CN113961738B - Multi-feature casting three-dimensional model retrieval method and device - Google Patents

Multi-feature casting three-dimensional model retrieval method and device Download PDF

Info

Publication number
CN113961738B
CN113961738B CN202111209575.0A CN202111209575A CN113961738B CN 113961738 B CN113961738 B CN 113961738B CN 202111209575 A CN202111209575 A CN 202111209575A CN 113961738 B CN113961738 B CN 113961738B
Authority
CN
China
Prior art keywords
casting
dimensional model
searched
operator
calculating
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202111209575.0A
Other languages
Chinese (zh)
Other versions
CN113961738A (en
Inventor
计效园
孙晓龙
周建新
潘徐政
王先飞
李宝辉
殷亚军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Shanghai Space Precision Machinery Research Institute
Original Assignee
Huazhong University of Science and Technology
Shanghai Space Precision Machinery Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology, Shanghai Space Precision Machinery Research Institute filed Critical Huazhong University of Science and Technology
Priority to CN202111209575.0A priority Critical patent/CN113961738B/en
Publication of CN113961738A publication Critical patent/CN113961738A/en
Application granted granted Critical
Publication of CN113961738B publication Critical patent/CN113961738B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5854—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using shape and object relationship
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
    • G06T2219/20—Indexing scheme for editing of 3D models
    • G06T2219/2016—Rotation, translation, scaling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
    • G06T2219/20—Indexing scheme for editing of 3D models
    • G06T2219/2021—Shape modification

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Computer Graphics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Architecture (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention provides a multi-feature casting three-dimensional model retrieval method and device, which belong to the field of casting three-dimensional model retrieval, and the method comprises the following steps: acquiring a three-dimensional model of a casting to be searched, and performing triangular faceting on the three-dimensional model; calculating the shape characteristics and the domain knowledge of the casting to be searched; calculating the similarity between the shape characteristics of the casting to be searched and each model in a model library, and screening out a model with a certain proportion from the model library; calculating the similarity between the field knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result; the field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and asperity features. The invention realizes the retrieval of the similar three-dimensional model in a robust way under the condition that the casting is subjected to the changes of translation, rotation and the like, and simultaneously shortens the period in the aspect of process design compared with the retrieval construction of the two-dimensional model due to the retrieval of the three-dimensional model.

