CN112819106A

CN112819106A - IFC component type identification method, device, storage medium and equipment

Info

Publication number: CN112819106A
Application number: CN202110408428.XA
Authority: CN
Inventors: 于雪; 郝海风; 曾江佑; 万旻; 杨佳东; 熊慧江
Original assignee: Jiangxi Booway New Technology Co ltd
Current assignee: Jiangxi Booway New Technology Co ltd
Priority date: 2021-04-16
Filing date: 2021-04-16
Publication date: 2021-05-18
Anticipated expiration: 2041-04-16
Also published as: CN112819106B

Abstract

The invention provides an IFC component type identification method, an IFC component type identification device, a storage medium and IFC component type identification equipment, wherein the method comprises the following steps: training a training sample set to obtain an SVM model, wherein the training sample is an IFC component of a known type; acquiring a target IFC component of a type to be identified, and extracting preset characteristic information from the target IFC component; and inputting the preset characteristic information of the target IFC component into the SVM model, and outputting to obtain the type of the target IFC component. The IFC component type automatic identification method based on the SVM model can completely automatically and accurately identify the type of the IFC component, so that the type of the IFC component can be automatically marked subsequently, excessive effort of a user on type marking is avoided, the workload of the user is greatly reduced, the efficiency of the user using three-dimensional cost software is effectively improved, and meanwhile, compared with a manual identification and marking mode, the error rate can be greatly reduced.

Description

IFC component type identification method, device, storage medium and equipment

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to an IFC component type identification method, apparatus, storage medium, and device.

Background

The GIM file is a power grid information model standard file, and the IFC file contained in the GIM file is three-dimensional model data of a power grid building part. The individual three-dimensional models within an IFC file are referred to as IFC components, which are divided into different types. In practice, due to the need for three-dimensional construction operations (e.g., calculating work volume, price, etc.), the specific types of these IFC components need to be labeled.

Some of the types of the IFC components can be determined by parsing and reading key information from the IFC file and then establishing a fixed mapping rule, but this method for determining the type through rule matching has many errors, and most of the IFC components cannot know the type according to the information in the IFC file, which requires manually marking the types of the IFC components one by one. However, in a GIM project, there are thousands of IFC components, and marking component types one by one undoubtedly brings huge workload to software users.

Disclosure of Invention

Based on this, the invention aims to provide an IFC component type identification method, an apparatus, a storage medium and a device, so as to solve the technical problem of large workload of the existing IFC component type labeling.

An IFC component type identification method according to an embodiment of the invention comprises the following steps:

training a training sample set to obtain an SVM model, wherein the training sample is an IFC component of a known type;

acquiring a target IFC component of a type to be identified, and extracting preset characteristic information from the target IFC component;

and inputting the preset characteristic information of the target IFC component into the SVM model, and outputting to obtain the type of the target IFC component.

In addition, the method for identifying the type of the IFC component according to the above embodiment of the present invention may further have the following additional technical features:

further, the step of training the training sample set to obtain the SVM model includes:

extracting preset characteristic information from the training samples;

and training to obtain the SVM model by taking the preset characteristic information of the training sample and the type of the training sample as input.

Further, the step of training the training sample set to obtain the SVM model further includes:

and extracting the type of the training sample from the file name of the training sample.

Further, the preset feature information includes one or more of a bounding box feature, a convex hull feature and a discrete contour feature.

Further, the types of IFC constituents include a base type and a non-base type, the SVM model includes a first SVM model for distinguishing the base type from the non-base type;

the training of the training sample set to obtain the SVM model comprises the following steps:

and training the training sample set by taking the basic type and the non-basic type as training indexes to obtain the first SVM model.

Further, the basic type is divided into sub-types, and the SVM model further comprises a second SVM model for identifying the basic type;

the training of the training sample set to obtain the SVM model further includes:

and training the training sample set by taking the sub-types divided by the basic types as training indexes to obtain the second SVM model.

