An automatic recognition method, system, device and medium for a typical structure of a complex product based on instance library matching

Abstract

The present application relates to the technical field of computer-aided design, in particular to a complex product typical structure automatic identification method, system, device and medium based on instance library matching; the method constructs a pre-defined typical structure instance library, comprehensively considers part shape information and spatial neighborhood characteristics between parts, performs part matching in a product three-dimensional model with a typical structure instance as a target, and finds structures similar in shape and structure through depth-first search, finally realizes identification of the typical structure, and effectively avoids the problem that it is difficult to obtain part connection relationships in a complex product.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer-aided design, in particular to a complex product typical structure automatic identification method, system, device and medium based on instance library matching. BACKGROUND

[0002] Process procedure is a production process document prepared by process department according to design requirements, process technical requirements and quality requirements. Process procedure not only includes process information, but also production information and quality information, which guides workers to operate the specific work instructions of the specified assembly process flow, including operation instructions, process, assembly sequence, change record and other information.

[0003] The content of process procedure is closely related to the structural features of the product. Specifically, the content of the process procedure of some typical local structures has a strong reference value. If these information can be obtained and reused, the preparation efficiency can be greatly accelerated. In order to achieve this goal, the effective identification of typical structures is needed first.

[0004] The Chinese invention patent with the title of "Assembly body retrieval method based on spatial connection skeleton descriptor" and the authorized announcement of "CN108628965B" proposes an assembly body retrieval method based on spatial connection skeleton descriptor. The spatial connection skeleton is established according to the position of the part center, the position of the fitting surface center and the connection relationship between the part and the fitting surface in the assembly body. The spatial distance distribution of all random sampling points on the surface of two parts through the skeleton is calculated as the descriptor input for assembly body retrieval. On this basis, the assembly body similarity comparison is carried out through the optimal sub-sequence bijective matching method, and the comprehensive retrieval of the shape and connection relationship of the assembly body model is realized. However, this method needs to consider the skeleton distance of any two part pairs in the assembly body. For complex product models with too many parts, the description process of this method is relatively cumbersome.

[0005] A typical structure discovery method of fusion local difference for assembly is proposed in the document "Wang P, Zhang J, Li Y, et al. Reuse-oriented common structured discovery in assembly models[J]. Journal of Mechanical Science and Technology, 2017, 31: 297-307", similar parts and connection relationships are clustered on the basis of part similarity analysis, and GSpan algorithm is used to discover assembly graph descriptors, and the obtained frequent subgraph corresponds to the typical structure in the assembly. The method relies on the algorithm to discover the typical structure automatically, and the object and range of the identification result cannot be effectively controlled, so there is still certain limitation in actual use.

[0006] Therefore, the existing identification method has the problem that the typical structure in the complex product is difficult to effectively obtain. SUMMARY

[0007] The present application proposes a complex product typical structure automatic identification method, system, device and medium based on instance library matching, which can effectively solve the problem that the typical structure in the complex product is difficult to effectively obtain by the existing identification method. The method constructs a pre-defined typical structure instance library, comprehensively considers the part shape information and the spatial neighborhood characteristics between parts, matches the parts in the product three-dimensional model with the typical structure instance as the target, and finds the structure similar in shape and structure through depth first search, finally realizes the identification of the typical structure, and effectively avoids the problem that the connection relationship of the parts in the complex product is difficult to obtain.

[0008] The specific implementation content of the present application is as follows:

[0009] A complex product typical structure automatic identification method based on instance library matching, first constructs a typical structure instance library according to the obtained STG model; secondly, according to the constructed typical structure instance library, obtains the part shape information and the spatial neighborhood characteristics between parts; then, according to the part shape information and the spatial neighborhood characteristics between parts, obtains the matching candidate set of the target product three-dimensional model; finally, according to the matching candidate set, obtains the spatial distribution aggregation of the matching parts, and identifies the typical structure.

