Methods, devices, electronic equipment, and storage media for measuring insulator creepage distance

Abstract

This application provides a method, apparatus, electronic device, and storage medium for measuring the creepage distance of insulators. The measurement method includes: acquiring a scanned image of the insulator to be measured; inputting the scanned image into a pre-trained image extraction model and outputting a marker image; calculating the length of the marker after dividing the marker image; and determining the length of the marker as the creepage distance of the insulator to be measured. The technical solution provided in this application enables the acquisition of a scanned image of the insulator using a three-dimensional scanning device, the extraction of markers attached to the insulator using a pre-trained image extraction model, the calculation of the marker length, and the determination of the marker length as the creepage distance. This achieves non-contact creepage distance measurement of insulators, improving the efficiency and accuracy of creepage distance measurement.

Description

Technical Field

[0001] This application relates to the technical field of three-dimensional measurement, and in particular to a method, apparatus, electronic device, and storage medium for measuring the creepage distance of insulators. Background Technology

[0002] Insulator creepage distance refers to the shortest distance or the sum of the shortest distances along the surface between conductive parts under normal operating voltage. Under different operating conditions, the insulating material surrounding the conductor becomes polarized, causing the insulator to become charged. Whether the insulator creepage distance meets the standard has a profound impact on the safety of operators.

[0003] The smooth surface of insulators, composed of multiple arc segments, presents challenges in measuring creepage distance. Currently, insulator creepage distance is primarily measured manually, which is difficult and inefficient. Therefore, developing a non-contact method for measuring insulator creepage distance has become an urgent problem to be solved. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, device, electronic device and storage medium for measuring the creepage distance of insulators, which can acquire scanned images of insulators through a three-dimensional scanning device, extract markers attached to the insulators using a pre-trained image extraction model, calculate the length of the markers, and determine the length of the markers as the creepage distance, thereby realizing non-contact measurement of insulator creepage distance and improving the efficiency and accuracy of creepage distance measurement.

[0005] This application mainly includes the following aspects:

[0006] In a first aspect, embodiments of this application provide a method for measuring the creepage distance of an insulator, the method comprising:

[0007] A scanned image of the insulator to be measured is acquired; wherein, a marker is attached to the surface of the insulator, and the marker extends from the center of the insulator along the surface of the insulator to the edge;

[0008] The scanned image is input into a pre-trained image extraction model, which outputs a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker.

[0009] The length of the marker is obtained by dividing and calculating the image of the marker, and the length of the marker is determined as the creepage distance of the insulator to be measured.

[0010] Furthermore, the image extraction model includes a downsampling layer, an attention extraction layer, and an upsampling layer, which are sequentially connected; the step of inputting the scanned image into the pre-trained image extraction model and outputting a marker image includes:

[0011] The scanned image is input into a downsampling layer and undergoes an encoding process to reduce the image size, thereby obtaining a reduced feature image.

[0012] The reduced feature image is input into the attention extraction layer, and feature extraction is performed based on the feature information of the marker to obtain the target feature image;

[0013] The target feature image is input into the upsampling layer and then processed through a decoding process to enlarge the image, thereby obtaining the marker image.

[0014] Furthermore, the attention extraction layer includes a channel attention extraction layer and a spatial attention extraction layer; the step of inputting the reduced feature image into the attention extraction layer and extracting features based on the feature information of the markers to obtain the target feature image includes:

[0015] The reduced feature image is input into the channel attention extraction layer, and feature extraction is performed based on the correlation between feature information in the channels to obtain the channel attention image;

[0016] The channel attention image is input into the spatial attention extraction layer, and feature extraction is performed based on the spatial relationship of feature information to obtain the target feature image.

[0017] Furthermore, the step of inputting the reduced feature image into the channel attention extraction layer, and extracting features based on the correlation between feature information in the channels to obtain the channel attention image includes:

[0018] After performing max pooling and average pooling on the reduced feature image, two first descriptive feature images are obtained;

[0019] Based on the correlation between the feature information in the two first descriptive feature maps, feature extraction is performed to obtain two feature images;

[0020] After image processing of the two feature images, the first weight coefficient is determined based on a preset activation function;

[0021] Based on the first weighting coefficient and the reduced feature image, a channel attention image is obtained.

