Infrared-ultraviolet dual-mode partial discharge detection method based on lightweight network

Abstract

The invention relates to an infrared-ultraviolet dual-mode partial discharge detection method based on a lightweight network. The method comprises the following steps: acquiring a discharge area in an infrared image corresponding to a to-be-detected area, and acquiring a discharge area in an ultraviolet image corresponding to the to-be-detected area; inputting the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain partial discharge probability information of the infrared image and partial discharge probability information of the ultraviolet image; determining a partial discharge detection result of the to-be-detected area according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; the partial discharge detection result represents whether partial discharge exists in the to-be-detected area. By adopting the method, whether partial discharge exists in the to-be-detected area can be analyzed by using the lightweight model, the discharge detection process is simplified, the discharge detection time is shortened, and the detection efficiency of partial discharge detection is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power equipment detection, and particularly to an infrared-ultraviolet dual-mode partial discharge detection method, device, computer device, computer-readable storage medium, and computer program product based on a lightweight network. Background Art

[0002] In the insulation system of power equipment, due to uneven electric field distribution or insulation material defects, a small-scale, non-complete breakdown discharge phenomenon occurs in local areas. Partial discharge usually occurs in high-voltage equipment such as transformers, cables, switch cabinets, and generators. Partial discharge may cause damage to the insulation of power equipment and affect the safe and stable operation of the power system.

[0003] When traditional techniques are used for partial discharge detection, complex analysis of power equipment images is required, which is cumbersome to operate and consumes a large amount of analysis time, and is not conducive to improving the detection efficiency of partial discharge detection. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide an infrared-ultraviolet dual-mode partial discharge detection method, device, computer device, computer-readable storage medium, and computer program product based on a lightweight network that can improve the detection efficiency of partial discharge detection.

[0005] In a first aspect, the present application provides an infrared-ultraviolet dual-mode partial discharge detection method based on a lightweight network, including:

[0006] Obtaining a discharge area in an infrared image corresponding to a detection area to be detected, and obtaining a discharge area in a ultraviolet image corresponding to the detection area to be detected;

[0007] Inputting the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain local discharge probability information of the infrared image and local discharge probability information of the ultraviolet image;

[0008] Determining a partial discharge detection result of the detection area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image; the partial discharge detection result indicates whether there is partial discharge in the detection area to be detected.

[0009] In one embodiment, the obtaining a discharge area in an infrared image corresponding to a detection area to be detected includes:

[0010] Obtaining the temperature of each pixel point in the infrared image;

[0011] For each pixel point in the infrared image, when the temperature corresponding to the pixel point is greater than a preset temperature threshold, use the pixel point as a candidate pixel point;

[0012] Based on the candidate pixel points, determine the discharge area in the infrared image corresponding to the area to be detected.

[0013] In one embodiment, obtaining the discharge area in the ultraviolet image corresponding to the area to be detected includes:

[0014] Obtain the number of photons corresponding to each pixel point in the ultraviolet image;

[0015] For each pixel point in the ultraviolet image, when the number of photons corresponding to the pixel point is greater than a preset photon number threshold, use the pixel point as a candidate pixel point;

[0016] Based on the candidate pixel points, determine the discharge area in the ultraviolet image corresponding to the area to be detected.

[0017] In one embodiment, determining the partial discharge detection result of the area to be detected based on the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image includes:

[0018] Based on the pixel values of the pixel points corresponding to the discharge area in the infrared image, determine a first pixel threshold;

[0019] Based on the first pixel threshold, screen out first target pixel points from the pixel points corresponding to the discharge area in the infrared image;

[0020] Based on the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image, determine a second pixel threshold;

[0021] Based on the second pixel threshold, screen out second target pixel points from the pixel points corresponding to the discharge area in the ultraviolet image;

[0022] Based on the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image, determine the partial discharge detection result of the area to be detected.

[0023] In one embodiment, determining the partial discharge detection result of the area to be detected based on the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image includes:

[0024] Take the product of the number of the first target pixel points and the partial discharge probability information of the infrared image as the first product;

[0025] Take the product of the number of the second target pixel points and the partial discharge probability information of the ultraviolet image as the second product;

[0026] Determine a first sum according to the sum of the first product and the second product;

[0027] Determine a second sum according to the sum of the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image;

[0028] Determine the partial discharge detection result of the area to be detected according to the ratio between the first sum and the second sum.

[0029] In one embodiment, the determining the first sum according to the sum of the first product and the second product includes:

[0030] Obtain the weight of the infrared image and the weight of the ultraviolet image;

[0031] Determine the first sum according to the sum of the product of the weight of the infrared image and the first product and the product of the weight of the ultraviolet image and the second product.

[0032] In one embodiment, the method further includes:

[0033] Perform gray-scale processing on the initial infrared image corresponding to the area to be detected to obtain a processed infrared image, and perform gray-scale processing on the initial ultraviolet image corresponding to the area to be detected to obtain a processed ultraviolet image;

[0034] Perform denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected;

[0035] Perform denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected.

[0036] In one embodiment, the method further includes:

[0037] Input a sample infrared image and a sample ultraviolet image into a lightweight image discharge detection model to be trained, and obtain the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image;

[0038] Train the lightweight image discharge detection model to be trained according to the differences between the partial discharge information and the partial discharge probability information of the sample infrared images, and the differences between the partial discharge information and the partial discharge probability information of the sample ultraviolet images.

