CN108573020B - Three-dimensional assembly model retrieval method integrating assembly information - Google Patents
Three-dimensional assembly model retrieval method integrating assembly information Download PDFInfo
- Publication number
- CN108573020B CN108573020B CN201810115326.7A CN201810115326A CN108573020B CN 108573020 B CN108573020 B CN 108573020B CN 201810115326 A CN201810115326 A CN 201810115326A CN 108573020 B CN108573020 B CN 108573020B
- Authority
- CN
- China
- Prior art keywords
- assembly
- code
- model
- retrieval
- graph
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
Abstract
The invention relates to the technical field of three-dimensional model retrieval, in particular to a three-dimensional assembly model retrieval method integrating assembly information. The technical scheme provided by the invention is as follows: a three-dimensional assembly model retrieval method fusing assembly information comprises the steps of firstly, retrieving the assembly information, representing the assembly information by symbolic codes, and searching the codes of parts to extract a three-dimensional model meeting the assembly design intention; on the basis, because the assembly parts have conjugate relation, conjugate subgraph exists between every two attribute adjacency graphs with the assembly relation, the geometric retrieval of the assembly parts is converted into the attribute adjacency graph which is searched and accords with the conjugate subgraph, and the model retrieval is carried out through the attribute adjacency graph of the parts; and finally, extracting the attribute adjacency graphs meeting the conjugate subgraphs by using a frequent subgraph mining method so as to reuse the model. The invention has high retrieval accuracy and high retrieval efficiency.
Description
Technical Field
The invention relates to the technical field of three-dimensional model retrieval, in particular to a three-dimensional assembly model retrieval method fusing assembly information.
Background
The application of the three-dimensional model retrieval method can solve the problem of reusing a large number of complex models of enterprises, and has important significance for improving the product development efficiency and quality and shortening the product life cycle. The three-dimensional model retrieval technology is based on a model library accumulated by an enterprise, and a system can extract similar part models by inputting required models so as to be convenient for a designer to select, reduce the workload and further achieve the purpose of model reuse. As the demand of enterprises for informatization increases, reusing three-dimensional assembly model information is a hot problem in current research, so that the demand of enterprises for model retrieval technology is more based on assembly model retrieval technology, and when a user uses a single part model as a retrieval condition, the system can find out a three-dimensional model capable of being assembled with the part in a model library. This is also a difficulty in three-dimensional model retrieval.
A three-dimensional model retrieval method respectively represents topological relation and qualitative spatial relation according to a functional surface adjacency graph and a qualitative geometric constraint graph of a part, establishes a part assembly structure qualitative model, and represents a part assembly structure by using a linear sequence of letters and numbers. The assembled structure is then encoded with reference to linear symbologies of compounds in a chemical database to establish a retrieval mechanism for the assembled structure of parts. The limitations of this approach are: the search condition is only considered from the geometrical structure, and the influence of the semantic hierarchy on the search result is ignored. When the designer considers the model reuse, the model is searched according to the assembly design intention, the complex part model can be effectively extracted, and the searching efficiency is accelerated. Meanwhile, more model choices can be provided for designers, and the product design process is accelerated.
A three-dimensional model retrieval method considers the functional semantics of a model, introduces the concept of an ontology and constructs a functional semantics ontology of a three-dimensional assembly model. The model is subjected to standard standardized labeling, and a functional semantic ontology of the three-dimensional assembly model is used for performing functional semantic-based three-dimensional model retrieval. And an algorithm supporting simultaneous retrieval of a plurality of functions is provided to calculate the functional semantic similarity between the models. The limitations of this approach are: the search results cannot emphasize that the geometric shapes are similar and a three-dimensional model conforming to the assembly interface is lacked.
Disclosure of Invention
The invention aims to provide a three-dimensional assembly model retrieval method integrating assembly information, aiming at the problems of semantic inconsistency, low accuracy and low retrieval efficiency of the existing three-dimensional assembly model retrieval method.
