in the CAD data based on a value of the shape feature parameter; and detecting a dimension value that defines a dimension of each shape feature, thereby forming each of the shape feature parameters. Detecting the shape feature from the CAD data. 2. A step of, after extracting the shape data, dividing each drawing data from various drawing data including a front view, a plan view, and a side view included in the extracted shape data, 2. The method according to claim 1, further comprising the steps of: 3. After the drawing data is divided, a simplified expression in design drawing including an expression in which half of the center line of a symmetrical figure is omitted from the divided drawing data is detected.
3. The method according to claim 2, further comprising the step of automatically correcting the detected simplified expression. 4. A CAD for detecting, from a two-dimensional data for computer-aided design, a shape feature which is a partial unit in the shape of an object and has some meaning for the function of the object, based on a given model. A storage medium storing a shape feature extraction program used when extracting shape features from data, comprising: a step of inputting CAD data as the two-dimensional data; and a step of inputting dimension lines and character information in the CAD data. A step of extracting shape data relating to only the shape of the object by deleting non-shape data including the shape data, and a model defining shape feature parameters including the shape feature, geometric constraint conditions, and dimensional tolerance range from the shape data Based on the value of the shape feature parameter defined in the model, A procedure for detecting various shape features from the shape data; and a procedure for detecting each of the shape feature parameters by detecting a dimension value defining a dimension of each shape feature. Medium. 5. After the step of extracting the shape data, each drawing data is divided from various drawing data including a front view, a top view, and a side view included in the extracted shape data. The recording medium according to claim 4, further comprising a step of performing the shape feature extraction program. 6. After the step of dividing the drawing data, a simplified expression in design drawing including an expression in which half of the symmetrical figure is omitted from the center line is detected from the divided drawing data. The recording medium according to claim 5, further comprising a step of automatically correcting the detected simplified expression.