Further, inputting the preset feature information of the target IFC component into the SVM model, and outputting the type of the target IFC component comprises:

inputting preset feature information of the target IFC component into the first SVM model;

judging whether the output of the first SVM model is of a non-basic type or not;

if yes, determining the type of the target IFC component as a non-basic type;

and if not, inputting the preset characteristic information of the target IFC component into the second SVM model so as to identify the specific basic type of the target IFC component.

An IFC component type identification apparatus according to an embodiment of the present invention, the apparatus comprising:

the model training module is used for training a training sample set to obtain an SVM model, wherein the training sample is an IFC component of a known type;

the data processing module is used for acquiring a target IFC component of a type to be identified and extracting preset characteristic information from the target IFC component;

and the type identification module is used for inputting the preset characteristic information of the target IFC component into the SVM model and outputting to obtain the type of the target IFC component.

The present invention also proposes a computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, implements the IFC component type identification method as described above.

The present invention also provides an IFC component type identification apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the IFC component type identification method as described above when executing the program.

Compared with the prior art, the IFC component type automatic identification method based on the SVM model can completely automatically and accurately identify the type of the IFC component, so that the subsequent automatic marking of the type of the IFC component can be realized, excessive energy of a user on type marking is avoided, the workload of the user is greatly reduced, the efficiency of the user using three-dimensional cost software is effectively improved, and meanwhile, compared with a manual identification and marking mode, the error rate can be greatly reduced.

Drawings

FIG. 1 is a flow chart of an IFC component type identification method in a first embodiment of the invention;

FIG. 2 is a flow chart of an IFC component type identification method in a second embodiment of the present invention;

FIG. 3 is a schematic diagram of an IFC component type identification arrangement in accordance with a third embodiment of the invention.

The following detailed description will further illustrate the invention in conjunction with the above-described figures.

Detailed Description

To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

It will be understood that when an element is referred to as being "secured to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. The terms "vertical," "horizontal," "left," "right," and the like as used herein are for illustrative purposes only.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

Example one

Referring to FIG. 1, a method for identifying a type of an IFC member according to a first embodiment of the present invention is shown, the method comprising steps S01 through S03.

And step S01, training a training sample set to obtain an SVM model, wherein the training sample is an IFC component of a known type.

Specifically, step S01 may specifically include:

extracting preset characteristic information from the training samples, and extracting the types of the training samples from the file names of the training samples;

An SVM (Support Vector Machine) is a typical supervised learning method in a Machine learning algorithm, and the method is realized on the premise that a large number of training samples are prepared, the samples are subjected to feature extraction to obtain feature vectors, feature Vector data and type indexes of the samples are used as input, and an SVM model is obtained through training.

In this embodiment, the training samples are selected as a known type of IFC component, and the type of training sample may be manually determined and marked. During specific implementation, before the model is trained, a large number of training samples can be collected and labeled to form a training sample set, and the greater the number of samples, the better the accuracy of the model is improved. Specifically, the smallest IFC entity, that is, a single IFC component, can be parsed from the IFC file in the GIM file, and the single IFC component is saved as an obj file, where each IFC component has a unique ID number in the IFC file. For example, 133 GIM files and 17000 other obj files are marked in the study, and all IFC components are manually checked one by one and marked with the types of the files. For the convenience of training, the type of the label is directly embodied in the file name. By way of example and not limitation, the format of the obj filename may be: obj, e.g., # 6074-independent basis obj, where "6074" is the ID number and "independent basis" is the component type.

Meanwhile, after the training sample set is collected, preset feature information can be extracted from each training sample to obtain a feature vector, and the SVM model is obtained through training by taking feature vector data and the type of the training sample (namely the type index of the sample) as input.

Step S02 acquires a target IFC component of a type to be identified and extracts preset feature information from among the target IFC component.