The three-dimensional assembly model searching method of the integration assembly information, carry on the assembly information search at first, represent the assembly information with the symbolic code, find the code of the part in order to find out and accord with the three-dimensional model of the assembly design intention; then, converting the geometric retrieval of the assembled parts into an attribute adjacency graph which is matched with the conjugate subgraph, and performing model retrieval through the attribute adjacency graph of the parts; and finally, extracting the attribute adjacency graphs meeting the conjugate subgraphs by using a frequent subgraph mining method so as to reuse the model.
Further, the method specifically comprises the following steps:
(1) formulating an assembly information coding scheme according to the assembly design intention of the three-dimensional assembly model;
(2) establishing an assembly information similarity evaluation model, converting the same code number obtained by comparing the assembly part model code with the target model code into a mathematical model capable of being qualitatively analyzed, wherein the similarity is as follows:
ω1,ω2＞0，ω1+ω2＝1；
similarity of the assembled part model to the target model, S (C)i∩Ck) -the assembly part code is the same number of bits as the target model code, NiThe number of bits, ω, encoded by the model ii-a weight coefficient;
(3) retrieving assembly information;
(a) obtaining a code to be retrieved through an interactive mode, transmitting the code into an identification Agent, and if the code does not accord with a system rule, ending the program;
(b) the recognition Agent judges the codes to obtain the number of the coded bits and the assembly semantic types corresponding to the codes, and transmits the result to the query Agent;
(c) the query Agent matches the codes in the model library according to the result input by the identification Agent;
(d) calculating the similarity between the models according to the matching result;
(f) the execution Agent lists the corresponding models one by one according to the size of the similarity;
(4) extracting the assembly information retrieval results according to a threshold value set by the similarity, and representing the assembly information retrieval results as attribute adjacency graphs;
(5) and (3) carrying out graph isomorphism judgment on the attribute adjacent graphs conforming to the conjugate subgraphs: with an a x b mapping Ma×bMatrix representation of graph G1And G2Vertex correspondence between mijIs to map i rows and j columns of elements in the matrix M if V1And V2Is associated, then m ij1, otherwise mij0; for matrix Ma×bIf there is a mapping such that there is only one 1 per row and no more than one 1 per column, then the matrix M isa×bIs shown as G1And G2The isomorphic mapping between the two steps is as follows:
(a) initializing mapping matrix M and setting two empty sets V1And V2Using them to store the matched vertexes of two graphs in the searching process;
(b) searching from the first row of the matrix M, and searching a column with the value of 1 from left to right; for any ith row, if the value of jth column is 1 and the column is not occupied, then it means that a possible vertex association is found: drawing G1The ith vertex of (1) and graph G2Corresponds to the jth vertex of (1);
(c) adding the two vertexes of the previous step into the set V respectively1And V2；
(d) If V1And V2The newly added vertex is the vertex pair corresponding to the matching in the final isomorphic mapping, and G can be obtained1Set of vertices V of matched1Constructed subgraph and G2Set of vertices V of matched2The constructed subgraphs are isomorphic;
(6) on the basis of a graph isomorphic matching method, an attribute adjacency graph which accords with an assembly characteristic conjugate subgraph is searched from a retrieval model base by adopting a frequent subgraph mining-based algorithm;
(7) and outputting the three-dimensional model according to the retrieval result.
Further, the step (1) comprises the following specific steps:
(a) taking four kinds of assembly information of the connection relation of the assembly parts, the motion relation of the assembly parts, the space positioning constraint of the assembly parts and the materials of the assembly parts as four code bits of the code;
(b) adopt the chain coding scheme, each assembly information is independent in every code bit, and the code bit is irrelevant around with: numbers 1,2, 3 and 4 are used for respectively representing connection relation, motion relation, space constraint and material type and are called classification codes;
(c) the connection relation is represented by an letter code and a number code, the motion relation is represented by an letter code, the space constraint is represented by two letter codes, and the material type is represented by an letter code and a number code, which are called expression codes; when writing the code, the following steps are used: "separate the classification code from the expression code, and when there is no corresponding semantic relationship, the corresponding classification code can be nulled.