Description

Multi-feature casting three-dimensional model retrieval method and device
Technical Field
The invention belongs to the field of three-dimensional model retrieval of castings, and particularly relates to a multi-feature three-dimensional model retrieval method and device for castings.
Background
The casting is widely applied to various fields of national economy such as aviation, aerospace, rail transit, engineering machinery and the like, wherein a typical casting represented by a box body, a steering axle, a casing, a space motor blade and the like has complex process structures such as multidimensional distortion, special-shaped curved surfaces and the like, so that the problems of large process design difficulty, long period and the like are caused, and the realization of 'mature process reuse' is one of the keys for solving the problems.
At present, casting enterprises often store mature casting process design schemes such as a pouring system, a riser, a chill and the like of historical casting products in the form of characters, pictures and the like in a casting process card, and in the searching process, the process card can only be searched by searching information such as product numbers, batch numbers and the like. When a new process design of a new cast product is performed, the similarity between the product and a certain historical batch and model of the product is often determined based on experience of a designer, and then the batch number and the product number of the batch of products are used for searching a historical mature process. Obviously, the process of the new product-manual judgment-batch/product number-historical product maturation process greatly depends on knowledge reserves of designers and familiarity degree of historical products, and the problems of long retrieval flow, low efficiency, great dependence on manual experience, low intelligent degree and the like exist, so that the defects of low retrieval result query accuracy, low query result and the like are caused, and the complex casting process design difficulty and long period of the box body, steering bridge and the like are difficult to improve through the maturation process multiplexing.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a multi-feature casting three-dimensional model retrieval method and device, and aims to solve the problems that the existing casting retrieval method is mainly based on two-dimensional retrieval to realize process multiplexing, but the two-dimensional retrieval result needs to be reconstructed into a three-dimensional model, so that the retrieval result obtained by the existing casting retrieval method has larger difficulty and longer period in process design.
In order to achieve the above purpose, in one aspect, the present invention provides a multi-feature casting three-dimensional model retrieval method, which comprises the following steps:
acquiring a three-dimensional model of a casting to be searched, and performing triangle faceting and posture normalization treatment on the three-dimensional model;
calculating the shape characteristics and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after triangle surface-mount;
Calculating the similarity between the shape characteristics of the casting to be searched and each model in a model library, and screening out a model with a certain proportion from the model library;
Calculating the similarity between the field knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and asperity features.
Preferably, the shape features include a D2 operator, an N2 operator, and a NaN operator of the three-dimensional model;
The calculation method of the D2 operator comprises the following steps:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
The calculation method of the N2 operator comprises the following steps:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
The calculation method of the NaN operator comprises the following steps:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
Preferably, the domain knowledge comprises Mod operator, sym operator and Env operator of the three-dimensional model;
the method for acquiring the Env operator comprises the following steps:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
The three-dimensional feature vector is obtained by calculating the combination of the width-to-length ratio, the aspect ratio and the height-to-length ratio of the envelope size, and is an Env operator;
the method for obtaining the Mod operator comprises the following steps:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the method for acquiring the Sym operator comprises the following steps:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
And calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane to serve as symmetry degrees of the coordinate planes, and sequentially calculating the symmetry degrees of the three coordinate planes to form feature vectors of the three-dimensional model, wherein the feature vectors are Sym operators.
Preferably, the method for normalizing the three-dimensional model of the casting to be retrieved comprises the following steps:
analyzing and reading a three-dimensional model file of the casting to be retrieved by utilizing three-dimensional modeling software;
Acquiring a triangular flaking file of a three-dimensional model of a casting to be searched by utilizing a flaking function of the three-dimensional model in software;
calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to a center coordinate formula, and further calculating the gravity center of the three-dimensional model of the casting to be searched;
calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
Calculating covariance matrixes of all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining a rotation matrix of the casting to be searched;
and (3) acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete the posture normalization of the three-dimensional model.
Preferably, the method for acquiring the NaN operator comprises the following steps:
S4.1: according to the sea-land formula, calculating the areas of all triangular patches on the three-dimensional model to be searched to obtain an ordered area array;
S4.2: searching all adjacent patches of each triangle patch on the condition that whether the patch adjacent points are contained or not, and creating a key-value dictionary by using a 'patch sequence number-adjacent patch sequence number set';
S4.3: searching a key-value dictionary, finding all adjacent patches of all triangular patches of the three-dimensional model of the casting to be searched, calculating the cosine values of included angles of normal vectors of each patch and all the adjacent patches, and averaging to obtain a normal vector included angle cosine value list;
S4.4: generating a plurality of random numbers in the ordered area array by utilizing a random number function, determining subscripts of the random numbers in the area array by utilizing binary search, and determining a plurality of sampling patches; inquiring a cosine value list of the normal vector included angles in S4.3 to obtain the cosine values of the normal vector included angles of a plurality of sampling patches;
s4.5: and determining the number of the statistical intervals, determining the range of each statistical interval, counting the occurrence times of the cosine value of the normal vector included angle of the chip in each interval range, obtaining corresponding frequency distribution, dividing the frequency distribution by the sampling number, and taking the frequency distribution as a NaN operator of the three-dimensional model.
Preferably, the method for acquiring the Sym operator specifically comprises the following steps:
According to the gravity center coordinate formula, the gravity centers of all triangular patches of the three-dimensional model of the casting to be searched after the posture normalization are recalculated;
Dividing all the patches of the progressive three-dimensional model into two parts according to the position relation between the gravity center and the coordinate plane of each patch, and respectively calculating the area sum of the patches of the two parts;
obtaining the surface area error of the two parts of the three-dimensional model segmentation according to the area sum of the two parts of the patches, and taking the surface area error as the symmetry degree of the three-dimensional model about a coordinate plane;
