Cable fault determination method and apparatus, and electronic device

Abstract

The invention discloses a cable fault determination method and device and electronic equipment. Relates to the field of artificial intelligence, and the method comprises the steps: obtaining image information of a target cable collected by a collection device through M collection modes, and obtaining M target images, M being a positive integer; the M target images are input into a fault recognition model, M recognition results are obtained, and the fault recognition model is used for predicting the fault type of the target cable according to the target images; and determining the fault type of the target cable according to the M identification results. According to the invention, the problem of low efficiency and accuracy of manual cable inspection in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular, to a method, device and electronic device for determining cable faults. Background Art

[0002] With the development of economy and the acceleration of urbanization, cables have gradually become important equipment for power transmission. However, in the process of using cable equipment to transmit power, cable faults frequently occur, which will affect the operation of the entire power system and may also cause large-scale power outages. Therefore, fault inspection of cable equipment is very important.

[0003] In order to check whether the cable is faulty, manual inspections are usually carried out on the underground cables. The inspectors observe the appearance of the cable equipment to check whether there are short circuits, aging or open circuits, and record and analyze the problems found.

[0004] However, manual inspection requires inspectors to check each cable and its equipment one by one, which consumes a lot of time and energy, and the judgment of the inspection results depends on the experience and skills of the inspectors. Therefore, manual cable inspection has the problem of low efficiency and accuracy.

[0005] Currently, no effective solution has been proposed to address the problem of low efficiency and accuracy of manual cable inspection in related technologies. Summary of the invention

[0006] The present application provides a method, device and electronic device for determining cable faults to solve the problem of low efficiency and accuracy of manual cable inspection in the related art.

[0007] According to one aspect of the present application, a method for determining a cable fault is provided. The method includes: acquiring image information of a target cable acquired by an acquisition device through M acquisition modes to obtain M target images, where M is a positive integer; inputting the M target images into a fault recognition model respectively to obtain M recognition results, where the fault recognition model is used to predict the fault type of the target cable according to the target image; and determining the fault type of the target cable according to the M recognition results.

[0008] Optionally, obtaining image information of the target cable acquired by an acquisition device through M acquisition methods to obtain M target images includes: acquiring M initial images sent by the acquisition device; adjusting the image size of each image to a preset size to obtain M first images; performing grayscale conversion on each first image to obtain M second images; inputting the M second images into an image processing model to obtain M target images, wherein the image processing model is used to improve image resolution.

[0009] Optionally, the image processing model is trained in the following manner: acquiring images of the target cable at multiple historical moments to obtain multiple groups of first historical images, wherein each group of first historical images includes historical images with a first resolution and historical images with a second resolution, and the first resolution is greater than the second resolution; determining each group of first historical images as a group of sample data to obtain multiple groups of first sample data, and using the multiple groups of first sample data to train the initial image processing model to obtain the image processing model.

[0010] Optionally, the fault recognition model is trained in the following manner: acquiring images of the target cable at multiple historical moments to obtain multiple second historical images; adding a fault type identifier to each second historical image in turn, and determining the second historical image carrying the fault type identifier as a group of sample data to obtain multiple groups of second sample data; using the multiple groups of second sample data to train the initial fault recognition model to obtain a fault recognition model.

[0011] Optionally, the identification result includes a fault type and a prediction accuracy rate, and determining the fault type of the target cable based on M identification results includes: obtaining the fault type with the highest prediction accuracy rate among the M identification results to obtain M fault types; grouping the M identification results according to the fault type to obtain multiple groups of identification results; obtaining the number of identification results in each group of identification results in turn to obtain multiple quantity values; determining the number of maximum quantity values ​​among the multiple quantity values ​​to obtain a target number; when the target number is a preset value, determining the fault type corresponding to a group of identification results to which the maximum quantity value belongs as the fault type of the target cable; when the target number is greater than the preset value, determining the fault type of the target cable according to the prediction accuracy rates in each identification result.

[0012] Optionally, determining the fault type of the target cable based on the prediction accuracy in each identification result includes: obtaining the target group identification result corresponding to the maximum number value to obtain multiple target group identification results; obtaining the accuracy of the fault type corresponding to the group in each target group identification result to obtain multiple groups of accuracy rates; calculating the average of the accuracy rates of each group to obtain multiple average accuracy rates; and determining the fault type corresponding to the target group identification result with the highest average accuracy rate as the fault type of the target cable.

[0013] Optionally, the identification result includes a fault type and a prediction accuracy rate, and determining the fault type of the target cable based on M identification results includes: combining the fault type and the corresponding prediction accuracy rate in each identification result to obtain a fault type identification result table; inputting the fault type identification result table into an evidence synthesis model for calculation to obtain probability values ​​of each fault type, wherein the evidence synthesis model includes an evidence synthesis algorithm; and determining the fault type with the largest probability value as the fault type of the target cable.