[0022] Furthermore, the step of inputting the channel attention image into the spatial attention extraction layer, and extracting features based on the spatial relationship of feature information to obtain the target feature image includes:

[0023] After performing max pooling and average pooling on the channel attention image, two second descriptive feature images are obtained;

[0024] The two second descriptive feature images are spliced ​​together according to the spatial relationship of the feature information, and then processed by convolution and activation function to determine the second weight coefficients;

[0025] The target feature image is obtained based on the second weighting coefficient and the channel attention image.

[0026] Furthermore, the step of dividing and calculating the marker image to obtain the marker length, and determining the marker length as the creepage distance, includes:

[0027] The marker image is cut through a cutting surface in a preset direction to obtain multiple segments of marker point cloud data between each two adjacent cutting surfaces;

[0028] For each segment of marker point cloud data, the segment of marker point cloud data after filtering is determined as a marker strip;

[0029] For each marker strip, the marker strip is projected onto the cutting surface, and noise reduction, sparsity reduction, and interpolation processing are performed to determine the length of the marker strip.

[0030] The length of the marker is obtained by adding the lengths of the determined multiple marker strips, and the length of the marker is determined as the creepage distance.

[0031] Secondly, embodiments of this application also provide a measuring device for insulator creepage distance, the measuring device comprising:

[0032] An acquisition module is used to acquire a scanned image of the insulator to be measured; wherein, a marker is attached to the surface of the insulator, and the marker covers the surface of the insulator from the center to the edge;

[0033] The processing module is used to input the scanned image into a pre-trained image extraction model and output a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker.

[0034] The determination module is used to obtain the length of the marker after dividing and calculating the marker image, and to determine the length of the marker as the creepage distance of the insulator to be measured.

[0035] Furthermore, the image extraction model includes a downsampling layer, an attention extraction layer, and an upsampling layer, which are sequentially connected. When the processing module inputs the scanned image into the pre-trained image extraction model and outputs a marker image, the processing module is specifically used for:

[0036] The scanned image is input into a downsampling layer and undergoes an encoding process to reduce the image size, thereby obtaining a reduced feature image.

[0037] The reduced feature image is input into the attention extraction layer, and feature extraction is performed based on the feature information of the marker to obtain the target feature image;

[0038] The target feature image is input into the upsampling layer and then processed through a decoding process to enlarge the image, thereby obtaining the marker image.

[0039] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the insulator creepage distance measurement method described above are performed.

[0040] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the insulator creepage distance measurement method described above.

[0041] This application provides a method, apparatus, electronic device, and storage medium for measuring the creepage distance of an insulator. The measurement method includes: acquiring a scanned image of an insulator to be measured, wherein a marker is attached to the surface of the insulator, and the marker extends from the center of the insulator along the surface to the edge; inputting the scanned image into a pre-trained image extraction model and outputting a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker; dividing and calculating the marker image to obtain the length of the marker, and determining the length of the marker as the creepage distance of the insulator to be measured.

[0042] Thus, by adopting the technical solution provided in this application, a scanned image of the insulator can be obtained through a three-dimensional scanning device, the marker attached to the insulator can be extracted using a pre-trained image extraction model, the length of the marker can be calculated, and the length of the marker can be determined as the creepage distance, thereby realizing non-contact insulator creepage distance measurement and improving the efficiency and accuracy of creepage distance measurement.

[0043] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart illustrating a method for measuring the creepage distance of an insulator according to an embodiment of this application is shown;

[0046] Figure 2 A flowchart illustrating another method for measuring the creepage distance of an insulator provided in an embodiment of this application is shown;

[0047] Figure 3 This invention provides a schematic diagram of the structure of an insulator creepage distance measuring device according to an embodiment of the present application.