combining symmetry degrees of three coordinate planes to form a three-dimensional feature vector serving as a Sym operator of the three-dimensional model;
wherein the coordinate plane includes a YOZ plane, an XOZ plane, and an XOY plane.
In another aspect, a multi-feature casting three-dimensional model retrieval apparatus includes:
The three-dimensional model processing module is used for acquiring a three-dimensional model of the casting to be searched, and performing triangle facing and posture normalization processing on the three-dimensional model;
the feature extraction module is used for calculating the shape features and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after the triangle surface-mount; wherein, the triangle surface masquerade is subjected to gesture normalization processing;
the model screening module is used for calculating the similarity between the shape characteristics of the casting to be searched and each model in the model library, and screening out a model with a certain proportion from the model library;
the method is used for calculating the similarity between the domain knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and relief features.
Preferably, the shape features include a D2 operator, an N2 operator, and a NaN operator of the three-dimensional model;
the feature extraction module comprises a shape feature extraction unit and a domain knowledge extraction unit; the shape feature extraction unit comprises a D2 operator calculator, an N2 operator calculator and a NaN operator calculator;
the D2 operator calculator is used for obtaining a D2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
the N2 operator calculator is used for acquiring an N2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
the NaN operator calculator is used for acquiring a NaN operator, and the specific implementation process is as follows:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
Preferably, the domain knowledge extraction unit includes: a Mod operator calculator, a Sym operator calculator, and an Env operator calculator;
the Mod operator calculator is used for obtaining the Mod operator, and the specific implementation process is as follows:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the Sym operator calculator is used for acquiring a Sym operator, and the specific implementation process is as follows:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
Calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane, and taking the surface area errors as symmetry of the coordinate planes;
Sequentially calculating symmetry degrees of the three coordinate planes to form a feature vector of the three-dimensional model, wherein the feature vector is a Sym operator;
The Env operator calculator is used for obtaining an Env operator, and the specific implementation process is as follows:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
and obtaining a three-dimensional feature vector by calculating the combination of the aspect ratio, the aspect ratio and the aspect ratio of the envelope size, wherein the three-dimensional feature vector is an Env operator.
Preferably, the three-dimensional model processing module includes:
The file analysis unit is used for analyzing and reading the three-dimensional model file of the casting to be searched by utilizing the three-dimensional modeling software;
The triangular surface-slicing unit is used for acquiring a triangular surface-slicing file of the three-dimensional model of the casting to be searched by utilizing the surface-slicing function of the three-dimensional model in software;
The area calculation unit is used for calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
the center of gravity calculation unit is used for calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to the center coordinate formula, so as to calculate the center of gravity of the three-dimensional model of the casting to be searched;
the translation matrix calculation unit is used for calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
The rotation matrix calculation unit is used for calculating covariance matrices of all triangular patch vertexes of the three-dimensional model of the casting to be searched and obtaining a rotation matrix of the casting to be searched;
and the gesture normalization unit is used for acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete gesture normalization of the three-dimensional model.
In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:
Aiming at the problem of 'history process reuse' of complex castings in the typical casting application field, the invention provides a multi-feature casting three-dimensional model retrieval method; geometric shape characteristics of the three-dimensional model are extracted through shape distribution operators such as a D2 operator and an N2 operator, and meanwhile, technological characteristics such as casting modulus, a casting symmetry plane and the like which are strongly related with technological design in the casting field are innovatively combined (namely, the shape characteristics and the field knowledge of the casting to be searched are calculated based on the three-dimensional model). Experimental results show that the method can effectively extract the information of the geometric shapes (D2 operator and N2 operator), concave-convex degree (NaN operator), hotspots (Mod operator), symmetrical planes (Sym operator), envelope sizes (Env operator) and the like of the three-dimensional model. Finally, the similar three-dimensional model retrieval is robustly carried out under the condition that the casting is subjected to the changes of translation, rotation and the like, and meanwhile, compared with the retrieval construction of a two-dimensional model, the three-dimensional model retrieval method shortens the period in the aspect of process design.
When the method is used for combining the characteristics of a D2 operator, an N2 operator, a NaN operator, a Mod operator and the like, a step-by-step elimination mode is adopted (similarity between the shape characteristics of the search casting and each model in a model library is calculated and searched, a model with a certain proportion is screened, domain knowledge is used for screening a three-dimensional model, and finally a search result is obtained), and compared with the traditional weight combination mode, the three-dimensional model has better search effect and higher flexibility in searching castings in multiple fields.
Drawings
FIG. 1 is a schematic diagram of a multi-feature casting three-dimensional model retrieval method provided by an embodiment of the invention;
FIG. 2 is a three-dimensional model of a casting to be retrieved provided by an embodiment of the present invention;
FIG. 3 (a) is a diagram of the initial effect of a three-dimensional model of a casting to be retrieved, provided by an embodiment of the present invention;
FIG. 3 (b) is a diagram of the effect of normalizing the pose of a three-dimensional model of a casting to be retrieved, provided by an embodiment of the invention;
FIG. 4 is a schematic view of an Env operator extraction flow of a three-dimensional model of a casting to be retrieved, provided by an embodiment of the invention;
FIG. 5 is a schematic illustration of a Sym operator extraction flow for a three-dimensional model of a casting to be retrieved, provided by an embodiment of the invention;
FIG. 6 is a multi-domain, multi-class casting model test set provided by an embodiment of the present invention;
FIG. 7 is a test set of casting models in the rail transit field provided by an embodiment of the present invention;
FIG. 8 is a test set of casting models in the field of construction machinery provided by an embodiment of the present invention;
FIG. 9 is a test set of casting models for the fluid field provided by an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
As shown in fig. 1, in one aspect, the present invention provides a multi-feature casting three-dimensional model retrieval method, including the following steps:
acquiring a three-dimensional model of a casting to be searched, and performing triangle faceting and posture normalization treatment on the three-dimensional model;
calculating the shape characteristics and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after triangle surface-mount;
Calculating the similarity between the shape characteristics of the casting to be searched and each model in a model library, and screening out a model with a certain proportion from the model library;