[0014] According to another aspect of the present application, a device for determining a cable fault is provided. The device includes: a first acquisition unit, used to acquire image information of a target cable acquired by an acquisition device through M acquisition modes, and obtain M target images, where M is a positive integer; an identification unit, used to input the M target images into a fault identification model respectively, and obtain M identification results, where the fault identification model is used to predict the fault type of the target cable according to the target image; and a first determination unit, used to determine the fault type of the target cable according to the M identification results.

[0015] According to another aspect of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, a method for determining a cable fault provided in the aforementioned embodiment of the present application is implemented.

[0016] According to another aspect of the present invention, an electronic device is provided, comprising one or more processors and a memory; the memory stores computer-readable instructions, and the processor is used to execute the computer-readable instructions, wherein when the computer-readable instructions are executed, a method for determining a cable fault provided in the aforementioned embodiment is executed.

[0017] Through this application, the following steps are adopted: obtain the image information of the target cable collected by the acquisition device through M acquisition methods to obtain M target images, where M is a positive integer; input the M target images into the fault recognition model respectively to obtain M recognition results, where the fault recognition model is used to predict the fault type of the target cable according to the target image; determine the fault type of the target cable according to the M recognition results. The problem of low efficiency and accuracy of manual cable inspection in the related technology is solved, the initial image is processed by using the image processing model to obtain the target image, the resolution of the target image is improved, and the target image is input into the fault recognition model, and the fault type of the target cable is determined according to the recognition result, which ensures the accuracy and reliability of the final recognition result, thereby achieving the effect of improving the efficiency and accuracy of fault recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 is a flow chart of a method for determining a cable fault according to an embodiment of the present application;

[0020] Figure 2 is a flowchart of an image processing process provided according to an embodiment of the present application;

[0021] Figure 3is a schematic diagram of a device for determining a cable fault according to an embodiment of the present application;

[0022] Figure 4 It is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that the cable fault determination method, device and electronic device determined in the present disclosure can be used in the field of artificial intelligence, and can also be used in any field other than the field of artificial intelligence. The application field of the cable fault determination method, device and electronic device determined in the present disclosure is not limited.

[0027] It should be noted that the collected information, user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) used in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with the relevant laws, regulations and standards of the relevant regions, necessary confidentiality measures are taken, and public order and good customs are not violated. Corresponding operation portals are provided for users to choose to authorize or refuse to use. For example, an interface is set up between this system and relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0028] The embodiments or examples of the present disclosure are not exhaustive, but are only illustrative of some embodiments or examples, and are not intended to be specific limitations on the scope of protection of the present disclosure. In the absence of contradiction, each step in a certain embodiment or example can be implemented as an independent example, and the steps can be combined arbitrarily. For example, the scheme after removing some steps in a certain embodiment or example can also be implemented as an independent example, and the order of the steps in a certain embodiment or example can be arbitrarily exchanged. In addition, the optional methods or optional examples in a certain embodiment or example can be combined arbitrarily; in addition, the various embodiments or examples can be combined arbitrarily, for example, some or all steps of different embodiments or examples can be combined arbitrarily, and a certain embodiment or example can be combined arbitrarily with the optional methods or optional examples of other embodiments or examples.

[0029] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0030] SRCNN model: SRCNN (Super-Resolution Convolutional Neural Network) is a deep learning model for image super-resolution reconstruction. The SRCNN model can convert blurry or low-resolution images into clear and high-resolution images by learning the mapping relationship from low-resolution images to high-resolution images.

[0031] ResNet model: ResNet (Residual Network) is a deep convolutional neural network model. ResNet introduces residual connections, which allows the network to learn residual functions by adding cross-layer connections in the network, making the network easier to train and optimize. It is widely used in the field of artificial intelligence, such as image recognition, object detection, semantic segmentation, etc.

[0032] Evidence synthesis algorithm: is a computational method used to integrate and combine evidence from multiple different sources, aiming to improve the accuracy and reliability of the model. By combining information from multiple sources of evidence, a more comprehensive understanding of the data can be achieved, providing more accurate predictions and decisions.

[0033] According to an embodiment of the present application, a method for determining a cable fault is provided.

[0034] Figure 1 FIG. 1 is a flow chart of a method for determining a cable fault according to an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0035] Step S101, acquiring image information of a target cable acquired by an acquisition device through M acquisition modes, and obtaining M target images, wherein M is a positive integer.