[0010] In order to better realize the present application, further, the complex product typical structure automatic identification method based on instance library matching specifically includes the following steps:

[0011] Step S1: according to the obtained STG model, a typical structure instance library is constructed, the STG model is a model with local features of a three-dimensional model as vertices and adjacency relations between the local features as edges, and the instances in the typical structure instance library are assemblies;

[0012] Step S2: according to the constructed typical structure instance library, part shape information and spatial proximal characteristics between parts are obtained, the obtained part shape information is converted into a k-dimensional vector, and a proximal vector set is calculated according to the obtained spatial proximal characteristics between parts;

[0013] Step S3: similarity between parts is calculated according to the k-dimensional vector of the parts, an initial set of similar parts is obtained according to the similarity, proximal set similarity between parts in the initial set is calculated according to the proximal vector set, and a matching candidate set corresponding to each part is obtained according to the proximal set similarity;

[0014] Step S4: matching parts are identified according to the matching candidate set of each part in the typical structure instance and the spatial distribution aggregation of the matching parts, and a typical structure composed of the matching parts is obtained.

[0015] In order to better realize the present application, further, the step S2 specifically comprises the following steps:

[0016] Step S21: according to the constructed typical structure instance library, parts of the typical structure instance and the product model are obtained, and a shape distribution algorithm is called to describe the shape information of the parts as a k-dimensional vector;

[0017] Step S22: according to the constructed typical structure instance library, the typical structure instance is obtained, proximal characteristics between parts are calculated, and a proximal vector set is obtained.

[0018] In order to better realize the present application, further, the step S22 specifically comprises the following steps:

[0019] Step S221: according to the constructed typical structure instance library, the typical structure instance is obtained, and an OBB bounding box of the parts is constructed; the typical structure instance includes a maximum length, a maximum width and a maximum height of the parts;

[0020] Step S222: according to the interference condition of the bounding box, spatial distribution relations between the parts are judged, and a proximal vector set is obtained;

[0021] Step S223: steps S221-S222 are repeated until each part in the product model and the typical structure instance library is traversed.

[0022] In order to better realize the present application, further, the step S3 specifically comprises the following steps:

[0023] Step S31: based on the set shape similarity threshold and the similarity between part p and part q, coarsely filter part p to obtain an initial set;

[0024] Step S32: Based on the initial set, calculate the similarity between the parts in the initial set and the neighboring set of part p;

[0025] Step S33: Match the initial set according to the set neighbor similarity threshold and neighbor similarity to obtain a matching candidate set.

[0026] In order to better implement the present invention, further, step S4 specifically includes the following steps:

[0027] Step S41: constructing an ordered set according to the size of the part neighbor set in the typical structure instance;

[0028] Step S42: Initialize the set list, set Ca, set Flag, and set Tar;

[0029] Step S43: assign the ordered set to the set list. If part p is the first element of the set list, assign the matching candidate set to the set Ca.

[0030] Step S44: randomly select part q1 from the set Ca and determine whether it meets the set criteria;

[0031] Step S45: If not satisfied, assign Ca-{q1} to the set Ca, and assign Ca p -{q1} is assigned to the matching candidate set Ca of part p p ; If the collection , then select the next part q2 in the set Ca and return to step S44; if the set , then the search ends; Ca-{q1}, Ca p -{q1} is respectively from Ca, Ca p The set after removing the element {q1};

[0032] Step S46: If the set criteria are met, then add the part p to the set Flag and create the neighbor vector set Nei of the part p. p An ordered set list p ;

[0033] Step S47: If , then the set of parts corresponding to the output set Tar is taken as the typical structure, list ins — Flag Assign to the collection list, Assign the value to the set Ca and return to step S44, listins as a typical structure instance ins an ordered set;

[0034] Step S48: if , then is assigned to the set Tar, and is assigned to the set Ca, and list p is assigned to the set list, and the process returns to step S44, a neighbor vector set of the part q1;

[0035] Step S49: the output set Tar is taken as a matched typical structure.

[0036] To better implement the present application, further, the criterion condition set in the step S44 is:

[0037] ;

[0038] wherein p is the first element of the initialized set list, Nei p represents a neighbor vector set of the part p , Nei q1 represents a neighbor vector set of the part q 1 , and u and v are respectively one neighbor vector in Nei p and Nei q1 .

[0039] To better implement the present application, further, the process of establishing the STG model in the step S1 is: first, a vertex adjacency graph is established according to the topological relationship between the acquired B-rep data geometric elements; second, a maximal clique in the adjacency graph is searched and the geometric region corresponding to the maximal clique is taken as a local feature of the solid model; then, a statistical method is called to convert the shape information of the local feature into a feature vector, form a feature space, and suppress the difference of the local features by using an unsupervised learning algorithm; finally, the STG model is established, taking the local features as vertices and the adjacency relationship between the local features as edges.

