JP5667818B2 - Three-dimensional shape skeleton model creation method and apparatus, and three-dimensional shape dimension measurement method and apparatus - Google Patents

Description

  The present invention relates to a method and apparatus for creating a skeleton model based on image data obtained by photographing a three-dimensional shape, and a method and apparatus for measuring a dimension of a three-dimensional shape using the skeleton model.

  When a three-dimensional shape is quantitatively grasped, it is performed on the basis of an image obtained by photographing a three-dimensional shape as the captured image becomes higher in definition and accuracy, and the image processing technology is increased in speed and capacity. It is coming.

  For example, it has been proposed to measure measurement data quantitatively based on image data obtained by photographing a human body from a plurality of directions. In Patent Document 1, a lattice pattern is projected onto an object to be measured to photograph a deformed lattice image observed on the object surface, and the photographed deformed lattice image and the three-dimensional coordinates of the surface of the measurement object are represented by a relational expression. It describes that by processing, the three-dimensional coordinates of the surface of the measurement object are extracted from the coordinates of the lattice lines of the deformed lattice image, and the human body is measured three-dimensionally. Further, in Patent Document 2, an object is photographed from a plurality of directions to create a three-dimensional surface shape, a measurement part is designated for the photographed two-dimensional image data, and a three-dimensional position corresponding to the designated measurement part is designated. It is described that the three-dimensional shape data in the vicinity of the three-dimensional position is calculated and the perimeter of the obtained three-dimensional shape data is calculated to calculate the size of the measurement site.

Japanese Patent No. 2868985 JP 2003-269921 A

  In the above-mentioned patent documents, it is possible to quantitatively grasp the three-dimensional shape by calculating the three-dimensional coordinates of the human body surface using images obtained by photographing the three-dimensional shape of the human body from a plurality of directions. Since the characteristic part of the three-dimensional shape is not clearly defined based on the calculated three-dimensional coordinates, the measurement position in the characteristic part such as the bust and the waist varies depending on the user. Therefore, there is a problem that accurate dimension measurement cannot be performed at a characteristic part of the three-dimensional shape.

  Therefore, the present invention provides a method and apparatus for creating a three-dimensional shape skeleton model capable of accurately setting a characteristic part based on three-dimensional shape data of a target object, and a three-dimensional shape dimension measurement method using the skeleton model. And an object of the present invention.

A three-dimensional skeleton model creation method according to the present invention is a three-dimensional skeleton model creation method for creating a skeleton model from three-dimensional shape data of a target object by a programmed computer. Performing center axis conversion processing on the contour shape data extracted corresponding to the contour of the target object viewed from a plurality of different directions using the three-dimensional shape data consisting of coordinate positions, and extracting the center axis data; And causing the computer to execute a step of creating skeleton model data in which a plurality of extracted central axis data are associated with feature points corresponding to nodal points of the central axis data .

The 3D-shaped skeleton model creation device according to the present invention uses a 3D shape data composed of 3D coordinate positions on the surface of the target object to extract a contour shape corresponding to the contours of the target object viewed from a plurality of different directions. A central axis extraction unit that performs central axis conversion processing on the data and extracts the central axis data expressed as a set of points, and associates the extracted multiple central axis data with feature points corresponding to the nodal points of the central axis data And a skeleton creation unit for creating skeleton model data.

A program according to the present invention is a program for causing a skeletal model creation device having a three-dimensional shape to create a skeleton model from three-dimensional shape data of a target object. Means for extracting center axis data by performing center axis conversion processing on contour shape data extracted corresponding to the contours of the target object viewed from a plurality of different directions using the three-dimensional shape data consisting of the three-dimensional coordinate positions; It is made to function as a means for creating skeleton model data in which a plurality of extracted central axis data are associated with feature points corresponding to nodal points of the central axis data .

  A dimension measuring method for a three-dimensional shape according to the present invention is a dimension measuring method for a three-dimensional shape in which a dimension of a predetermined part is measured from three-dimensional shape data of a target object by a programmed computer. A step of calculating the principal axis of the three-dimensional shape data composed of coordinate positions by principal component analysis, performing coordinate conversion so that the principal axis coincides with a predetermined coordinate axis direction of the measurement space, and performing the normalization process 3 A step of extracting contour shape data corresponding to the contour of the target object viewed from a plurality of different directions using the dimensional shape data, and performing central axis conversion processing on each extracted contour shape data to extract the central axis data A step of creating skeleton model data in which a plurality of extracted central axis data are associated with feature points; and the contour shape data and the skeleton model Identifying a predetermined portion located on the basis of the data, characterized in that and a step of calculating the dimension data of the predetermined portion based on the identified predetermined portion position and the 3-dimensional shape data to the computer.