[0036] Specifically, the device for collecting the target cable image can be a panoramic camera, inertial navigation or lidar and other devices. By collecting images of the target cables in a certain area at different angles and orientations, M initial images of the target cables can be obtained. After obtaining the initial image, the collected image data can be pre-processed, and then the image resolution can be improved using the image processing model to obtain M target images, thereby ensuring that the subsequent model can normally recognize the content in the image.

[0037] It should be noted that due to the different devices and methods for collecting the target cable image, there may be differences in the size of the initial image and the size of the storage space. In order to ensure that the image processing model can correctly process the image, the size of the collected image can be adjusted so that the size of the initial image meets the processing requirements of the image processing model. The adjusted image is then gray-processed to reduce the image memory, thereby reducing the computational burden of the image processing model and improving the model running speed.

[0038] It should be noted that when determining cable faults, in order to improve the accuracy of fault identification, high-resolution images are usually required to accurately identify whether the cable has minor damage. However, due to the limitations of the tunnel environment, it is difficult to obtain high-resolution images. Therefore, the image processing model can be used to improve the resolution of the image. By inputting the preprocessed initial image of the target cable into the image processing model, the target image is obtained after model processing, thereby improving the resolution of the target image, and then the details of the target cable can be displayed more clearly, which is convenient for the detection and analysis of cable faults.

[0039] Step S102, inputting M target images into a fault recognition model respectively to obtain M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable according to the target image.

[0040] Specifically, M target images are respectively input into the fault recognition model for fault recognition, and M recognition results are finally obtained. The recognition results include the fault types corresponding to the M target images and the predicted probability of each fault type. For example, the recognition result of image 1 is type A fault with a probability of 70%, type B fault with a probability of 20%, and type C fault with a probability of 10%. Among them, the fault types can be multiple and the number is unlimited. The recognition results that the fault recognition model may obtain are as follows:

[0041] Table 1

[0042]

[0043] It should be noted that the fault recognition model extracts and distinguishes the feature information of the target image. The feature information may include cracks, damage, color, shape, etc. of the target cable. The fault recognition model compares and analyzes the feature information to predict the fault type of the target cable and obtain the recognition result.

[0044] Step S103: determining the fault type of the target cable according to the M identification results.

[0045] Specifically, after M recognition results of the target cable image are obtained according to the fault recognition model, since the predicted probabilities of the fault types in the recognition results of different target images are different, the fault type of the target cable can be finally determined by combining the M recognition results.

[0046] For example, the fault type with the highest prediction accuracy in each identification result can be obtained, M identification results can be grouped according to the fault type, and then the number of identification results corresponding to the fault type in each group of identification results can be counted. The fault type with the most and unique identification results is the fault type of the target cable. If the fault type with the most identification results is not unique, the fault type of the target cable is determined by comparing the prediction accuracy in each identification result, thereby ensuring the accuracy and reliability of the final identification result.

[0047] The cable fault determination method provided in the embodiment of the present application obtains M target images by acquiring image information of the target cable acquired by the acquisition device through M acquisition methods, wherein M is a positive integer; the M target images are respectively input into the fault recognition model to obtain M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable according to the target image; the fault type of the target cable is determined according to the M recognition results. The problem of low efficiency and accuracy of manual cable inspection in the related art is solved, the target image is obtained by processing the initial image using the image processing model, the resolution of the target image is improved, and the target image is input into the fault recognition model, the fault type of the target cable is determined according to the recognition result, the accuracy and reliability of the final recognition result are ensured, and the effect of improving the efficiency and accuracy of fault recognition is achieved.

[0048] Optionally, in the method for determining a cable fault provided in an embodiment of the present application, Figure 2 is a flowchart of an image processing process provided in an embodiment of the present application, such as Figure 2 As shown, the image information of the target cable collected by the collection device through M collection methods is obtained, and the M target images obtained include:

[0049] Step S201, obtaining M initial images sent by a collection device.

[0050] Step S202: adjusting the image size of each image to a preset size to obtain M first images.

[0051] Step S203: grayscale conversion is performed on each first image to obtain M second images.

[0052] Step S204, inputting the M second images into an image processing model to obtain M target images, wherein the image processing model is used to improve image resolution.

[0053] Specifically, since the size and memory of the initial images obtained from the acquisition device may differ, some initial image information is not suitable for model training. Therefore, the image size of each initial image can be adjusted to a preset size to obtain an adjusted first image, wherein the method of adjusting the image size to the preset size is not limited, for example, it can be an online image processing tool, local image processing software, or a programming language implementation.