[0040] Based on the above-mentioned complex product typical structure automatic recognition method based on instance library matching, to better implement the present application, further, a complex product typical structure automatic recognition system based on instance library matching is proposed, which is used for executing the above-mentioned complex product typical structure automatic recognition method based on instance library matching; comprising an instance library construction unit, a feature acquisition unit, a matching candidate unit, and a recognition unit.

[0041] The instance library construction unit is used for constructing a typical structure instance library according to the acquired STG model.

[0042] The feature acquisition unit is configured to acquire part shape information and spatial proximate characteristics between parts according to the constructed typical structure instance library.

[0043] The matching candidate unit is configured to obtain a matching candidate set of the three-dimensional model of the target product according to the part shape information and the spatial proximate characteristics between parts.

[0044] The recognition unit is configured to acquire spatial distribution aggregation of matching parts according to the matching candidate set, and recognize a typical structure.

[0045] Based on the above-mentioned instance library matching-based complex product typical structure automatic recognition method, in order to better realize the present application, further, an electronic device is provided, comprising a memory and a processor; the memory stores a computer program; when the computer program is executed on the processor, the above-mentioned instance library matching-based complex product typical structure automatic recognition method is realized.

[0046] Based on the above-mentioned instance library matching-based complex product typical structure automatic recognition method, in order to better realize the present application, further, a computer readable storage medium is provided, the computer readable storage medium stores computer instructions; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned instance library matching-based complex product typical structure automatic recognition method is realized.

[0047] The present application has the following advantages:

[0048] (1) The present application replaces the part connection relationship in the existing method with the spatial distribution relationship between parts, which can effectively avoid the problem of difficulty in obtaining the part connection relationship in a complex product.

[0049] (2) The present application is realized by a pre-constructed instance library, and the recognition object is controlled by increasing, deleting or modifying the instance library, which has good expansibility. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 The complex product typical structure automatic recognition method provided by the present application is shown in the overall flowchart.

[0051] Figure 2 The typical structure instance library constructed in the method embodiment of the present application is shown in the example diagram.

[0052] Figure 3 The proximate set of the ear part connected by the first typical structure instance and the shape vector of the part in the method embodiment of the present application are shown in the example diagram.

[0053] Figure 4 The three-dimensional model instance diagram of the card plate product in the method embodiment of the present application is shown in the example diagram.

[0054] Figure 5 An example diagram of a matching candidate set of rivet parts in the method embodiment of the present application.

[0055] Figure 6 A flow chart of searching a typical structure using matching parts in the method embodiment of the present application.

[0056] Figure 7 An example diagram of a part set corresponding to a typical structure identified from a card product in the method embodiment of the present application. EMBODIMENT

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. It should be understood that the described embodiments are only some of the embodiments of the present application, not all the embodiments, and therefore should not be regarded as limiting the scope of protection. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0058] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "set", "connected", "connected" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0059] Embodiment 1:

[0060] The embodiment proposes a complex product typical structure automatic identification method based on instance library matching. Firstly, according to the obtained STG model, a typical structure instance library is constructed. Secondly, according to the constructed typical structure instance library, the part shape information and the spatial neighborhood characteristics between parts are obtained. Then, according to the part shape information and the spatial neighborhood characteristics between parts, a matching candidate set of the target product three-dimensional model is obtained. Finally, according to the matching candidate set, the spatial distribution aggregation of the matching parts is obtained, and the typical structure is identified.

[0061] Working principle: the embodiment constructs a pre-defined typical structure instance library, comprehensively considers part shape information and spatial neighborhood characteristics between parts, performs part matching in a product three-dimensional model by taking typical structure instances as targets, and finds structures similar in shape and structure by depth-first search, so that the typical structure recognition is finally realized; the spatial distribution relationship between parts is used to replace the part connection relationship in the existing method, so that the problem of difficulty in obtaining the part connection relationship in a complex product can be effectively avoided.

[0062] Embodiment 2

[0063] The embodiment is based on the above-mentioned embodiment 1 and is described in the form of steps.