Calculating the similarity between the field knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and asperity features.
Preferably, the shape features include a D2 operator, an N2 operator, and a NaN operator of the three-dimensional model;
The calculation method of the D2 operator comprises the following steps:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
The calculation method of the N2 operator comprises the following steps:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
The calculation method of the NaN operator comprises the following steps:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
Preferably, the domain knowledge comprises Mod operator, sym operator and Env operator of the three-dimensional model;
the method for acquiring the Env operator comprises the following steps:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
The three-dimensional feature vector is obtained by calculating the combination of the width-to-length ratio, the aspect ratio and the height-to-length ratio of the envelope size, and is an Env operator;
the method for obtaining the Mod operator comprises the following steps:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the method for acquiring the Sym operator comprises the following steps:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
And calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane to serve as symmetry degrees of the coordinate planes, and sequentially calculating the symmetry degrees of the three coordinate planes to form feature vectors of the three-dimensional model, wherein the feature vectors are Sym operators.
Preferably, the method for normalizing the three-dimensional model of the casting to be retrieved comprises the following steps:
analyzing and reading a three-dimensional model file of the casting to be retrieved by utilizing three-dimensional modeling software;
Acquiring a triangular flaking file of a three-dimensional model of a casting to be searched by utilizing a flaking function of the three-dimensional model in software;
calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to a center coordinate formula, and further calculating the gravity center of the three-dimensional model of the casting to be searched;
calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
Calculating covariance matrixes of all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining a rotation matrix of the casting to be searched;
and (3) acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete the posture normalization of the three-dimensional model.
Preferably, the method for acquiring the NaN operator comprises the following steps:
S4.1: according to the sea-land formula, calculating the areas of all triangular patches on the three-dimensional model to be searched to obtain an ordered area array;
S4.2: searching all adjacent patches of each triangle patch on the condition that whether the patch adjacent points are contained or not, and creating a key-value dictionary by using a 'patch sequence number-adjacent patch sequence number set';
S4.3: searching a key-value dictionary, finding all adjacent patches of all triangular patches of the three-dimensional model of the casting to be searched, calculating the cosine values of included angles of normal vectors of each patch and all the adjacent patches, and averaging to obtain a normal vector included angle cosine value list;
S4.4: generating a plurality of random numbers in the ordered area array by utilizing a random number function, determining subscripts of the random numbers in the area array by utilizing binary search, and determining a plurality of sampling patches; inquiring a cosine value list of the normal vector included angles in S4.3 to obtain the cosine values of the normal vector included angles of a plurality of sampling patches;
s4.5: and determining the number of the statistical intervals, determining the range of each statistical interval, counting the occurrence times of the cosine value of the normal vector included angle of the chip in each interval range, obtaining corresponding frequency distribution, dividing the frequency distribution by the sampling number, and taking the frequency distribution as a NaN operator of the three-dimensional model.
Preferably, the method for acquiring the Sym operator specifically comprises the following steps:
According to the gravity center coordinate formula, the gravity centers of all triangular patches of the three-dimensional model of the casting to be searched after the posture normalization are recalculated;
Dividing all the patches of the progressive three-dimensional model into two parts according to the position relation between the gravity center and the coordinate plane of each patch, and respectively calculating the area sum of the patches of the two parts;
obtaining the surface area error of the two parts of the three-dimensional model segmentation according to the area sum of the two parts of the patches, and taking the surface area error as the symmetry degree of the three-dimensional model about a coordinate plane;
combining symmetry degrees of three coordinate planes to form a three-dimensional feature vector serving as a Sym operator of the three-dimensional model;
wherein the coordinate plane includes a YOZ plane, an XOZ plane, and an XOY plane.
In another aspect, a multi-feature casting three-dimensional model retrieval apparatus includes:
The three-dimensional model processing module is used for acquiring a three-dimensional model of the casting to be searched, and performing triangle facing and posture normalization processing on the three-dimensional model;
the feature extraction module is used for calculating the shape features and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after the triangle surface-mount; wherein, the triangle surface masquerade is subjected to gesture normalization processing;
the model screening module is used for calculating the similarity between the shape characteristics of the casting to be searched and each model in the model library, and screening out a model with a certain proportion from the model library;
the method is used for calculating the similarity between the domain knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and relief features.
Preferably, the shape features include a D2 operator, an N2 operator, and a NaN operator of the three-dimensional model;
the feature extraction module comprises a shape feature extraction unit and a domain knowledge extraction unit; the shape feature extraction unit comprises a D2 operator calculator, an N2 operator calculator and a NaN operator calculator;
the D2 operator calculator is used for obtaining a D2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
the N2 operator calculator is used for acquiring an N2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
the NaN operator calculator is used for acquiring a NaN operator, and the specific implementation process is as follows:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
Preferably, the domain knowledge extraction unit includes: a Mod operator calculator, a Sym operator calculator, and an Env operator calculator;
the Mod operator calculator is used for obtaining the Mod operator, and the specific implementation process is as follows:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the Sym operator calculator is used for acquiring a Sym operator, and the specific implementation process is as follows:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
Calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane, and taking the surface area errors as symmetry of the coordinate planes;
Sequentially calculating symmetry degrees of the three coordinate planes to form a feature vector of the three-dimensional model, wherein the feature vector is a Sym operator;
The Env operator calculator is used for obtaining an Env operator, and the specific implementation process is as follows:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
and obtaining a three-dimensional feature vector by calculating the combination of the aspect ratio, the aspect ratio and the aspect ratio of the envelope size, wherein the three-dimensional feature vector is an Env operator.
Preferably, the three-dimensional model processing module includes:
The file analysis unit is used for analyzing and reading the three-dimensional model file of the casting to be searched by utilizing the three-dimensional modeling software;