[0054] It should be noted that, since the grayscale conversion operation can make the memory occupied by the image smaller, after obtaining M first images, each first image can be grayscale converted to obtain a second image, wherein the grayscale conversion can adopt a weighted average method, based on the sensitivity of the human eye to different colors, different weights are given to red, green, and blue, and then weighted average is performed, which can be converted by the following formula:

[0055] Gray(i,j)=0.299*R(i,j)+0.578*G(i,j)+0.114*B(i,j);

[0056] Among them, Gray(i,j) represents the grayscale value at position (i,j), R(i,j) represents the red channel value at position (i,j), G(i,j) represents the green channel value at position (i,j), and B(i,j) represents the blue channel value at position (i,j). Three commonly used weights of 0.299, 0.578 and 0.114 are selected as the weights of red, green and blue.

[0057] It should be noted that after grayscale conversion is performed on the first image to obtain the second image, the second image may be further subjected to a resolution enhancement operation, thereby improving the recognition accuracy when the content in the processed second image is subsequently recognized. When the image resolution is enhanced, the second image may be processed using an image processing model to enhance the resolution of the second image to a preset resolution, thereby completing the image processing operation and obtaining M target images.

[0058] This embodiment preprocesses the initial image by adjusting the image size, converting the image grayscale, and adjusting the image resolution, thereby processing the images of the target cable acquired by different acquisition methods, thereby ensuring the accuracy of subsequent operations for identifying the image content.

[0059] Optionally, in the cable fault determination method provided in an embodiment of the present application, the image processing model is trained in the following manner: images of the target cable at multiple historical moments are acquired to obtain multiple groups of first historical images, wherein each group of first historical images includes historical images with a first resolution and historical images with a second resolution, and the first resolution is greater than the second resolution; each group of first historical images is determined as a group of sample data to obtain multiple groups of first sample data, and the multiple groups of first sample data are used to train the initial image processing model to obtain the image processing model.

[0060] Specifically, before using the image processing model to process the resolution of the image, the image processing model needs to be trained first to ensure the accuracy of the model's image processing. First, the images of the target cable at multiple historical moments can be acquired through the acquisition device to obtain multiple groups of first historical images, wherein each group of first historical images can include a group of historical images with the same image content but different resolutions. There is no limit to the method of acquiring historical images with different resolutions, but the parameters such as the image content and shooting angle of each group of first historical images need to be the same. For example, the first historical image of the target cable at a certain position can be acquired at a certain shooting angle through different acquisition devices, thereby obtaining a group of first historical images with different resolutions.

[0061] It should be noted that after obtaining multiple groups of first historical images, each group of first historical images can be determined as a group of sample data to obtain multiple groups of first sample data, and the initial image processing model can be trained using the multiple groups of first sample data until the accuracy of the model reaches the expected value, thereby obtaining the image processing model.

[0062] For example, the image processing model can be an SRCNN model. In the process of training the initial image processing model, a SRCNN neural network structure can be established first. Then, in the process of selecting cable images, it is necessary to obtain multiple groups of image files with expected resolutions and those whose resolutions need to be improved, and divide the above multiple groups of image files into training sets and test sets. The training set is used to train the model, and the test set is used to test the model, thereby completing the training operation of the image processing model.

[0063] In this embodiment, the initial image processing model is trained by using first historical images of different resolutions to obtain the image processing model, thereby ensuring the accuracy of the image processing model in performing resolution enhancement operations on the image.

[0064] Optionally, in the cable fault determination method provided in an embodiment of the present application, the fault identification model is trained in the following manner: obtaining images of the target cable at multiple historical moments to obtain multiple second historical images; adding a fault type identifier to each second historical image in turn, and determining the second historical image carrying the fault type identifier as a group of sample data to obtain multiple groups of second sample data; using the multiple groups of second sample data to train the initial fault identification model to obtain a fault identification model.

[0065] Specifically, in the process of training the fault recognition model through sample data, first, images of the target cable at multiple historical moments can be obtained through the acquisition equipment to obtain multiple second historical images, and then a fault type identification can be added to each second historical image, such as short circuit, aging, open circuit, etc. Each second historical image carrying a fault type identification can be used as a group of sample data. By inputting each group of second sample data into the initial fault recognition model for training, through the training process, the model can learn and establish a mapping relationship between the fault image feature information and the fault type, and obtain a fault recognition model that can efficiently and accurately identify the fault type of the target cable image.

[0066] For example, the initial fault recognition model can be a ResNet neural network model. When the ResNet neural network model is trained with multiple sets of sample data, the model parameters and network structure are initialized first, and the number of batch images is set to batch_size=12, that is, the model can process 12 cable images at the same time during each training process, and the total number of tunnel cable states is n_class=5, that is, the model needs to identify 5 types of cable faults, the number of iterations is epoch=500, the learning rate learn_rate=0.0001, and a 3×3 convolutional layer is used for feature extraction to improve the model calculation efficiency while ensuring the area of ​​the neural network batch processing image; then the ResNet neural network is trained with sample data, the total number of tunnel cable maintenance state classifications is n_class=5, and the correct recognition rate R is used. CR As an evaluation index, it is calculated by the following formula:

[0067] R CR = Total correct classifications / classification behavior execution*100%;

[0068] Therefore, whether the training operation of the initial fault recognition model is completed is judged according to the correct recognition rate, thereby ensuring the recognition accuracy of the fault recognition model.