Further, the step (6) comprises the following specific steps:
(a) connecting k-order frequent subgraphs to generate k + 1-order candidate subgraphs;
(b) shearing a candidate K + 1-order sub-graph set, and deleting all candidate K + 1-order sub-graphs containing K-order infrequent sub-graphs (the algorithm is shown in the article: automatic extraction of public reusable local structures of three-dimensional CAD models, academic bulletin of computer aided design and graphics, 2011,23(9)1512 and 1519.);
(c) calculating the frequency of all sub-graphs in the candidate k + 1-order sub-graph set;
(d) deleting the candidate K + 1-order subgraph set with the frequency less than SminThe candidate subgraph of (1).
Compared with the prior art, the invention has the beneficial effects that:
and a retrieval mode combining semantic retrieval and geometric retrieval is adopted. Firstly, expressing the assembly information in a coding mode by utilizing the characteristic of high coding retrieval efficiency, and eliminating a model irrelevant to the assembly information in a model library to reduce the workload for the next geometric retrieval; and then, carrying out geometric structure retrieval based on graph isomorphism, and extracting an attribute adjacency graph conforming to the assembled conjugate subgraph by adopting a frequent subgraph mining algorithm. The searching method improves the performance of the three-dimensional assembly model searching method by utilizing the high efficiency of coding searching and the accuracy of graph searching.
The semantic retrieval adopts a coding mode, the coding retrieval is divided into an identification Agent, a query Agent and an execution Agent, and the three parts can independently execute respective functions and can also carry out data transmission mutually. The modular programming method simplifies the program redundancy and provides convenience for later repair. Therefore, the computer can efficiently process the coding rule, accelerate the retrieval efficiency and reduce the workload for the subsequent geometric retrieval; and then geometric retrieval based on graph isomorphism is carried out, and retrieval is carried out aiming at a matching interface between the assembly parts by utilizing the conjugate subgraphs, so that the retrieval accuracy is improved. In addition, in the searching process, after one vertex pair which is possibly matched is found in each step, two sub-graphs which are formed by the vertex sets which are matched currently in the two graphs are isomorphic, the matching relation of the vertex pair is rapidly judged, the centralized judgment after all the vertex pairs are completely corresponding can be avoided, and the retrieval efficiency is greatly improved.
Drawings
Fig. 1 is a schematic of the retrieval method.
Fig. 2 is a general flow chart.
Fig. 3 is a coding scheme.
FIG. 4 is an assembled part attribute adjacency graph.
FIG. 5 is a frequent subgraph mining algorithm.
Fig. 6 is a diagram illustrating the result of the assembly information retrieval.
Fig. 7 is a schematic diagram of a result of geometric information retrieval.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous details are set forth in order to provide a better understanding of the present application. However, those skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in detail in order to avoid obscuring the core of the present application from excessive description, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art.
The technical scheme provided by the invention is as follows: a three-dimensional assembly model retrieval method fusing assembly information comprises the steps of firstly, carrying out assembly information retrieval, representing the assembly information by symbolic codes, and searching the codes of parts to find out a three-dimensional model meeting assembly design intent; on the basis, because conjugate relation exists between the assembly parts, conjugate subgraphs exist between every two attribute adjacency graphs with the assembly relation, geometric retrieval of the assembly parts is converted into the attribute adjacency graph which is searched and accords with the conjugate subgraphs, and model retrieval is carried out through the attribute adjacency graphs of the parts. And finally, extracting the attribute adjacency graphs meeting the conjugate subgraphs by using a frequent subgraph mining method so as to reuse the model.