[0039] In a second aspect, the present application further provides an infrared-ultraviolet dual-modal partial discharge detection device based on a lightweight network, including:

[0040] A determination module, configured to obtain the discharge area in the infrared image corresponding to the area to be detected, and obtain the discharge area in the ultraviolet image corresponding to the area to be detected;

[0041] A prediction module, configured to input the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image;

[0042] A detection module, configured to determine the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image; the partial discharge detection result indicates whether there is partial discharge in the area to be detected.

[0043] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0044] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0045] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0046] The above infrared-ultraviolet dual-modal partial discharge detection method, device, computer device, computer-readable storage medium, and computer program product based on a lightweight network obtain the discharge area in the infrared image corresponding to the area to be detected and the discharge area in the ultraviolet image corresponding to the area to be detected, so as to accurately analyze the discharge positions in the infrared image and the ultraviolet image; input the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image, so as to accurately analyze the probability of partial discharge in the infrared image and the ultraviolet image by using the lightweight image discharge detection model; determine the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image; the partial discharge detection result indicates whether there is partial discharge in the area to be detected. Therefore, by combining the discharge area in the infrared image, the partial discharge probability of the infrared image, the discharge area in the ultraviolet image, and the partial discharge probability of the ultraviolet image, it can accurately analyze whether there is partial discharge in the area to be detected, can analyze the probability of partial discharge in the image by using a lightweight model, and combine the discharge area in the image to determine whether there is partial discharge in the area to be detected, without complex analysis of the power equipment image, but simplify the discharge detection process by combining a lightweight model, thereby shortening the discharge detection analysis time and improving the detection efficiency of partial discharge detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0048] Figure 1 It is an application environment diagram of an infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network in an embodiment;

[0049] Figure 2 It is a flowchart of an infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network in an embodiment;

[0050] Figure 3 It is a flowchart of an infrared-ultraviolet dual-modal adaptive fusion partial discharge detection based on a lightweight network in an embodiment;

[0051] Figure 4 It is a structural block diagram of an infrared-ultraviolet dual-modal partial discharge detection device based on a lightweight network in an embodiment;

[0052] Figure 5 It is the internal structure diagram of a computer device in an embodiment. Specific implementation manners

[0053] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] The infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network provided by an embodiment of the present application can be applied to, for example Figure 1 the application environment shown in the figure. Among them, the terminal 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed in the cloud or on other network servers. The terminal 102 acquires the discharge area in the infrared image corresponding to the area to be detected, and acquires the discharge area in the ultraviolet image corresponding to the area to be detected; the terminal 102 inputs the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; the terminal 102 determines the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image; the partial discharge detection result indicates whether there is partial discharge in the area to be detected. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The server 104 can be an independent physical server, can also be a server cluster or a distributed system composed of multiple physical servers, or can also be a cloud server providing cloud computing services.

[0055] In an exemplary embodiment, as Figure 2 shown in the figure, a method for infrared-ultraviolet dual-modal partial discharge detection based on a lightweight network is provided. Taking the application of this method to a terminal as an example, it includes the following steps S202 to step S206. Among them:

[0056] Step S202, acquire the discharge area in the infrared image corresponding to the area to be detected, and acquire the discharge area in the ultraviolet image corresponding to the area to be detected.

[0057] Among them, the area to be detected may refer to a part / all area of any power equipment in the power system.

[0058] Among them, the infrared image may refer to an image obtained after preprocessing an image formed by an infrared remote sensor receiving infrared rays reflected by an object (such as a power device).

[0059] Among them, the ultraviolet image may refer to an image obtained after preprocessing an image obtained by using an ultraviolet imaging detection device to image and collect a power device.

[0060] Among them, the discharge area may refer to an area where partial discharge exists in the area to be detected.

[0061] As an example, the terminal can analyze each pixel point in the infrared image corresponding to the area to be detected to determine the discharge area in the infrared image corresponding to the area to be detected; the terminal can also analyze each pixel point in the ultraviolet image corresponding to the area to be detected to determine the discharge area in the ultraviolet image corresponding to the area to be detected.

[0062] Step S204, input the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image.

[0063] Among them, the lightweight image discharge detection model may refer to a model used to determine the probability of the existence of partial discharge in an image. In practical applications, the lightweight image discharge detection model (such as the FalconNet model) may include a siamese neural network and two independent fully connected layers.

[0064] Among them, the partial discharge probability information may refer to the information output by the lightweight image discharge detection model and used to characterize the probability of the existence of partial discharge in the image.

[0065] As an example, the terminal can input the infrared image and the ultraviolet image corresponding to the area to be detected into a pre-trained lightweight image discharge detection model. Since the lightweight image discharge detection model includes a siamese neural network and two independent fully connected layers, the output result of the siamese neural network after processing the infrared image corresponding to the area to be detected is input into the first fully connected layer in the lightweight image discharge detection model, and the result output by the first fully connected layer can be used as the partial discharge probability information of the infrared image. The output result of the siamese neural network after processing the ultraviolet image corresponding to the area to be detected is input into the second fully connected layer in the lightweight image discharge detection model, and the result output by the second fully connected layer can be used as the partial discharge probability information of the ultraviolet image. Among them, the first fully connected layer and the second fully connected layer are independent of each other.