[0069] This embodiment generates sample data by adding a fault type identifier to the second historical image, and then trains an initial fault recognition model based on the sample data to obtain a fault recognition model, thereby ensuring the accuracy of identifying the target cable fault type through the fault recognition model.

[0070] Optionally, in the cable fault determination method provided in the embodiment of the present application, the identification result includes a fault type and a prediction accuracy rate, and determining the fault type of the target cable based on M identification results includes: obtaining the fault type with the highest prediction accuracy rate among the M identification results to obtain M fault types; grouping the M identification results according to the fault type to obtain multiple groups of identification results; obtaining the number of identification results in each group of identification results in turn to obtain multiple quantity values; determining the number of maximum quantity values ​​among the multiple quantity values ​​to obtain a target number; when the target number is a preset value, determining the fault type corresponding to a group of identification results to which the maximum quantity value belongs as the fault type of the target cable; when the target number is greater than the preset value, determining the fault type of the target cable according to the prediction accuracy rates in each identification result.

[0071] It should be noted that by inputting M target images into the fault recognition model respectively, M recognition results will be obtained. Since the prediction probabilities of different fault types in the recognition results of different target images are different, the M recognition results can be combined for group judgment to ultimately determine the fault type of the target cable.

[0072] Specifically, we first need to obtain the fault type with the highest prediction accuracy from the M recognition results to obtain M fault types. For example, the recognition result of image 1 is that the probability of type A fault is 70%, the probability of type B fault is 20%, and the probability of type C fault is 10%. Then the fault type corresponding to image 1 is type A fault. If there is more than one fault type with the highest prediction accuracy, the fault types with the highest probabilities are all fault types corresponding to the image. For example, in the recognition result of image 2, the probabilities of fault types B and C are both 40%, and the probability of fault type A is 20%. Then the fault types corresponding to image 2 are type B fault and type C fault.

[0073] Furthermore, the M recognition results can be grouped according to the fault type, where each fault type corresponds to multiple different target images. For example, the fault type corresponding to image 1 is type A fault, image 2 corresponds to type B and type C faults, and image 3 corresponds to type A fault. The fault types are grouped to obtain type A: image 1 and image 3, type B: image 2, and type C: image 2. Then, the quantity value in each group of recognition results is obtained, that is, the number of target images corresponding to the three fault types A, B, and C is obtained, and the quantity value of type A faults is 2, and the quantity values ​​of type B faults and type C faults are both 1.

[0074] Furthermore, the number of the maximum number of multiple quantity values ​​is taken as the target number. If the target number is the same as the preset value, which is usually 1, the fault type corresponding to the group of identification results to which the maximum number value belongs is the fault type of the target cable.

[0075] For example, in the recognition results of the following three target images, the fault types with the highest accuracy are type A fault in the first group of recognition results, type B and type C faults in the second group of recognition results, and type A fault in the third group of recognition results. They are grouped according to the three fault types of A, B, and C. Type A fault corresponds to the first and third groups, type B fault corresponds to the second group, and type C fault corresponds to the second group. The number values ​​of the three fault types are 2, 1, and 1, respectively. The maximum number value is 2, which corresponds to type A fault, and the target number is 1, which is the same as the preset value. Therefore, type A fault is finally determined to be the fault type of the target cable.

[0076] Table 2

[0077]

[0078] It should be noted that when the target number is greater than the preset value, the fault type of the target cable can be determined based on the prediction accuracy in each identification result, and the fault type of the target cable can be determined by comparing the prediction accuracy in multiple groups of identification results to which the maximum number value belongs.

[0079] For example, in the identification results in the following table, the number values ​​of type A faults and type C faults are both 2, the number value of type B faults is 1, and the target number is 2, which is greater than the preset value. In this case, the two highest prediction accuracies of type A faults in the first and third groups of identification results can be combined with the two highest prediction accuracies of type C faults in the second and third groups of identification results for comparison. By comparing the average values ​​of the two groups of prediction accuracies, that is, calculating the average values ​​of 70% and 45% and comparing them with the average values ​​of 40% and 45%, type A fault can be finally determined as the fault type of the target cable.

[0080] Table 3

[0081]

[0082] This embodiment groups and compares the identification results according to the fault type, thereby accurately determining the fault type of the target cable.

