CN114445845A

CN114445845A - Method and device for identifying components in drawing, electronic equipment and storage medium

Info

Publication number: CN114445845A
Application number: CN202111661641.8A
Authority: CN
Inventors: 彭帆
Original assignee: Wanyi Technology Co Ltd
Current assignee: Wanyi Technology Co Ltd
Priority date: 2021-12-31
Filing date: 2021-12-31
Publication date: 2022-05-06

Abstract

The application relates to a method and a device for identifying a component in a drawing, electronic equipment and a storage medium, which are applied to the technical field of data identification, wherein the method comprises the following steps: acquiring a drawing to be identified; acquiring the identification requirement of a user; determining a target recognition model from at least two initial recognition models trained in advance according to the recognition requirement, wherein the recognition requirement comprises the category of the component to be recognized, the at least two initial recognition models are obtained by training based on different training samples, and the target recognition model comprises at least one initial recognition model; and identifying the component to be identified based on the target identification model to obtain a target identification result. In order to solve prior art, adopt single identification model to discern the component in the drawing, but this kind of mode has great limitation, and identification model often can not satisfy the problem of user's demand to the recognition result of drawing.

Description

Method and device for identifying components in drawing, electronic equipment and storage medium

Technical Field

The present application relates to the field of data identification technologies, and in particular, to a method and an apparatus for identifying a component in a drawing, an electronic device, and a storage medium.

Background

With the rapid development of information technology, CAD electronic drawings are widely used in many fields such as construction, design, and manufacture. In general, geometric primitives such as points, lines and faces are used in electronic drawings with the addition of text labels to describe components in the drawings, and after the drawings are drawn, the components in the drawings are often required to be identified for further auditing or viewing.

In the related art, a single recognition model is adopted to recognize components in a drawing, but the method has great limitation, and the recognition result of the recognition model on the drawing often cannot meet the requirement of a user.

Disclosure of Invention

The application provides a method and a device for identifying a component in a drawing, electronic equipment and a storage medium, which are used for solving the problems that in the prior art, a single identification model is adopted to identify the component in the drawing, but the mode has great limitation, and the identification result of the identification model on the drawing cannot meet the requirement of a user frequently.

In a first aspect, an embodiment of the present application provides a method for identifying a component in a drawing, including:

acquiring a drawing to be identified;

acquiring identification requirements of a user, wherein the identification requirements comprise the category of a component to be identified;

determining a target recognition model from at least two pre-trained initial recognition models according to the recognition requirements, wherein the at least two initial recognition models are obtained by training based on different training samples, and the target recognition model comprises at least one initial recognition model;

and identifying the component to be identified based on the target identification model to obtain a target identification result.

Optionally, the identifying requirement further includes an identification index of the category of the component to be identified, and the determining, according to the identifying requirement, a target identifying model from at least two initial identifying models trained in advance includes:

acquiring a recognition performance set, wherein the recognition performance set is obtained based on recognition results of the at least two initial recognition models;

determining the target recognition performance satisfying the recognition index in the recognition performance set;

and determining the initial recognition model corresponding to the target recognition performance as the target recognition model.

Optionally, obtaining the recognition performance set based on the recognition results of the at least two initial recognition models includes:

acquiring the recognition result of each initial recognition model on the same sample drawing;

combining the recognition results to obtain at least two combined results;

acquiring a real result corresponding to each recognition result;

determining the identification performance corresponding to each combined result according to the real result and the combined result;

and determining the set of the identification performances corresponding to the at least two combined results as the identification performance set.

Optionally, the identification result includes component classification results of components of different classes, and the determining, according to the real result and the combined result, the identification performance corresponding to each combined result includes:

determining a combined classification result of each type of component classification result from the combined results;

and determining the identification performance according to the combined classification result and the real result.

Optionally, the combining the recognition results to obtain at least two combined results includes:

grouping the identification results to obtain at least one grouping result, wherein the grouping result comprises at least one identification result;

and combining the identification results in each grouping result to obtain at least two combination results.