The method specifically comprises the following steps:
(1) formulating an assembly information coding scheme according to the assembly design intention of the three-dimensional assembly model;
(a) taking four kinds of assembly information of the connection relation of the assembly parts, the motion relation of the assembly parts, the space positioning constraint of the assembly parts and the materials of the assembly parts as four code bits of the code;
(b) and by adopting a chain type coding scheme, the assembly information in each code bit is independent and is irrelevant to the front code bit and the rear code bit. Numbers 1,2, 3 and 4 are used for respectively representing connection relation, motion relation, space constraint and material type and are called classification codes;
(c) the connection relation is represented by an letter code and a numeral code, the motion relation is represented by an letter code, the space constraint is represented by two letter codes, and the material type is represented by an letter code and a numeral code, which are called expression codes. When writing the code, the following steps are used: "separate the classification code from the expression code, when there is no corresponding semantic relation, can set the corresponding classification code empty;
(2) establishing an assembly information similarity evaluation model, converting the same code number obtained by comparing the assembly part model code with the target model code into a mathematical model capable of being qualitatively analyzed, wherein the similarity is as follows:
ω1,ω2＞0，ω1+ω2＝1。
similarity of the assembled part model to the target model, S (C)i∩Ck) -the assembly part code is the same number of bits as the target model code, NiThe number of bits, ω, encoded by the model iiThe weight coefficients can be appropriately valued according to different conditions, in general, when the coding bit numbers of the two models are close, the weight coefficients are closer, and conversely, the weight coefficient with a larger coding bit number is smaller.
(3) Retrieving assembly information;
(a) obtaining a code to be retrieved through an interactive mode, transmitting the code into an identification Agent, and if the code does not accord with a system rule, ending the program;
(b) the recognition Agent judges the codes to obtain the number of the coded bits and the assembly semantic types corresponding to the codes, and transmits the result to the query Agent;
(c) the query Agent matches the codes in the model library according to the result input by the identification Agent;
(d) calculating the similarity between the models according to the matching result;
(f) the execution Agent lists the corresponding models one by one according to the size of the similarity;
(4) extracting the assembly information retrieval results according to a threshold value set by the similarity, and representing the assembly information retrieval results as attribute adjacency graphs;
(5) And (3) carrying out graph isomorphism judgment on the attribute adjacent graphs conforming to the conjugate subgraphs: with an a x b mapping Ma×bMatrix representation of graph G1And G2Vertex correspondence between mijIs to map i rows and j columns of elements in the matrix M if V1And V2Is associated, then m ij1, otherwise mij0; for matrix Ma×bIf there is a mapping such that there is only one 1 per row and no more than one 1 per column, then the matrix M isa×bIs shown as G1And G2The isomorphic mapping between the two steps is as follows:
(a) initializing mapping matrix M and setting two empty sets V1And V2Using them to store the matched vertexes of two graphs in the searching process;
(b) the search is performed from the first row of the matrix M, looking up the column with a value of 1 from left to right. For any ith row, if the value of jth column is 1 and the column is not occupied, then it means that a possible vertex association is found: drawing G1The ith vertex of (1) and graph G2Corresponds to the jth vertex of (1);
(c) adding the two vertexes of the previous step into the set V respectively1And V2；
(d) If V1And V2The newly added vertex is the vertex pair corresponding to the matching in the final isomorphic mapping, and G can be obtained1Set of vertices V of matched1Constructed subgraph and G2Set of vertices V of matched2The constructed subgraphs are isomorphic;
(6) on the basis of a graph isomorphic matching method, an attribute adjacency graph which accords with an assembly characteristic conjugate subgraph is searched from a retrieval model base by adopting a frequent subgraph mining-based algorithm:
(a) connecting k-order frequent subgraphs to generate k + 1-order candidate subgraphs;
(b) and shearing the candidate K +1 order sub-graph set, and deleting all candidate K +1 order sub-graphs containing K order infrequent sub-graphs. (the algorithm is shown in an article, namely, the public reusable local structure automatic extraction of the three-dimensional CAD model, and the study and report of computer aided design and graphics, 2011,23(9)1512 and 1519.);
(c) calculating the frequency of all sub-graphs in the candidate k + 1-order sub-graph set;
(d) deleting the candidate K + 1-order subgraph set with the frequency less than SminThe candidate subgraph of (1);
(7) and outputting the three-dimensional model according to the retrieval result.