[0066] Step S206, determine the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image.

[0067] Among them, the partial discharge detection result can refer to the information indicating whether there is partial discharge in the area to be detected.

[0068] As an example, the terminal can combine the characteristics of partial discharge in the infrared image and the characteristics of partial discharge in the ultraviolet image, and analyze whether there is partial discharge in the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information in the infrared image, and the partial discharge probability information in the ultraviolet image, so as to determine the partial discharge detection result of the area to be detected. In practical applications, the terminal can first calculate the discharge degree information used to represent the probability of partial discharge in the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information in the infrared image, and the partial discharge probability information in the ultraviolet image, and then combine the discharge degree information and a preset threshold to determine the partial discharge detection result of the area to be detected. For example: when the discharge degree information is greater than the preset threshold, the terminal can determine that the partial discharge detection result of the area to be detected is that there is partial discharge in the area to be detected; when the discharge degree information is less than or equal to the preset threshold, the terminal can determine that the partial discharge detection result of the area to be detected is that there is no partial discharge in the area to be detected.

[0069] In the above infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network, by obtaining the discharge area in the infrared image corresponding to the area to be detected and obtaining the discharge area in the ultraviolet image corresponding to the area to be detected, the discharge positions in the infrared image and the ultraviolet image can be accurately analyzed; the infrared image and the ultraviolet image are input into a pre-trained lightweight image discharge detection model to obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image, so as to accurately analyze the probability of partial discharge in the infrared image and the ultraviolet image by using the lightweight image discharge detection model; according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image, the partial discharge detection result of the area to be detected is determined; the partial discharge detection result indicates whether there is partial discharge in the area to be detected, so as to accurately analyze whether there is partial discharge in the area to be detected by combining the discharge area in the infrared image, the partial discharge probability in the infrared image, the discharge area in the ultraviolet image, and the partial discharge probability in the ultraviolet image, and it is possible to analyze the probability of partial discharge in the image by using the lightweight model and combine the discharge area in the image to determine whether there is partial discharge in the area to be detected, without complex analysis of the power equipment image, but simplifying the discharge detection process by combining the lightweight model, thereby shortening the discharge detection analysis time and further improving the detection efficiency of partial discharge detection.

[0070] In an exemplary embodiment, obtaining the discharge area in the infrared image corresponding to the area to be detected includes: obtaining the temperature of each pixel point in the infrared image; for each pixel point in the infrared image, when the temperature corresponding to the pixel point is greater than a preset temperature threshold, taking the pixel point as a candidate pixel point; and determining the discharge area in the infrared image corresponding to the area to be detected according to the candidate pixel points.

[0071] Wherein, the preset temperature threshold may include 1.5 times the room temperature, and the preset temperature threshold can be flexibly adjusted and set based on the actual situation.

[0072] As an example, the terminal can determine the temperature of each pixel point in the infrared image according to the infrared image corresponding to the area to be detected. For each pixel point in the infrared image, the terminal can analyze the magnitude relationship between the temperature corresponding to each pixel point and the preset temperature threshold. When the temperature corresponding to the pixel point is greater than the preset temperature threshold, the terminal can take the pixel point as a candidate pixel point, traverse each pixel point in the infrared image, and analyze whether the pixel point can be used as a candidate pixel point. Then, the terminal can determine the discharge area in the infrared image corresponding to the area to be detected according to each candidate pixel point. In practical applications, the distribution area of each candidate pixel point in the infrared image can be used as the discharge area in the infrared image corresponding to the area to be detected.

[0073] In this embodiment, by obtaining the temperature of each pixel point in the infrared image; for each pixel point in the infrared image, when the temperature corresponding to the pixel point is greater than the preset temperature threshold, taking the pixel point as a candidate pixel point; and determining the discharge area in the infrared image corresponding to the area to be detected according to the candidate pixel points, it is possible to analyze whether the pixel point belongs to the discharge area based on the temperature of the pixel point in the infrared image, so as to accurately analyze the discharge area in the infrared image corresponding to the area to be detected, improve the accuracy of the discharge area in the infrared image corresponding to the area to be detected, and facilitate subsequent local discharge detection in combination with the discharge area in the infrared image corresponding to the area to be detected, thereby improving the accuracy of local discharge detection.

[0074] In some embodiments, obtaining the discharge area in the ultraviolet image corresponding to the area to be detected includes: obtaining the number of photons corresponding to each pixel point in the ultraviolet image; for each pixel point in the ultraviolet image, when the number of photons corresponding to the pixel point is greater than a preset photon number threshold, taking the pixel point as a candidate pixel point; and determining the discharge area in the ultraviolet image corresponding to the area to be detected according to the candidate pixel points.

[0075] Wherein, the number of photons corresponding to the pixel point may refer to the number of ultraviolet photons received by the pixel point.

[0076] Among them, the preset photon number threshold may include a calibrated photon number. The calibrated photon number may refer to performing a partial discharge experiment on the current device before the inspection tour. By creating a known standard partial discharge and regarding the number of photons generated by this partial discharge on the device as the calibrated photon number.