[0064] The complex product typical structure automatic recognition method based on the instance library matching specifically includes the following steps.

[0065] Step S1: constructing a typical structure instance library according to the obtained STG model.

[0066] The STG model in step S1 takes local features as vertices and adjacency relationships as edges; the instances in the instance library are stored in the form of assemblies.

[0067] Step S2: converting the obtained part shape information into a k-dimensional vector according to the constructed typical structure instance library, and calculating a neighbor vector set according to the obtained spatial neighborhood characteristics between parts.

[0068] The step S2 specifically includes the following steps.

[0069] Step S21: obtaining parts p of the typical structure instance library and the product model according to the constructed typical structure instance library, and calling a shape distribution algorithm to describe the shape information of the parts as a k-dimensional vector.

[0070] Step S22: obtaining a typical structure instance according to the constructed typical structure instance library, calculating the neighborhood characteristics between parts, and obtaining a neighbor vector set.

[0071] In order to better realize the present application, further, the step S22 specifically includes the following steps.

[0072] Step S221: obtaining a typical structure instance according to the constructed typical structure instance library, and constructing an OBB bounding box of the part; the typical structure instance includes the maximum length, width and height of the part p.

[0073] Step S222: judging the spatial distribution relationship between parts according to the interference condition of the bounding box, and obtaining a neighbor vector set.

[0074] Step S223: repeat step S221-step S222 until each part in the product model and the typical structure instance library is traversed.

[0075] Step S3: search for a matching candidate set of the target product three-dimensional model according to the k-dimensional vector and the set of near neighbor vectors.

[0076] The step S3 specifically includes the following steps:

[0077] Step S31: according to the set shape similarity threshold and the part p and part q similarity, the initial set is obtained by rough filtering part p;

[0078] Step S32: according to the initial set, the near neighbor set similarity of the parts in the initial set and part p is calculated;

[0079] Step S33: according to the set near neighbor similarity threshold and the near neighbor similarity, the initial set is matched to obtain the matching candidate set.

[0080] Step S4: according to the matching candidate set, the spatial distribution aggregation of the matching parts is obtained, and the typical structure composed of the matching parts is identified.

[0081] The step S4 specifically includes the following steps:

[0082] Step S41: according to the size of the part near neighbor set in the typical structure instance, an ordered set is constructed;

[0083] Step S42: initialize the set list, the set Ca, the set Flag, and the set Tar;

[0084] Step S43: assign the ordered set to the set list, and if part p is the first element of the set list, assign the matching candidate set to the set Ca;

[0085] Step S44: randomly select part q1 from the set Ca, and judge whether the set criteria is met;

[0086] Step S45: if not, assign Ca-{q1} to the set Ca, and assign Cap-{q1} to the matching candidate set Ca p ; if the set , select the next part q2 in the set Ca, and return to step S44; if the set , end the search; Ca-{q1}, Ca p -{q1} are the sets obtained by removing the element {q1} from Ca, Ca p , respectively;

[0087] Step S46: If the set criterion is met, the part p is added to the set Flag, and a near neighbor vector set Nei is established p ; p ;

[0088] Step S47: If , the part set corresponding to the output set Tar is taken as the typical structure, the value of is assigned to the set list, the value of is assigned to the set Ca, and the process returns to step S44;

[0089] Step S48: If , the value of is assigned to the set Tar, the value of is assigned to the set Ca, the value of is assigned to the set list, and the process returns to step S44;

[0090] Step S49: The output set Tar is taken as the matched typical structure.

[0091] Working principle: In this embodiment, the process personnel selects part of the typical structures in the product according to the work experience and business needs to construct an instance library; then, taking the typical structure instance as input, the shape similarity is calculated to preliminarily filter the parts in the typical structure and the target product three-dimensional model, and the near neighbor similarity between the parts is used for further optimization to obtain the candidate matching set of each part in the typical structure instance. Finally, the depth-first search algorithm is used to combine the same or similar structure as the target structure from all the matched parts according to the spatial distribution characteristics of the matched parts in the product and output, so as to realize the recognition of the typical structure.

[0092] The other parts of this embodiment are the same as those of the above-mentioned embodiment 1, and thus will not be described again.

[0093] Embodiment 3

[0094] This embodiment is based on any one of the above-mentioned embodiments 1-2, and is described with a specific embodiment as shown in Figure 1 .