  The dimension measuring apparatus for a three-dimensional shape according to the present invention calculates a principal axis by principal component analysis of three-dimensional shape data composed of three-dimensional coordinate positions on the surface of a target object so that the principal axis coincides with a predetermined coordinate axis direction of a measurement space. A normalization processing unit that performs coordinate transformation processing, a contour extraction unit that extracts contour shape data corresponding to the contour of the target object viewed from a plurality of different directions using the normalized three-dimensional shape data, and A central axis extraction unit that performs central axis conversion processing on each contour shape data to extract central axis data, a skeleton generation unit that generates skeleton model data in which a plurality of extracted central axis data are associated with feature points, A position specifying unit for specifying a predetermined part position based on the contour shape data and the skeleton model data; a predetermined part position based on the specified predetermined part position and the three-dimensional shape data; Characterized in that it includes a dimension calculation unit for calculating the data.

  Another program according to the present invention is a program for causing a three-dimensional shape dimension measuring device for measuring a dimension of a predetermined part from three-dimensional shape data of a target object to function as the target object surface. Means for calculating the principal axis of the three-dimensional shape data consisting of the three-dimensional coordinate positions by principal component analysis and performing coordinate conversion so that the principal axis coincides with a predetermined coordinate axis direction of the measurement space, and is normalized. Means for extracting contour shape data corresponding to the contour of the target object viewed from a plurality of different directions using the three-dimensional shape data, and extracting the central axis data by performing center axis conversion processing on each extracted contour shape data Means for creating skeleton model data in which a plurality of extracted central axis data are associated with feature points, predetermined based on the contour shape data and the skeleton model data It means for identifying the position location, characterized in that to function as means for calculating the size data of the predetermined portion on the basis of a predetermined portion position and the 3-dimensional shape data specified.

  According to the present invention having the above-described configuration, the characteristic part can be accurately set based on the three-dimensional shape data of the target object. In particular, when the target object is a human body or a three-dimensional shape similar to the human body, objective characteristic parts can be set based on the extracted skeleton model data, and there is no setting variation by the user. Thus, the position of the characteristic part can be objectively specified.

It is the whole processing flow regarding the skeleton model preparation method of the three-dimensional shape which concerns on this invention. It is a block block diagram regarding the skeleton model creation apparatus of the three-dimensional shape for enforcing the method shown in FIG. It is a perspective view which displays the acquired three-dimensional shape data. It is explanatory drawing shown typically about a normalization process. It is explanatory drawing regarding the extraction process of outline shape data at the time of seeing a human body from the front. It is explanatory drawing regarding the extraction process of outline shape data at the time of seeing a human body from the side. It is explanatory drawing regarding the state cut | disconnected by the two orthogonal planes which pass along a principal axis with respect to three-dimensional shape data. It is explanatory drawing regarding the extraction process of central axis data by a Voronoi division process, and a feature point. It is explanatory drawing regarding the extracted central axis data and a feature point. It is a whole processing flow regarding the dimension measuring method of the three-dimensional shape concerning the present invention. It is a block block diagram regarding the three-dimensional-shaped dimension measuring apparatus for enforcing the method shown in FIG. It is explanatory drawing regarding the case where a chest position and a waist | hip | lumbar part position are specified. It is explanatory drawing regarding the case where a trunk | drum part position is specified. It is explanatory drawing regarding the dimension calculation in a chest position.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments described below are preferable specific examples for carrying out the present invention, and thus various technical limitations are made. However, the present invention is particularly limited in the following description. Unless otherwise specified, the present invention is not limited to these forms.

  FIG. 1 is an overall processing flow relating to a three-dimensional shape skeleton model creation method according to the present invention, and FIG. 2 is a block configuration diagram relating to a three-dimensional shape skeleton model creation device for carrying out the method shown in FIG. is there.