[0077] As an example, the terminal can determine the number of photons corresponding to each pixel point in the ultraviolet image according to the ultraviolet image corresponding to the area to be detected. For each pixel point in the ultraviolet image, the terminal can analyze the magnitude relationship between the number of photons corresponding to each pixel point and the preset photon number threshold. When the number of photons corresponding to the pixel point is greater than the preset photon number threshold, the terminal can use this pixel point as a candidate pixel point, traverse each pixel point in the ultraviolet image, and analyze whether the pixel point can be used as a candidate pixel point. Then the terminal can determine the discharge area in the ultraviolet image corresponding to the area to be detected according to each candidate pixel point. In practical applications, the distribution area of each candidate pixel point in the ultraviolet image can be used as the discharge area in the ultraviolet image corresponding to the area to be detected.

[0078] In this embodiment, by obtaining the number of photons corresponding to each pixel point in the ultraviolet image; for each pixel point in the ultraviolet image, when the number of photons corresponding to the pixel point is greater than the preset photon number threshold, using the pixel point as a candidate pixel point; and determining the discharge area in the ultraviolet image corresponding to the area to be detected according to the candidate pixel points, it is possible to analyze whether a pixel point belongs to the discharge area based on the number of photons of the pixel point in the ultraviolet image, thereby accurately analyzing the discharge area in the ultraviolet image corresponding to the area to be detected, improving the accuracy of the discharge area in the ultraviolet image corresponding to the area to be detected, so as to subsequently perform partial discharge detection in combination with the discharge area in the ultraviolet image corresponding to the area to be detected, and further improving the accuracy of partial discharge detection.

[0079] In some embodiments, determining the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image includes: determining a first pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the infrared image; screening out first target pixel points from the pixel points corresponding to the discharge area in the infrared image according to the first pixel threshold; determining a second pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image; screening out second target pixel points from the pixel points corresponding to the discharge area in the ultraviolet image according to the second pixel threshold; and determining the partial discharge detection result of the area to be detected according to the number of first target pixel points, the number of second target pixel points, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image.

[0080] Among them, the first pixel threshold may include a preset ratio (such as 20%) of the maximum pixel value among the pixel values of the pixel points corresponding to the discharge area in the infrared image.

[0081] Among them, the first target pixel point may refer to the pixel points among the pixel points corresponding to the discharge area in the infrared image whose pixel values are greater than the first pixel threshold.

[0082] Among them, the second pixel threshold may include a preset ratio (such as 20%) of the maximum pixel value among the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image.

[0083] Among them, the second target pixel point may refer to the pixel points among the pixel points corresponding to the discharge area in the ultraviolet image whose pixel values are greater than the second pixel threshold.

[0084] As an example, the terminal may first determine the maximum pixel value among the pixel values of the pixel points corresponding to the discharge area in the infrared image according to the pixel values of the pixel points corresponding to the discharge area in the infrared image. Then, the terminal may determine the first pixel threshold e1 according to the maximum pixel value. Then, the terminal may screen out the pixel points whose pixel values are greater than the first pixel threshold e1 from the pixel points corresponding to the discharge area in the infrared image as the first target pixel points. The terminal may also first determine the maximum pixel value among the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image according to the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image. Then, the terminal may determine the second pixel threshold e2 according to the maximum pixel value. Then, the terminal may screen out the pixel points whose pixel values are greater than the second pixel threshold e2 from the pixel points corresponding to the discharge area in the ultraviolet image as the second target pixel points. Then, the terminal may determine the discharge degree information for characterizing the probability of partial discharge in the area to be detected according to the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image. Then, in combination with the discharge degree information and a preset threshold, the terminal may determine the partial discharge detection result of the area to be detected.

[0085] In this embodiment, by determining the first pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the infrared image; screening out the first target pixel points from the pixel points corresponding to the discharge area in the infrared image according to the first pixel threshold; determining the second pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image; screening out the second target pixel points from the pixel points corresponding to the discharge area in the ultraviolet image according to the second pixel threshold; according to the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image, it is possible to first screen out the target pixel points whose pixel values meet the preset requirements from the infrared image and the ultraviolet image respectively, and combine the number of the target pixel points and the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image output by the lightweight model to accurately generate the partial discharge detection result of the area to be detected, without performing complex analysis on the power equipment image, but simplifying the discharge detection process by combining the lightweight model, shortening the discharge detection analysis time, and improving the detection efficiency of partial discharge detection.

[0086] In some embodiments, determining the partial discharge detection result of the area to be detected according to the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image includes: taking the product of the number of the first target pixel points and the partial discharge probability information of the infrared image as the first product; taking the product of the number of the second target pixel points and the partial discharge probability information of the ultraviolet image as the second product; determining the first sum according to the sum of the first product and the second product; determining the second sum according to the sum of the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; determining the partial discharge detection result of the area to be detected according to the ratio between the first sum and the second sum.

[0087] As an example, the terminal can calculate the product between the number S1 of first target pixels and the partial discharge probability information P1 of the infrared image. At this time, the first product can be expressed as S1 * P1; the terminal can calculate the product between the number S2 of second target pixels and the partial discharge probability information P2 of the ultraviolet image. At this time, the second product can be expressed as S2 * P2; determine the first sum according to the sum of the first product and the second product. At this time, the first sum can be expressed as S1 * P1 + S2 * P2; the terminal can calculate the sum of the partial discharge probability information P1 of the infrared image and the partial discharge probability information P2 of the ultraviolet image. At this time, the second sum can be expressed as P1 + P2; the terminal can calculate the discharge degree information for characterizing the probability of partial discharge in the area to be detected according to the ratio (S1 * P1 + S2 * P2) / (P1 + P2) between the first sum and the second sum. Then, in combination with the discharge degree information and a preset threshold, determine the partial discharge detection result of the area to be detected. In practical applications, the number S1 of first target pixels and the number S2 of second target pixels can be used as information for adjusting the confidence of the partial discharge probability information output by the lightweight image discharge detection model.