[0095] Step S1: According to the acquired STG model, the instance library of the typical structure is constructed according to the process characteristics, wherein each instance is stored in the form of an assembly.

[0096] Step S21: For each part p in the instance library and the product model, the shape information of the part is described as an k dimensional vector by using the shape distribution algorithm.

[0097] Step S22: Calculate the near-neighbor characteristics of parts by using OBB for each typical structure instance and product model,

[0098] The specific steps are as follows:

[0099] Step S221: Obtain parts p Construct OBB of parts p;

[0100] Step S222: Determine the spatial distribution relationship between parts according to the OBB interference condition, and obtain the near-neighbor vector set of parts p :

[0101] (1);

[0102] Wherein: A represents the instance or product model in which part p is located, q represents a part model in A other than p, and and respectively represent the OBB of parts p and q, represents the overlapping region of the OBB in three-dimensional space.

[0103] Step S223: Repeat steps S221-S222 until each part in the product and instance library is traversed.

[0104] Step S3: For a product model A and each typical structure instance Ins , search for its matching candidate set in the product Ins according to the part shape and near-neighbor similarity in A , and the specific steps are as follows:

[0105] Step S31: Set a shape similarity threshold , and perform coarse filtering on parts according to the size of part similarity to obtain an initial set :

[0106] (2);

[0107] Wherein, s p represents the shape vector of part p, s q represents the shape vector of part q, k represents the dimension of the k-dimensional shape vector of part, and l i p represents the i-dimensional shape vector of part p, and l i q represents the i-dimensional shape vector of part q.

[0108] Step S32: For the initial set after coarse filtering, calculate the near-neighbor vector set similarity of each part to part p :

[0109] (3);

[0110] wherein, denotes a part p the size of the neighborhood vector set.

[0111] Step S33: setting the shape neighborhood similarity threshold , according to the size of the neighborhood similarity to perform fine matching, obtaining the matching candidate set of part p ;

[0112] (4).

[0113] Step S4: according to the matching candidate set of each part in Ins , according to the spatial distribution of the matching parts, the typical structure recognition is realized:

[0114] Step S41: according to the size of the neighborhood set of each part in Ins , an ordered set is constructed:

[0115] (5);

[0116] Step S42: initializing the set list , Ca , Flag , Tar ;

[0117] Step S43: assigning , assuming p is the first element of list , then assigning ;

[0118] Step S44: assuming list is the first element of p , selecting one part Ca in , the following judgment is made:

[0119] (6);

[0120] Step S45: if the criterion is not true, then assigning , , if the set , selecting the next part q2 in Ca and returning to step S44; if the set , the search is ended;

[0121] ​Step S46: If the criterion is satisfied, the part p is added to the set Flag , an ordered set is established :

[0122] (7) ;

[0123] Step S47: If , the corresponding part set is output as the typical structure, and the following assignments are made, and the process returns to step S44; Tar

[0124] (8) ;

[0125] Step S48: If , the following assignments are made, and the process returns to step S44;

[0126] (9) ;

[0127] Step S49: The above recursive process is completed, and all output Tar is the matching typical structure.

[0128] Working principle: The typical structure recognition of this embodiment is realized through a pre-defined instance library. Process personnel can control the recognition objects by increasing, deleting, and modifying the instance library, which has good scalability. The spatial distribution relationship between parts is used to replace the part connection relationship in existing methods, which can effectively avoid the problem of difficulty in obtaining the part connection relationship in complex products.

[0129] The other parts of this embodiment are the same as any one of the above embodiments 1-2, and thus will not be described again.

[0130] Embodiment 4

[0131] This embodiment is based on any one of the above embodiments 1-3, as shown in Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , taking a clamp commonly used in the mechanical industry with clamping function as an example, the specific steps of the complex product typical structure automatic recognition method based on instance library matching are as follows:

[0132] Step S1: A process personnel constructs an instance library of typical structures according to process characteristics, wherein each instance is stored in the form of an assembly. In this embodiment, two typical structure instances are defined, as shown in Figure 2 ;

[0133] ​Step S2: For each part in the instance library and product model, the shape distribution algorithm is used to describe the shape information of the part as a 50-dimensional vector , the description results of some parts are as follows Figure 3 As shown;

[0134] For the product model and each typical structural instance, the bounding box is used to calculate the neighbor characteristics between parts. The specific steps are as follows:

[0135] Step S221: Get a part in the model p The maximum length, width and height of the part are used to construct the OBB bounding box;

[0136] Step S222: Determine the spatial distribution relationship between parts based on the bounding box interference and obtain a set of neighbor vectors:

[0137] (10);

[0138] Taking the connecting lug in the first typical structure example as an example, the parts included in its neighbor vector set are as follows Figure 3 shown.