  The skeletal model creation device includes a control unit 1 that performs control processing of the entire device and information processing for creating a skeletal model, and a transmission / reception unit that transmits and receives information such as a captured image of a target object and three-dimensional shape data to and from an external device 2. Storage unit 3 for storing a program for creating a skeleton model together with basic software (OS), display unit 4 for displaying output results of various information processing, operation input for inputting data necessary for various information processing and control processing Part 5 is provided. You may make it provide the connection part for connecting peripheral devices, such as an imaging device, as needed.

  The control unit 1 includes a data acquisition unit 10 that acquires three-dimensional shape data including the three-dimensional coordinate position on the surface of the target object, calculates a principal axis by principal component analysis of the acquired three-dimensional shape data, and sets in a predetermined coordinate axis direction. A normalization processing unit 11 that performs coordinate transformation processing so that the main axes coincide with each other, and contour extraction that extracts contour shape data corresponding to the contour of the target object viewed from a plurality of different directions using the normalized three-dimensional shape data Unit 12, a central axis extraction unit 13 that performs central axis conversion processing on each extracted contour shape data to extract central axis data expressed by a set of points, and a skeleton that associates a plurality of extracted central axis data with feature points A skeleton creation unit 14 for creating model data is provided.

  An apparatus having such a block configuration can be realized using a known computer. For example, a CPU and a memory corresponding to the control unit 1, a LAN board corresponding to the transmission / reception unit 2, a hard disk corresponding to the storage unit 3, a display such as a liquid crystal display panel corresponding to the display unit 4, a keyboard corresponding to the input unit 5, and A computer in which known hardware such as a mouse is connected to each other through a data transmission path may be used.

  When creating a skeleton model, first, the three-dimensional shape data of the target object is acquired (S100). Here, the three-dimensional shape data is data including three-dimensional coordinate positions of the entire surface of the target object, and includes a three-dimensional coordinate position of the inner surface when there is a cavity inside the target object. The three-dimensional shape data may be acquired from an external device via the transmission / reception unit 2, or a target object can be obtained from a photographed image by connecting a photographing device such as a 3D scanner or a camera to the skeleton model creating device. Three-dimensional shape data corresponding to the surface shape may be acquired.

  For example, three-dimensional shape data corresponding to the surface shape of the target object may be directly obtained by using a general 3D scanner, or the surface of the target object is photographed using slit light and a camera. The three-dimensional shape data may be created from the above. Alternatively, the target object may be photographed with a camera from a plurality of directions, and a three-dimensional shape data may be generated by extracting a three-dimensional position of the target object surface from the obtained two-dimensional captured images.

  The acquisition of the three-dimensional shape data is not particularly limited as long as the three-dimensional shape data of the target object surface can be acquired. FIG. 3 is a perspective view displaying the acquired three-dimensional shape data. In FIG. 3, a human body is used as a target object, and three-dimensional shape data of the human body surface is acquired for a human body in an upright state.

  Such acquisition processing of three-dimensional shape data is performed by the data acquisition unit 10.

  Normalization processing is performed on the acquired three-dimensional shape data (S101). By performing the normalization process, it is possible to remove variable elements such as the posture and installation position of the target object from the three-dimensional shape data. As a normalization process, the principal axis in the three-dimensional shape data is set by performing principal component analysis on the coordinate position of the three-dimensional shape data, and the coordinate conversion is performed so that the directions of the principal axis and the preset coordinate axes coincide. Do. By performing such normalization processing, various three-dimensional shape data can be set in a unified coordinate space, which is effective when comparatively analyzing different three-dimensional shape data.

  FIG. 4 is an explanatory diagram schematically showing the normalization process. A three-dimensional coordinate position of the three-dimensional shape data 200 is plotted against a coordinate space 100 represented by an orthogonal coordinate system including xyz axes orthogonal to each other, and a principal component analysis of the three-dimensional shape data 200 is performed to obtain a first principal component. The principal axis A is determined based on the first principal component (FIG. 4A). Then, coordinate conversion of the three-dimensional shape data 200 is performed so that the principal axis A coincides with the z-axis of the orthogonal coordinate system (FIG. 4B). In this case, since the three-dimensional shape data of the human body is targeted, the lowest position of the three-dimensional shape data may be set to the same height as the origin O with z = 0.