[0088] In this embodiment, by taking the product between the number of first target pixels and the partial discharge probability information of the infrared image as the first product; taking the product between the number of second target pixels and the partial discharge probability information of the ultraviolet image as the second product; determining the first sum according to the sum of the first product and the second product; determining the second sum according to the sum of the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; and determining the partial discharge detection result of the area to be detected according to the ratio between the first sum and the second sum, it is possible to combine the number of first target pixels, the number of second target pixels, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image to generate an accurate partial discharge detection result and improve the accuracy of the partial discharge detection result.

[0089] In some embodiments, determining the first sum according to the sum of the first product and the second product includes: obtaining the weight of the infrared image and the weight of the ultraviolet image; determining the first sum according to the sum of the product of the weight of the infrared image and the first product and the product of the weight of the ultraviolet image and the second product.

[0090] Among them, the weight of the infrared image can be information indicating the importance degree of the first product in the first sum. The weight of the ultraviolet image can be information indicating the importance degree of the second product in the first sum. The sum of the weight C1 of the infrared image and the weight C2 of the ultraviolet image can be 1. In practical applications, when the partial discharge detection is performed on the area to be detected during the day or the ambient temperature of the power equipment is higher than the preset ambient temperature threshold, C1 < C2; when the partial discharge detection is performed on the area to be detected at night or the ambient temperature of the power equipment is lower than the preset ambient temperature threshold, C1 > C2.

[0091] As an example, when calculating the first sum, the terminal can also obtain the weight of the infrared image and the weight of the ultraviolet image, and then the terminal can perform weighted summation on the first product and the second product according to the weight of the infrared image and the weight of the ultraviolet image to obtain the first sum. For example: the first product can be expressed as S1 * P1, the second product can be expressed as S2 * P2, the weight of the infrared image can be expressed as C1, and the weight of the ultraviolet image can be expressed as C2. At this time, the first sum can be expressed as: C1 * S1 * P1 + C2 * S2 * P2.

[0092] In this embodiment, by obtaining the weight of the infrared image and the weight of the ultraviolet image; determining the first sum according to the sum of the product between the weight of the infrared image and the first product and the product between the weight of the ultraviolet image and the second product, it is possible to accurately calculate the first sum by using the corresponding weights in different situations in combination with the detection time of partial discharge / ambient temperature of the equipment, providing a data basis for determining the partial discharge detection result subsequently, thereby improving the accuracy of the partial discharge detection result.

[0093] In some embodiments, the above method further includes: performing gray processing on the initial infrared image corresponding to the area to be detected to obtain the processed infrared image, and performing gray processing on the initial ultraviolet image corresponding to the area to be detected to obtain the processed ultraviolet image; performing denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected; performing denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected.

[0094] Among them, the initial infrared image can be an image formed by an infrared remote sensor receiving infrared rays reflected by an object (such as a power equipment).

[0095] Among them, the processed infrared image can be an image obtained by performing gray processing on the initial infrared image.

[0096] Among them, the initial ultraviolet image can be an image obtained by imaging and collecting a power equipment using an ultraviolet imaging detection device.

[0097] Among them, the processed ultraviolet image may refer to the image obtained after performing gray-scale processing on the initial ultraviolet image.

[0098] Among them, the histogram mean information of the processed infrared image may refer to the average pixel value of the processed infrared image calculated based on the statistical chart of the number of occurrences of each gray level in the processed infrared image.

[0099] Among them, the histogram mean information of the processed ultraviolet image may refer to the average pixel value of the processed ultraviolet image calculated based on the statistical chart of the number of occurrences of each gray level in the processed ultraviolet image.

[0100] As an example, before performing partial discharge detection based on the initial infrared image and the initial ultraviolet image of the area to be detected, the terminal can preprocess the initial infrared image and the initial ultraviolet image. For example: the terminal can perform gray-scale processing on the initial infrared image corresponding to the area to be detected to obtain the processed infrared image, and the terminal can also perform gray-scale processing on the initial ultraviolet image corresponding to the area to be detected to obtain the processed ultraviolet image. Then, the terminal can perform denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected, and perform denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected.

[0101] In this embodiment, by performing gray-scale processing on the initial infrared image corresponding to the area to be detected to obtain the processed infrared image, and performing gray-scale processing on the initial ultraviolet image corresponding to the area to be detected to obtain the processed ultraviolet image; performing denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected; performing denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected, it is possible to perform preprocessing such as gray-scale processing and denoising processing on the infrared image and the ultraviolet image, avoid the influence of noise on subsequent partial discharge detection, and thus improve the accuracy of the partial discharge detection result.

[0102] In some embodiments, the above method further includes: inputting the sample infrared image and the sample ultraviolet image into the lightweight image discharge detection model to be trained to obtain the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image; training the lightweight image discharge detection model to be trained according to the difference between the partial discharge information of the sample infrared image and the partial discharge probability information of the sample infrared image, and the difference between the partial discharge information of the sample ultraviolet image and the partial discharge probability information of the sample ultraviolet image.