[0139] Step S223: repeating steps S221 and S222 until every part in the product and instance library is traversed;

[0140] Step S3: Given Figure 4 A pallet product model shown A ,for Figure 2 Each typical structure instance in Ins ,according to Ins The similarity of part shape and neighbors in the product A Search for its matching candidate set in the following steps:

[0141] Step S31: Set the shape similarity threshold to 0.95, based on the part similarity The size of the parts is roughly filtered to obtain the initial set :

[0142] (11);

[0143] Step S32: Initial set after coarse filtering , calculate each part and part p The similarity of the neighbor set of :

[0144] (12);

[0145] in, Indicates parts p The size of the neighbor set.

[0146] Step S33: Set the shape proximity similarity threshold value as 0.80, and perform fine matching according to the size of the proximity similarity to obtain a matching candidate set ;

[0147] (13) ;

[0148] Take the rivet of the first typical structure as an example. The matching candidate set after fine matching is shown in Figure 5 .

[0149] Step S4: According to the matching candidate set of each part in Ins , consider the spatial distribution aggregation of the matching parts to realize the recognition of the typical structure. The specific process of the recognition process is shown in Figure 6 , and specifically includes the following steps:

[0150] Step S41: According to the size of the proximity vector set of each part in Ins , construct an ordered set:

[0151] (14) ;

[0152] Step S42: Initialize the set list , Ca , Flag , Tar ;

[0153] Step S43: Assign . Assuming that p is the first element of list , assign ;

[0154] Step S44: Select one part in Ca , and make the following judgments:

[0155] (15) ;

[0156] Step S45: If the criterion is not met, assign , . If the set , select the next part in , and return to step S44; if the set , end the search;

[0157] Step S46: If the criterion is met, add the part p to the set Flag , and establish ​ordered set of parts:

[0158] (16);

[0159] Step S47: if , output Tar the corresponding part set as a typical structure, while making the following assignments, and return to step S44;

[0160] (17);

[0161] Step S48: if , make the following assignments, and return to step S44;

[0162] (18);

[0163] Step S49: complete the above recursive process, at which time the output Tar set is output once, and contains 13 part models, as shown in Figure 7 , which represent a typical structure in the product model A .

[0164] This implementation example shows that the complex product typical structure automatic recognition method based on example library matching proposed by the present application can be used for automatic recognition of pre-defined structures, and can achieve good results.

[0165] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary, and should not be construed as limiting the present application, and those of ordinary skill in the art can make changes, modifications, replacements and variations to the above embodiments without departing from the principles and purposes of the embodiments within the scope of the present application.

[0166] The other parts of this embodiment are the same as any one of the above embodiments 1-3, and thus will not be described again.

[0167] Example 5

[0168] This embodiment is based on any one of the above embodiments 1-4, and a specific embodiment is used to illustrate the establishment process of the STG model.

[0169] The process of establishing the STG model is as follows: first, a vertex adjacency graph is established according to the topological relationship between the obtained B-rep data geometric elements; second, a maximum clique in the adjacency graph is searched, and the geometric region corresponding to the maximum clique is taken as a local feature of the entity model; then, a statistical method is called to convert the shape information of the local feature into a feature vector, form a feature space, and suppress the local features by using an unsupervised learning algorithm; finally, an STG model is established, in which the local features are taken as vertices and the adjacency relationship is taken as edges.

[0170] The other parts of the embodiment are the same as any one of Embodiment 1-Embodiment 4 described above, and thus will not be described again.

[0171] Embodiment 6:

[0172] The embodiment is based on any one of Embodiment 1-Embodiment 5 described above, and proposes an automatic recognition system of a typical structure of a complex product based on instance library matching, which is used to execute the automatic recognition method of the typical structure of the complex product based on instance library matching described above; and includes an instance library construction unit, a feature acquisition unit, a matching candidate unit, and a recognition unit.