In a specific implementation, the preset feature information includes one or more of a bounding box feature, a convex hull feature and a discrete contour feature. In particular, bounding boxes are algorithms for solving an optimal bounding space of a discrete set of points, the basic idea being to approximately replace complex geometric objects with slightly larger and characteristically simple geometries (called bounding boxes). By way of example and not limitation, the bounding box selected in this embodiment is specifically an AABB bounding box, which is the earliest bounding box applied. It is defined as the smallest hexahedron (i.e., a cuboid) that contains the object with sides parallel to the coordinate axes. Therefore, only six scalars (Xmin, Xmax, Ymin, Ymax, Zmin, Zmax) are required to describe an AABB. In the present embodiment, the bounding box of the IFC component is characterized by 4 scalars, namely [ Xmax-Xmin, Ymax-Ymin, Zmin, Zmax ], respectively, where the length, width and Z-axis of the bounding box lie.

In addition, Convex Hull (Convex Hull) is a concept in computing geometry (graphics). In a real vector space V, for a given set X, the intersection S of all convex sets containing X is called the convex hull of X. The convex hull of X may be constructed with a convex combination of all points (X1.. Xn) within X. Obviously, the convex hull can embody the outer contour characteristics of the three-dimensional model more than the bounding box. Specifically, in the three-dimensional figure, the convex hull is composed of a plurality of triangular plates with different sizes, so that the whole three-dimensional figure is surrounded. In the present embodiment, the convex hull characteristic of the IFC member takes the area of the largest 10 triangular pieces of the convex hull as the characteristic value, and the characteristics of less than 10 triangular pieces are supplemented with 0.

Meanwhile, a discrete contour feature is embodied as Alpha Shapes, which is a method for abstracting an intuitive shape from a discrete set of spatial points (point sets), and in short, a rough contour is obtained from a pile of unordered points. Alpha Shapes can embody concave-convex profile details compared to convex hulls. Alpha Shapes are composed of several tetrahedrons, and the discrete contour feature of the IFC component takes the volume of 100 tetrahedrons with the largest volume as the feature value, and the features of less than 100 tetrahedrons are supplemented with 0.

It should be noted that the preset feature information selected in the model training process and the preset feature information selected in this step should be kept corresponding to each other. For example, when the bounding box feature, the convex hull feature and the discrete contour feature of the IFC component (training sample) are selected as input for model training during model training, the bounding box feature, the convex hull feature and the discrete contour feature of the target IFC component should also be selected and input into the SVM model when the subsequently trained SVM model is used for identifying the type of the target IFC component, so as to ensure that the SVM model can accurately output the type of the target IFC component.

In the present embodiment, the preset feature information includes a bounding box feature, a convex hull feature and a discrete contour feature at the same time, that is, the bounding box feature, the convex hull feature and the discrete contour feature of the IFC component are extracted at the same time. Specifically, the present embodiment extracts the features of the IFC component as follows:

4 characteristic values [ Xmax-Xmin, Ymax-Ymin, Zmin, Zmax ] of the AABB bounding box;

the area of the 10 largest triangular faces of the convex hull; and

volume of 100 largest tetrahedrons of Alpha Shapes. A total of 114 data, namely each IFC component is subjected to feature extraction, and a feature vector with 1 x 114 dimensions is formed.

Step S03, inputting the preset feature information of the target IFC component into the SVM model, and outputting to obtain the type of the target IFC component.

In summary, in the IFC component type identification method in the above embodiments of the present invention, by providing an IFC component type automatic identification method based on an SVM model, the type of the IFC component can be completely automatically and accurately identified, so that the subsequent automatic identification of the type of the IFC component can be achieved, excessive effort of a user on type marking is avoided, the workload of the user is greatly reduced, the efficiency of the user in using three-dimensional cost software is effectively improved, and meanwhile, compared with a manual identification and marking method, the error rate can be greatly reduced.

Example two

Referring to fig. 2, a method for identifying IFC component types according to a second embodiment of the present invention is shown, wherein the IFC component types include a basic type and a non-basic type, and the basic type is further divided into several sub-types, such as a stepwise single basis, a third-order single basis, an independent basis, a structural support basis, a capacitor basis, etc., so that the SVM model trained in this embodiment includes a first SVM model for distinguishing the basic type from the non-basic type, and a second SVM model for identifying a specific basic type, and the method specifically includes steps S11 to S17.