[0103] Among them, whether there is partial discharge in the sample infrared image (i.e., the partial discharge information of the sample infrared image) is known information, and whether there is partial discharge in the sample ultraviolet image (i.e., the partial discharge information of the sample ultraviolet image) is known information.

[0104] Among them, the sample infrared image and the sample ultraviolet image can be obtained through pre-calibration settings in the laboratory. The partial discharge probability information of the sample infrared image can refer to the information representing the probability of partial discharge in the sample infrared image, and the partial discharge probability information of the sample ultraviolet image can refer to the information representing the probability of partial discharge in the sample ultraviolet image.

[0105] As an example, in order to obtain a pre-trained lightweight image discharge detection model, the terminal can input the sample infrared image and the sample ultraviolet image into the lightweight image discharge detection model to be trained, and obtain the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image. Then, the terminal can train the lightweight image discharge detection model to be trained according to the difference between the partial discharge information of the sample infrared image and the partial discharge probability information of the sample infrared image, and the difference between the partial discharge information of the sample ultraviolet image and the partial discharge probability information of the sample ultraviolet image. In practical applications, when training the lightweight image discharge detection model to be trained, the structures in the siamese neural network of the lightweight image discharge detection model that process the infrared image and the ultraviolet image share the same set of weights and biases. The terminal can determine the loss function value of the lightweight image discharge detection model according to the difference between the partial discharge information of the sample infrared image and the partial discharge probability information of the sample infrared image. If the loss function value of the lightweight image discharge detection model does not meet the preset requirements, the model parameters of the lightweight image discharge detection model are adjusted, and the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image are re-determined, and the new loss function value is calculated until the loss function value meets the preset requirements. At this time, the terminal can determine that the training of the lightweight image discharge detection model is completed.

[0106] In this embodiment, by inputting the sample infrared image and the sample ultraviolet image into the lightweight image discharge detection model to be trained, the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image are obtained; according to the difference between the partial discharge information of the sample infrared image and the partial discharge probability information of the sample infrared image, and the difference between the partial discharge information of the sample ultraviolet image and the partial discharge probability information of the sample ultraviolet image, the lightweight image discharge detection model to be trained is trained, so that the lightweight image discharge detection model can be trained using training samples, the model performance of the lightweight image discharge detection model can be optimized, and the accuracy of the partial discharge probability information output by the lightweight image discharge detection model can be improved, so as to determine the partial discharge detection result in combination with the partial discharge probability information subsequently, thereby improving the accuracy of the partial discharge detection result.

[0107] In some embodiments, as Figure 3 shown, a schematic flowchart of infrared-ultraviolet dual-modal adaptive fusion partial discharge detection based on a lightweight network is provided. The terminal can perform gray-scale processing on the infrared image and the ultraviolet image collected for the area to be detected to obtain the gray-scale processed infrared image and the gray-scale processed ultraviolet image, and then perform denoising processing on the gray-scale processed images based on the histogram mean. For example, the terminal can perform denoising processing on the gray-scale processed infrared image based on the histogram mean information of the gray-scale processed infrared image to obtain the infrared image processed based on the histogram mean; the terminal can perform denoising processing on the gray-scale processed ultraviolet image based on the histogram mean information of the gray-scale processed ultraviolet image to obtain the ultraviolet image processed based on the histogram mean. Then the terminal can input the infrared image processed based on the histogram mean and the ultraviolet image processed based on the histogram mean into the pre-trained lightweight image discharge detection model (such as the FalconNet model). The structure in the lightweight image discharge detection model for processing the infrared image can be used as the FalconNet for the infrared channel, and the structure in the lightweight image discharge detection model for processing the ultraviolet image can be used as the FalconNet for the ultraviolet channel. The FalconNet for the infrared channel and the FalconNet for the ultraviolet channel share parameters. The FalconNet for the infrared channel can analyze the infrared image processed based on the histogram mean and output the partial discharge probability information of the infrared image representing the probability P1 of judging partial discharge in the infrared channel. The FalconNet for the ultraviolet channel can analyze the ultraviolet image processed based on the histogram mean and output the partial discharge probability information of the ultraviolet image representing the probability P2 of judging partial discharge in the ultraviolet channel.

[0108] The terminal can determine the discharge area from the infrared image processed based on the histogram mean according to the temperature of each pixel point in the infrared image processed based on the histogram mean. The terminal can also determine the discharge area from the ultraviolet image processed based on the histogram mean according to the number of photons of each pixel point in the ultraviolet image processed based on the histogram mean. The terminal can further calculate the number S1 of pixel points with pixel values greater than the threshold e according to the pixel values of each pixel point in the infrared image processed based on the histogram mean, where the threshold e for determining S1 can be the maximum pixel value among the pixel values of the pixel points in the discharge area in the infrared image processed based on the histogram mean. The terminal can also calculate the number S2 of pixel points with pixel values greater than the threshold e according to the pixel values of each pixel point in the ultraviolet image processed based on the histogram mean, where the threshold e for determining S2 can be the maximum pixel value among the pixel values of the pixel points in the discharge area in the ultraviolet image processed based on the histogram mean.

[0109] Then the terminal can calculate the discharge degree of the area to be detected based on the partial discharge probability information P1, the partial discharge probability information P2, the number S1 of pixel points, the number S2 of pixel points, the weight C1 of the infrared image, and the weight C2 of the ultraviolet image. The discharge degree can be expressed as: (C1 * S1 * P1 + C2 * S2 * P2) / (S1 + S2). The terminal can analyze whether there is partial discharge in the area to be detected according to the discharge degree of the area to be detected, so as to determine the partial discharge detection result of the area to be detected. In practical applications, if the discharge degree of the area to be detected is 80%, the terminal can combine the discharge degree threshold (such as 50%) to determine that there is partial discharge in the area to be detected at this time.

[0110] In this embodiment, the collected infrared image and ultraviolet image are subjected to gray-scale processing; then the gray-scale processed images are subjected to denoising processing based on the histogram mean to obtain infrared and ultraviolet discharge detection images containing the target positions of partial discharge; then the pixel areas of the target discharge positions in the denoised infrared and ultraviolet images are calculated respectively; finally, a dual-modal detection model with parameter sharing is designed based on the lightweight network FalconNet, and the output confidence of the model is adjusted by using the calculated pixel area to enhance the robustness of the algorithm, so as to judge the degree of discharge. Compared with the traditional detection model, it is more lightweight, and at the same time, by using the method of histogram mean denoising under different discharge intensities, the pixel areas of the discharge positions in the infrared and ultraviolet images are different, enabling the model to have the ability to recognize the degree of discharge and improving the efficiency and accuracy of partial discharge detection.

[0111] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0112] Based on the same inventive concept, an embodiment of the present application further provides a lightweight network-based infrared-ultraviolet dual-modal partial discharge detection device for implementing the above-mentioned lightweight network-based infrared-ultraviolet dual-modal partial discharge detection method. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the lightweight network-based infrared-ultraviolet dual-modal partial discharge detection device provided below can refer to the limitations on the lightweight network-based infrared-ultraviolet dual-modal partial discharge detection method in the above text, and will not be repeated here.

[0113] In an exemplary embodiment, as Figure 4 shown, a lightweight network-based infrared-ultraviolet dual-modal partial discharge detection device is provided, including: an acquisition module 402, a prediction module 404, and a detection module 406, where:

[0114] The acquisition module 402 is configured to acquire a discharge area in an infrared image corresponding to a to-be-detected area, and acquire a discharge area in a ultraviolet image corresponding to the to-be-detected area.

[0115] The prediction module 404 is configured to input the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain local discharge probability information of the infrared image and local discharge probability information of the ultraviolet image.

[0116] The detection module 406 is configured to determine a local discharge detection result of the to-be-detected area according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image; the local discharge detection result indicates whether there is local discharge in the to-be-detected area.

[0117] In one exemplary embodiment, the obtaining module 402 is further specifically configured to obtain the temperature of each pixel point in the infrared image; for each pixel point in the infrared image, when the temperature corresponding to the pixel point is greater than a preset temperature threshold, use the pixel point as a candidate pixel point; and determine a discharge area in the infrared image corresponding to the area to be detected according to the candidate pixel points.

[0118] In one exemplary embodiment, the obtaining module 402 is further specifically configured to obtain the number of photons corresponding to each pixel point in the ultraviolet image; for each pixel point in the ultraviolet image, when the number of photons corresponding to the pixel point is greater than a preset photon number threshold, use the pixel point as a candidate pixel point; and determine a discharge area in the ultraviolet image corresponding to the area to be detected according to the candidate pixel points.

[0119] In one exemplary embodiment, the detecting module 406 is further specifically configured to determine a first pixel threshold according to the pixel value of the pixel point corresponding to the discharge area in the infrared image; screen out first target pixel points from the pixel points corresponding to the discharge area in the infrared image according to the first pixel threshold; determine a second pixel threshold according to the pixel value of the pixel point corresponding to the discharge area in the ultraviolet image; screen out second target pixel points from the pixel points corresponding to the discharge area in the ultraviolet image according to the second pixel threshold; and determine a partial discharge detection result of the area to be detected according to the number of the first target pixel points, the number of the second target pixel points, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image.

[0120] In one exemplary embodiment, the detecting module 406 is further specifically configured to use the product of the number of the first target pixel points and the partial discharge probability information of the infrared image as a first product; use the product of the number of the second target pixel points and the partial discharge probability information of the ultraviolet image as a second product; determine a first sum according to the sum of the first product and the second product; determine a second sum according to the sum of the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; and determine a partial discharge detection result of the area to be detected according to the ratio between the first sum and the second sum.

[0121] In one exemplary embodiment, the detecting module 406 is further specifically configured to obtain the weight of the infrared image and the weight of the ultraviolet image; and determine the first sum according to the sum of the product of the weight of the infrared image and the first product and the product of the weight of the ultraviolet image and the second product.

[0122] In one exemplary embodiment, the device further includes a preprocessing module, which is specifically configured to perform grayscale processing on the initial infrared image corresponding to the area to be detected to obtain a processed infrared image, and perform grayscale processing on the initial ultraviolet image corresponding to the area to be detected to obtain a processed ultraviolet image; perform denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected; perform denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected.

