A method, device, electronic device and storage medium for segmenting damages of damaged structures

Abstract

The present invention relates to the technical field of seismic damage identification of concrete structures, and particularly to a method, device, electronic device and storage medium for damage segmentation of damaged structures. Among them, the method includes: obtaining the original images of damaged RC structures to generate a training data set; performing annotation processing on the training data set to obtain an annotated data set; performing data augmentation on the annotated data set to obtain an augmented data set; constructing a damage segmentation model for damaged structures, and using the augmented data set to train the damage segmentation model for damaged structures to obtain a trained damage segmentation model for damaged structures; using the trained damage segmentation model for damaged structures to perform damage segmentation on the image to be segmented to obtain a segmentation result. By using a weighted multi-class cross-entropy loss function and an adaptive learning rate automatic scheduler for the learning rate scheduler, the efficiency of post-earthquake assessment work is greatly improved. Moreover, the established model architecture is concise, with a small number of parameters, suitable for local deployment, and has high practical engineering significance.

Description

Technical Field

[0001] The present application relates to the technical field of seismic damage identification of concrete structures, and particularly relates to a method, device, electronic device and storage medium for damage segmentation of damaged structures. Background Art

[0002] Performing accurate visual inspection on damaged RC structures is a rather cumbersome task and places high requirements on the professional qualities of assessors. In addition, different assessors may give different assessment results for the same damaged component. These problems limit the efficiency and accuracy of existing post-earthquake assessment work.

[0003] In recent years, the rapid development of computer vision technology has provided many new paradigms for many problems in the engineering field and has great development potential in the field of post-earthquake assessment. However, the current application of computer vision in the field of civil engineering is still limited to the early health detection of structures mainly based on crack identification. Different from health monitoring research, the post-earthquake assessment of RC structures requires people to accurately identify more extensive visible visual damage information such as concrete spalling and bare reinforcement. Although there are currently some damage location models for damaged RC structures based on object detection methods, these models can only achieve damage location and cannot achieve damage segmentation, and cannot meet the need for quantifying the degree of damage in post-earthquake assessment. However, due to the lack of public datasets, there are still very few models available for multi-class damage segmentation of damaged RC structures. Summary of the Invention

[0004] The present application aims to solve at least one of the technical problems in the related art to some extent.

[0005] To this end, the first object of the present application is to propose a method for damage segmentation of damaged structures to solve problems such as the inability to achieve damage segmentation by existing technical means and the inability to meet the need for quantifying the degree of damage in post-earthquake assessment.

[0006] The second object of the present application is to propose a device.

[0007] The third object of the present application is to propose an electronic device.

[0008] The fourth object of the present application is to propose a computer-readable storage medium.

[0009] To achieve the above object, the first aspect embodiment of the present application proposes a method for damage segmentation of damaged structures, including:

[0010] Obtain the original image of the damaged RC structure and generate a training dataset;

[0011] Perform annotation processing on the training dataset to obtain an annotated dataset;

[0012] Perform data augmentation on the labeled dataset to obtain an augmented dataset;

[0013] Construct a damage segmentation model for damaged structures, and use the augmented dataset to train the damage segmentation model for damaged structures to obtain a trained damage segmentation model for damaged structures;

[0014] Use the trained damage segmentation model for damaged structures to perform damage segmentation on the image to be segmented to obtain a segmentation result.

[0015] Preferably, the obtaining of the original pictures of damaged RC structures and generating the training dataset includes:

[0016] Obtain the test pictures taken in the tests of RC beams, columns and joints, the paper illustrations of the experimental research on RC components, the actual damage photos of damaged RC structures in earthquake-stricken areas, and the RC component images taken in different environments and backgrounds, and generate a training dataset.

[0017] Preferably, the performing of the labeling process on the training dataset to obtain the labeled dataset includes:

[0018] Label the pixels in the training dataset, including background, concrete spalling, concrete cracking, concrete crushing, longitudinal bar exposure, longitudinal bar bare exposure, longitudinal bar buckling, stirrup bare exposure, concrete fragments. Each pixel point in the feature map exported after labeling has a unique category, and generate a labeled dataset.

[0019] Preferably, the performing of the data augmentation on the labeled dataset to obtain the augmented dataset includes: performing data augmentation processing on the labeled dataset by using mirror flipping, color distortion and random Gaussian noise to obtain an augmented dataset.

[0020] Preferably, the constructing of the damage segmentation model for damaged structures includes:

[0021] Construct a damage segmentation model for damaged structures by using a network architecture with pyramid multi-scale feature fusion, Haar wavelet downsampling and an attention module's U-shaped network with skip connections.

[0022] Preferably, the training of the damage segmentation model for damaged structures by using the augmented dataset to obtain a trained damage segmentation model for damaged structures includes:

[0023] Adopt an adaptive learning rate automatic scheduler to dynamically adjust the learning rate according to the change of the loss during the training process. Use weighted multi-class cross-entropy as the loss function. When the change of the loss is small, that is, when the model tends to be stable, reduce the learning rate for more refined adjustment; when the change of the loss is large, increase the learning rate to accelerate convergence.

[0024] Preferably, the calculation formula of the multi-class cross-entropy loss function is as follows:

[0025]

[0026] where E is the weighted cross-entropy loss, α is the weight, k is the total number of classes, α s is the scaled weight, t k is the label, and y k is the predicted value.

[0027] To achieve the above object, an embodiment of the second aspect of the present application provides a damaged structure damage segmentation device, including:

[0028] A data acquisition module, which acquires the original pictures of the damaged RC structure and generates a training data set;

[0029] A data annotation module, which performs annotation processing on the training data set to obtain an annotated data set;

[0030] A data augmentation module, which augments the annotated data set to obtain an augmented data set;

[0031] A model training module, which constructs a damaged structure damage segmentation model and uses the augmented data set to train the damaged structure damage segmentation model to obtain a trained damaged structure damage segmentation model;

[0032] A model prediction module, which uses the trained damaged structure damage segmentation model to perform damage segmentation on the image to be segmented to obtain a segmentation result.

[0033] To achieve the above object, an embodiment of the third aspect of the present application provides an electronic device, including: a processor and a memory communicatively connected to the processor;

[0034] The memory stores computer execution instructions;

[0035] The processor executes the computer execution instructions stored in the memory to implement the method described in any one of the above.

[0036] To achieve the above object, an embodiment of the fourth aspect of the present application provides a computer-readable storage medium, including computer execution instructions stored in the computer-readable storage medium, and the computer execution instructions are used to implement the method described in any one of the above when executed by a processor.

[0037] A method for segmenting damage to damaged structures provided by this application overcomes the defect of insufficient publicly available datasets in the task of segmenting damage to damaged RC structures based on computer vision, greatly improving the efficiency of post-earthquake assessment work. The deep learning model structure parameters proposed in this application are small, and the established model architecture is concise with few parameters, suitable for local deployment, enabling rapid positioning and accurate segmentation of visible seismic damage to damaged RC structures, improving the efficiency of post-earthquake assessment work, and having high practical engineering significance.

[0038] Additional aspects and advantages of this application will be given in part in the following description, become apparent in part from the following description, or be understood through the practice of this application. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and / or additional aspects and advantages of this application will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:

[0040] Figure 1 is a flowchart of the first specific embodiment of a method for segmenting damage to damaged structures provided by the present invention;

[0041] Figure 2 are some pictures in the dataset;

[0042] Figure 3 is the deep learning network model architecture;

[0043] Figure 4 is a structural block diagram of a device for segmenting damage to damaged structures provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The core of the present invention is to provide a method, device, electronic device, and storage medium for segmenting damage to damaged structures, which can achieve rapid positioning and accurate segmentation of visible seismic damage to damaged RC structures through a deep learning model, improving the efficiency of post-earthquake assessment work.

[0045] To enable those skilled in the art to better understand the solution of the present invention, the present invention will be further described in detail below in conjunction with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Please refer to Figure 1 , Figure 1 is a flowchart of the first specific embodiment of a method for segmenting damage to damaged structures provided by the present invention; the specific operation steps are as follows:

[0047] Step S101: Obtain the original images of damaged RC structures and generate a training dataset;

[0048] Obtain the test images taken during the tests of RC beams, columns and joints, the paper illustrations of the experimental research on RC components, the actual damage photos of damaged RC structures in earthquake-stricken areas, and the images of RC components taken in different environments and backgrounds, and generate a training dataset.

[0049] Step S102: Perform annotation processing on the training dataset to obtain an annotated dataset;

[0050] Annotate the pixels in the training dataset, including background, concrete spalling, concrete cracking, concrete crushing, longitudinal reinforcement exposure, longitudinal reinforcement bare exposure, longitudinal reinforcement buckling, stirrup exposure, concrete fragments. Each pixel point in the feature map exported after annotation has a unique category, and an annotated dataset is generated.

[0051] Step S103: Augment the annotated dataset to obtain an augmented dataset;

[0052] Perform data augmentation on the annotated dataset using mirror flipping, color distortion, and random Gaussian noise to obtain an augmented dataset.

[0053] Step S104: Construct a damaged structure damage segmentation model, and use the augmented dataset to train the damaged structure damage segmentation model to obtain a trained damaged structure damage segmentation model;

[0054] Use a network architecture with pyramid multi-scale feature fusion, Haar wavelet downsampling, and an attention module's U-shaped network with skip connections to construct a damaged structure damage segmentation model.

[0055] Adopt an adaptive learning rate automatic scheduler to dynamically adjust the learning rate according to the change of loss during the training process. Use weighted multi-class cross-entropy as the loss function. When the loss change is small, that is, when the model tends to be stable, reduce the learning rate for finer adjustment; when the loss change is large, increase the learning rate to accelerate convergence.

[0056] The calculation formula of the multi-class cross-entropy loss function is:

[0057]

[0058] Among them, E is the weighted cross-entropy loss, α is the weight, k is the total number of categories, α s is the scaled weight, t k is the label, u k is the predicted value.

[0059] Step S105: Use the trained seismic damage structure damage segmentation model to segment the damage of the image to be segmented to obtain a segmentation result.

[0060] Construct a seismic damage structure damage segmentation model using a network architecture with pyramid multi-scale feature fusion, Haar wavelet downsampling, and an attention module in a U-shaped structure with skip connections.

[0061] This embodiment provides a method for segmenting seismic damage structures, which overcomes the defect of insufficient publicly available datasets in the task of segmenting seismic damage of RC structures based on computer vision, greatly improving the efficiency of post-earthquake assessment work. The deep learning model proposed in this application has small structural parameters, and the established model architecture is concise with few parameters, suitable for local deployment, enabling rapid localization and accurate segmentation of visible seismic damage to RC structures, improving the efficiency of post-earthquake assessment work, and having high practical engineering significance.

[0062] Based on the above embodiment, this embodiment describes the method for segmenting seismic damage structures in detail as follows:

[0063] The purpose of this application is to propose a deep learning model and a training dataset construction method for segmenting seismic damage of RC structures to improve the efficiency of post-earthquake assessment work and increase the quantity and quality of input samples for the seismic damage recognition model of RC structures based on computer vision. The constructed dataset contains 1,400 pictures with different structural member types and different damage degrees, and different structural damage categories in the pictures have been manually annotated at the pixel level. The proposed deep learning model for damage segmentation includes multi-scale feature fusion, Haar wavelet downsampling, and an attention module to address the challenges of small sample size, low image resolution, and complex background environment in the seismic damage segmentation task.

[0064] Obtain and screen the original pictures of seismic damaged RC structures that can be used to construct the model training dataset from existing RC structure experimental studies and publicly available picture datasets; the sources of the pictures mainly include the following: experimental pictures taken in recent years in RC beam, column, joint, etc. experiments carried out by the applicant's research team; paper illustrations in existing RC component experimental studies; actual damage pictures of seismic damaged RC structures in actual earthquake disaster areas shown in existing earthquake damage reports; search engines (e.g., Google Image). The pictures in the dataset mainly include various different RC structural components (beams, columns, joints, walls, etc.). The pictures in the dataset were taken in many different environments (laboratories, environments with dim light, earthquake disaster areas, etc.), and contain various complex background information and random occlusions (plant occlusions, concrete fragments, etc.).

[0065] Manually perform pixel-level annotation on different damage categories in the collected images to obtain the corresponding damage feature maps of the images; as Figure 2 shown, a total of 8 types of pixels are marked in the image, namely background, concrete spalling, concrete cracking, concrete crushing, exposed longitudinal reinforcement, buckling of longitudinal reinforcement, exposed stirrups, and concrete fragments. The manual annotation process is based on the labelme annotation tool, and each pixel point in the exported feature map after annotation has a unique category.

[0066] Perform local data augmentation on the original images and the annotated damage feature maps; three data augmentation methods are applied to each image, namely mirror flipping, color distortion, and random Gaussian noise;

[0067] Input the image dataset after data augmentation into a deep learning network for training, as Figure 3 shown, the model network structure used is a U-shaped network architecture with skip connections that includes pyramid multi-scale feature fusion, Haar wavelet downsampling, and an attention module;

[0068] The loss function used in training is a weighted multi-class cross-entropy loss function, and the learning rate scheduler adopts the ALRS learning rate automatic scheduler; the loss function used in training is a weighted multi-class cross-entropy loss function, and its calculation formula is:

[0069]

[0070] where E is the weighted cross-entropy loss, α is the weight, k is the total number of classes, α s is the scaled weight, t k is the label, and y k is the predicted value.

[0071] A method for segmenting seismic damage to damaged structures provided by an embodiment of the present invention overcomes the defect of insufficient public datasets in the task of segmenting seismic damage to damaged RC structures based on computer vision. In addition, it also overcomes the deficiency that existing semantic segmentation models are not applicable to the task of segmenting seismic damage to damaged RC structures. The constructed dataset includes various different RC structural components (beams, columns, joints, walls, etc.). The images in the dataset are taken in many different environments (laboratories, environments with dim light, earthquake disaster areas, etc.), and contain various complex background information and random occlusions (plant occlusions, concrete fragments, etc.), and are applicable to various different computer vision tasks. The deep learning network structure proposed in the present invention has a small number of parameters, which helps the model to be quickly deployed on local detection devices and is convenient for popularization and use in actual engineering. Generally speaking, the present invention can achieve rapid positioning and accurate segmentation of visible seismic damage to damaged RC structures, improving the efficiency of post-earthquake assessment work.

[0072] Please refer to Figure 4 , Figure 4 which is a structural block diagram of a damaged structure damage segmentation device provided by an embodiment of the present invention; the specific device may include:

[0073] A data acquisition module 100, which acquires the original pictures of the damaged RC structure and generates a training data set;

[0074] A data annotation module 200, which performs annotation processing on the training data set to obtain an annotated data set;

[0075] A data enhancement module 300, which enhances the annotated data set to obtain an enhanced data set;

[0076] A model training module 400, which constructs a damaged structure damage segmentation model and uses the enhanced data set to train the damaged structure damage segmentation model to obtain a trained damaged structure damage segmentation model;

[0077] A model prediction module 500, which uses the trained damaged structure damage segmentation model to perform damage segmentation on the image to be segmented to obtain a segmentation result.

[0078] A damaged structure damage segmentation device in this embodiment is used to implement the foregoing damaged structure damage segmentation method. Therefore, the specific implementation manners in a damaged structure damage segmentation device can be seen in the embodiment part of the foregoing damaged structure damage segmentation method. For example, the data acquisition module 100, the data annotation module 200, the data enhancement module 300, the model training module 400, and the model prediction module 500 are respectively used to implement steps S101, S102, S103, S104, and S105 in the foregoing damaged structure damage segmentation method. Therefore, its specific implementation manners can refer to the descriptions of the corresponding various part embodiments and will not be elaborated herein.

[0079] To implement the above embodiment, the present application also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided in the foregoing embodiment.

[0080] To implement the above embodiment, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method provided in the foregoing embodiment.

[0081] To implement the above embodiment, the present application also proposes a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the method provided in the foregoing embodiment.

[0082] The collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information involved in this application all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0083] It should be noted that personal information from users should be collected for legal and reasonable purposes and should not be shared or sold outside of these legal uses. In addition, such collection / sharing should be carried out after obtaining the informed consent of the user, including but not limited to notifying the user to read the user agreement / user notice and signing an agreement / authorization that authorizes relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and to ensure that others with access to the personal information data comply with their privacy policies and procedures.

[0084] This application is expected to provide an implementation plan for users to selectively block the use or access of personal information data. That is, this disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by restricting data collection and deleting the data. In addition, when applicable, personal identifiers are removed from such personal information to protect the privacy of the user.

[0085] In the description of the foregoing embodiments, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0086] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0087] Any process or method description represented in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logic function or process. The scope of the preferred embodiments of the present application includes additional implementations, where functions may be performed in an order not shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved, which should be understood by those skilled in the technical field to which the embodiments of the present application pertain.

[0088] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered a sequenced list of executable instructions for implementing a logical function, and can be embodied specifically 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, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device 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 include the following: an electrical connection having one or more wires (electronic device), 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 can even be paper or other suitable medium on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0089] It should be understood that various parts of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0090] Those of ordinary skill in the art can understand that all or part of the steps carried out in the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0091] In addition, in each of the embodiments of the present application, the functional units can be integrated in a processing module, or each unit can exist physically alone, or two or more units can be integrated in a module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0092] The storage medium mentioned above can be a read-only memory, a magnetic disk, an optical disc, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for segmenting damage of a damaged structure, characterized in that, Including: Obtain the original images of damaged RC structures and generate a training dataset; Perform annotation processing on the training dataset to obtain an annotated dataset; Perform data augmentation on the annotated dataset to obtain an augmented dataset; Construct a damaged structure damage segmentation model, and use the augmented dataset to train the damaged structure damage segmentation model to obtain a trained damaged structure damage segmentation model. Among them, the construction of the damaged structure damage segmentation model includes using a network architecture with pyramid multi-scale feature fusion, Haar wavelet downsampling, and an attention module's U-shaped network with skip connections to construct the damaged structure damage segmentation model; Use the trained damaged structure damage segmentation model to perform damage segmentation on the image to be segmented to obtain a segmentation result.

2. The method for segmenting damage of a damaged structure according to claim 1, wherein, The obtaining of the original images of damaged RC structures and generating a training dataset includes: Obtain the test images taken in RC beam, column, and joint tests, the paper illustrations of RC component test studies, the actual damage photos of damaged RC structures in earthquake-stricken areas, and RC component images taken in different environments and backgrounds, and generate a training dataset.

3. The method for segmenting damage of a damaged structure according to claim 1, wherein The performing of annotation processing on the training dataset to obtain an annotated dataset includes: Annotate the pixels in the training dataset, including background, concrete spalling, concrete cracking, concrete crushing, longitudinal bar exposure, longitudinal bar bare exposure, longitudinal bar buckling, stirrup bare exposure, and concrete fragments. Each pixel point in the feature map exported after annotation has a unique category, and an annotated dataset is generated.

4. The damage segmentation method for damaged structures according to claim 1, characterized in that The performing of data augmentation on the annotated dataset to obtain an augmented dataset includes: Perform data augmentation processing on the annotated dataset using mirror flipping, color distortion, and random Gaussian noise to obtain an augmented dataset.

5. The method for segmenting damage of a damaged structure according to claim 1, characterized in that, The training of the damaged structure damage segmentation model using the augmented dataset to obtain a trained damaged structure damage segmentation model includes: Adopt an adaptive learning rate automatic scheduler to dynamically adjust the learning rate according to the loss change situation during the training process. Use weighted multi-class cross-entropy as the loss function. When the loss change is small, that is, when the model tends to be stable, reduce the learning rate for finer adjustment; when the loss change is large, increase the learning rate to accelerate convergence.

6. The method for segmenting damage of a damaged structure according to claim 5, characterized in that, The calculation formula of the multi-class cross-entropy loss function is: Among them, E is the weighted cross-entropy loss, α is the weight, k is the total number of categories, and α s is the scaled weight, t k is the label, and y k is the predicted value.

7. A damage segmentation device for damaged structures, characterized in that, Including: A data acquisition module that obtains the original images of damaged RC structures and generates a training dataset; A data annotation module that performs annotation processing on the training dataset to obtain an annotated dataset; A data augmentation module that performs data augmentation on the annotated dataset to obtain an augmented dataset; A model training module that constructs a damaged structure damage segmentation model and uses the augmented dataset to train the damaged structure damage segmentation model to obtain a trained damaged structure damage segmentation model. Among them, the construction of the damaged structure damage segmentation model includes using a network architecture with pyramid multi-scale feature fusion, Haar wavelet downsampling, and an attention module's U-shaped network with skip connections to construct the damaged structure damage segmentation model; A model prediction module that uses the trained damaged structure damage segmentation model to perform damage segmentation on the image to be segmented to obtain a segmentation result.

8. An electronic device, characterized in that, Including: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1-6.

Images