The triangular surface-slicing unit is used for acquiring a triangular surface-slicing file of the three-dimensional model of the casting to be searched by utilizing the surface-slicing function of the three-dimensional model in software;
The area calculation unit is used for calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
the center of gravity calculation unit is used for calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to the center coordinate formula, so as to calculate the center of gravity of the three-dimensional model of the casting to be searched;
the translation matrix calculation unit is used for calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
The rotation matrix calculation unit is used for calculating covariance matrices of all triangular patch vertexes of the three-dimensional model of the casting to be searched and obtaining a rotation matrix of the casting to be searched;
and the gesture normalization unit is used for acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete gesture normalization of the three-dimensional model.
Examples
The embodiment provides a three-dimensional model retrieval method of a multi-feature casting, wherein the three-dimensional model of the casting shown in fig. 2 is adopted as a casting to be retrieved for explanation, and the number of statistical intervals of a D2 operator is 128; the number of statistical intervals of the N2 operator and the NaN operator is 180; the number of the statistical intervals of the D2 operator is 127 sampling points (sampling patches) which are 100000; the method comprises the following steps:
s1: three-dimensional models of various castings are obtained, and triangular facing is carried out on the three-dimensional models;
Carrying out posture normalization processing on the model, wherein the posture normalization processing mainly comprises the following steps of: (1) Translating the origin of the coordinate system to the gravity center of the three-dimensional model to realize the translation invariance of the three-dimensional model; (2) Determining a coordinate system main shaft to realize the rotation invariance of the three-dimensional model; the method comprises the following steps:
s1.1: analyzing and reading the casting three-dimensional model file by utilizing three-dimensional modeling software;
s1.2: obtaining a triangular flaking file of a casting three-dimensional model by utilizing a flaking function of the three-dimensional model in software, and storing the triangular flaking file by using an STL text file;
Specifically, after three-dimensional modeling software is utilized to realize analysis reading and triangle surface patch formation of a casting three-dimensional model, a list data structure is used for respectively storing three-dimensional vertex coordinates and three-dimensional normal vector coordinates of the molded surface patches, and the subscript of the list is utilized to correspond to the sequence number of the triangle surface patches;
S1.3: according to the sea-renformula, the areas of all triangular patches on the three-dimensional model of the casting are calculated and stored in a list S [ S 1,S2,S3…Sn ], and the sea-renformula is as follows:
p＝(a+b+c)/2
S1.4: calculating the gravity centers of all triangular patches of the three-dimensional model of the casting according to a central coordinate formula, wherein a list Gm[G1[x1,y1,z1],G2[x2,y2,z2],……,Gn[xn,yn,zn]], exists to further calculate the gravity centers of the model, and the gravity centers are marked as G M[Gxm,Gym,Gzm, and a translation matrix for realizing translation invariance of the three-dimensional model is marked as M T; the calculation formula is as follows:
s1.5: recording the vertex set of all triangle patches of the three-dimensional model as [[x1 o,y1 o,z1 o],[x2 o,y2 o,z2 o],……,[xp o,yp o,zp o]],, wherein p is the total number of vertices of the three-dimensional model; the covariance matrix of the vertex is M, the rotation matrix M R for realizing rotation invariance of the three-dimensional model is composed of three unitized eigenvectors [ a i,bi,ci ] of the covariance matrix M, in particular, three eigenvectors in the rotation matrix are arranged according to the ascending order of eigenvalues, and the calculation formula is as follows:
S1.6: and applying the translation matrix M T and the rotation matrix M R to each Vertex of the three-dimensional model to obtain a Vertex coordinate set Vertex of the three-dimensional model after posture normalization, wherein the calculation formula is as follows:
specifically, for the three-dimensional model of the casting to be searched as shown in fig. 2, fig. 3 (a) is an effect diagram of the casting to be searched under the initial coordinate system; after the actions of the translation matrix and the rotation matrix, the effect of posture normalization is shown in fig. 3 (b);
S2: randomly selecting a plurality of point pairs on the surface of the casting three-dimensional model, calculating the distance between two points of each point pair to obtain an array containing the distances of all the point pairs, normalizing the array, setting a statistical interval, counting the number of times of the point pair distances in the array in each interval to form a feature vector, namely a D2 operator of the casting three-dimensional model, and extracting the overall geometric distance distribution information of all vertexes of the casting three-dimensional model;
s3: randomly selecting a plurality of surface patch pairs on the casting three-dimensional model, calculating cosine values of angles of normal vectors of two surface patches of each surface patch pair, obtaining an array containing all cosine values, setting a statistical interval, counting the times of occurrence of cosine values of angles in the array in each interval, forming a feature vector, namely an N2 operator of the casting three-dimensional model, and extracting overall geometric angle distribution information of all surface patches of the casting three-dimensional model;
S4: using the adjacent points of the patches as clues to find out all adjacent patches of each triangular patch; randomly selecting a plurality of patches on the casting three-dimensional model, calculating cosine values of normal vector included angles of each patch and adjacent patches, averaging to obtain an array containing all average cosine values, setting a statistical interval, counting the occurrence times of the average cosine values in each interval in the array to form a feature vector, namely a NaN operator of the casting three-dimensional model, and extracting local concave-convex degree distribution information of all patches of the casting three-dimensional model; the method specifically comprises the following steps:
S4.1: according to the sea-land formula, calculating the areas of all triangular patches on the three-dimensional model of the casting, storing the areas in a list S 1,S2,S3…Sn, and further obtaining an ordered area array T [ T 1,T2,T3,……,Tn ] according to the list S, wherein x=1, 2,3, … and n;
S4.2: searching all adjacent patches of each triangle patch on the condition that whether the patch adjacent points are contained or not, and creating a key-value dictionary by using a 'patch sequence number-adjacent patch sequence number set';
s4.3: searching a key-value dictionary created in the step S4.2, finding all adjacent patches of all triangular patches of the three-dimensional model, calculating the cosine values of included angles of normal vectors of each patch and all the adjacent patches, and averaging to obtain a normal vector included angle cosine value list;
S4.4: generating a plurality of random numbers between 0 and T n by utilizing a random number function, determining subscripts of the random numbers in an area array by utilizing binary search, and determining a plurality of sampling patches; inquiring a cosine value list of the normal vector included angles in S4.3 to obtain the cosine values of the normal vector included angles of a plurality of sampling patches;
S4.5: determining the number of the statistical intervals, determining the range of each statistical interval, counting the occurrence times of the cosine value of the normal vector included angle of the chip in each interval range, obtaining corresponding frequency distribution, dividing the frequency distribution by the sampling number, and taking the frequency distribution as a NaN operator of the three-dimensional model;