The method of the present invention is further illustrated with reference to the accompanying drawings and examples.
Implementing a model library: a set of machine tool accessory CAD model library has 10 sets of assembly fixture, and more than 100 parts.
The piston rod capable of being assembled with the piston cylinder is searched in the model base by taking the piston cylinder as a target searching object.
The technical scheme for solving the technical problem is as follows: firstly, parts which accord with the assembly design intention are searched out in an assembly information search mode through code search. The model subjected to semantic retrieval eliminates irrelevant assembly part models, and reduces the number of models for retrieval based on graph isomorphism. And based on graph isomorphism retrieval, extracting the attribute adjacency graphs conforming to the assembled conjugate subgraphs by adopting a frequent subgraph mining method. For verification at different degrees of frequency SminIn the case of (3), the search is performed at Smin1.00 and SminThe experiment was carried out under 0.60. Which comprises the following steps:
(1) the overall retrieval method is schematically shown in fig. 1, and the overall flow chart is shown in fig. 2. Firstly, according to the three-dimensional assembly model assembly design intention, an assembly information coding scheme is formulated, as shown in FIG. 3;
(a) taking four kinds of assembly information of the connection relation of the assembly parts, the motion relation of the assembly parts, the space positioning constraint of the assembly parts and the materials of the assembly parts as four code bits of the code;
(b) and by adopting a chain type coding scheme, the assembly information in each code bit is independent and is irrelevant to the front code bit and the rear code bit. Numbers 1,2, 3 and 4 are used for respectively representing connection relation, motion relation, space constraint and material type and are called classification codes;
(c) the connection relation is represented by an letter code and a numeral code, the motion relation is represented by an letter code, the space constraint is represented by two letter codes, and the material type is represented by an letter code and a numeral code, which are called expression codes. When writing the code, the following steps are used: "separate the classification code from the expression code, when there is no corresponding semantic relation, can set the corresponding classification code empty;
(2) establishing an assembly information similarity evaluation model, converting the same code number obtained by comparing the assembly part model code with the target model code into a mathematical model capable of being qualitatively analyzed, wherein the similarity is as follows:
ω1,ω2＞0，ω1+ω2＝1。
four code bits are set for the assembly semantics, each code bit having 2 codes for a total of 8 codes. In this example, the piston rod of the hydraulic cylinder is coded as 1: Z2-2: Z-3: PP-4: G2, and if the output model is coded as 1: Z2-2: Z-3: PP-4: G2, the value is omega1＝0.5，ω20.5; if the output model code is 1: Z2-3: PP-4: G2, omega1＝0.43，ω2＝0.57。
(3) Retrieving assembly information;
(a) obtaining a code to be retrieved through an interactive mode, transmitting the code into an identification Agent, and if the code does not accord with a system rule, ending the program;
(b) the recognition Agent judges the codes to obtain the number of the coded bits and the assembly semantic types corresponding to the codes, and transmits the result to the query Agent;
(c) the query Agent matches the codes in the model library according to the result input by the identification Agent;
(d) calculating the similarity between the models according to the matching result;
(f) the execution Agent lists the corresponding models one by one according to the size of the similarity;
(4) extracting the assembly information retrieval results according to the threshold set by the similarity, as shown in fig. 6, and representing the assembly information retrieval results as attribute adjacency graphs;
(5) carrying out graph isomorphism judgment on attribute adjacency graphs conforming to conjugate subgraphs, wherein the attribute adjacency graphs of the assembled parts are shown asAs shown in fig. 4. With an a x b mapping Ma×bMatrix representation of graph G1And G2Vertex correspondence between them. m isijIs to map i rows and j columns of elements in the matrix M if V1And V2Is associated, then m ij1, otherwise mij0. For matrix Ma×bIf there is a mapping such that there is only one 1 per row and no more than one 1 per column, then the matrix M isa×bIs shown as G1And G2Isomorphic mapping between;
(a) initializing mapping matrix M and setting two empty sets V1And V2Using them to store the matched vertexes of two graphs in the searching process;
(b) the search is performed from the first row of the matrix M, looking up the column with a value of 1 from left to right. For any ith row, if the value of jth column is 1 and the column is not occupied, then it means that a possible vertex association is found: drawing G1The ith vertex of (1) and graph G2Corresponds to the jth vertex of (1);
(c) adding the two vertexes of the previous step into the set V respectively1And V2；
(d) If V1And V2The newly added vertex is the vertex pair corresponding to the matching in the final isomorphic mapping, and G can be obtained1Set of vertices V of matched1Constructed subgraph and G2Set of vertices V of matched2The constructed subgraphs are isomorphic;