[0048] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0050] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0051] To enable those skilled in the art to use the content of this application, and in conjunction with the specific application scenario of "measurement of insulator creepage distance", the following implementation method is provided. For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application.

[0052] The methods, apparatus, electronic devices, or computer-readable storage media described in this application can be applied to any scenario where insulator creepage distance needs to be measured. This application does not limit the specific application scenario. Any scheme that uses an insulator creepage distance measurement method, apparatus, electronic device, and storage medium provided in this application is within the protection scope of this application.

[0053] It is worth noting that insulator creepage distance refers to the shortest distance or the sum of the shortest distances along the surface between conductive parts under normal operating voltage. Under different operating conditions, the insulation material around the conductor can become polarized, causing the insulator to exhibit a charged phenomenon. Whether the insulator creepage distance meets the standard has a profound impact on the safety of operators.

[0054] Because the surface of insulators is smooth and composed of multiple arc segments, measuring creepage distance presents certain difficulties. Currently, measurements are mainly taken manually, which is both difficult and inefficient. Therefore, finding a non-contact method to measure the creepage distance of insulators has become an urgent problem to be solved.

[0055] Based on this, this application proposes a method, apparatus, electronic device, and storage medium for measuring the creepage distance of insulators. The measurement method includes: acquiring a scanned image of the insulator to be measured, wherein a marker is attached to the surface of the insulator, and the marker extends from the center of the insulator along the surface to the edge; inputting the scanned image into a pre-trained image extraction model and outputting a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker; calculating the length of the marker after dividing the marker image, and determining the length of the marker as the creepage distance of the insulator to be measured. The technical solution provided in this application enables the acquisition of a scanned image of the insulator using a three-dimensional scanning device, the extraction of markers attached to the insulator using a pre-trained image extraction model, the calculation of the marker length, and the determination of the marker length as the creepage distance, thus realizing non-contact insulator creepage distance measurement and improving the efficiency and accuracy of creepage distance measurement.

[0056] To facilitate understanding of this application, the technical solutions provided in this application will be described in detail below with reference to specific embodiments.

[0057] Please see Figure 1 , Figure 1 A flowchart illustrating a method for measuring the creepage distance of an insulator provided in an embodiment of this application is shown below. Figure 1 As shown, the measurement method includes:

[0058] S101. Obtain a scanned image of the insulator to be measured;

[0059] In this step, the insulator is scanned in three dimensions using a calibrated 3D scanning device to obtain a scanned image of the insulator's 3D point cloud data. The calibration and standardization methods for the 3D scanning device are not limited; for example, calibration and standardization can be performed using a checkerboard target. It should be noted that a long strip-shaped marker with a certain thickness is attached to the surface of the insulator and is tightly and continuously pasted along a radial direction of the insulator, covering the surface from the center to the edge. The line connecting any point on the surface between the center and the edge of the insulator can be selected as the direction of the marker's attachment. The marker is black, red, or another color with a certain contrast to the insulator. The color and thickness of the markers should not be too thick or too thin. The thickness should be greater than the scanning capability of the 3D scanning equipment, so that the relative height of the insulator in the marked area is higher than that in other areas. If the markers are too thin, the 3D scanning equipment will not be able to obtain its point cloud information, and if the markers are too thick, the measured insulator creep distance error will be large. When the insulator is suspended by a suspension device and scanned by the 3D scanning equipment, the top, bottom and side surfaces of the insulator should be scanned from multiple angles to obtain more detailed point cloud data of the insulator and the markers, and generate a scanned image containing the 3D point cloud data of the insulator and the markers.