s5: calculating the volume V (cm 3) of the three-dimensional model of the casting, taking the sum of all triangular areas as the heat dissipation area A (cm 2) of the three-dimensional model, calculating the overall modulus of the three-dimensional model of the casting by using the sum as a Mod operator of the three-dimensional model, and extracting the wall thickness distribution information, namely the hot spot information, of the overall three-dimensional model of the casting;
S6: determining the maximum value and the minimum value of the model vertexes on X, Y and a Z axis by traversing all triangle patch vertexes of the three-dimensional model after posture normalization, and finally obtaining the envelope size of the three-dimensional model along the direction of a principal axis of a coordinate system; obtaining a three-dimensional feature vector, namely an Env operator of the casting three-dimensional model, by calculating the combination of the width-to-length ratio, the width-to-length ratio and the height-to-length ratio of the envelope dimension, and extracting the envelope dimension distribution information of the whole casting three-dimensional model; the method extraction flow is shown in figure 4; the method specifically comprises the following steps:
s6.1: traversing all vertex coordinates of the casting three-dimensional model, respectively finding out the maximum value and the minimum value of the vertexes in X, Y and Z directions, and obtaining the range distribution of the vertexes of the three-dimensional model in a coordinate system main shaft, wherein the X-axis direction distribution size is used as the width (width) of the casting three-dimensional model, the Y-axis direction distribution size is used as the height (height) of the casting three-dimensional model, and the Z-axis direction distribution size is used as the length (length) of the casting three-dimensional model;
S6.2: in order to eliminate the influence caused by the different sizes of different three-dimensional models, three characteristic values of width/length, width/height and height/length are respectively calculated to form a three-dimensional characteristic vector, and the three characteristic vectors are used as Env operators of the three-dimensional models;
S7: comparing the position relation between the gravity center of each surface piece of the three-dimensional model and the coordinate plane, dividing the three-dimensional model along the YOZ plane, the XOZ plane and the XOY plane respectively, calculating the surface area errors of the two parts of the divided model as the symmetry degree of the coordinate plane, sequentially calculating the model symmetry degrees of the three coordinate planes to form three-dimensional feature vectors, namely Sym operator of the casting three-dimensional model, extracting the information of the whole symmetry plane of the casting three-dimensional model, and extracting the information of the whole symmetry plane of the casting three-dimensional model by the method, wherein the extraction flow is shown in figure 5; the method comprises the following specific steps:
S7.1: according to the gravity center coordinate formula, the gravity centers of all triangular patches of the cast three-dimensional model after the posture normalization are recalculated Gm a[G1 a[x1,y1,z1],G2 a[x2,y2,z2],……,Gn a[xn,yn,zn]];
S7.2: firstly, dividing all the patches of the gradual three-dimensional model into two parts according to the position relation between the gravity centers of the patches and the XOY coordinate plane, namely, the coordinate value of the gravity center Z axis of the patches is more than or equal to 0, and considering that the patches are positioned on the coordinate plane; when the coordinate value is smaller than 0, the surface patch is considered to be positioned below the coordinate plane, and the segmentation is completed through the method; then, calculating the area sum AreaSumA, areaSumB of the two parts of the dough sheets respectively; then, the surface area error of the two parts of the three-dimensional model segmentation can be obtained through the following formula and used as the symmetry degree of the three-dimensional model about the XOY plane;
s7.3: sequentially calculating symmetry degrees of the three-dimensional model on the YOZ plane and the XOZ plane according to the S7.2 process;
S7.4: combining three symmetry degrees of the three-dimensional model obtained in the S7.2 and the S7.3 about the XOY plane, the XOZ plane and the YOZ plane to form a three-dimensional feature vector, and taking the three-dimensional feature vector as a Sym operator of the three-dimensional model;
s8: firstly, sequentially calculating the similarity between a model to be searched under a D2 operator and each model in a model library, and selecting the first 25-30 models with larger similarity for the next feature screening; then, sequentially calculating the similarity between the model to be searched under the N2 operator and each screening model of the previous round, and selecting the first 20-25 models with larger similarity; then, eliminating a model with a certain proportion by operators such as NaN, mod, sym in sequence; finally, 5-8 models reserved after the sixth screening by Env operator are the retrieval models for providing process reference; the method comprises the following specific steps:
S8.1: sequentially calculating the similarity between the three-dimensional model to be searched and all models in the model library under the D2 operator, searching 25-30 three-dimensional models with large similarity according to the similarity, and entering a round of model feature elimination; the operator similarity measure is as follows:
Wherein D is a feature vector D= [ D1, D2, D3, …, di ] of the model to be retrieved obtained by a certain operator, and i represents the dimension of D; d n is a feature vector dn= [ D1, D2, D3, …, di ] n obtained by the same operator for the model in the model library, and n is the nth model;
S8.2: sequentially calculating the similarity between the model to be searched and 25-30 three-dimensional models obtained in the step S8.1 under the N2 operator, searching the three-dimensional models with 20-25 large similarity according to the similarity, and eliminating the characteristics of the model of the next round;
S8.3: sequentially calculating the similarity between the model to be searched and 20-25 three-dimensional models obtained in S8.2 under the NaN operator, searching the three-dimensional models with 15-20 high similarity according to the similarity, and eliminating the characteristics of the model of the next round;
S8.4: sequentially calculating the similarity between the model to be searched and 15-20 three-dimensional models obtained in S8.3 under the Mod operator, searching 10-15 three-dimensional models with large similarity according to the similarity, and eliminating the characteristics of the model of the next round;
S8.5: sequentially calculating the similarity between the model to be searched and 10-15 three-dimensional models obtained in S8.4 under a Sym operator, searching 5-8 three-dimensional models with large similarity according to the similarity, and eliminating the characteristics of the model of the next round;
S8.6: and under the Env operator, sequentially calculating the similarity between the model to be searched and 8-12 three-dimensional models obtained in S8.5, and searching 5-8 three-dimensional models with large similarity according to the similarity, namely obtaining a final three-dimensional model searching result.
In order to better embody the classifying capability of the method for the common types of the steel castings, the searching effect of the method is concentrated in the casting field casting model test of the figures 6, 7, 8 and 9 and the like is given in the table 1;
TABLE 1
The invention provides a method for searching a multi-feature casting three-dimensional model by combining shape features with domain knowledge and a step-by-step elimination mechanism, and provides a method for extracting information such as geometric shapes, concave-convex degrees, hot spots, symmetrical planes, envelope sizes and the like of the three-dimensional model from two dimensions of the shape features and the casting domain knowledge on the basis of 'surface-slicing' of the model so as to realize similar searching of the three-dimensional model. The method has good searching effect on typical casting application fields such as aviation, aerospace, rail transit, engineering machinery, fluid machinery and the like.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (10)