(6) on the graph isomorphic matching method, a frequent subgraph mining-based algorithm is adopted, the algorithm flow is shown in FIG. 5, and an attribute adjacency graph which accords with an assembly characteristic conjugate subgraph is searched from a retrieval model library;
(a) connecting k-order frequent subgraphs to generate k + 1-order candidate subgraphs;
(b) shearing a candidate K +1 order sub-graph set, and deleting all candidate K +1 order sub-graphs containing K order infrequent sub-graphs;
(c) calculating the frequency of all sub-graphs in the candidate k + 1-order sub-graph set;
(d) deleting the candidate subgraph with the frequency less than 1 in the candidate K + 1-order subgraph set;
(f) circulating a to c, and deleting the candidate subgraphs with the frequency less than 0.6 in the candidate K + 1-order subgraph set;
(7) based on the search result, a three-dimensional model is output, as shown in fig. 7.
The present invention has been described in terms of specific examples, which are provided to aid understanding of the invention and are not intended to be limiting. For a person skilled in the art to which the invention pertains, several simple deductions, modifications or substitutions may be made according to the idea of the invention.
Claims (3)
1. The three-dimensional assembly model retrieval method integrating the assembly information is characterized in that the assembly information is retrieved, the symbolized codes are used for representing the assembly information, and the codes of parts are searched to find out the three-dimensional model which accords with the assembly design intention; then, converting the geometric retrieval of the assembled parts into an attribute adjacency graph which is matched with the conjugate subgraph, and performing model retrieval through the attribute adjacency graph of the parts; finally, extracting the attribute adjacency graphs meeting the conjugate subgraphs by using a frequent subgraph mining method so as to reuse the model;
the method specifically comprises the following steps:
(1) formulating an assembly information coding scheme according to the assembly design intention of the three-dimensional assembly model;
(2) establishing an assembly information similarity evaluation model, converting the same code number obtained by comparing the assembly part model code with the target model code into a mathematical model capable of being qualitatively analyzed, wherein the similarity is as follows:
ω1,ω2＞0，ω1+ω2＝1；
similarity of the assembled part model to the target model, S (C)i∩Ck) -the assembly part code is the same number of bits as the target model code, NiThe number of bits, ω, encoded by the model ii-weight coefficients；
(3) Retrieving assembly information;
(a) obtaining a code to be retrieved through an interactive mode, transmitting the code into an identification Agent, and if the code does not accord with a system rule, ending the program;
(b) the recognition Agent judges the codes to obtain the number of the coded bits and the assembly semantic types corresponding to the codes, and transmits the result to the query Agent;
(c) the query Agent matches the codes in the model library according to the result input by the identification Agent;
(d) calculating the similarity between the models according to the matching result;
(f) the execution Agent lists the corresponding models one by one according to the size of the similarity;
(4) extracting the assembly information retrieval results according to a threshold value set by the similarity, and representing the assembly information retrieval results as attribute adjacency graphs;
(5) and (3) carrying out graph isomorphism judgment on the attribute adjacent graphs conforming to the conjugate subgraphs: with an a x b mapping Ma×bMatrix representation of graph G1And G2Vertex correspondence between mijIs to map i rows and j columns of elements in the matrix M if V1And V2Is associated, then mij1, otherwise mij0; for matrix Ma×bIf there is a mapping such that there is only one 1 per row and no more than one 1 per column, then the matrix M isa×bIs shown as G1And G2The isomorphic mapping between the two steps is as follows:
(a) initializing mapping matrix M and setting two empty sets V1And V2Using them to store the matched vertexes of two graphs in the searching process;
(b) searching from the first row of the matrix M, and searching a column with the value of 1 from left to right; for any ith row, if the value of jth column is 1 and the column is not occupied, then it means that a possible vertex association is found: drawing G1The ith vertex of (1) and graph G2Corresponds to the jth vertex of (1);
(c) adding the two vertexes of the previous step into the set V respectively1And V2；
(d) If V1And V2The newly added vertex is the vertex pair corresponding to the matching in the final isomorphic mapping, and G can be obtained1Set of vertices V of matched1Constructed subgraph and G2Set of vertices V of matched2The constructed subgraphs are isomorphic;
(6) on the basis of a graph isomorphic matching method, an attribute adjacency graph which accords with an assembly characteristic conjugate subgraph is searched from a retrieval model library which extracts an assembly information retrieval result by adopting a frequent subgraph mining-based algorithm;
(7) and outputting the three-dimensional model according to the retrieval result.
2. The method according to claim 1, wherein the step (1) comprises the following specific steps:
(a) taking four kinds of assembly information of the connection relation of the assembly parts, the motion relation of the assembly parts, the space positioning constraint of the assembly parts and the materials of the assembly parts as four code bits of the code;
(b) adopt the chain coding scheme, each assembly information is independent in every code bit, and the code bit is irrelevant around with: numbers 1,2, 3 and 4 are used for respectively representing connection relation, motion relation, space constraint and material type and are called classification codes;
(c) the connection relation is represented by an letter code and a number code, the motion relation is represented by an letter code, the space constraint is represented by two letter codes, and the material type is represented by an letter code and a number code, which are called expression codes; when writing the code, the following steps are used: "separate the classification code from the expression code, and when there is no corresponding semantic relationship, the corresponding classification code can be nulled.
3. The method according to claim 1, wherein the step (6) comprises the following specific steps:
(a) connecting k-order frequent subgraphs to generate k + 1-order candidate subgraphs;
(b) shearing a candidate K +1 order sub-graph set, and deleting all candidate K +1 order sub-graphs containing K order infrequent sub-graphs;
(c) calculating the frequency of all sub-graphs in the candidate k + 1-order sub-graph set;
(d) deleting the candidate K + 1-order subgraph set with the frequency less than SminThe candidate subgraph of (1).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810115326.7A CN108573020B (en) | 2018-02-06 | 2018-02-06 | Three-dimensional assembly model retrieval method integrating assembly information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810115326.7A CN108573020B (en) | 2018-02-06 | 2018-02-06 | Three-dimensional assembly model retrieval method integrating assembly information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108573020A CN108573020A (en) | 2018-09-25 |
| CN108573020B true CN108573020B (en) | 2021-07-02 |
Family
ID=63576588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810115326.7A Active CN108573020B (en) | 2018-02-06 | 2018-02-06 | Three-dimensional assembly model retrieval method integrating assembly information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108573020B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399657B (en) * | 2019-07-09 | 2022-12-13 | 西安工业大学 | Excavation method based on three-dimensional assembly model common structure |
| CN112699518A (en) * | 2019-10-22 | 2021-04-23 | 晟通科技集团有限公司 | Template preassembling-free method |
| CN111507007B (en) * | 2020-04-22 | 2023-07-07 | 西安工业大学 | Method for constructing general structure of three-dimensional assembly model based on conjugate subgraph |
| CN112989518B (en) * | 2021-03-16 | 2023-03-21 | 华南理工大学 | Intelligent assembly modeling method based on design intention |
| CN113806891B (en) * | 2021-09-23 | 2023-08-04 | 山东大学深圳研究院 | Quick design method of clamp suitable for workpiece change |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004164503A (en) * | 2002-11-15 | 2004-06-10 | Olympus Corp | Three-dimensional model retrieval method, three-dimensional model retrieval device, three-dimensional model retrieval program and three-dimensional model retrieval system |
| CN101446958A (en) * | 2008-12-12 | 2009-06-03 | 大连理工大学 | Method for searching CAD model based on topological adjacency approximation |
| CN103810299A (en) * | 2014-03-10 | 2014-05-21 | 西安电子科技大学 | Image retrieval method on basis of multi-feature fusion |
