Composite insulator infrared image core rod temperature curve extraction method and device

Abstract

The invention discloses a composite insulator infrared image core rod temperature curve extraction method and device. The method comprises the following steps: collecting an infrared image of the composite insulator of the power transmission line and preprocessing the infrared image to obtain a preprocessed infrared image; carrying out central axis position labeling in the preprocessed infrared image by using a target detection model based on deep learning; extracting a preliminary temperature curve by using the marked central axis position; filtering the initial temperature curve to obtain a filtered temperature curve; extracting each local minimum value on the filtered temperature curve, and taking each local minimum value as the temperature of the composite insulator core rod; and performing spline interpolation on each extracted local minimum value to obtain a final core rod temperature curve. Therefore, the temperature of the insulator core rod can be efficiently and intelligently extracted from the infrared image of the composite insulator, and a more accurate temperature curve is generated.

Description

Technical Field

[0001] The present application relates to the technical field of insulator fault detection, and in particular to a method and device for extracting a temperature curve of a composite insulator core rod using an infrared image. Background Art

[0002] Composite insulator core rods are the key position that needs to be paid attention to during power inspections. Their abnormal heating is one of the important factors leading to power system failures. Timely and accurate monitoring and diagnosis of the thermal state of the center axis of composite insulator core rods is of great significance for preventing power accidents and improving power grid reliability.

[0003] In recent years, aerial inspections using drones equipped with infrared thermal imagers have become an effective technical means to quickly capture infrared images of large-area composite insulators.

[0004] The existing method for detecting the thermal status of the centerline of the insulator core rod relies heavily on the manual marking of the core rod position by professionals. This reliance makes the process time-consuming and easily affected by personal experience and judgment criteria, which may cause the analysis results to be inconsistent with the actual situation.

[0005] Therefore, it is necessary to provide a new technical solution for extracting the temperature curve of the core rod of the composite insulator infrared image to reduce manual intervention and improve the processing speed and efficiency of the infrared inspection image. Summary of the invention

[0006] In view of this, the present application proposes a method and device for extracting the temperature curve of the core rod of a composite insulator by infrared image, aiming to solve the problem in the prior art that the process of extracting the temperature curve of the core rod of a composite insulator by infrared image is time-consuming and easily affected by personal experience and judgment criteria.

[0007] In a first aspect, the present application provides a method for extracting a temperature curve of a composite insulator core rod from an infrared image, comprising:

[0008] Collect infrared images of composite insulators of transmission lines and pre-process the infrared images to obtain pre-processed infrared images;

[0009] Use a deep learning-based object detection model to mark the central axis position in the preprocessed infrared image;

[0010] Extract a preliminary temperature curve using the marked central axis position; filter the preliminary temperature curve to obtain a filtered temperature curve;

[0011] Extracting each local minimum value on the filtered temperature curve, and taking each local minimum value as the core rod temperature of the composite insulator;

[0012] The extracted local minimum values ​​are subjected to spline interpolation to obtain the final core rod temperature curve.

[0013] Furthermore, the preprocessing of the infrared image includes:

[0014] Image denoising and image size unification processing;

[0015] The image denoising includes: using Gaussian filtering and / or bilateral filtering.

[0016] Furthermore, the method of using a deep learning-based target detection model to mark the central axis position in the preprocessed infrared image includes:

[0017] On the composite insulator infrared image with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; the Yolo series target detection model based on deep learning is used to mark the central axis position in the composite insulator area.

[0018] Furthermore, the extracting of a preliminary temperature curve using the marked central axis position includes:

[0019] Using the marked central axis position, the temperature value of each pixel on the central axis of the infrared image is extracted, and the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position to obtain a preliminary temperature curve.

[0020] In a second aspect, the present application provides a composite insulator infrared image core rod temperature curve extraction device, comprising:

[0021] An acquisition unit, used for collecting infrared images of composite insulators of transmission lines and preprocessing the infrared images to obtain preprocessed infrared images;

[0022] A labeling unit, used for labeling the central axis position in the preprocessed infrared image using a deep learning-based target detection model;

[0023] The central axis temperature extraction unit is used to extract a preliminary temperature curve using the marked central axis position; filter the preliminary temperature curve to obtain a filtered temperature curve;

[0024] A core rod temperature extraction unit is used to extract each local minimum value on the filtered temperature curve, and use each local minimum value as the core rod temperature of the composite insulator;

[0025] The temperature curve generating unit is used to perform spline interpolation on each local minimum value extracted to obtain a final core rod temperature curve.

[0026] Furthermore, the acquisition unit is used to pre-process the infrared image, including:

[0027] Image denoising and image size unification processing;

[0028] The image denoising includes: using Gaussian filtering and / or bilateral filtering.

[0029] Furthermore, the labeling unit is used to label the central axis position in the preprocessed infrared image using a deep learning-based target detection model, including:

[0030] On the composite insulator infrared image with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; the Yolo series target detection model based on deep learning is used to mark the central axis position in the composite insulator area.

[0031] Furthermore, the central axis temperature extraction unit is used to extract a preliminary temperature curve using the marked central axis position, including:

[0032] Using the marked central axis position, the temperature value of each pixel on the central axis of the infrared image is extracted, and the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position to obtain a preliminary temperature curve.

[0033] In a third aspect, the present application provides a terminal, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0034] In a fourth aspect, the present application provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the method described in the first aspect.