[0060] S102. Input the scanned image into a pre-trained image extraction model and output the marker image;

[0061] In this step, the marker image is extracted from the scanned image based on the marker's feature information. For example, the marker image can be extracted from the scanned image based on the color characteristics of the marker and the insulator, using a color comparison between the marker and the insulator. The image extraction model, such as the U-Net model, includes downsampling layers, upsampling layers, and skip connections. An attention mechanism is added between the downsampling and upsampling layers, including channel attention and spatial attention modules, to perform attention marking in the image channels and spatial dimensions, making the feature extraction network focus more on the marker during feature extraction. The scanned image is encoded and then decoded by the image extraction model to obtain the target region, i.e., the marker image. Encoding is achieved using downsampling, with convolutional layers, pooling layers, and activation functions as the basic units of the feature extractor. After several feature extractions on the scanned image to reduce its size, feature information from the insulator point cloud data is obtained. Further, using upsampling layer decoding and skip connections, the reduced image is restored to its original size, and the segmentation result, i.e., the marker image, is obtained. For detailed steps on extracting the marker image, please refer to [link to relevant documentation]. Figure 2 , Figure 2 A flowchart illustrating another method for measuring the creepage distance of an insulator provided in this application embodiment is shown below. Figure 2 As shown, the steps for outputting the marker image include:

[0062] S201. The scanned image is input into the downsampling layer and processed by the encoding process to reduce the image size, thereby obtaining a reduced feature image;

[0063] In this step, the scanned image obtained in step S101 is input into the downsampling layer and processed by the encoding process to reduce the image size, thereby obtaining a reduced feature image.

[0064] S202. Input the reduced feature image into the attention extraction layer, and perform feature extraction based on the feature information of the marker to obtain the target feature image;

[0065] In this step, the attention extraction layer includes a channel attention extraction layer and a spatial attention extraction layer; the step of inputting the reduced feature image obtained in step S201 into the attention extraction layer, and performing feature extraction based on the feature information of the markers to obtain the target feature image includes:

[0066] S2021. Input the reduced feature image into the channel attention extraction layer, and perform feature extraction based on the correlation between feature information in the channels to obtain the channel attention image;

[0067] In this step, the reduced feature image obtained in step S201 is input to the channel attention extraction layer. Here, the steps of feature extraction to obtain the channel attention image based on the correlation between the feature information of the markers in the channels include:

[0068] (1) After performing max pooling and average pooling on the reduced feature image, two first descriptive feature images are obtained;

[0069] In this step, the input feature map, i.e., the reduced feature image obtained in step S201, is passed through a global max pooling layer and a global average pooling layer with unchanged width and height, respectively, to obtain two first descriptive feature images of the same size. For example, the input reduced feature image is a feature F of size H×W×C (H×W represents pixel size), and then a spatial global max pooling and average pooling are performed to obtain two 1×1×C channel descriptions.

[0070] (2) Based on the correlation between the feature information in the channels, the two first descriptive feature maps are used to extract features to obtain two feature images;

[0071] In this step, two neural network layers and activation functions are used as shared layers to extract features from the two first descriptive feature maps obtained in step (1) based on the feature information of the markers, resulting in two feature images. For example, the two first descriptive feature maps are fed into a two-layer neural network layer, with the first layer having C / r neurons and ReLU activation function, and the second layer having C neurons. The two neural networks are shared, and two feature images are output.

[0072] (3) After processing the two feature images, determine the first weight coefficient based on the preset activation function;

[0073] In this step, after image processing of the two feature images obtained in step (2) above, the first weight coefficient is determined based on a preset activation function; for example, the two feature images are added together and then passed through a Sigmoid activation function to obtain the first weight coefficient Mc.

[0074] (4) Based on the first weight coefficient and the reduced feature image, obtain the channel attention image;

[0075] In this step, the channel attention image is determined based on the first weight coefficient determined in step (3) above and the reduced feature image of the input channel attention extraction layer. For example, the input reduced feature image is a feature F of H×W×C (H×W represents the pixel size). The first weight coefficient is multiplied by the reduced feature image, i.e., Mc×F, to obtain the channel attention image.

[0076] S2022. Input the channel attention image into the spatial attention extraction layer, and perform feature extraction based on the spatial relationship of feature information to obtain the target feature image.

[0077] In this step, the channel attention image determined in step S2021 is input into the spatial attention extraction layer. Here, the step of extracting features based on the spatial relationship of the feature information of the markers to obtain the target feature image includes:

[0078] A. After performing max pooling and average pooling on the channel attention image, two second descriptive feature images are obtained;

[0079] In this step, the output of the channel attention extraction layer, i.e., the channel attention image, is used as the input feature of the spatial attention extraction layer. First, channel-based global max pooling and global average pooling operations are performed on the input feature channel attention image to obtain two single-channel feature maps, i.e., two second descriptive feature images. For example, similar to the processing of the channel attention extraction layer, the channel attention image is a feature F' of H×W×C, and max pooling and average pooling are performed on one channel dimension to obtain two second descriptive feature images of H×W×1.

[0080] B. After splicing the two second descriptive feature images according to the spatial relationship of the feature information, the second weight coefficients are determined by convolution and activation function processing.

[0081] In this step, the two second descriptive feature images obtained in step A above are spliced ​​together according to the spatial relationship of the feature information of the markers, and then processed by convolution and activation function to determine the second weight coefficients; for example, the two second descriptive feature images are spliced ​​together according to channels, and then processed by a 7×7 convolutional layer with Sigmoid activation function to obtain the second weight coefficient Ms.

[0082] C. Based on the second weighting coefficient and the channel attention image, obtain the target feature image.

[0083] In this step, the target feature image is determined based on the second weight coefficient determined in step B above and the input of the spatial attention extraction layer, namely the channel attention image. For example, the target feature image can be determined by multiplying the second weight coefficient with the channel attention image; for example, Ms×F' yields the target feature image.

[0084] S203. The target feature image is input to the upsampling layer and subjected to image magnification processing through the decoding process to obtain the marker image.

[0085] In this step, the target feature image determined in step C above is input into the upsampling layer and filled through the decoding process. The target feature image is then enlarged to restore it to the size of the original input image of the image extraction model, i.e., the scanned image, to obtain the marker image.

[0086] S103. The length of the marker is obtained by dividing and calculating the image of the marker, and the length of the marker is determined as the creepage distance of the insulator to be measured.

[0087] This step, which involves dividing and calculating the marker image to obtain the marker length and then determining the marker length as the creepage distance, includes:

[0088] S1031. Cut the marker image through a cutting surface in a preset direction to obtain multiple segments of marker point cloud data between each pair of adjacent cutting surfaces;

[0089] In this step, for the extracted insulator marker point cloud, i.e. the marker image obtained in step S102, the normal direction of the plane where the insulator is located is taken as the normal direction of the marker. Two parallel planes that are coplanar with the normal and have a fixed distance between them are randomly selected as cutting planes to cut the marker image and obtain multiple segments of marker point cloud data between each two adjacent cutting planes.

[0090] S1032. For each segment of marker point cloud data, the segment of marker point cloud data after filtering is determined as a marker strip;

[0091] In this step, the point cloud data of all markers between every two cutting surfaces obtained in step S1031 are screened out, and points at a fixed distance from the normal to the center of the insulator are screened out. The point cloud data of the markers after screening is determined as the marker strip.

[0092] S1033. For each segment of the marker strip, project the segment of the marker strip onto the cutting surface, perform noise reduction, sparsity and interpolation processing, and determine the length of the segment of the marker strip.

[0093] In this step, for the extracted marker strip, it is projected onto the cutting surface. The projected marker point cloud data is then downsampled, including denoising and sparsification. Linear splines are used to interpolate the downsampled marker point cloud data to calculate the length of the marker strip. Gaussian smoothing can be used to denoise the projected marker point cloud data, with the specific formula as follows:

[0094]

[0095] in, The image is Gaussian smoothed and denoised, z(x) is a strip of marker image projected onto the cutting plane, "*" indicates convolution, and σ is the standard deviation.

[0096] S1034. The length of the marker is obtained by adding the lengths of the determined multiple marker strips, and the length of the marker is determined as the creepage distance.

[0097] In this step, the lengths of the multiple marker strips determined in step S1033 are added together to obtain the total length of the markers, and the total length of the markers is determined as the creepage distance; the formula for calculating the total length of the markers is as follows:

[0098]

[0099] Where L is the total length of the marker, i.e. the creepage distance of the insulator, li represents the length of the i-th marker strip, and N represents the number of segments in the marker image.

[0100] This application provides a method for measuring the creepage distance of an insulator. The method includes: acquiring a scanned image of the insulator to be measured, wherein a marker is attached to the surface of the insulator, and the marker extends from the center of the insulator along the surface to the edge; inputting the scanned image into a pre-trained image extraction model and outputting a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker; dividing and calculating the marker image to obtain the length of the marker, and determining the length of the marker as the creepage distance of the insulator to be measured.

[0101] Thus, by adopting the technical solution provided in this application, a scanned image of the insulator can be obtained through a three-dimensional scanning device, the marker attached to the insulator can be extracted using a pre-trained image extraction model, the length of the marker can be calculated, and the length of the marker can be determined as the creepage distance, thereby realizing non-contact insulator creepage distance measurement and improving the efficiency and accuracy of creepage distance measurement.

[0102] Based on the same application concept, this application also provides an insulator creepage distance measuring device corresponding to the insulator creepage distance measuring method provided in the above embodiment. Since the principle of the device in this application is similar to the insulator creepage distance measuring method in the above embodiment, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0103] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of an insulator creepage distance measuring device provided in an embodiment of this application, as shown below. Figure 3 As shown, the measuring device 310 includes:

[0104] The acquisition module 311 is used to acquire a scanned image of the insulator to be measured; wherein, the surface of the insulator is covered with a marker, which extends from the center of the insulator along the surface of the insulator to the edge;

[0105] Processing module 312 is used to input the scanned image into a pre-trained image extraction model and output a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker;

[0106] The determination module 313 is used to obtain the length of the marker after dividing and calculating the marker image, and to determine the length of the marker as the creepage distance of the insulator to be measured.

[0107] Optionally, the image extraction model includes a downsampling layer, an attention extraction layer, and an upsampling layer, which are connected sequentially. When the processing module 312 inputs the scanned image into the pre-trained image extraction model and outputs a marker image, the processing module 312 is specifically used for:

[0108] The scanned image is input into a downsampling layer and undergoes an encoding process to reduce the image size, thereby obtaining a reduced feature image.

[0109] The reduced feature image is input into the attention extraction layer, and feature extraction is performed based on the feature information of the marker to obtain the target feature image;

[0110] The target feature image is input into the upsampling layer and then processed through a decoding process to enlarge the image, thereby obtaining the marker image.

[0111] Optionally, the attention extraction layer includes a channel attention extraction layer and a spatial attention extraction layer; when the processing module 312 inputs the reduced feature image into the attention extraction layer and performs feature extraction based on the feature information of the marker to obtain the target feature image, the processing module 312 is specifically used for:

[0112] The reduced feature image is input into the channel attention extraction layer, and feature extraction is performed based on the correlation between feature information in the channels to obtain the channel attention image;

[0113] The channel attention image is input into the spatial attention extraction layer, and feature extraction is performed based on the spatial relationship of feature information to obtain the target feature image.

[0114] Optionally, when the processing module 312 inputs the reduced feature image into the channel attention extraction layer and performs feature extraction based on the correlation between feature information in the channels to obtain the channel attention image, the processing module 312 is specifically used for:

[0115] After performing max pooling and average pooling on the reduced feature image, two first descriptive feature images are obtained;

[0116] Based on the correlation between the feature information in the two first descriptive feature maps, feature extraction is performed to obtain two feature images;

[0117] After image processing of the two feature images, the first weight coefficient is determined based on a preset activation function;

[0118] Based on the first weighting coefficient and the reduced feature image, a channel attention image is obtained.

[0119] Optionally, when the processing module 312 inputs the channel attention image into the spatial attention extraction layer and performs feature extraction based on the spatial relationship of feature information to obtain the target feature image, the processing module 312 is specifically used for:

[0120] After performing max pooling and average pooling on the channel attention image, two second descriptive feature images are obtained;

[0121] The two second descriptive feature images are spliced ​​together according to the spatial relationship of the feature information, and then processed by convolution and activation function to determine the second weight coefficients;

[0122] The target feature image is obtained based on the second weighting coefficient and the channel attention image.

[0123] Optionally, when the determining module 313 is used to obtain the length of the marker after dividing and calculating the marker image, and to determine the length of the marker as the creepage distance, the determining module 313 is specifically used for:

[0124] The marker image is cut through a cutting surface in a preset direction to obtain multiple segments of marker point cloud data between each two adjacent cutting surfaces;

[0125] For each segment of marker point cloud data, the segment of marker point cloud data after filtering is determined as a marker strip;

[0126] For each marker strip, the marker strip is projected onto the cutting surface, and noise reduction, sparsity reduction, and interpolation processing are performed to determine the length of the marker strip.

[0127] The length of the marker is obtained by adding the lengths of the determined multiple marker strips, and the length of the marker is determined as the creepage distance.

[0128] This application provides an insulator creepage distance measuring device, comprising: an acquisition module for acquiring a scanned image of an insulator to be measured, wherein a marker is attached to the surface of the insulator, the marker extending from the center of the insulator along the surface to the edge; a processing module for inputting the scanned image into a pre-trained image extraction model and outputting a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker; and a determination module for dividing and calculating the marker image to obtain the length of the marker, and determining the length of the marker as the creepage distance of the insulator to be measured.

[0129] Thus, by adopting the technical solution provided in this application, a scanned image of the insulator can be obtained through a three-dimensional scanning device, the marker attached to the insulator can be extracted using a pre-trained image extraction model, the length of the marker can be calculated, and the length of the marker can be determined as the creepage distance, thereby realizing non-contact insulator creepage distance measurement and improving the efficiency and accuracy of creepage distance measurement.

[0130] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0131] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 as well as Figure 2 The steps of the insulator creepage distance measurement method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0132] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 as well as Figure 2 The steps of the insulator creepage distance measurement method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0133] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0134] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0135] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0136] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0137] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0138] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for measuring the creepage distance of an insulator, characterized in that, The measurement method includes: A scanned image of the insulator to be measured is acquired; wherein, a marker is attached to the surface of the insulator, and the marker extends from the center of the insulator along the surface of the insulator to the edge; The scanned image is input into a pre-trained image extraction model, which outputs a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker. The length of the marker is obtained by dividing and calculating the image of the marker, and the length of the marker is determined as the creepage distance of the insulator to be measured. The step of dividing and calculating the length of the marker image to obtain the marker length, and determining the marker length as the creepage distance, includes: The marker image is cut through a cutting surface in a preset direction to obtain multiple segments of marker point cloud data between each two adjacent cutting surfaces; wherein, the normal direction of the plane where the insulator is located is taken as the normal direction of the marker, and two parallel planes that are coplanar with the normal direction of the marker and have a fixed distance between them are arbitrarily determined as the cutting surfaces; For each segment of marker point cloud data, the segment of marker point cloud data after filtering is determined as a marker strip; wherein, the segment of marker point cloud data after filtering is the marker point cloud data that has been filtered out from a fixed distance from the normal to the center of the insulator. For each marker strip, the marker strip is projected onto the cutting surface, and noise reduction, sparsity reduction, and interpolation processing are performed to determine the length of the marker strip. The length of the marker is obtained by adding the lengths of the determined multiple marker strips, and the length of the marker is determined as the creepage distance.

2. The measurement method according to claim 1, characterized in that, The image extraction model includes a downsampling layer, an attention extraction layer, and an upsampling layer, which are connected sequentially. The step of inputting the scanned image into a pre-trained image extraction model and outputting a marker image includes: The scanned image is input into a downsampling layer and undergoes an encoding process to reduce the image size, thereby obtaining a reduced feature image. The reduced feature image is input into the attention extraction layer, and feature extraction is performed based on the feature information of the marker to obtain the target feature image; The target feature image is input into the upsampling layer and then subjected to image magnification processing through a decoding process to obtain the marker image.

3. The measurement method according to claim 2, characterized in that, The attention extraction layer includes a channel attention extraction layer and a spatial attention extraction layer; the step of inputting the reduced feature image into the attention extraction layer, and extracting features based on the feature information of the markers to obtain the target feature image includes: The reduced feature image is input into the channel attention extraction layer, and feature extraction is performed based on the correlation between feature information in the channels to obtain the channel attention image; The channel attention image is input into the spatial attention extraction layer, and feature extraction is performed based on the spatial relationship of feature information to obtain the target feature image.

4. The measurement method according to claim 3, characterized in that, The step of inputting the reduced feature image into the channel attention extraction layer, and extracting features based on the correlation between feature information in the channels to obtain the channel attention image includes: After performing max pooling and average pooling on the reduced feature image, two first descriptive feature images are obtained; Based on the correlation between the feature information in the two first descriptive feature maps, feature extraction is performed to obtain two feature images; After image processing of the two feature images, the first weight coefficient is determined based on a preset activation function; Based on the first weighting coefficient and the reduced feature image, a channel attention image is obtained.

5. The measurement method according to claim 3, characterized in that, The step of inputting the channel attention image into the spatial attention extraction layer, and extracting features based on the spatial relationship of feature information to obtain the target feature image includes: After performing max pooling and average pooling on the channel attention image, two second descriptive feature images are obtained; The two second descriptive feature images are spliced ​​together according to the spatial relationship of the feature information, and then processed by convolution and activation function to determine the second weight coefficients; The target feature image is obtained based on the second weighting coefficient and the channel attention image.

6. A measuring device for insulator creepage distance, characterized in that, The measuring device includes: An acquisition module is used to acquire a scanned image of the insulator to be measured; wherein, a marker is attached to the surface of the insulator, and the marker covers the surface of the insulator from the center to the edge; The processing module is used to input the scanned image into a pre-trained image extraction model and output a marker image; wherein the marker image is extracted from the scanned image based on the feature information of the marker. The determination module is used to obtain the length of the marker after dividing and calculating the marker image, and to determine the length of the marker as the creepage distance of the insulator to be measured; When the determining module obtains the length of the marker after segmenting and calculating the marker image, and determines the length of the marker as the creepage distance, the determining module is specifically used for: The marker image is cut through a cutting surface in a preset direction to obtain multiple segments of marker point cloud data between each two adjacent cutting surfaces; wherein, the normal direction of the plane where the insulator is located is taken as the normal direction of the marker, and two parallel planes that are coplanar with the normal direction of the marker and have a fixed distance between them are arbitrarily determined as the cutting surfaces; For each segment of marker point cloud data, the segment of marker point cloud data after filtering is determined as a marker strip; wherein, the segment of marker point cloud data after filtering is the marker point cloud data that has been filtered out from a fixed distance from the normal to the center of the insulator. For each marker strip, the marker strip is projected onto the cutting surface, and noise reduction, sparsity reduction, and interpolation processing are performed to determine the length of the marker strip. The length of the marker is obtained by adding the lengths of the determined multiple marker strips, and the length of the marker is determined as the creepage distance.

7. The measuring device according to claim 6, characterized in that, The image extraction model includes a downsampling layer, an attention extraction layer, and an upsampling layer, which are sequentially connected. When the processing module inputs the scanned image into the pre-trained image extraction model and outputs a marker image, the processing module is specifically used for: The scanned image is input into a downsampling layer and undergoes an encoding process to reduce the image size, thereby obtaining a reduced feature image. The reduced feature image is input into the attention extraction layer, and feature extraction is performed based on the feature information of the marker to obtain the target feature image; The target feature image is input into the upsampling layer and then subjected to image magnification processing through a decoding process to obtain the marker image.

8. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the insulator creepage distance measurement method as described in any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the insulator creepage distance measurement method as described in any one of claims 1 to 5.

Images