1. The multi-feature casting three-dimensional model retrieval method is characterized by comprising the following steps of:
acquiring a three-dimensional model of a casting to be searched, and performing triangle faceting and posture normalization treatment on the three-dimensional model;
Calculating the shape characteristics and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after the triangle surface-mounting; wherein, the triangle surface masquerade is subjected to gesture normalization processing;
Calculating the similarity between the shape characteristics of the casting to be searched and each model in a model library, and screening out a model with a certain proportion from the model library;
Calculating the similarity between the domain knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and relief features.
2. The multi-feature casting three-dimensional model retrieval method according to claim 1, wherein the shape features include D2 operators, N2 operators, and NaN operators of the three-dimensional model;
The calculation method of the D2 operator comprises the following steps:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
the calculation method of the N2 operator comprises the following steps:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
The calculation method of the NaN operator comprises the following steps:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
3. The multi-feature casting three-dimensional model retrieval method according to claim 2, wherein the domain knowledge includes Mod operator, sym operator, and Env operator of the three-dimensional model;
The method for acquiring the Env operator comprises the following steps:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
The three-dimensional feature vector is obtained by calculating the combination of the width-to-length ratio, the aspect ratio and the height-to-length ratio of the envelope size, and is an Env operator;
The method for obtaining the Mod operator comprises the following steps:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the method for acquiring the Sym operator comprises the following steps:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
Calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane, and taking the surface area errors as symmetry of the coordinate planes;
and sequentially calculating symmetry degrees of the three coordinate planes to form a feature vector of the three-dimensional model, wherein the feature vector is a Sym operator.
4. A multi-feature casting three-dimensional model retrieval method according to any one of claims 1 to 3, characterized in that the method of posture normalization of the three-dimensional model of the casting to be retrieved comprises the steps of:
analyzing and reading a three-dimensional model file of the casting to be retrieved by utilizing three-dimensional modeling software;
Acquiring a triangular flaking file of a three-dimensional model of a casting to be searched by utilizing a flaking function of the three-dimensional model in software;
calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to a center coordinate formula, and further calculating the gravity center of the three-dimensional model of the casting to be searched;
calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
Calculating covariance matrixes of all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining a rotation matrix of the casting to be searched;
and (3) acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete the posture normalization of the three-dimensional model.
5. The multi-feature casting three-dimensional model retrieval method according to claim 2, wherein the NaN operator acquisition method comprises the following steps:
S4.1: according to the sea-land formula, calculating the areas of all triangular patches on the three-dimensional model to be searched to obtain an ordered area array;
S4.2: searching all adjacent patches of each triangle patch on the condition that whether the patch adjacent points are contained or not, and creating a key-value dictionary by using a 'patch sequence number-adjacent patch sequence number set';
S4.3: searching a key-value dictionary, finding all adjacent patches of all triangular patches of the three-dimensional model of the casting to be searched, calculating the cosine values of included angles of normal vectors of each patch and all the adjacent patches, and averaging to obtain a normal vector included angle cosine value list;
S4.4: generating a plurality of random numbers in the ordered area array by utilizing a random number function, determining subscripts of the random numbers in the area array by utilizing binary search, and determining a plurality of sampling patches; inquiring a cosine value list of the normal vector included angles in S4.3 to obtain the cosine values of the normal vector included angles of a plurality of sampling patches;
s4.5: and determining the number of the statistical intervals, determining the range of each statistical interval, counting the occurrence times of the cosine value of the normal vector included angle of the chip in each interval range, obtaining corresponding frequency distribution, dividing the frequency distribution by the sampling number, and taking the frequency distribution as a NaN operator of the three-dimensional model.
6. A multi-feature casting three-dimensional model retrieval method according to claim 3, wherein the method of obtaining Sym operators comprises the steps of:
According to the gravity center coordinate formula, the gravity centers of all triangular patches of the three-dimensional model of the casting to be searched after the posture normalization are recalculated;
Dividing all the patches of the progressive three-dimensional model into two parts according to the position relation between the gravity center and the coordinate plane of each patch, and respectively calculating the area sum of the patches of the two parts;
obtaining the surface area error of the two parts of the three-dimensional model segmentation according to the area sum of the two parts of the patches, and taking the surface area error as the symmetry degree of the three-dimensional model about a coordinate plane;
combining symmetry degrees of three coordinate planes to form a three-dimensional feature vector serving as a Sym operator of the three-dimensional model;
wherein the coordinate plane includes a YOZ plane, an XOZ plane, and an XOY plane.
7. A multi-feature casting three-dimensional model retrieval device, comprising:
The three-dimensional model processing module is used for acquiring a three-dimensional model of the casting to be searched, and performing triangle facing and posture normalization processing on the three-dimensional model;
the feature extraction module is used for calculating the shape features and the domain knowledge of the casting to be searched based on the three-dimensional model of the casting to be searched after the triangle surface-mount; wherein, the triangle surface masquerade is subjected to gesture normalization processing;
the model screening module is used for calculating the similarity between the shape characteristics of the casting to be searched and each model in the model library, and screening out a model with a certain proportion from the model library;
the method is used for calculating the similarity between the domain knowledge of the casting to be searched and the screened model, and screening out a three-dimensional model search result;
The field knowledge comprises hot joint characteristics, symmetry plane characteristics and envelope dimensions of the casting to be searched; the shape features include geometric features of the casting to be retrieved and relief features.
8. The multi-feature casting three-dimensional model retrieval device of claim 7, wherein the shape features include D2 operators, N2 operators, and NaN operators of the three-dimensional model;
the feature extraction module comprises a shape feature extraction unit and a domain knowledge extraction unit; the shape feature extraction unit comprises a D2 operator calculator, an N2 operator calculator and a NaN operator calculator;
the D2 operator calculator is used for obtaining a D2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of point pairs on the surface of the casting to be searched, calculating the distance between two points of each point pair, obtaining an array containing the distances of all the point pairs, and normalizing the array;
Setting statistical intervals, and counting the times of occurrence of the distance of the point pairs in the array in each interval after normalization, wherein the formed feature vector is a D2 operator of the casting to be searched;
The N2 operator calculator is used for acquiring an N2 operator, and the specific implementation process is as follows:
randomly selecting a plurality of surface patch pairs on a three-dimensional model of a casting to be searched, and calculating cosine values of normal vector included angles of two surface patches of each surface patch pair to obtain an array containing all cosine values;
setting statistical intervals, and counting the occurrence times of included angle cosine values in the array in each interval, wherein the formed feature vector is an N2 operator of the casting to be searched;
the NaN operator calculator is used for acquiring a NaN operator, and the specific implementation process is as follows:
Using the adjacent points of the patches as clues to find all the adjacent patches of each triangular patch;
Randomly selecting a plurality of patches on a three-dimensional model of the casting to be searched, and calculating the cosine value of the normal vector included angle between each patch and the adjacent patch;
Averaging the cosine values to obtain an array containing all average cosine values;
setting a statistical interval, and counting the occurrence times of the cosine value of the included angle in the array in each interval, wherein the formed feature vector is a NaN operator of the casting to be searched.
9. The multi-feature casting three-dimensional model retrieval apparatus according to claim 8, wherein the domain knowledge extraction unit includes: a Mod operator calculator, a Sym operator calculator, and an Env operator calculator;
the Mod operator calculator is used for acquiring the Mod operator, and the specific implementation process is as follows:
Taking the sum of the triangular areas of the castings to be searched as the heat dissipation area of the three-dimensional model of the castings to be searched;
Dividing the volume of the casting three-dimensional model to be searched by the heat dissipation area to obtain a Mod operator;
the Sym operator calculator is used for acquiring a Sym operator, and the specific implementation process is as follows:
Comparing the relation between the gravity center of each surface piece of the casting to be searched and the coordinate surface, and dividing the three-dimensional model of the casting to be searched along the YOZ surface, the XOZ surface and the XOY surface respectively;
Calculating surface area errors of two parts of the three-dimensional model after being divided along each coordinate plane, and taking the surface area errors as symmetry of the coordinate planes;
Sequentially calculating symmetry degrees of the three coordinate planes to form a feature vector of the three-dimensional model, wherein the feature vector is a Sym operator;
the Env operator calculator is used for acquiring Env operators, and the specific implementation process is as follows:
Determining maximum values and minimum values of the three-dimensional model on X, Y and a Z axis under a Cartesian coordinate system by traversing all triangular patch vertexes of the three-dimensional model of the casting to be searched, and obtaining envelope sizes of the three-dimensional model of the casting to be searched along the direction of a coordinate main axis;
and obtaining a three-dimensional feature vector by calculating the combination of the aspect ratio, the aspect ratio and the aspect ratio of the envelope size, wherein the three-dimensional feature vector is an Env operator.
10. The multi-feature casting three-dimensional model retrieval apparatus according to any one of claims 7 to 9, wherein the three-dimensional model processing module includes:
The file analysis unit is used for analyzing and reading the three-dimensional model file of the casting to be searched by utilizing the three-dimensional modeling software;
The triangular surface-slicing unit is used for acquiring a triangular surface-slicing file of the three-dimensional model of the casting to be searched by utilizing the surface-slicing function of the three-dimensional model in software;
The area calculation unit is used for calculating the areas of all triangular patches on the casting to be searched according to the sea-land formula;
the center of gravity calculation unit is used for calculating the centers of all triangular patches of the three-dimensional model of the casting to be searched according to the center coordinate formula, so as to calculate the center of gravity of the three-dimensional model of the casting to be searched;
the translation matrix calculation unit is used for calculating a translation matrix of the three-dimensional model of the casting to be searched according to the gravity centers of the triangular patches and the gravity centers of the three-dimensional model of the casting to be searched;
The rotation matrix calculation unit is used for calculating covariance matrices of all triangular patch vertexes of the three-dimensional model of the casting to be searched and obtaining a rotation matrix of the casting to be searched;
and the gesture normalization unit is used for acting the translation matrix and the rotation matrix on each vertex of the three-dimensional model of the casting to be searched to complete gesture normalization of the three-dimensional model.
CN202111209575.0A 2021-10-18 2021-10-18 Multi-feature casting three-dimensional model retrieval method and device Active CN113961738B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111209575.0A CN113961738B (en) 2021-10-18 2021-10-18 Multi-feature casting three-dimensional model retrieval method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111209575.0A CN113961738B (en) 2021-10-18 2021-10-18 Multi-feature casting three-dimensional model retrieval method and device