  Such normalization processing of the three-dimensional shape data is performed by the normalization processing unit 11.

  Next, contour shape data is extracted from the normalized three-dimensional shape data (S102). Examples of the contour shape data include a silhouette shape obtained by parallel projection of a target object from a predetermined direction or a cross-sectional shape obtained by cutting the target object along a predetermined direction by a point set. When the inside of the target object is hollow, the cross-sectional shape is preferably used as the contour shape data. Moreover, you may extract both silhouette shape and cross-sectional shape as outline shape data.

  When extracting the silhouette shape, for example, the target object from a plurality of directions such as the front surface and the side surface of the target object is parallel-projected on a plane, and the silhouette shape is extracted from the outer shape of the projected image. A plurality of contour shape data viewed from the direction are extracted.

  FIG. 5 is an explanatory diagram regarding processing for extracting contour shape data when viewed from the front of the human body that is the target object. The normalized three-dimensional shape data is set in a state viewed from the direction set as the front direction of the human body, and the silhouette shape is extracted as contour shape data 210.

  FIG. 6 is an explanatory diagram regarding the extraction processing of contour shape data when the human body is viewed from the side as in FIG. The normalized three-dimensional shape data is set in a state viewed from the side surface direction orthogonal to the front direction of the human body, and the silhouette shape is extracted as contour shape data 220.

  In addition to the above-described extraction processing, the contour shape data may be acquired by extracting the contour line of the target object using, for example, captured images obtained by capturing the target object with a camera from a plurality of directions.

  In addition, when the target object has a hollow formed inside, such as a cup, for example, accurate contour shape data of the target object cannot be obtained with the above-described silhouette shape. By extracting the cross-sectional shape as the contour shape data from the dimensional shape data, it is possible to obtain contour shape data that accurately represents the internal cavity. As the cross-sectional shape of the target object, for example, a cross-sectional shape cut by a plane including the main axis of the three-dimensional shape data may be used. FIG. 7 is an explanatory diagram relating to a state in which the three-dimensional shape data 200 is cut by two orthogonal planes F1 and F2 passing through the main axis A. What is necessary is just to extract the cross-sectional shape cut | disconnected by such two planes as outline shape data.

  When the target object has a simple shape (for example, a shape close to a columnar shape), the camera is not used without acquiring the three-dimensional shape data in step S100 and the normalization process in step S101. A shape obtained by directly extracting the contour line of the target object from the used captured image may be extracted as contour shape data.

  Also, subtle differences in shape due to noise and direction differences in the extracted contour shape data are removed when the center axis extraction process described later is performed, so it is necessary to accurately match the direction to be extracted with the determined direction. Absent.

  The contour shape data extraction process from the three-dimensional shape data is performed by the contour extraction unit 12.

  Next, center axis conversion processing is performed on the extracted contour shape data to extract center axis data (S103). A central axis conversion process is performed on a plurality of contour shape data extracted from the above-described three-dimensional shape data, and central axis data of each contour shape data is extracted. For the central axis conversion processing, a known processing method such as a method using Voronoi division or a method using thinning may be used.

  In this example, center axis data is extracted by a method using Voronoi division. Voronoi division is a method of dividing the nearest neighbor region of each generating point by drawing a perpendicular bisector to a straight line connecting adjacent generating points. In this example, the point set constituting the contour shape data is divided into points. Perform Voronoi division as a generating point.

  For example, FIG. 8A shows the contour shape data 210 extracted when the human body as the target object is viewed from the front, and FIG. 8B shows the Voronoi division processing performed on the contour shape data 210. It is the figure divided | segmented by the side 211. FIG.

  When extracting the central axis data of the contour shape data, a pruning process is performed to leave only the Voronoi sides that do not intersect the contour line corresponding to the silhouette shape based on the contour shape data among the Voronoi sides 211 shown in FIG. Do. FIG. 8C shows the central axis data 212 obtained after the pruning process.

  FIG. 9 shows the central axis data 222 extracted by performing the central axis conversion process on the contour shape data 220 extracted from the side of the human body that is the target object, as in FIG.

  As described above, the central axis data is extracted from the contour shape data viewed from a plurality of different directions. Such central axis conversion processing is performed by the central axis extraction unit 13.