[0083] Optionally, in the cable fault determination method provided in the embodiment of the present application, determining the fault type of the target cable according to the prediction accuracy in each identification result includes: obtaining the target group identification result corresponding to the maximum number value, and obtaining multiple target group identification results; obtaining the accuracy of the fault type corresponding to the group in each target group identification result, and obtaining multiple groups of accuracy rates; calculating the average of the accuracy rates of each group, and obtaining multiple average accuracy rates; and determining the fault type corresponding to the target group identification result with the highest average accuracy rate as the fault type of the target cable.

[0084] It should be noted that when the target number is greater than the preset value, the fault type corresponding to a group of identification results with a larger average prediction accuracy can be determined as the fault type of the target cable by comparing the prediction accuracy of multiple groups of identification results to which the maximum number value belongs.

[0085] Specifically, taking the following recognition results as an example, the quantity values ​​of the three fault types are 2, 1, and 2 respectively. First, the target group recognition result corresponding to the maximum quantity value can be obtained, that is, type A faults correspond to the first and third groups, and type C faults correspond to the second and third groups. Then, the accuracy of the fault type corresponding to the group in each target group recognition result is obtained to obtain multiple groups of accuracy. The accuracy rates of type A faults are 70% and 45% respectively, and the accuracy rates of type C faults are 40% and 45% respectively. Finally, the average of the accuracy rates of each group is calculated to obtain multiple average accuracy rates. The fault type corresponding to the target group recognition result with the highest average accuracy rate is determined as the fault type of the target cable. In this example, the average accuracy rate of type A fault is the highest. Therefore, type A fault is finally determined to be the fault type of the target cable.

[0086] Table 4

[0087]

[0088] This embodiment calculates the average of the prediction accuracy of the corresponding fault types in the target group identification results, and determines the cable fault type based on the average value, thereby achieving the technical effect of accurately determining the fault type of the target cable.

[0089] Optionally, in the cable fault determination method provided in the embodiment of the present application, the identification result includes a fault type and a prediction accuracy rate, and determining the fault type of the target cable based on M identification results includes: combining the fault type and the corresponding prediction accuracy rate in each identification result to obtain a fault type identification result table; inputting the fault type identification result table into an evidence synthesis model for calculation to obtain probability values ​​of each fault type, wherein the evidence synthesis model includes an evidence synthesis algorithm; and determining the fault type with the largest probability value as the fault type of the target cable.

[0090] Specifically, the recognition results output by the fault recognition model include fault type and prediction accuracy. The fault type and the corresponding prediction accuracy in each recognition result can be combined to obtain a fault type recognition result table, in which the rows and columns are the fault type and the target image, respectively.

[0091] For example, the fault type identification results table may be as follows:

[0092] Table 5

[0093]

[0094] Furthermore, the fault type identification result table can be input into the evidence synthesis model to calculate the normalization constant, i.e., the conflict coefficient, and then the probability value of each fault type can be obtained according to the formula. The evidence synthesis model includes an evidence synthesis algorithm. According to the obtained probability value of the fault type, the fault type with the largest probability value is selected as the final result. The fault result obtained by the evidence synthesis model integrates multiple independent identification results, so the identification result is more accurate and reliable.

[0095] This embodiment calculates the fault type identification result table through the evidence synthesis model, and determines the fault type with the largest probability value as the fault type of the target cable, thereby ensuring that the identification result is more accurate and reliable.

[0096] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0097] The present application also provides a device for determining a cable fault. It should be noted that the device for determining a cable fault in the present application can be used to execute the method for determining a cable fault provided in the present application. The following introduces the device for determining a cable fault provided in the present application.

[0098] Figure 3 is a schematic diagram of a device for determining a cable fault according to an embodiment of the present application. Figure 3 As shown, the device comprises:

[0099] The first acquisition unit 31 is used to acquire image information of the target cable acquired by the acquisition device through M acquisition modes to obtain M target images, where M is a positive integer.

[0100] The recognition unit 32 is used to input the M target images into the fault recognition model respectively to obtain M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable according to the target image.

[0101] The first determining unit 33 is configured to determine the fault type of the target cable according to the M identification results.

[0102] The cable fault determination device provided in the embodiment of the present application obtains the image information of the target cable collected by the collection device through M collection methods through the first acquisition unit 31, and obtains M target images, wherein M is a positive integer; the recognition unit 32 inputs the M target images into the fault recognition model respectively, and obtains M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable according to the target image; the first determination unit 33 determines the fault type of the target cable according to the M recognition results. The target image is obtained by processing the initial image using the image processing model, the resolution of the target image is improved to ensure the accuracy of the recognition result, and the target image is input into the fault recognition model, and the fault type of the target cable is determined according to the recognition result, which ensures the accuracy and reliability of the final recognition result, solves the problem of low efficiency and accuracy of manual cable inspection in the related technology, and thus achieves the effect of improving the efficiency and accuracy of fault recognition.