Publications (2)

Publication Number Publication Date
CN113961738A CN113961738A (en) 2022-01-21
CN113961738B true CN113961738B (en) 2024-07-09

Family

ID=79465027

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111209575.0A Active CN113961738B (en) 2021-10-18 2021-10-18 Multi-feature casting three-dimensional model retrieval method and device

Country Status (1)

Country Link
CN (1) CN113961738B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115423947B (en) * 2022-11-03 2023-03-24 成都飞机工业（集团）有限责任公司 Three-dimensional model retrieval method, device, equipment and medium
CN115496865B (en) * 2022-11-18 2023-03-14 南京智程信息科技有限公司 Similar model searching method, device and storage medium
CN116402988B (en) * 2023-05-11 2023-12-19 北京冰河起源科技有限公司 Three-dimensional model processing method, device and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109299301A (en) * 2018-09-17 2019-02-01 北京工业大学 A kind of method for searching three-dimension model based on distribution of shapes and curvature
CN112164144A (en) * 2020-09-24 2021-01-01 华中科技大学 Casting three-dimensional model classification method combining D2 operator and normal operator

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101084498A (en) * 2004-12-31 2007-12-05 富士通株式会社 Search method of 3d model and device thereof
EP2093714A1 (en) * 2006-11-10 2009-08-26 National University Corporation Toyohashi Universisty of Technology Three-dimensional model search method, computer program, and three-dimensional model search system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109299301A (en) * 2018-09-17 2019-02-01 北京工业大学 A kind of method for searching three-dimension model based on distribution of shapes and curvature
CN112164144A (en) * 2020-09-24 2021-01-01 华中科技大学 Casting three-dimensional model classification method combining D2 operator and normal operator

Also Published As

Publication number Publication date
CN113961738A (en) 2022-01-21

Similar Documents

Publication Publication Date Title
CN113961738B (en) Multi-feature casting three-dimensional model retrieval method and device
CN109887015B (en) Point cloud automatic registration method based on local curved surface feature histogram
Caelli et al. An eigenspace projection clustering method for inexact graph matching
Kazmi et al. A survey of 2D and 3D shape descriptors
CN108376408B (en) Three-dimensional point cloud data rapid weighting registration method based on curvature features
CN111914112B (en) Part CAD model reuse method based on point cloud classification network
CN111080684A (en) Point cloud registration method for point neighborhood scale difference description
US7940263B2 (en) Curved-surface generating method and program, and three-dimensional shape processing apparatus
CN110163863B (en) Three-dimensional object segmentation method, apparatus, and medium
CN105654483B (en) The full-automatic method for registering of three-dimensional point cloud
CN114677418B (en) Registration method based on point cloud feature point extraction
US10580114B2 (en) Methods and systems for real time 3D-space search and point-cloud registration using a dimension-shuffle transform
CN109299301B (en) Three-dimensional model retrieval method based on shape distribution and curvature
CN115409886B (en) Part geometric feature measuring method, device and system based on point cloud
CN101201845A (en) Method for searching three-dimensional model based on virgula point-set delamination helix information
CN116452583B (en) Point cloud defect detection method, device and system and storage medium
CN115797418A (en) Complex mechanical part measurement point cloud registration method and system based on improved ICP
CN116008671A (en) Lightning positioning method based on time difference and clustering
CN111597367B (en) Three-dimensional model retrieval method based on view and hash algorithm
CN108010114A (en) The Geometric Shape Recognition method and characteristic recognition method of element figure point cloud surface
CN115147433A (en) Point cloud registration method
CN115147471A (en) Laser point cloud automatic registration method based on curvature density characteristics
CN112509018B (en) Quaternion space optimized three-dimensional image registration method
CN110945499B (en) Method and system for real-time three-dimensional space search and point cloud registration by applying dimension shuffling transformation
KR101839687B1 (en) Apparatus and method for retrieving model considering differences of level-of-detail between query and retrieval target models

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant