Insulation Component Breakage Identification Method and Related Device

Abstract

The present application discloses a method for identifying damage to an insulating component and related devices. The method includes: obtaining an image region of the insulating component to be processed from the image to be analyzed; determining first integrity information of the insulating component to be processed based on the image features of the image region; in response to the first integrity information satisfying a first preset integrity condition, determining second integrity information of the insulating component to be processed based on the similarity between the image features and multiple reference image features, where the multiple reference image features are obtained by feature extraction based on reference images of insulating components with different degrees of integrity; determining damage information of the insulating component corresponding to the image to be analyzed based on the second integrity information, where the damage information of the insulating component is based on the degree of damage of the insulating component to be processed in the image to be analyzed. This can reduce the probability of missed detection of damaged insulating components caused by errors in the first detection result, thereby improving the accuracy of identifying damage to insulating components.

Description

Technical Field

[0001] This application relates to the technical field of power identification, and particularly to a method for identifying damage to insulating components and related devices. Background Art

[0002] Due to atmospheric pollutants and weather effects, the insulating components of transmission lines are prone to defects such as cracks and damage, which can lead to failures in transmission lines. Therefore, regularly inspecting the insulating components is of great significance for ensuring the safe and stable operation of transmission lines. The method of manual inspection is difficult and requires a large amount of work, which greatly increases the input costs of manpower and material resources. With the development of technology and the progress of artificial intelligence technology, video surveillance has been widely used and has gradually been applied to fields such as transmission line monitoring and identification of damaged insulating components. Since the monitoring equipment is not affected by the external environment, it can achieve all-round and uninterrupted monitoring of the insulating components of transmission lines. By using image processing and deep learning technologies to analyze the monitoring images in real time, once damage to the insulating components is detected, relevant personnel can be immediately reminded to handle it in a timely manner, minimizing the harm. It is an efficient and reliable method for identifying damaged insulating components of transmission lines.

[0003] Currently, the commonly used method is to perform object detection on the collected images of transmission lines and use the average distance between the centroids of adjacent insulators to determine whether there is a burst between two insulating components. Using only the centroid distance threshold to judge whether the insulating components are burst easily leads to incorrect discrimination, and then there will be a situation of missed detection of damaged insulating components. Therefore, there is an urgent need for a new method for identifying damage to insulating components to solve the above problems. Summary of the Invention

[0004] The main technical problem to be solved by this application is to provide a method for identifying damage to insulating components and related devices, which can reduce the probability of missed detection of damaged insulating components caused by incorrect first detection results.

[0005] To solve the above technical problem, one technical solution adopted by this application is: to provide a method for identifying damage to insulating components, including: obtaining the image region of the insulating component to be processed from the image to be analyzed; determining the first integrity information of the insulating component to be processed based on the image features of the image region; in response to the first integrity information meeting the first preset integrity condition, determining the second integrity information of the insulating component to be processed based on the similarity between the image features and multiple reference image features; wherein, the multiple reference image features are obtained by feature extraction based on reference images containing insulating components with different degrees of integrity; determining the damage information of the insulating component corresponding to the image to be analyzed based on the second integrity information; wherein, the damage information of the insulating component is based on the degree of damage of the insulating component to be processed in the image to be analyzed.

[0006] Wherein, the first integrity information includes one of the damaged component and the intact component; the first preset integrity condition includes that the first integrity information is the intact component; or the first integrity information includes a first integrity reference value representing the integrity degree of the insulating component to be processed; the first preset integrity condition includes that the first integrity reference value is greater than the first integrity reference value threshold; or the first integrity information includes a first damage reference value representing the damage degree of the insulating component to be processed; the first preset integrity condition includes that the first damage reference value is less than the first damage reference value threshold.

[0007] Wherein, the step of determining the second integrity information of the insulating component to be processed based on the similarity between the image feature and multiple reference image features includes: determining, based on the similarity between the image feature and each of the reference image features, the reference image features whose similarity is greater than or equal to the threshold; determining the second integrity information of the insulating component to be processed based on the reference integrity information corresponding to each of the determined reference image features; the reference integrity information is determined based on the integrity degree of the insulating component included in the reference image from which the corresponding reference image feature is obtained.

[0008] Wherein, the step of determining the damage information of the insulating component corresponding to the image to be analyzed based on the second integrity information includes: in response to the second integrity information satisfying the second preset integrity condition, determining that the damage information of the insulating component corresponding to the image to be analyzed is the damaged component; wherein, the second integrity information includes one of the damaged component and the intact component; the second preset integrity condition includes that the second integrity information is the damaged component; or, the second integrity information includes a second integrity reference value representing the integrity degree of the insulating component to be processed; the second preset integrity condition includes that the second integrity reference value is less than the second integrity reference value threshold; or, the second integrity information includes a second damage reference value representing the damage degree of the insulating component to be processed; the second preset integrity condition includes that the second damage reference value is greater than the second damage reference value threshold.

[0009] Wherein, after the step of determining the first integrity information of the insulating component to be processed based on the image feature of the image region, it further includes: in response to the first integrity information not satisfying the first preset integrity condition, determining that the damage information of the insulating component corresponding to the image to be analyzed is the damaged component.

[0010] Among them, the image to be analyzed includes video frames in a monitoring video collected for the insulating component to be processed, and the method further includes: in response to the number of target video frames in the monitoring video being greater than a threshold number, performing an alarm operation on the insulating component to be processed; where the target video frames include video frames with damaged component information being damaged components.

[0011] Among them, the step of obtaining the image region of the insulating component to be processed from the image to be analyzed includes: performing semantic segmentation on the image to be analyzed to obtain at least one first image, and each of the first images contains one insulating component to be processed; obtaining the boundary region corresponding to the insulating component to be processed from the first image, and obtaining a plurality of image regions containing the insulating component to be processed according to the boundary region.

[0012] Among them, the step of performing semantic segmentation on the image to be analyzed to obtain at least one first image, and each of the first images contains at least one insulating component to be processed includes: performing plant segmentation and sky segmentation on the image to be analyzed to obtain plant background information and sky background information; using a rotating object detection model to detect the image to be analyzed to obtain at least one insulating component region; filtering the plant regions containing the plant background information and the sky regions containing the sky background information in the insulating component region to obtain a first image containing at least one insulating component to be processed.

[0013] Among them, the step of determining the first integrity information of the insulating component to be processed based on the image features of the image region includes: determining the first integrity information of the insulating component to be processed using a classification model based on the image features of the image region; before the step of determining the first integrity information of the insulating component to be processed using a classification model based on the image features of the image region, it includes: obtaining a plurality of reference images containing insulating components with different degrees of integrity to form a classification image set; training the classification model using the classification image set.

[0014] To solve the above technical problems, another technical solution adopted by this application is: to provide an electronic device, including a memory and a processor coupled to each other, the memory stores program instructions, and the processor is used to execute the program instructions to implement the insulating component damage recognition method mentioned in any of the above embodiments.

[0015] To solve the above technical problems, another technical solution adopted by this application is: to provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is used to implement the insulating component damage recognition method mentioned in any of the above embodiments.

[0016] Distinct from the prior art, the beneficial effects of this application are as follows: The method for identifying damage to an insulating component provided in this application includes: obtaining an image area of the insulating component to be processed from the image to be analyzed, determining first integrity information of the insulating component to be processed based on the image features of the image area, when the first integrity information meets the first preset integrity condition, determining second integrity information of the insulating component to be processed based on the similarity between the image features and multiple reference image features, and determining damage information of the insulating component corresponding to the image to be analyzed based on the second integrity information. Through this design method, combining the judgment of the second integrity information can reduce the probability of missed detection of damaged insulating components caused by errors in the first detection result, thereby improving the accuracy of identifying damage to insulating components. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0018] Figure 1 is a schematic flowchart of an embodiment of the method for identifying damage to an insulating component in this application;

[0019] Figure 2 is Figure 1 a schematic flowchart of an embodiment of step S1 therein;

[0020] Figure 3 is Figure 2 a schematic flowchart of an embodiment of step S10 therein;

[0021] Figure 4 is Figure 3 a schematic flowchart of an embodiment before step S101 therein;

[0022] Figure 5 is Figure 2 a schematic flowchart of an embodiment before step S11 therein;

[0023] Figure 6 is Figure 1 a schematic flowchart of an embodiment before step S2 therein;

[0024] Figure 7 is Figure 1 a schematic flowchart of an embodiment of step S4 therein;

[0025] Figure 8 is Figure 1 a schematic flowchart of an embodiment of step S6 therein;

[0026] Figure 9 It is a schematic flow chart of another embodiment of the method for identifying damage to an insulating component in this application;

[0027] Figure 10 It is a schematic structural diagram of an embodiment of the insulator damage identification device in this application;

[0028] Figure 11 It is a schematic framework diagram of an embodiment of an electronic device in this application;

[0029] Figure 12 It is a schematic framework diagram of an embodiment of a computer-readable storage medium in this application. Detailed implementation manners

[0030] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts shall fall within the scope of protection of this application.

[0031] Glossary:

[0032] Insulating component: An electrical isolation device, such as but not limited to including insulators. An insulator is used to fix the wire and should have sufficient electrical insulation ability and mechanical strength to insulate between live wires or between a wire and the ground.

[0033] Please refer to Figure 1 , Figure 1 It is a schematic flow chart of an embodiment of the method for identifying damage to an insulating component in this application. The method includes:

[0034] S1: Obtain the image area of the insulating component to be processed from the image to be analyzed.

[0035] Specifically, the image to be analyzed can be a camera monitoring screen or an image obtained through other means, which is not limited in this application. Specifically, in this embodiment, please refer to Figure 2 , Figure 2 is Figure 1 a schematic flow chart of an embodiment of step S1 in

[0036] S10: Perform semantic segmentation on the image to be analyzed to obtain at least one first image, and each first image contains at least one insulating component to be processed.

[0037] In one embodiment, please refer to Figure 3 , Figure 3 is Figure 2Schematic flowchart of an implementation of step S10. Step S10 specifically includes:

[0038] S100: Perform plant segmentation and sky segmentation on the image to be analyzed to obtain plant background information and sky background information.

[0039] In addition, in this embodiment, before executing step S100, a plant data set and a sky data set are acquired and labeled, and a semantic segmentation network model is trained using the plant data set and the sky data set. Specifically, the semantic segmentation network model is used to perform plant segmentation and sky segmentation on the image to be analyzed to obtain plant background information and sky background information.

[0040] S101: Use a rotating object detection model to detect the image to be analyzed to obtain at least one insulating component area.

[0041] Specifically, the rotating object detection model obtained in step S100 is used to detect the image to be analyzed to obtain the insulating component area. Since the rotating object detection model can detect and identify insulating components captured from different angles and different environments, the insulating components in the image to be analyzed in any direction and any environment can be detected and identified, and the insulating component area in the image to be analyzed can be accurately obtained, thereby improving the accuracy and efficiency of detection and identification.

[0042] In one embodiment, please refer to Figure 4 , Figure 4 is Figure 3 Schematic flowchart of an implementation before step S101 in

[0043] S20: Construct a training sample set.

[0044] Specifically, the training sample set contains training sample images of multiple insulating components captured from different angles and different environments.

[0045] S21: Use the training sample set to train a rotating object detection model.

[0046] Specifically, the training sample set obtained in step S20 is used to train the rotating object detection model. Since the training sample set contains training sample images of multiple insulating components captured from different angles and different environments, the rotating object detection model trained thereby is more accurate and comprehensive, and can detect and identify insulating components in any direction and any environment in the high-voltage transmission line image, thereby improving the accuracy and efficiency of detection and identification. In addition, the rotating object detection model can be a convolutional neural network model, etc., and the present application does not limit this here.

[0047] S102: Filter the plant area containing plant background information and the sky area containing sky background information in the insulation component area to obtain a first image containing at least one insulation component to be processed.

[0048] Specifically, the insulation component area obtained in step S101 also includes a plant area and a sky area. At this time, the plant area can be determined using the vegetation background information obtained in step S100, and the sky area can be determined using the sky background information obtained in step S200. Then, the determined plant area and sky area are filtered, and finally, a first image containing only the insulation component to be processed is obtained. Through this design method, the background information in the first image that does not belong to the insulation component to be processed can be reduced, so as to obtain a more accurate insulation component area, thereby improving the accuracy and efficiency of insulation component detection and recognition.

[0049] S11: Obtain the boundary area corresponding to the insulation component to be processed from the first image, and obtain an image area containing the insulation component to be processed based on the boundary area.

[0050] In one embodiment, please refer to Figure 5 , Figure 5 is Figure 2 a schematic flowchart of an embodiment of step S11 in

[0051] S110: Use filtering and noise reduction and edge detection to obtain the boundary area corresponding to the insulation component to be processed in the first image.

[0052] In this embodiment, filtering and noise reduction and edge detection are used to obtain the boundary area corresponding to the insulation component to be processed in the first image. Of course, in other embodiments, other methods may also be used to obtain the boundary area corresponding to the insulation component to be processed in the first image, and the present application does not limit this here.

[0053] S111: Obtain the insulation component to be processed whose boundary area in the first image is a closed contour.

[0054] Specifically, the boundary area of the insulation component is a closed contour. Therefore, the target whose boundary area in the first image is a closed contour can be obtained from the first image, and this target is the insulation component to be processed. This can more accurately detect the insulation component to be processed, thereby improving the accuracy and efficiency of detection and recognition.

[0055] S2: Determine the first integrity information of the insulation component to be processed based on the image features of the image area.

[0056] Specifically, in this embodiment, step S2 specifically includes: based on the image features obtained from the image region in step S111, using a classification model to detect and identify the insulating component to be processed in the image region, so as to determine the first integrity information of the insulating component to be processed. Additionally, in this embodiment, please refer to Figure 6 , Figure 6 is Figure 1 a schematic flowchart of a previous embodiment before step S2. Before step S2, it specifically includes:

[0057] S30: Obtain a plurality of reference images containing insulating components with different degrees of integrity to form a classification image set.

[0058] Specifically, before step S2, obtain and collect a plurality of reference images containing insulating components with different degrees of integrity to form a classification image set.

[0059] S31: Train a classification model using the classification image set.

[0060] Specifically, using the classification model obtained by training with a classification image set formed by reference images of insulating components with different degrees of integrity to detect and identify insulating components can improve the accuracy and efficiency of insulating component detection and identification.

[0061] S3: Determine whether the first integrity information meets the first preset integrity condition.

[0062] Specifically, the first integrity information includes one of the information of damaged components and complete components. The first preset integrity condition includes that the first integrity information is a complete component; or, the first integrity information includes a first complete reference value representing the integrity degree of the insulating component to be processed. At this time, the first preset integrity condition includes that the first complete reference value is greater than the first complete reference value threshold; or, the first integrity information includes a first damage reference value representing the damage degree of the insulating component to be processed. At this time, the first preset integrity condition includes that the first damage reference value is less than the first damage reference value threshold, which can be set according to the actual situation and is not limited in this application.

[0063] S4: If so, based on the similarity between the image features and the features of multiple reference images, determine the second integrity information of the insulating component to be processed.

[0064] Specifically, if the first integrity information meets the first preset integrity condition, then enter the step of determining the second integrity information of the insulating component to be processed based on the similarity between the image features and the features of multiple reference images, which can avoid the missed detection of damaged insulating components caused by incorrect discrimination in the first detection result. In this embodiment, the multiple reference image features are obtained by feature extraction based on reference images of insulating components with different degrees of integrity. These reference images are relatively representative, so that the credibility of the second integrity information obtained after comparing the features of the insulating component to be processed in the image area with the reference images can be ensured to be higher, thereby improving the accuracy and efficiency of the detection and recognition of insulating components.

[0065] In addition, in this embodiment, representative intact insulator samples and damaged insulator samples can be selected from the classified image set formed in step S30. Of course, in other embodiments, the representative reference images can also be representative reference images separately selected from multiple insulating component images. This application does not make any limitations here. In addition, the number of reference images can be set according to the actual situation, as long as it includes intact component samples and damaged component samples. This application does not make any limitations here.

[0066] Through such a design method, combined with the second integrity information, the probability of missed detection of damaged insulators caused by the error of the first detection result can be reduced, thereby improving the accuracy and efficiency of the identification of damaged insulating components.

[0067] In one embodiment, please refer to Figure 7 , Figure 7 is Figure 1 a schematic flowchart of an implementation manner of step S4 in

[0068] S40: Based on the similarity between the image features and each reference image, determine the reference image features whose similarity is greater than or equal to the threshold.

[0069] Specifically, extract the image features of the image area obtained in step S111, compare them with each reference image to obtain the similarity, and determine the reference image features whose similarity is greater than or equal to the threshold.

[0070] S41: Based on the reference integrity information corresponding to the determined reference image features, determine the second integrity information of the insulating component to be processed.

[0071] Specifically, the reference integrity information is determined based on the integrity of the insulating components included in the reference image from which the corresponding reference image features are obtained. In this way, through the representative reference image, it is possible to more accurately know whether the insulating component to be processed in the image area is a damaged component or a complete component, thereby improving the accuracy of insulating component detection and recognition.

[0072] S5: Otherwise, determine that the damage information of the insulating component corresponding to the image to be analyzed is a damaged component.

[0073] Specifically, if the first integrity information does not meet the first preset integrity condition, determine that the damage information of the insulating component corresponding to the image to be analyzed is a damaged component, and enter the step of performing an alarm operation on the insulating component to be processed in response to the number of target video frames in the monitoring video being greater than the number threshold, so as to further judge the insulating component to be processed.

[0074] S6: Based on the second integrity information, determine the damage information of the insulating component corresponding to the image to be analyzed.

[0075] Specifically, the damage information of the insulating component is based on the degree of damage of the insulating component to be processed in the image to be analyzed.

[0076] Specifically, in this embodiment, please refer to Figure 8 , Figure 8 is Figure 1 a schematic flowchart of an implementation manner of step S6 in

[0077] S50: Judge whether the second integrity information meets the second preset integrity condition.

[0078] Specifically, the second integrity information includes one of the information of the damaged component and the complete component. The second preset integrity condition includes that the second integrity information is a damaged component; or, the second integrity information includes a second complete reference value representing the integrity of the insulating component to be processed. At this time, the second preset integrity condition includes that the second complete reference value is less than the second complete reference value threshold; or, the second integrity information includes a second damage reference value representing the degree of damage of the insulating component to be processed. At this time, the second preset integrity condition includes that the second damage reference value is greater than the second damage reference value threshold.

[0079] S51: If so, determine that the damage information of the insulating component corresponding to the image to be analyzed is a damaged component.

[0080] Specifically, if the second integrity information meets the second preset integrity condition, it is determined that the damaged information of the insulating component corresponding to the image to be analyzed is a damaged component, and the step of performing an alarm operation on the insulating component to be processed in response to the number of target video frames in the monitoring video being greater than the number threshold is entered, so as to further judge the insulating component to be processed.

[0081] S52: Otherwise, it is determined that the damaged information of the insulating component corresponding to the image to be analyzed is a complete component.

[0082] Specifically, if the second integrity information does not meet the second preset integrity condition, it is determined that the damaged information of the insulating component corresponding to the image to be analyzed is a complete component, and the process returns to the step of obtaining the image area of the insulating component to be processed from the image to be analyzed. This can save the detection time and thus improve the detection efficiency.

[0083] In this way, it is possible to more accurately know whether the insulating component to be processed in the image area is a damaged component or a complete component, thereby improving the accuracy of detecting and identifying the insulating component.

[0084] In one embodiment, the image to be analyzed includes a video frame in a monitoring video collected for the insulating component to be processed. Specifically, the monitoring video is a video collected over a continuous period of time, and the application does not limit the time T of the collected video. Please refer to Figure 9 , Figure 9 is a schematic flowchart of another embodiment of the method for identifying damaged insulating components in this application. Specifically, the method for identifying damaged insulating components further includes:

[0085] S60: Judge whether the number of target video frames in the monitoring video is greater than the number threshold.

[0086] Specifically, the value of the number threshold N (N≥1) can be set according to the actual situation, and the application does not limit this here.

[0087] S61: If so, perform an alarm operation on the insulating component to be processed.

[0088] Specifically, the above-mentioned target video frame includes a video frame with the insulation component damage information being the damaged component. Specifically, if the number of target video frames in the monitoring video is greater than the number threshold N, it indicates that the insulation component is truly a damaged component, then the position of the damaged component is output and an alarm prompt is issued to accurately remind the staff of the position of the damaged component and promptly process the damaged component. In this embodiment, when it is determined that one insulation component to be processed is a damaged component, an alarm prompt is given for this damaged component. Of course, it is also possible to give an alarm prompt for these damaged components when multiple insulation components to be processed are cumulatively determined to be damaged components. This application does not make a limitation here. Additionally, in this embodiment, the alarm prompt can be made by means such as sound, light, or other means, and this application does not make a limitation here.

[0089] S62: Otherwise, return to the step of obtaining the image area of the insulation component to be processed from the image to be analyzed.

[0090] Specifically, if the number of target video frames in the monitoring video is less than or equal to the number threshold N, it indicates that the insulation component is truly an intact component, then return to the step of obtaining the image area of the insulation component to be processed from the image to be analyzed. This can save the detection time and thus improve the detection efficiency.

[0091] Through this method, it can be known whether the insulation component is truly a damaged component, which can enhance the credibility of the detection result.

[0092] Through this design method, combined with the judgment of the second integrity information, the probability of missing the detection of damaged insulation components due to the error of the first detection result can be reduced, thereby improving the accuracy of the identification of insulation component damage.

[0093] Please refer to Figure 10 , Figure 10 which is a schematic structural diagram of an embodiment of the insulator damage identification device of this application. The insulator damage identification device specifically includes:

[0094] An obtaining module 10, configured to obtain the image area of the insulation component to be processed from the image to be analyzed.

[0095] The processing module 12 is coupled to the obtaining module 10 and is configured to determine the first integrity information of the insulating component to be processed based on the image features of the image region. The processing module 12 is further configured to, in response to the first integrity information satisfying the first preset integrity condition, determine the second integrity information of the insulating component to be processed based on the similarity between the image features and a plurality of reference image features; wherein, the plurality of reference image features are obtained by feature extraction based on reference images including insulating components with different degrees of integrity. In addition, the processing module 12 is further configured to determine the damage information of the insulating component corresponding to the image to be analyzed based on the second integrity information; wherein, the damage information of the insulating component is based on the degree of damage of the insulating component to be processed in the image to be analyzed.

[0096] Please refer to Figure 11 , Figure 11 FIG. is a schematic framework diagram of an embodiment of the electronic device of the present application. The electronic device includes a memory 20 and a processor 22 that are coupled to each other. Specifically, in this embodiment, program instructions are stored in the memory 20, and the processor 22 is configured to execute the program instructions to implement the insulating component damage recognition method mentioned in any of the above embodiments.

[0097] Specifically, the processor 22 may also be referred to as a CPU (Central Processing Unit). The processor 22 may be an integrated circuit chip with signal processing capabilities. The processor 22 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. Additionally, the processor 22 may be implemented by a plurality of integrated circuit chips together.

[0098] Please refer to Figure 12 , Figure 12It is a schematic framework diagram of an embodiment of the computer-readable storage medium of the present application. The computer-readable storage medium 30 stores a computer program 300 that can be read by a computer. The computer program 300 can be executed by a processor to implement the insulation component breakage recognition method mentioned in any of the above embodiments. Among them, the computer program 300 can be stored in the above computer-readable storage medium 30 in the form of a software product, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The computer-readable storage medium 30 with a storage function can be various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc, or a terminal device such as a computer, a server, a mobile phone, or a tablet.

[0099] In summary, different from the prior art, the insulation component breakage recognition method provided by the present application includes: obtaining an image area of the insulation component to be processed from the image to be analyzed, determining the first integrity information of the insulation component to be processed based on the image features of the image area, when the first integrity information meets the first preset integrity condition, determining the second integrity information of the insulation component to be processed based on the similarity between the image features and multiple reference image features, and determining the insulation component breakage information corresponding to the image to be analyzed based on the second integrity information. Through this design method, combining the judgment of the second integrity information can reduce the probability of missed detection of damaged insulation components caused by errors in the first detection result, thereby improving the accuracy of insulation component breakage recognition.

[0100] The above is only the embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for identifying damage to an insulating component, characterized in that, Including: Obtaining an image region of a to-be-processed insulating component from an image to be analyzed, including: performing plant segmentation and sky segmentation on the image to be analyzed to obtain plant background information and sky background information; using a rotating object detection model to detect the image to be analyzed to obtain at least one insulating component region; filtering the plant region containing the plant background information and the sky region containing the sky background information in the insulating component region to obtain a first image containing at least one to-be-processed insulating component; obtaining a boundary region corresponding to the to-be-processed insulating component from the first image, and obtaining an image region containing the to-be-processed insulating component according to the boundary region; wherein, the rotating object detection model is used to detect and identify the to-be-processed insulating component in the image to be analyzed in different directions and different environments; the boundary region of the insulating component is a closed contour; Based on the image features of the image region, using a classification model to determine the first integrity information of the to-be-processed insulating component; In response to the first integrity information satisfying a first preset integrity condition, based on the similarity between the image features and multiple reference image features, determining the second integrity information of the to-be-processed insulating component; wherein, the multiple reference image features are obtained by feature extraction based on reference images containing insulating components with different integrity levels; the reference images are selected from a classification image set formed by reference images containing insulating components with different integrity levels, or the reference images are representative reference images separately selected from multiple insulating component images; Based on the second integrity information, determining the insulating component damage information corresponding to the image to be analyzed; wherein, the insulating component damage information is based on the damage degree of the to-be-processed insulating component in the image to be analyzed; Wherein, the first integrity information includes one of damaged components and intact components; the first preset integrity condition includes that the first integrity information is the intact component; or The first integrity information includes a first integrity reference value representing the integrity level of the to-be-processed insulating component; the first preset integrity condition includes that the first integrity reference value is greater than a first integrity reference value threshold; or The first integrity information includes a first damage reference value representing the damage degree of the to-be-processed insulating component; the first preset integrity condition includes that the first damage reference value is less than a first damage reference value threshold.

2. The insulation component breakage identification method according to claim 1, wherein The step of determining the second integrity information of the to-be-processed insulating component based on the similarity between the image features and multiple reference image features includes: Based on the similarity between the image features and each reference image feature, determining the reference image features whose similarity is greater than or equal to a threshold; Based on the reference integrity information corresponding to each determined reference image feature, determining the second integrity information of the to-be-processed insulating component; wherein, the reference integrity information is determined based on the integrity level of the insulating component contained in the reference image corresponding to the obtained reference image feature.

3. The method for identifying damage to an insulating component according to claim 1, wherein The step of determining the damaged information of the insulating component corresponding to the image to be analyzed based on the second integrity information includes: In response to the second integrity information satisfying the second preset integrity condition, determining that the damaged information of the insulating component corresponding to the image to be analyzed is a damaged component; wherein, The second integrity information includes one of the damaged component and the intact component; the second preset integrity condition includes that the second integrity information is the damaged component; or, the second integrity information includes a second integrity reference value characterizing the integrity degree of the insulating component to be processed; the second preset integrity condition includes that the second integrity reference value is less than the second integrity reference value threshold; or, the second integrity information includes a second damage reference value characterizing the damage degree of the insulating component to be processed; the second preset integrity condition includes that the second damage reference value is greater than the second damage reference value threshold.

4. The method for identifying damage to an insulating component according to claim 1, wherein After the step of determining the first integrity information of the insulating component to be processed based on the image features of the image region, it further includes: In response to the first integrity information not satisfying the first preset integrity condition, determining that the damaged information of the insulating component corresponding to the image to be analyzed is a damaged component.

5. The method for identifying damage to an insulating component according to any one of claims 1-4, characterized in that, The image to be analyzed includes a video frame in a monitoring video collected for the insulating component to be processed, and the method further includes: In response to the number of target video frames in the monitoring video being greater than the number threshold, performing an alarm operation on the insulating component to be processed; wherein, the target video frames include video frames with the damaged information of the insulating component being a damaged component.

6. The method for identifying damage to an insulating component according to claim 1, characterized in that, Before the step of determining the first integrity information of the insulating component to be processed by using a classification model based on the image features of the image region, it includes: Obtaining a plurality of reference images containing insulating components with different integrity degrees to form a classification image set; Training the classification model by using the classification image set.

7. An electronic device, characterized in that, Including a memory and a processor coupled to each other, the memory stores program instructions, and the processor is configured to execute the program instructions to implement the insulating component damage recognition method according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program is configured to be executed by a processor to implement the insulating component damage recognition method according to any one of claims 1 to 6.

Images