[0103] Optionally, in the cable fault determination device provided in the embodiment of the present application, the first acquisition unit 31 includes: a first acquisition module, used to acquire M initial images sent by the acquisition device; an adjustment module, used to adjust the image size of each image to a preset size, to obtain M first images; a conversion module, used to perform grayscale conversion on each first image, to obtain M second images; a processing module, used to input the M second images into an image processing model, to obtain M target images, wherein the image processing model is used to improve image resolution.

[0104] Optionally, in the cable fault determination device provided in an embodiment of the present application, the image processing model is trained in the following manner: a second acquisition unit is used to acquire images of the target cable at multiple historical moments to obtain multiple groups of first historical images, wherein each group of first historical images includes historical images with a first resolution and historical images with a second resolution, and the first resolution is greater than the second resolution; a first training unit is used to determine each group of first historical images as a group of sample data to obtain multiple groups of first sample data, and use the multiple groups of first sample data to train the initial image processing model to obtain an image processing model.

[0105] Optionally, in the cable fault determination device provided in an embodiment of the present application, the fault identification model is trained in the following manner: a third acquisition unit is used to acquire images of the target cable at multiple historical moments to obtain multiple second historical images; a second determination unit is used to add a fault type identifier to each second historical image in turn, and determine the second historical image carrying the fault type identifier as a group of sample data to obtain multiple groups of second sample data; a second training unit is used to train the initial fault identification model using multiple groups of second sample data to obtain a fault identification model.

[0106] Optionally, in the cable fault determination device provided in the embodiment of the present application, the identification result includes a fault type and a prediction accuracy, and the first determination unit 33 includes: a second acquisition module, used to obtain the fault type with the highest prediction accuracy among M identification results, and obtain M fault types; a grouping module, used to group the M identification results according to the fault type, and obtain multiple groups of identification results; a third acquisition module, used to obtain the number of identification results in each group of identification results in turn, and obtain multiple quantity values; a first determination module, used to determine the number of maximum quantity values ​​among multiple quantity values, and obtain a target number; a second determination module, used to determine the fault type corresponding to a group of identification results to which the maximum quantity value belongs as the fault type of the target cable when the target number is a preset value; a third determination module, used to determine the fault type of the target cable according to the prediction accuracy in each identification result when the target number is greater than the preset value.

[0107] Optionally, in the cable fault determination device provided in the embodiment of the present application, the third determination module includes: a first acquisition submodule, used to obtain the target group identification result corresponding to the maximum number value, and obtain multiple target group identification results; a second acquisition submodule, used to obtain the accuracy of the fault type corresponding to the group in each target group identification result, and obtain multiple groups of accuracy rates; a calculation submodule, used to calculate the average of the accuracy rates of each group, and obtain multiple average accuracy rates; a determination submodule, used to determine the fault type corresponding to the target group identification result with the highest average accuracy rate as the fault type of the target cable.

[0108] Optionally, in the cable fault determination device provided in the embodiment of the present application, the identification result includes the fault type and the prediction accuracy, and the first determination unit 33 includes: a combination module, used to combine the fault type and the corresponding prediction accuracy in each identification result to obtain a fault type identification result table; a calculation module, used to input the fault type identification result table into the evidence synthesis model for calculation to obtain the probability value of each fault type, wherein the evidence synthesis model includes an evidence synthesis algorithm; a fourth determination module, used to determine the fault type with the largest probability value as the fault type of the target cable.

[0109] The cable fault determination device provided in the embodiment of the present application processes the initial image by using an image processing model to obtain a target image, thereby improving the resolution of the target image and ensuring the accuracy of the recognition result, and inputs the target image into a fault recognition model, determines the fault type of the target cable according to the recognition result, thereby ensuring the accuracy and reliability of the final recognition result, solving the problem of low efficiency and accuracy of manual cable inspection in related technologies, and thereby achieving the effect of improving fault recognition efficiency and accuracy.

[0110] The above-mentioned cable fault determination device includes a processor and a memory. The above-mentioned first acquisition unit 31, identification unit 32, first determination unit 33, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0111] The processor includes a kernel, and the kernel calls the corresponding program unit from the memory. One or more kernels can be set, and the problem of low efficiency and accuracy of manual cable inspection in related technologies can be solved by adjusting kernel parameters.

[0112] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0113] An embodiment of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the method for determining a cable fault is implemented.

[0114] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for determining a cable fault is executed when the program is run.