  Next, skeleton model data in which the extracted central axis data is associated with feature points is created (S104).

  As shown in FIG. 8C, the extracted central axis data 212 includes nodal points P1 and P2 as feature points indicating nodules. The nodal points P1 and P2 are set at the positions of heights T1 and T2, respectively.

  Also for the central axis data 222 shown in FIG. 9, feature points P1 'and P2' corresponding to the nodal points P1 and P2 are set at the positions of the heights T1 and T2. In this way, skeleton model data in which a plurality of central axis data viewed from a plurality of different directions are associated with feature points is created. In this way, the skeleton model data creation process in which the extracted plurality of central axis data is associated with the feature points is performed by the skeleton creation unit 14.

  The skeleton model data obtained in this way can be analyzed three-dimensionally when the target object is viewed from a plurality of different directions, and can be analyzed accurately. For example, when identifying the shapes of two target objects, the shape identification process can be accurately performed at high speed by comparing the skeleton model data described above. Further, when the target object is a human body, the posture of the human body can be determined by analyzing the skeleton model data. For example, it can be determined whether it is an upright state or a forward tilt state based on the positional relationship between the nodal points P1 and P2. By analyzing the longitudinal relationship between the nodal point P2 and the central axis below it, it becomes possible to determine the bending posture of the leg at high speed.

  Moreover, objective measurement without variation can be performed by specifying the measurement site position based on the feature points of the skeleton model data and measuring the dimensions of the target object.

  FIG. 10 is an overall process flow relating to the three-dimensional shape dimension measuring method according to the present invention, and FIG. 11 is a block configuration diagram relating to the three-dimensional shape dimension measuring apparatus for carrying out the method shown in FIG.

  The dimension measuring apparatus has the same apparatus configuration as the skeleton model creating apparatus shown in FIG. The control unit 1 includes a data acquisition unit 10, a normalization processing unit 11, a contour extraction unit 12, a central axis extraction unit 13, and a skeleton creation unit 14 that are the same as those of the skeleton model creation device. A position specifying unit 15 that specifies a position of a predetermined part to be measured from the feature point, and a dimension calculating part 16 that calculates a dimension using three-dimensional shape data at the specified measurement part are provided.

  In the processing flow, steps S200 to S204 are the same as steps S100 to S104 shown in FIG. In the normalization process of step S201, a coordinate space represented by an orthogonal coordinate system composed of mutually orthogonal xyz axes is set as a measurement space. In step S204, as in step S104, the skeleton model data is created by associating a plurality of central axis data respectively extracted from contour shape data viewed from a plurality of different directions with the nodal points P1 and P2 as feature points. .

  Next, a predetermined part position is specified using skeleton model data and contour shape data (S205). In this example, the positions of the parts of the chest, torso, and lower back of the human body that are target objects are specified.

  FIG. 12 is an explanatory diagram regarding the case where the chest position and the waist position are specified. The chest position passes through the most protruding point in the horizontal direction (xy plane direction) on the front side of the silhouette shape using the contour shape data viewed from the side surface between the nodal points P1 and P2 which are characteristic points. A position where a plane orthogonal to the central axis and the central axis intersect is defined as a chest position.

  As shown in FIG. 12, the horizontal most protruding point B on the front side between the feature points P1 ′ and P2 ′ associated with the nodal points P1 and P2 on the central axis 222 of the contour shape data 220 viewed from the side. To decide. Then, the position at which the plane FB that passes through the most protruding point B and is orthogonal to the central axis 222 intersects the central axis is specified as the chest position (feature point P3 '). Further, the feature point P3 of the center axis 212 associated with the feature point P3 'intersecting the plane FB on the center axis 222 so as to coincide with the height is set, and the plane FB orthogonal to the center axis 212 is set.

  As for the waist position, between the nodal point P2 and the crotch point P4, which are characteristic points, the contour shape data viewed from the side surface is used to project the silhouette shape to the horizontal direction (xy plane direction) on the back side. A position where a plane passing through the point and perpendicular to the central axis intersects with the central axis is defined as a waist position.

  As shown in FIG. 12, in the contour shape data 210, the highest position of the contour shape data is set as the crotch point P4 between the central axes branched from the node P2. Then, the feature point P4 ′ of the central axis 222 associated with the crotch point P4 so as to coincide with the height is set, and the horizontal direction on the back side of the contour shape data 220 is set between the feature points P2 ′ and P4 ′. The most protruding point H is determined. Then, the position that passes through the most protruding point H and intersects the plane FH orthogonal to the central axis 222 is specified as the waist position. Further, the feature point P5 of the center axis 212 associated with the feature point P5 'intersecting the plane FH on the center axis 222 so as to coincide with the height is set, and the plane FH orthogonal to the center axis 212 is set.

  For the trunk position, the perimeter of the cross-sectional shape cut by a plane orthogonal to the central axes 212 and 222 is calculated using the three-dimensional shape data between the nodal points P1 and P2, and the perimeter is the shortest. The position where the plane and the central axis intersect is specified as the body position. FIG. 13 is an explanatory diagram regarding the case where the trunk position is specified. A plane including a normal line orthogonal to the central axis 212 and a normal line orthogonal to the central axis 222 between the nodal points P1 and P2 set on the central axis 212 is used as a cut surface of the three-dimensional shape data, and its cross-sectional shape The position where the perimeter of the crossing with the plane FW having the shortest perimeter is specified as the body position. Also, a feature point P6 that intersects the plane FW at the central axis 212 and a feature point P6 'that intersects the plane FW at the central axis 222 are set.

  As described above, since the position of the predetermined part is specified based on the feature point set on the central axis using the three-dimensional shape data, the contour shape data, and the skeleton model data, the measurement part is unique without variation. Can be decided. Therefore, even when a plurality of users perform dimension measurement, the dimension measurement can be performed by setting the measurement site position among the users without variation.

  As for the method for specifying the predetermined part position, in addition to the method described above, three-dimensional shape data of a reference object in which the predetermined part position is set in advance is prepared, and the contour shape data and the skeleton model data of the target object and the reference object to be measured are prepared. The position of the target object having the highest degree of coincidence with the predetermined part position set as the reference object can be specified as the predetermined part position by performing pattern matching processing.

  As described above, based on the contour shape data and the skeleton model data, the position specifying unit 15 performs the process of specifying the predetermined part position using the three-dimensional shape data as necessary.

  Next, a dimension is calculated at the specified predetermined position (S206). Although the calculation method of the dimension at the predetermined part position is different depending on the predetermined part, in this example, a case will be described in which a dimension necessary for creating clothing that matches the human body as the target object is calculated. In the calculation of the dimensions in this case, a method of calculating the perimeter at a predetermined part position of the three-dimensional shape data and a method of calculating a length between feature points or between predetermined part positions are used.

  As a method for calculating the perimeter of the predetermined part position, an example of calculating at the chest position will be described. As shown in FIG. 12, the three-dimensional shape data at the chest position intersecting the plane FB passing through the feature points P3 and P3 'is used. FIG. 14 is an explanatory diagram relating to dimension calculation at the chest position. First, as shown in FIG. 14A, a cross-sectional shape R1 along the plane FB is obtained based on the three-dimensional shape data at the chest position. Next, a shape R2 having the longest perimeter is extracted from the cross-sectional shape R1 (FIG. 14B), a convex hull is drawn on the point set forming the shape R2, and converted into a convex polygon R3. For the perimeter, the sum of the lengths of each side is calculated from the coordinates of the vertices of the convex polygon R3. The calculated perimeter is the dimension value of the chest circumference. By using the same calculation method, it is possible to calculate the waist circumference at the trunk position and the waist circumference at the waist position.

  As a method for calculating the length between the feature points or between the predetermined part positions, for example, in the case of the height of the human body, the height direction (z-axis direction) in the point set forming the contour shape data is used. The length in the height direction between the minimum value and the maximum value may be calculated. In the case of the crotch height, the crotch point P4 shown in FIG. 12 and the minimum value in the height direction are similarly obtained. What is necessary is just to calculate the length of the height direction between points. Also for feature parts such as shoulder width, feature points for specifying the part position are set based on the contour shape data and skeleton creation data, and the dimensions are calculated using the three-dimensional shape data based on the set feature points. That's fine.

  As described above, the process of calculating the dimension data of the predetermined part based on the specified predetermined part position and the three-dimensional shape data is performed by the dimension calculation unit 16.