And step S11, training a training sample set by taking the basic type and the non-basic type as training indexes to obtain the first SVM model.

In this step, the training sample set includes the basic type IFC component and the non-basic type IFC component, which are collectively referred to as training samples, that is, in this embodiment, before training the model, a large number of basic type IFC components and non-basic type IFC components are collected to form the training sample set, and then the training sample set is trained once, and during training, the basic type and the non-basic type are used as indexes for training.

Based on the training principle of the SVM model, the first SVM model is trained by using the basic type and the non-basic type as the type indexes, so that the IFC component of the basic type and the non-basic type can be accurately distinguished by the trained first SVM model.

And step S12, training the training sample set by taking the subtype divided by the basic type as a training index to obtain the second SVM model.

After the first SVM model is obtained through the training for the first time, the type index is changed, the specific sub-types (i.e., the stepwise single basis, the third-order single basis, the independent basis, the structural support basis, the capacitor basis, etc.) divided by the basic types are used as the training index, the training sample set is trained for the second time, and the second SVM model capable of accurately distinguishing the specific basic types is obtained through the training.

In addition, it should be further explained that the specific training means of the SVM model described in the first embodiment can be referred to in the embodiment for training both the first SVM model and the second SVM model.

Step S13, a target IFC component of the type to be identified is obtained and preset feature information is extracted from the target IFC component.

Step S14, inputting the preset feature information of the target IFC component into the first SVM model.

Step S15, judging whether the output of the first SVM model is of a non-basic type.

When the output of the first SVM model is judged to be of a non-basic type, executing step S16; when the output of the first SVM model is determined to be of a base type, step S17 is performed in order to further determine which base type the target IFC component specifically belongs to.

In step S16, the type of the target IFC component is determined to be a non-base type.

Step S17, inputting the preset feature information of the target IFC component into the second SVM model to identify a specific basic type to which the target IFC component belongs.

For example, the bounding box features, the convex hull features, and the discrete contour features of the target IFC construct are input into a second SVM model, which is capable of outputting that the target IFC construct is of a "free-standing" type.

It should be noted that the principle of the present embodiment is basically the same as that of the first embodiment, and the contents of the first embodiment can be referred to where the present embodiment is not mentioned.

EXAMPLE III

Referring to fig. 3, there is shown an IFC component type identification apparatus in a third embodiment of the present invention, the apparatus including:

the model training module 11 is configured to train a training sample set to obtain an SVM model, where the training sample is an IFC component of a known type;

a data processing module 12, configured to acquire a target IFC component of a type to be identified, and extract preset feature information from the target IFC component;

and the type identification module 13 is configured to input preset feature information of the target IFC component into the SVM model, and output the preset feature information to obtain the type of the target IFC component.

In some cases of the present embodiment, the model training module 11 includes:

the information extraction unit is used for extracting preset characteristic information from the training samples;

and the model training unit is used for training the preset characteristic information of the training samples and the types of the training samples as input to obtain the SVM model.

In some cases of the present embodiment, the information extracting unit is further configured to extract the type of the training sample from a file name of the training sample.

In some cases of the present embodiment, the preset feature information includes one or more of a bounding box feature, a convex hull feature and a discrete contour feature.

In some instances of the present embodiments, the types of IFC constituents include a base type and a non-base type, the SVM model includes a first SVM model for distinguishing the base type from the non-base type; the model training module 11 includes:

and the first training unit is used for training the training sample set by taking the basic type and the non-basic type as training indexes to obtain the first SVM model.

In some cases of the present embodiment, the basic type is divided into sub-types, and the SVM model further includes a second SVM model for identifying the basic type; the model training module 11 further comprises:

and the second training unit is used for training the training sample set by taking the sub-types divided by the basic types as training indexes to obtain the second SVM model.