[0115] Figure 4 is a schematic diagram of an electronic device provided according to an embodiment of the present application, such as Figure 4 As shown, an embodiment of the present invention provides an electronic device, the electronic device 40 includes a processor, a memory, and a program stored in the memory and can be run on the processor, and the processor implements the following steps when executing the program: acquiring image information of a target cable acquired by an acquisition device through M acquisition methods, obtaining M target images, where M is a positive integer; inputting the M target images into a fault recognition model respectively, obtaining M recognition results, where the fault recognition model is used to predict the fault type of the target cable according to the target image; determining the fault type of the target cable according to the M recognition results. The device in this article can be a server, PC, PAD, mobile phone, etc.

[0116] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: acquiring image information of a target cable acquired by an acquisition device through M acquisition methods to obtain M target images, wherein M is a positive integer; inputting the M target images into a fault recognition model respectively to obtain M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable based on the target image; and determining the fault type of the target cable based on the M recognition results.

[0117] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0118] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0121] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0122] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0123] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0124] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0125] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for determining a cable fault, characterized in that: include: Obtain image information of the target cable collected by the collection device through M collection modes to obtain M target images, where M is a positive integer; Inputting the M target images into a fault recognition model respectively to obtain M recognition results, wherein the fault recognition model is used to predict the fault type of the target cable according to the target image; The fault type of the target cable is determined according to the M identification results.

2. The method according to claim 1, characterized in that The image information of the target cable acquired by the acquisition device through M acquisition modes is obtained, and the M target images obtained include: Acquire M initial images sent by the acquisition device; Resize each image to a preset size to obtain M first images; Performing grayscale conversion on each first image respectively to obtain M second images; The M second images are input into an image processing model to obtain the M target images, wherein the image processing model is used to improve image resolution.

3. The method according to claim 2, characterized in that The image processing model is trained in the following way: Acquire images of the target cable at multiple historical moments to obtain multiple groups of first historical images, wherein each group of first historical images includes historical images with a first resolution and historical images with a second resolution, and the first resolution is greater than the second resolution; Each group of first historical images is determined as a group of sample data to obtain multiple groups of first sample data, and the initial image processing model is trained using the multiple groups of first sample data to obtain the image processing model.

4. The method according to claim 1, characterized in that: The fault identification model is trained in the following way: Acquire images of the target cable at multiple historical moments to obtain multiple second historical images; adding a fault type identifier to each second historical image in turn, and determining the second historical images carrying the fault type identifier as a group of sample data, to obtain multiple groups of second sample data; The initial fault identification model is trained using the multiple groups of second sample data to obtain the fault identification model.

5. The method according to claim 1, characterized in that The identification result includes a fault type and a prediction accuracy rate. Determining the fault type of the target cable according to the M identification results includes: Obtain the fault type with the highest prediction accuracy among the M recognition results to obtain M fault types; Grouping the M identification results according to the fault type to obtain multiple groups of identification results; Obtaining the number of recognition results in each group of recognition results in turn to obtain multiple quantity values; Determine the number of maximum quantity values ​​among the multiple quantity values ​​to obtain a target number; In the case where the target number is a preset value, determining the fault type corresponding to a group of identification results to which the maximum number value belongs as the fault type of the target cable; When the target number is greater than the preset value, the fault type of the target cable is determined according to the prediction accuracy in each identification result.

6. The method according to claim 5, characterized in that Determining the fault type of the target cable according to the prediction accuracy in each identification result includes: Obtaining the target group recognition result corresponding to the maximum number value to obtain multiple target group recognition results; Obtain the accuracy of the fault type corresponding to each target group in the recognition results of the target group, and obtain multiple groups of accuracy; Calculate the average of the accuracy of each group to obtain multiple average accuracy rates; The fault type corresponding to the target group identification result with the highest average accuracy is determined as the fault type of the target cable.

7. The method according to claim 1, characterized in that The identification result includes a fault type and a prediction accuracy rate. Determining the fault type of the target cable according to the M identification results includes: Combine the fault type and the corresponding prediction accuracy in each recognition result to obtain a fault type recognition result table; Inputting the fault type identification result table into the evidence synthesis model for calculation to obtain the probability value of each fault type, wherein the evidence synthesis model includes an evidence synthesis algorithm; The fault type with the largest probability value is determined as the fault type of the target cable.

8. A device for determining a cable fault, characterized in that: include: A first acquisition unit is used to acquire image information of a target cable acquired by an acquisition device through M acquisition modes to obtain M target images, where M is a positive integer; An identification unit is used to input the M target images into a fault identification model respectively to obtain M identification results, wherein the fault identification model is used to predict the fault type of the target cable according to the target image; a first determination unit is used to determine the fault type of the target cable according to the M identification results.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the computer-readable storage medium is located is controlled to execute the cable fault determination method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program executes the cable fault determination method according to any one of claims 1 to 7 when running.

Images