Optionally, the identifying result includes a component position and a component name of a component at the component position, and when the combining result includes at least two identifying results, the combining the identifying results in each grouping result to obtain the at least two combining results includes:

determining a target component name of the same component position in each identification result, wherein the target component name is any one of the component names;

determining each of the component positions and a target component name for each of the component positions as the combined result.

Optionally, the determining the target component name of the same component position in each of the recognition results includes:

judging whether the component names of the same component positions in each grouping result are the same or not;

if not, taking any one of the component names as a target component name of the component position;

if the component names are the same, determining the target component name of the component position.

Optionally, the training process of the initial recognition model includes:

acquiring a training sample set, wherein the training sample set comprises at least two training samples and a real result of each training sample, the training samples are obtained by analyzing sample drawings in a sample drawing set according to a target analysis mode, the training samples comprise sample components, and the target analysis mode is one of a preset analysis mode set;

performing the following training process separately for each training sample in the set of training samples:

inputting the training sample into an initial neural network model, performing feature extraction on the training sample based on the initial neural network model to obtain sample features, and obtaining an output result of the training sample according to the sample features;

calculating a loss function value according to the output result and the real result, updating parameters in the initial neural network model according to the loss function value, acquiring a next training sample from the training sample set, repeatedly executing the training process until the loss function is smaller than a preset value, and taking the initial neural network model as the initial recognition model.

In a second aspect, an embodiment of the present application provides an apparatus for identifying a component in a drawing, including:

the first acquisition module is used for acquiring the drawing to be identified;

the second acquisition module is used for acquiring the identification requirements of the user, wherein the identification requirements comprise the category of the component to be identified;

the determining module is used for determining a target recognition model from at least two pre-trained initial recognition models according to the recognition requirements, the at least two initial recognition models are obtained by training based on different training samples, and the target recognition model comprises at least one initial recognition model;

and the identification module is used for identifying the component to be identified based on the target identification model to obtain a target identification result.

In a third aspect, an embodiment of the present application provides an electronic device, including: the system comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; the memory for storing a computer program;

the processor is configured to execute the program stored in the memory to implement the method for identifying a component in a drawing according to the first aspect.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the method for identifying a component in a drawing according to the first aspect.

Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages: according to the method provided by the embodiment of the application, the drawing to be identified is obtained; acquiring the identification requirement of a user; determining a target recognition model from at least two initial recognition models trained in advance according to the recognition requirement, wherein the recognition requirement comprises the category of the component to be recognized, the at least two initial recognition models are obtained by training based on different training samples, and the target recognition model comprises at least one initial recognition model; and identifying the component to be identified based on the target identification model to obtain a target identification result. Therefore, the at least two initial recognition models are set to serve as the recognition models of the drawing to be recognized, so that the recognition convenience of the drawing to be recognized can be improved; in addition, the target recognition model meeting the user recognition requirements is determined, and the drawing to be recognized is recognized, so that the recognition result of the drawing to be recognized can meet the expectations of users, and the requirements of the users can be met better.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.

Fig. 1 is an application scenario diagram of a method for identifying a component in a drawing according to an embodiment of the present application;

FIG. 2 is a flow chart of a method for identifying a component in a drawing provided in accordance with an embodiment of the present application;

FIG. 3 is a block diagram of an apparatus for identifying components in a drawing provided in accordance with an embodiment of the present application;

fig. 4 is a block diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

According to one embodiment of the application, a method for identifying a component in a drawing is provided. Alternatively, in the embodiment of the present application, the method for identifying a component in the drawing may be applied to a hardware environment formed by the terminal 101 and the server 102 as shown in fig. 1. As shown in fig. 1, a server 102 is connected to a terminal 101 through a network, and may be used to provide services (such as application services) for the terminal or a client installed on the terminal, and a database may be provided on the server or separately from the server, and may be used to provide data storage services for the server 102, where the network includes but is not limited to: the terminal 101 is not limited to a PC, a mobile phone, a tablet computer, and the like.

The method for identifying the components in the drawings of the embodiment of the application can be executed by the server 102, the terminal 101, or both the server 102 and the terminal 101. The terminal 101 executes the method for identifying the component in the drawing of the embodiment of the present application, or may be executed by a client installed thereon.

Taking a terminal to execute the method for identifying a component in a drawing of the embodiment of the present application as an example, fig. 2 is a schematic flow chart of an optional method for identifying a component in a drawing according to the embodiment of the present application, and as shown in fig. 2, the flow of the method may include the following steps:

step 201, obtaining a drawing to be identified.

In some embodiments, the drawing to be identified may be uploaded to the terminal after the designer finishes designing and when the construction drawing needs to be identified; alternatively, the identification may be acquired from a database when the relevant person identifies the construction drawing. The database stores a plurality of building drawings to be identified, and related personnel can select one of the building drawings to be identified according to the drawing name of the building drawing to be identified.

By way of example, the construction drawing to be identified may be, but is not limited to, a drawing for an airport, train station, bus stop, office building, residential building, hospital, museum, tourist attraction, church, school, park, or the like.

Step 202, obtaining the identification requirement of the user, wherein the identification requirement comprises the category of the component to be identified.

In some embodiments, a user may input an identification requirement required by the user through a display interface on the terminal, so that the terminal acquires the identification requirement.

Wherein the identification requirement may be, but is not limited to, a category of the component to be identified.

Further, the identification requirement may further include a category identification index of the component to be identified. Wherein, the identification index can be that the accuracy exceeds a first value and/or the identification rate exceeds a second value.

The first value and the second value may be set according to actual conditions, for example, the first value may be set to 90%, and the second value may be set to 95%.

Step 203, according to the identification requirement, determining a target identification model from at least two pre-trained initial identification models, wherein the at least two initial identification models are obtained by training based on different training samples, and the target identification model comprises at least one initial identification model.

In some embodiments, the pre-trained initial recognition model is trained by using different training samples, so that after the drawing to be recognized is obtained, the target recognition model can be selected from at least two initial recognition models according to the recognition requirements of the user.

Wherein the target recognition model is an initial recognition model capable of recognizing a category of the member to be recognized.

After each initial recognition model is trained, the initial recognition model is tested through the test data, and therefore the recognition performance of each initial recognition model is obtained. The identification performance can comprise identification accuracy and identification rate.

The test data comprises the real result of the component, after the test data is identified through the initial identification model, the identification result can be compared with the real result to obtain the final identification accuracy, and the identification rate is calculated according to the quantity of the identification result and the quantity of the test data.

In an optional embodiment, the identification requirement further includes an identification index of the category of the component to be identified, and according to the identification requirement, determining a target identification model from at least two initial identification models trained in advance includes:

acquiring a recognition performance set, wherein the recognition performance set is obtained based on recognition results of the at least two initial recognition models; determining the target recognition performance satisfying the recognition index in the recognition performance set; and determining the initial recognition model corresponding to the target recognition performance as the target recognition model.

In some embodiments, after training of each initial recognition model is completed, each initial recognition model has its own recognition performance, and after the recognition requirements of the user are obtained, the recognition requirements can be compared with the recognition performances in the recognition performance set to determine the target recognition performance meeting the recognition requirements, so that the corresponding initial recognition model is determined according to the target recognition performance and is used as the target recognition model.

It is to be understood that the above-mentioned recognition performance may be a recognition performance determined by a recognition result of a single initial recognition model, or may be a recognition performance determined by recognition results of two or more initial recognition models in common.

In an optional embodiment, deriving the set of recognition performances based on the recognition results of the at least two initial recognition models comprises:

acquiring the recognition result of each initial recognition model on the same sample drawing; combining the recognition results to obtain at least two combined results; acquiring a real result corresponding to each recognition result; determining the identification performance corresponding to each combined result according to the real result and the combined result; and determining the set of the identification performances corresponding to the at least two combined results as the identification performance set.

In some embodiments, when the recognition performance is determined by the recognition results of two or more initial recognition models, each initial recognition model has its own recognition result, so that when the recognition performance of a plurality of initial recognition models is determined, the recognition results of each initial recognition model may be combined, and the recognition performance may be obtained according to the combined result and the real result.

Specifically, when the identification performance is accuracy, the combined result may be compared with the real result, and if the combined result is consistent with the real result, the identification is considered accurate, and if the combined result is inconsistent with the real result, the identification is considered inaccurate, so that the ratio of the number of accurate identifications to the number of real results in each combined result is calculated to obtain the accuracy. And when the identification performance is the identification rate, calculating the ratio of the number of different identification results in the combined result to the number of the real results to obtain the identification rate.

It is to be understood that the above-mentioned identification performance may be the identification performance of all the identification results in the combined result, or may be the identification performance of the identification of the members of different classes.

Specifically, the identification result includes member classification results of members of different categories; determining the identification performance corresponding to each combined result according to the real result and the combined result, including:

determining a combined classification result of each type of component classification result from the combined results; and determining the identification performance according to the combined classification result and the real result.

In some embodiments, the component identification categories are classified, and the identification accuracy or the identification completeness of the combined classification result in each combined result is determined, so that when a user needs to identify a certain component with high accuracy, the corresponding target initial identification model can be determined according to the identification performance determined in the step.

The process of obtaining the initial recognition result of each initial recognition model on the same sample drawing may be:

classifying each initial identification result to obtain at least one component classification result; determining the set of at least one component classification result as the recognition result.

In an optional embodiment, the combining the recognition results to obtain at least two combined results includes:

grouping the identification results to obtain at least one grouping result, wherein the grouping result comprises at least one identification result; and combining the identification results in each grouping result to obtain at least two combination results.

Illustratively, with the number of initial recognition models being 3, the recognition results are X, Y, Z, respectively. Then X, Y, Z is combined to obtain 7 combination results X, Y, Z, X ≡ Y, X ≡ Z, Y ≡ and xuy ≡ Z. Each combination result corresponds to an identification strategy.

In an optional embodiment, the combining the identification results in each grouping result to obtain the at least two combination results includes:

determining a target component name of the same component position in each identification result, wherein the target component name is any one of the component names; determining each of the component positions and a target component name for each of the component positions as the combined result.

In an optional embodiment, the determining the target component name of the same component position in each of the recognition results comprises:

judging whether the component names of the same component positions in each grouping result are the same or not; if not, taking any one of the component names as a target component name of the component position; if the component names are the same, determining the target component name of the component position.

In some embodiments, the step of using any one of the component names as the target component name of the component position may be comparing the component name with the real name, and when the identification accuracy of a certain initial identification model is higher than that of other initial identification models, using the identified component name of the initial identification model as the target component name of the component position.

In an alternative embodiment, the training process of the initial recognition model includes:

acquiring a training sample set, wherein the training sample set comprises at least two training samples and a real result of each training sample, the training samples are obtained by analyzing sample drawings in a sample drawing set according to a target analysis mode, the training samples comprise sample components, and the target analysis mode is one of a preset analysis mode set; performing the following training process separately for each training sample in the set of training samples:

inputting the training sample into an initial neural network model, performing feature extraction on the training sample based on the initial neural network model to obtain sample features, and obtaining an output result of the training sample according to the sample features; calculating a loss function value according to the output result and the real result, updating parameters in the initial neural network model according to the loss function value, acquiring a next training sample from the training sample set, repeatedly executing the training process until the loss function is smaller than a preset value, and taking the initial neural network model as the initial recognition model.

It can be understood that different target analysis modes are adopted when different initial recognition models are trained. Specifically, the parsing manners in the preset parsing manner set may be, but are not limited to, the following:

the first method comprises the following steps: analyzing the block picture elements in the sample drawing to obtain the text information of the block picture elements, and taking the text information of each block picture element as a training sample. Wherein the text information includes a size and an identifier of the tile.

And the second method comprises the following steps: and cutting the sample drawing according to the area of the component in the sample drawing, and taking the obtained cutting result as a training sample.

And the third is that: and (4) segmenting the sample drawing by adopting a region segmentation model, and taking an obtained segmentation result as a training sample.

The initial neural network model may be of various types, such as a convolutional neural network model or a deep neural network model.

It is understood that the above-mentioned initial recognition model may be obtained based on the above-mentioned training process, or may be a recognition model capable of recognizing the component in the prior art.

And step 204, identifying the component to be identified based on the target identification model to obtain a target identification result.

In some embodiments, after the target recognition model is determined, the drawing to be recognized is input into the target recognition model, so that the recognition result of the drawing to be recognized is obtained.

It can be understood that, when there is one target recognition model, the recognition result of the target recognition model is taken as the recognition result of the drawing to be recognized. And when the number of the target recognition models is at least two, combining the recognition results of the plurality of target recognition models, and taking the obtained at least two combined results as the recognition results of the drawing to be recognized.

In a specific embodiment, the method for identifying the components in the drawings inputs various drawings into a plurality of identification models to respectively identify the drawings, adopts a uniform calibration mode for a target component when the identification models identify the target component, and finally performs combined statistical analysis based on the identification results of the identification models of the target component to obtain a fusion identification strategy of the target component based on the characteristics of the drawings. The method specifically comprises the following steps:

the method comprises the steps of firstly, preprocessing a drawing to be identified, and extracting relevant characteristic information of the drawing.

Drawing type, picture frame name, floor and other information for identifying drawing characteristics can be extracted.

And secondly, inputting the drawing to be recognized into a plurality of recognition models to recognize the target component based on one or more types of target components to be recognized, and obtaining a component recognition result set of each type of target component.

Such as: the target component to be identified is a door, and the door includes a side hung door, a double door, a single door, and the like, and the component identification result set may include the identified component type, the component name, the component position (two points at the upper left and the lower right), the confidence of the component, and the like.

And thirdly, combining the component recognition result sets recognized by different recognition models based on each type of target components to obtain a plurality of component recognition result sets corresponding to each type of target components on each drawing.

For example, the recognition model includes A, B, C and the list of recognized building blocks is X, Y, Z. Then the combination of X, Y, Z can result in a combined result of X, Y, Z, X ≦ Y, X ≦ Z, Y ≦ Z and xouy ≦ Z in 7.

Assuming that the a model identifies a single door member at the point of coordinate P1, and the B model identifies a double door member at the point of coordinate P2, the combined result is the sum of the two, and assuming that the B model identifies a double door member at the point of coordinate P1 (which may contain a certain fault tolerance), the combined result is either one.

And step four, comparing a plurality of component identification result sets corresponding to each type of target component with the standard identification result of the drawing to be identified to obtain the precision ratio and the recognition rate of each component identification result set of each type of component under each drawing.

For each drawing to be recognized, a professional can manually calibrate the component information in the drawing to obtain a standard recognition result of each drawing to be recognized, and further obtain the recognition rate or the precision ratio of each type of components.

And step five, carrying out clustering analysis based on the characteristics of the drawings according to the recognition rate and the precision ratio of various target components in the plurality of drawings under different combinations to obtain a recognition strategy of each target component based on different drawing characteristics.

For example, through clustering analysis, the recognition rate of door recognition in the ground building plan by adopting the recognition model A and the recognition model B is the highest. For door identification in the underground building plan, the precision ratio identified by the identification model C is the highest.

When clustering analysis is performed, different components can be clustered based on different drawing characteristics.

In the above step, without limiting the type of the target component to be identified, all components in the drawing may be directly identified, all the identified components are combined together when calculating the combination, and the total recognition rate and the precision rate are counted for each component in the fourth step.

Referring to the following table 1, in table 1, the parking spaces are identified by using three different schemes, and it can be seen that the emphasis points of the different identification schemes are different, and the identification rate of a single scheme can be improved by fusing the three schemes.

TABLE 1

Based on the same concept, the embodiment of the present application provides an apparatus for identifying a component in a drawing, and specific implementation of the apparatus may refer to the description of the method embodiment section, and repeated details are not repeated, as shown in fig. 3, the apparatus mainly includes:

the first obtaining module 301 is used for obtaining a drawing to be identified;

a second obtaining module 302, configured to obtain an identification requirement of a user, where the identification requirement includes a category of a component to be identified;

a determining module 303, configured to determine, according to the identification requirement, a target identification model from at least two initial identification models trained in advance, where the at least two initial identification models are obtained by training based on different training samples, and the target identification model includes at least one initial identification model;

an identifying module 304, configured to identify the component to be identified based on the target identification model, so as to obtain a target identification result.

Based on the same concept, an embodiment of the present application further provides an electronic device, as shown in fig. 4, the electronic device mainly includes: a processor 401, a memory 402 and a communication bus 403, wherein the processor 401 and the memory 402 communicate with each other via the communication bus 403. The memory 402 stores a program executable by the processor 401, and the processor 401 executes the program stored in the memory 402, so as to implement the following steps:

acquiring a drawing to be identified;

The communication bus 403 mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus 403 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 4, but this does not indicate only one bus or one type of bus.

The Memory 402 may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Alternatively, the memory may be at least one memory device located remotely from the aforementioned processor 401.

The Processor 401 may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc., and may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic devices, discrete gates or transistor logic devices, and discrete hardware components.

In still another embodiment of the present application, there is also provided a computer-readable storage medium having stored therein a computer program which, when run on a computer, causes the computer to execute the method of identifying a member in a drawing described in the above-described embodiment.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wirelessly (e.g., infrared, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The available media may be magnetic media (e.g., floppy disks, hard disks, tapes, etc.), optical media (e.g., DVDs), or semiconductor media (e.g., solid state drives), among others.

It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying a component in a drawing is characterized by comprising the following steps:

acquiring a drawing to be identified;

2. The method for identifying the components in the drawing according to claim 1, wherein the identification requirement further includes an identification index of the category of the component to be identified, and the determining the target identification model from at least two initial identification models trained in advance according to the identification requirement comprises:

determining target identification performance meeting the identification index in the identification performance set;

3. The method for identifying the component in the drawing according to claim 2, wherein the obtaining of the identification performance set based on the identification results of the at least two initial identification models comprises:

combining the recognition results to obtain at least two combined results;

acquiring a real result corresponding to each recognition result;

4. The method for identifying components in drawings according to claim 3, wherein the identification results include component classification results of components of different classes, and the determining the identification performance corresponding to each combination result according to the real result and the combination result includes:

5. The method for identifying the component in the drawing according to claim 3, wherein the combining the identification results to obtain at least two combination results comprises:

6. The method for identifying the component in the drawing according to claim 4, wherein the identification result includes a component position and a component name of the component at the component position, and when the combination result includes at least two identification results, the combining the identification results in each grouping result to obtain the at least two combination results includes:

7. The method for identifying the component in the drawing according to claim 5, wherein the determining the target component name of the same component position in each identification result comprises:

8. An apparatus for identifying a member in a drawing, comprising:

9. An electronic device, comprising: the system comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; the memory for storing a computer program;

the processor is used for executing the program stored in the memory to realize the identification method of the component in the drawing according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for identifying a member in a drawing according to any one of claims 1 to 7.