DESCRIPTION OF SYMBOLS 1 Control part 2 Transmission / reception part 3 Storage part 4 Display part 5 Operation input part

Claims (6)

A method for creating a skeleton model of a three-dimensional shape in which a skeleton model is created from three-dimensional shape data of a target object by a programmed computer
Center axis data is obtained by performing center axis conversion processing on the contour shape data extracted corresponding to the contours of the target object viewed from a plurality of different directions using the three-dimensional shape data composed of the three-dimensional coordinate positions on the surface of the target object. Extracting the
Generating a skeletal model data in which a plurality of extracted central axis data are associated with feature points corresponding to nodal points of the central axis data. . A central axis conversion process is performed on the contour shape data extracted corresponding to the contours of the target object viewed from a plurality of different directions using the three-dimensional shape data composed of the three-dimensional coordinate positions on the surface of the target object. A central axis extraction unit that extracts the expressed central axis data, and a skeleton generation unit that generates skeleton model data in which the extracted plurality of central axis data are associated with feature points corresponding to the nodes of the central axis data. A skeleton model creation device having a three-dimensional shape. A program for causing a three-dimensional shape skeleton model creation device to create a skeleton model from three-dimensional shape data of a target object,
The skeleton model creation device,
Center axis data is obtained by performing center axis conversion processing on the contour shape data extracted corresponding to the contours of the target object viewed from a plurality of different directions using the three-dimensional shape data composed of the three-dimensional coordinate positions on the surface of the target object. Means for extracting,
A program that functions as means for creating skeleton model data in which a plurality of extracted central axis data are associated with feature points corresponding to nodal points of the central axis data . A method for measuring a dimension of a three-dimensional shape by measuring a dimension of a predetermined part from three-dimensional shape data of a target object by a programmed computer,
Performing a normalization process by calculating the principal axis of the three-dimensional shape data consisting of the three-dimensional coordinate position of the target object surface by principal component analysis and converting the coordinates so that the principal axis coincides with a predetermined coordinate axis direction of the measurement space;
Extracting contour shape data corresponding to the contour of the target object viewed from a plurality of different directions using the normalized three-dimensional shape data;
Performing center axis conversion processing on each extracted contour shape data to extract center axis data;
Creating skeletal model data in which a plurality of extracted central axis data are associated with feature points;
Identifying a predetermined site position based on the contour shape data and the skeleton model data;
A method for measuring a dimension of a three-dimensional shape, which causes the computer to execute a step of calculating dimension data of a predetermined part based on the specified predetermined part position and the three-dimensional shape data.   A normalization processing unit that calculates principal axes from principal component analysis of three-dimensional shape data composed of three-dimensional coordinate positions on the surface of the target object and performs coordinate transformation processing so that the principal axes coincide with predetermined coordinate axes in the measurement space; A contour extracting unit that extracts contour shape data corresponding to a contour of the target object viewed from a plurality of different directions using the three-dimensional shape data that has been processed, and a central axis conversion process for each extracted contour shape data A central axis extraction unit that extracts central axis data, a skeleton creation unit that creates skeleton model data in which the extracted plurality of central axis data are associated with feature points, and a predetermined value based on the contour shape data and the skeleton model data A position specifying unit for specifying a part position; and a dimension calculating unit for calculating dimension data of the predetermined part based on the specified predetermined part position and the three-dimensional shape data. Dimension measuring apparatus of a three-dimensional shape, characterized in that. A program for functioning a three-dimensional shape measuring device for measuring a dimension of a predetermined part from three-dimensional shape data of a target object,
The dimension measuring device;
Means for calculating a main axis by principal component analysis of three-dimensional shape data consisting of three-dimensional coordinate positions on the surface of the target object, performing coordinate processing so that the main axis coincides with a predetermined coordinate axis direction of the measurement space, and performing normalization processing;
Means for extracting contour shape data corresponding to a contour obtained by viewing the target object from a plurality of different directions using the normalized three-dimensional shape data;
Means for performing center axis conversion processing on each extracted contour shape data to extract center axis data;
Means for creating skeleton model data in which a plurality of extracted central axis data are associated with feature points;
Means for specifying a predetermined site position based on the contour shape data and the skeleton model data;
A program that functions as means for calculating dimension data of a predetermined part based on the specified predetermined part position and the three-dimensional shape data.

Images