| CN104462163A (en) * | 2014-03-06 | 2015-03-25 | 北京工商大学 | Three-dimensional model characterization method, search method and search system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8015125B2 (en) * | 2006-08-31 | 2011-09-06 | Drexel University | Multi-scale segmentation and partial matching 3D models |
-
2018
- 2018-02-06 CN CN201810115326.7A patent/CN108573020B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004164503A (en) * | 2002-11-15 | 2004-06-10 | Olympus Corp | Three-dimensional model retrieval method, three-dimensional model retrieval device, three-dimensional model retrieval program and three-dimensional model retrieval system |
| CN101446958A (en) * | 2008-12-12 | 2009-06-03 | 大连理工大学 | Method for searching CAD model based on topological adjacency approximation |
| CN104462163A (en) * | 2014-03-06 | 2015-03-25 | 北京工商大学 | Three-dimensional model characterization method, search method and search system |
| CN103810299A (en) * | 2014-03-10 | 2014-05-21 | 西安电子科技大学 | Image retrieval method on basis of multi-feature fusion |
Non-Patent Citations (2)
| Title |
|---|
| Automatic discovery of common design structures in CAD models;LujieMa;《Computers & Graphics》;20101030;全文 * |
| 支持产品快速设计的CAD模型重用技术研究;马铁强;《中国博士学位论文全文数据库信息科技辑》;20091015;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108573020A (en) | 2018-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108573020B (en) | Three-dimensional assembly model retrieval method integrating assembly information | |
| US11113431B2 (en) | MBD-based three-dimensional process designing method and platform for typical automobile machined part | |
| CN110597735B (en) | Software defect prediction method for open-source software defect feature deep learning | |
| CN105095204B (en) | The acquisition methods and device of synonym | |
| CN111090461B (en) | Code annotation generation method based on machine translation model | |
| CN106682343B (en) | Formal verification method of adjacency matrix based on graph | |
| CN112800172B (en) | Code searching method based on two-stage attention mechanism | |
| Zhang et al. | Zero-shot hashing with orthogonal projection for image retrieval | |
| CN114218389A (en) | Long text classification method in chemical preparation field based on graph neural network | |
| KR20110062870A (en) | Apparatus for search of 3d model data | |
| CN105205135A (en) | 3D (three-dimensional) model retrieving method based on topic model and retrieving device thereof | |
| Piccialli et al. | SOS-SDP: an exact solver for minimum sum-of-squares clustering | |
| Stephan et al. | Using mutation analysis for a model-clone detector comparison framework | |
| Hoseini et al. | A new algorithm for mining frequent patterns in can tree | |
| Song et al. | Interactive visual pattern search on graph data via graph representation learning | |
| CN111382333B (en) | Case element extraction method in news text sentence based on case correlation joint learning and graph convolution | |
| CN112148879B (en) | Computer readable storage medium for automatically labeling code with data structure | |
| CN110765781A (en) | Man-machine collaborative construction method for domain term semantic knowledge base | |
| CN110765276A (en) | Entity alignment method and device in knowledge graph | |
| Kordomatis et al. | Web object identification for web automation and meta-search | |
| CN101334793A (en) | Method for automatic recognition for dependency relationship of demand | |
| CN114579468A (en) | Source item selection software defect prediction method based on semantic metric value | |
| CN106156259A (en) | A kind of user behavior information displaying method and system | |
| CN117573096B (en) | Intelligent code completion method integrating abstract syntax tree structure information | |
| CN116680418B (en) | Big data retrieval method and system based on knowledge graph |
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 | ||
| GR01 | Patent grant |