[0123] In one exemplary embodiment, the device further includes a model training module, which is specifically configured to input a sample infrared image and a sample ultraviolet image into a lightweight image discharge detection model to be trained, and obtain the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image; train the lightweight image discharge detection model to be trained according to the difference between the partial discharge information and the partial discharge probability information of the sample infrared image, and the difference between the partial discharge information and the partial discharge probability information of the sample ultraviolet image.

[0124] Each module in the above infrared-ultraviolet dual-mode partial discharge detection device based on a lightweight network can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0125] In one exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 5As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0126] Those skilled in the art can understand that Figure 5 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0127] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0128] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0129] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0130] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0131] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., and are not limited thereto.

[0132] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0133] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application shall be subject to the appended claims.

Claims

1. An infrared-ultraviolet dual-modal partial discharge detection method based on a lightweight network, characterized in that The method includes: Obtaining a discharge area in an infrared image corresponding to a region to be detected, and obtaining a discharge area in an ultraviolet image corresponding to the region to be detected; Inputting the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model to obtain local discharge probability information of the infrared image and local discharge probability information of the ultraviolet image; Determining a local discharge detection result of the region to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image; the local discharge detection result indicates whether there is local discharge in the region to be detected.

2. The method according to claim 1, characterized in that, The obtaining of the discharge area in the infrared image corresponding to the region to be detected includes: Obtaining the temperature of each pixel point in the infrared image; For each pixel point in the infrared image, when the temperature corresponding to the pixel point is greater than a preset temperature threshold, taking the pixel point as a candidate pixel point; Determining the discharge area in the infrared image corresponding to the region to be detected according to the candidate pixel points.

3. The method according to claim 1, wherein The obtaining of the discharge area in the ultraviolet image corresponding to the region to be detected includes: Obtaining the number of photons corresponding to each pixel point in the ultraviolet image; For each pixel point in the ultraviolet image, when the number of photons corresponding to the pixel point is greater than a preset photon number threshold, taking the pixel point as a candidate pixel point; Determining the discharge area in the ultraviolet image corresponding to the region to be detected according to the candidate pixel points.

4. The method according to claim 1, characterized in that The determining of the local discharge detection result of the region to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image includes: Determining a first pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the infrared image; Screening out first target pixel points from the pixel points corresponding to the discharge area in the infrared image according to the first pixel threshold; Determining a second pixel threshold according to the pixel values of the pixel points corresponding to the discharge area in the ultraviolet image; Screening out second target pixel points from the pixel points corresponding to the discharge area in the ultraviolet image according to the second pixel threshold; Determining the local discharge detection result of the region to be detected according to the number of the first target pixel points, the number of the second target pixel points, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image.

5. The method according to claim 4, wherein The determining of the local discharge detection result of the region to be detected according to the number of the first target pixel points, the number of the second target pixel points, the local discharge probability information of the infrared image, and the local discharge probability information of the ultraviolet image includes: Taking the product between the number of the first target pixel points and the local discharge probability information of the infrared image as a first product; Taking the product between the number of the second target pixel points and the local discharge probability information of the ultraviolet image as a second product; Determining a first sum according to the sum of the first product and the second product; Determine a second sum based on the sum of the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; Determine the partial discharge detection result of the area to be detected according to the ratio between the first sum and the second sum.

6. The method according to claim 5, wherein The determining the first sum according to the sum of the first product and the second product includes: Obtain the weight of the infrared image and the weight of the ultraviolet image; Determine the first sum according to the sum of the product of the weight of the infrared image and the first product and the product of the weight of the ultraviolet image and the second product.

7. The method according to claim 1, characterized in that The method further includes: Perform gray-scale processing on the initial infrared image corresponding to the area to be detected to obtain a processed infrared image, and perform gray-scale processing on the initial ultraviolet image corresponding to the area to be detected to obtain a processed ultraviolet image; Perform denoising processing on the processed infrared image according to the histogram mean information of the processed infrared image to obtain the infrared image corresponding to the area to be detected; Perform denoising processing on the processed ultraviolet image according to the histogram mean information of the processed ultraviolet image to obtain the ultraviolet image corresponding to the area to be detected.

8. The method according to claim 1, wherein The method further includes: Input the sample infrared image and the sample ultraviolet image into the lightweight image discharge detection model to be trained, and obtain the partial discharge probability information of the sample infrared image and the partial discharge probability information of the sample ultraviolet image; Train the lightweight image discharge detection model to be trained according to the difference between the partial discharge information of the sample infrared image and the partial discharge probability information of the sample infrared image, and the difference between the partial discharge information of the sample ultraviolet image and the partial discharge probability information of the sample ultraviolet image.

9. An infrared-ultraviolet dual-modal partial discharge detection device based on a lightweight network, characterized in that, The device includes: An acquisition module, configured to acquire the discharge area in the infrared image corresponding to the area to be detected, and acquire the discharge area in the ultraviolet image corresponding to the area to be detected; A prediction module, configured to input the infrared image and the ultraviolet image into a pre-trained lightweight image discharge detection model, and obtain the partial discharge probability information of the infrared image and the partial discharge probability information of the ultraviolet image; A detection module, configured to determine the partial discharge detection result of the area to be detected according to the discharge area in the infrared image, the discharge area in the ultraviolet image, the partial discharge probability information of the infrared image, and the partial discharge probability information of the ultraviolet image; the partial discharge detection result indicates whether there is partial discharge in the area to be detected.

10. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images