[0173] The instance library construction unit is configured to construct a typical structure instance library according to the obtained STG model.

[0174] The feature acquisition unit is configured to acquire part shape information and spatial proximal characteristics between parts according to the constructed typical structure instance library.

[0175] The matching candidate unit is configured to obtain a matching candidate set of a target product three-dimensional model according to the part shape information and the spatial proximal characteristics between parts.

[0176] The recognition unit is configured to acquire spatial distribution aggregation of matching parts according to the matching candidate set, and recognize a typical structure.

[0177] The embodiment also proposes an electronic device including a memory and a processor; the memory has stored thereon a computer program; when the computer program is executed on the processor, the automatic recognition method of the typical structure of the complex product based on instance library matching described above is implemented.

[0178] The embodiment also proposes a computer readable storage medium having stored thereon computer instructions; when the computer instructions are executed on the electronic device described above, the automatic recognition method of the typical structure of the complex product based on instance library matching described above is implemented.

[0179] The other parts of the embodiment are the same as any one of Embodiment 1-Embodiment 5 described above, and thus will not be described again.

[0180] The processor involved in the embodiments of the present application can be a chip. For example, it can be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD) or other integrated chip.

[0181] The memory involved in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable ROM (EPROM), an electrically EPROM (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM) and direct rambus RAM (DR RAM). It should be noted that the memory of the system and method described herein is intended to include, but not limited to, these and any other suitable types of memory.

[0182] It should be understood that the size of the sequence number of the above processes does not mean the order of execution in various embodiments of the present application, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0183] Those skilled in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0184] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-described system, device and module can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0185] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interface, device or module, which can be electrical, mechanical or other forms.

[0186] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, that is, they can be located in one device, or can be distributed to multiple devices. According to actual needs, part or all of the modules can be selected to achieve the purpose of the present embodiment.

[0187] In addition, each functional module in each embodiment of the present application can be integrated in one device, or each module can exist physically, or two or more modules can be integrated in one device.

[0188] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by using a software program, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or data storage device including one or more servers, data centers, etc. integrated with the medium. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk (SSD)) and the like.

[0189] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for automatically identifying typical structures of complex products based on case library matching, characterized in that: The following steps are involved: Step S1: constructing a typical structure instance library based on the obtained STG model, wherein the STG model is a model established with local features of a three-dimensional model as vertices and adjacency relationships between local features as edges, and the instances in the typical structure instance library are assemblies; Step S2: Based on the constructed typical structure case library, obtain part shape information and spatial neighbor characteristics between parts, convert the obtained part shape information into a k-dimensional vector, and calculate a set of neighbor vectors based on the obtained spatial neighbor characteristics between parts; Step S3: Calculate the similarity between parts based on the k-dimensional vectors of the parts, and obtain an initial set of similar parts based on the similarity; calculate the neighbor set similarity between the parts in the initial set based on the neighbor vector set, and obtain a matching candidate set corresponding to each part based on the neighbor set similarity; Step S4: identifying matching parts based on the matching candidate set of each part in the typical structure instance and the spatial distribution aggregation of the matching parts, and obtaining a typical structure composed of the matching parts; In step S3, for a product model A And each typical structure instance Ins ,according to Ins The similarity of part shape and neighbors in the product A Search for its matching candidate set in the following steps: Step S31: Setting the shape similarity threshold , based on part similarity The size of the parts is roughly filtered to obtain the initial set : ； Among them, s p Represents the shape vector of part p, s q represents the shape vector of part q, k represents the dimension of the k-dimensional shape vector of part q, l i p Represents the i-th dimension shape vector of part p, l i q Represents the i-th dimension shape vector of part q; Step S32: Initial set after coarse filtering , calculate each part and part p Similarity of neighbor vector set , Nei p Indicates parts p Neighbor vector set, Nei q Indicates parts q The set of neighbor vectors of ; Step S33: Setting the shape neighbor similarity threshold , according to the neighbor similarity Size pair Perform fine matching to obtain the matching candidate set of part p ; ； Step S4 specifically includes the following steps: Step S41: constructing an ordered set according to the size of the part neighbor set in the typical structure instance; Step S42: Initialize the set list, set Ca, set Flag, and set Tar; Step S43: assign the ordered set to the set list. If part p is the first element of the set list, assign the matching candidate set to the set Ca. Step S44: randomly select part q1 from the set Ca and determine whether it meets the set criteria; Step S45: If not satisfied, assign Ca-{q1} to the set Ca, and assign Ca p -{q1} is assigned to the matching candidate set Ca of part p p ; If the collection , then select the next part q2 in the set Ca and return to step S44; if the set , then the search ends; Ca-{q1}, Ca p -{q1} is respectively from Ca, Ca p The set after removing the element {q1}; Step S46: If the set criteria are met, then add the part p to the set Flag and create the neighbor vector set Nei of the part p. p An ordered set list p ; Step S47: If , then the set of parts corresponding to the output set Tar is taken as the typical structure, list ins — Flag Assign to the collection list, Assign the value to the set Ca and return to step S44, list ins A typical structural example ins An ordered set of Step S48: If , then Assign to the collection Tar, Assign to the set Ca, list p Assign the value to the set list and return to step S44; is the set of neighbor vectors of part q1; Step S49: Use the output set Tar as the typical matching structure.