In some cases of the present embodiment, the type identification module 13 includes:

a first recognition unit configured to input preset feature information of the target IFC component into the first SVM model;

the judging unit is used for judging whether the output of the first SVM model is of a non-basic type or not;

a determining unit, configured to determine that the type of the target IFC component is a non-base type when it is determined that the output of the first SVM model is a non-base type;

and the second identification unit is used for inputting the preset characteristic information of the target IFC component into the second SVM model to identify the specific basic type of the target IFC component when the output of the first SVM model is judged not to be the non-basic type.

In summary, the IFC component type identification apparatus in the above embodiment of the present invention can completely automatically and accurately identify the type of the IFC component by providing the IFC component type automatic identification method based on the SVM model, so that the subsequent automatic identification of the type of the IFC component can be achieved, the user is prevented from investing too much effort in type marking, the workload of the user is greatly reduced, the efficiency of the user in using the three-dimensional cost software is effectively improved, and meanwhile, compared with the manual identification and marking method, the error rate can be greatly reduced.

Embodiments of the present invention also provide a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the IFC component type identification method as described above.

Embodiments of the present invention also provide an IFC component type identification apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the IFC component type identification method as described above.

By way of example and not limitation, the IFC component type identification device may be a scanning device (e.g., a printer), or a computer device (e.g., a computer, a notebook, etc.) capable of processing a scanned image. In particular, the processor may be, in some embodiments, a Central Processing Unit (CPU), controller, microcontroller, microprocessor or other IFC component type identification chip for executing program code stored in memory or Processing data.

Wherein the memory includes at least one type of readable storage medium including flash memory, hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory may, in some embodiments, be an internal storage unit of the IFC component type identification device, such as a hard disk of the IFC component type identification device. The memory may also be an external storage device of the IFC component type identification device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the IFC component type identification device. Further, the memory may also include both an internal storage unit of the IFC component type identification device and an external storage device. The memory may be used not only to store application software installed in the IFC component type identification device and various types of data, but also to temporarily store data that has been output or is to be output.

Optionally, the IFC component type identification device may further include a user interface, a network interface, a communication bus, etc., the user interface may include a Display (Display), an input unit such as a remote controller, physical keys, etc., and the optional user interface may further include a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the IFC component type identification device and for displaying a visual user interface. The network interface may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), typically used to establish a communication link between the IFC component type identification device and other electronic devices. The communication bus is used to enable connection communication between these components.

In summary, the IFC component type identification device in the above embodiment of the present invention can completely automatically and accurately identify the type of the IFC component, so that the subsequent automatic marking of the type of the IFC component can be achieved, excessive effort of the user on type marking is avoided, the workload of the user is greatly reduced, the efficiency of the user using the three-dimensional cost software is effectively improved, and meanwhile, compared with a manual identification and marking method, the error rate can be greatly reduced.

Those of skill in the art will understand that the logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be viewed as implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

The above examples only show some embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of IFC component type identification, the method comprising:

inputting preset feature information of the target IFC component into the SVM model, and outputting to obtain the type of the target IFC component;

the types of IFC components include a base type and a non-base type, and the SVM model comprises a first SVM model for distinguishing the base type from the non-base type;

training the training sample set by taking the basic type and the non-basic type as training indexes to obtain the first SVM model;

the basic types are divided into sub-types, and the SVM model further comprises a second SVM model for identifying the basic types;

2. The IFC component type identification method of claim 1, wherein the step of training a set of training samples to obtain an SVM model comprises:

extracting preset characteristic information from the training samples;

3. The IFC component type identification method of claim 2, wherein the step of training the training sample set to obtain the SVM model further comprises:

4. The IFC component type identification method of any one of claims 1-3, wherein the pre-set characterization information includes one or more of bounding box characterization, convex hull characterization, and discrete contour characterization.

5. The method of identifying a type of IFC component of claim 1, wherein inputting pre-determined characterization information for said target IFC component into said SVM model, and outputting said type of target IFC component comprises:

if yes, determining the type of the target IFC component as a non-basic type;

6. An IFC component type identification apparatus, the apparatus comprising:

7. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 5.

8. An IFC component type identification device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the method of any one of claims 1-5.