2. The method for automatically identifying typical structures of complex products based on case library matching according to claim 1 is characterized in that: The step S2 specifically includes the following steps: Step S21: according to the constructed typical structure instance library, obtain the parts of the typical structure instance and the product model, and call the shape distribution algorithm to describe the shape information of the parts as a k-dimensional vector; Step S22: According to the constructed typical structure instance library, typical structure instances are obtained, and the neighbor characteristics between parts are calculated to obtain a set of neighbor vectors.

3. The method for automatically identifying typical structures of complex products based on case library matching according to claim 2 is characterized in that: The step S22 specifically includes the following steps: Step S221: Obtain a typical structure instance based on the constructed typical structure instance library and construct an OBB bounding box of the part; the typical structure instance includes the maximum length, maximum width, and maximum height of the part; Step S222: Determine the spatial distribution relationship between the parts based on the bounding box interference situation and obtain a set of neighbor vectors; Step S223: Repeat steps S221 and S222 until every part in the product model and typical structure instance library is traversed.

4. The method for automatically identifying typical structures of complex products based on case library matching according to claim 1 is characterized in that: The criteria conditions set in step S44 are: ； Among them, p is the first element of the initialization set list, Nei p Indicates parts p Neighbor vector set, Nei q1 Indicates parts q 1 The neighbor vector set of u and v are Nei p and Nei q1 A nearest neighbor vector in .

5. The method for automatically identifying typical structures of complex products based on case library matching according to claim 1 is characterized in that: The process of establishing the STG model in step S1 is as follows: first, a vertex adjacency graph is established based on the topological relationship between the geometric elements of the acquired B-rep data; secondly, the maximum clique in the adjacency graph is searched and the geometric area corresponding to the maximum clique is used as the local feature of the entity model; then, a statistical method is called to convert the shape information of the local feature into a feature vector to form a feature space, and an unsupervised learning algorithm is used to differentially suppress the local features; finally, an STG model is established with local features as vertices and the adjacency relationship between local features as edges.

6. A system for automatically identifying typical structures of complex products based on case library matching, used to execute the method for automatically identifying typical structures of complex products based on case library matching as claimed in claim 1; characterized in that: It includes instance library construction unit, feature acquisition unit, matching candidate unit, and recognition unit; The case library construction unit is used to construct a typical structure case library based on the acquired STG model; The feature acquisition unit is used to acquire part shape information and spatial neighbor characteristics between parts based on the constructed typical structure instance library; The matching candidate unit is used to obtain a matching candidate set of the target product three-dimensional model based on the part shape information and the spatial neighbor characteristics between parts; The recognition unit is used to obtain the spatial distribution aggregation of matching parts based on the matching candidate set and identify the typical structure.

7. An electronic device, characterized in that: It comprises a memory and a processor; a computer program is stored in the memory; when the computer program is executed on the processor, the method for automatically identifying typical structures of complex products based on case library matching as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device as described in claim 7, the method for automatically identifying typical structures of complex products based on case library matching as described in any one of claims 1 to 5 is implemented.

Images