[0035] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0037] Figure 1 A schematic diagram of the process of extracting the temperature curve of the composite insulator core rod from the infrared image of an embodiment of the present application;

[0038] Figure 2 A schematic flow chart of a method for extracting a composite insulator core rod temperature curve from an infrared image according to another embodiment of the present application;

[0039] Figure 3AAn infrared image of a composite insulator generated by an infrared thermal imager carried by a drone in an embodiment of the present application;

[0040] Figure 3B For Figure 3A A schematic diagram of a temperature curve L(x) of the center axis of an insulator obtained by extracting an infrared image of a composite insulator shown in FIG.

[0041] Figure 3C For Figure 3B The schematic diagram of the temperature curve B(x) after filtering the temperature curve L(x) of the center axis of the insulator shown;

[0042] Figure 3D For Figure 3C The schematic diagram of filtering the local minimum point of the temperature curve after filtering is shown;

[0043] Figure 3E For Figure 3D Schematic diagram of the final temperature curve S(x) obtained by spline interpolation of the local minimum point after screening;

[0044] Figure 4 It is a schematic diagram of the composition of the composite insulator infrared image core rod temperature curve extraction device according to an embodiment of the present application;

[0045] Figure 5 A schematic diagram of the components of a terminal to which the method for extracting a temperature curve of a composite insulator core rod by infrared image according to an embodiment of the present application is applied;

[0046] Figure 6 It is a schematic diagram of a program product using the partial discharge type identification capability detection method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0048] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0049] Composite insulators usually include core rods, sheaths, sheds, hardware and other parts. The core rod can be made of epoxy resin impregnated glass fiber and drawn and cured, and the sheath and shed materials can be high-temperature vulcanized silicone rubber. The core rod is an insulating component inside the composite insulator to ensure its mechanical properties.

[0050] Composite insulators play a vital role in the power system. They support and isolate high-voltage power lines to ensure the safety and reliability of power transmission. As the scale of the power grid expands and the transmission capacity increases, the stability of composite insulators directly affects the operating efficiency and safety of the power system.

[0051] Abnormal heating of composite insulator core rods is one of the important factors leading to power system failures. Composite insulator core rod heating may be caused by factors such as material aging, mechanical stress, environmental corrosion or excessive electrical load.

[0052] At present, drone inspections can collect a large amount of infrared image data in a short period of time, but the processing and analysis of this data and the extraction of temperature curves are very complex and time-consuming. At present, when thermal monitoring on the center axis of the insulator core rod is performed to identify potential overheating problems, it is necessary to detect the temperature on the center axis of the insulator core rod and generate a temperature curve. In addition to the core rod, there are also a large number of shed areas on the center axis of the composite insulator, and the shed temperature is quite different from the core rod temperature, which requires high precision to accurately extract the core rod temperature when manually positioning the core rod.

[0053] Therefore, the current monitoring method lacks automation and intelligent processing capabilities, making it difficult to effectively process and analyze massive data, difficult to quickly and efficiently separate the insulator core rod temperature and shed temperature, and difficult to quickly and efficiently extract the accurate core rod temperature.

[0054] The present invention proposes a technical solution for extracting the temperature curve of the composite insulator core rod from the infrared image, which can efficiently and intelligently extract the temperature of the insulator core rod from the composite insulator infrared image and generate a more accurate temperature curve. In this way, manual intervention can be greatly reduced and the efficiency of abnormal temperature rise detection can be improved, so that maintenance or repair measures can be taken in time, thereby better ensuring the safety and stable operation of the power system.

[0055] like Figure 1 As shown, the composite insulator infrared image core rod temperature curve extraction method proposed by the present invention comprises the following steps:

[0056] S100: Collect infrared images of composite insulators of transmission lines and pre-process the infrared images to obtain pre-processed infrared images.

[0057] Specifically, the infrared image is preprocessed, including image denoising and image size unification.

[0058] Specifically, image denoising includes: using Gaussian filtering or bilateral filtering to reduce noise in the image and improve image quality. Specifically, Gaussian filtering can blur the image by weighted averaging the neighboring pixels around each pixel. Specifically, bilateral filtering can consider the spatial distance and grayscale difference between pixels to remove noise while retaining edge information.

[0059] Specifically, the image size unification process includes: unifying the size of each infrared image into 640×512, that is, there are 640 pixels in the width direction and 512 pixels in the height direction.

[0060] In this way, the preprocessed infrared images have a uniform image size.

[0061] S200: Using a deep learning-based target detection model to mark the central axis position in the preprocessed infrared image.

[0062] Specifically, on the infrared image of the composite insulator with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; and the central axis position is marked in the composite insulator area using a deep learning-based target detection model.

[0063] Specifically, in step S200, a deep learning-based target detection model is used to train and test the labeled image of the central axis of the composite insulator, such as the Yolo series target detection model. In this way, the deep learning-based target detection model that has been trained and has met the test performance requirements can automatically find the position of the central axis in the infrared image of the composite insulator, thereby realizing automatic positioning of the central axis.

[0064] S300: Using the marked central axis position, extract the temperature value of each pixel on the central axis of the infrared image (such as M temperature values, M is a natural number greater than 1), arrange the temperature values ​​from the low-voltage end to the high-voltage end of the insulator according to the pixel position, and obtain a preliminary temperature curve L(x); filter the preliminary temperature curve to obtain a filtered temperature curve.

[0065] In some embodiments, in step S300, different filtering methods are selected according to the waveform structure of the current temperature curve L(x). For example, low-pass filtering is used when high-frequency noise is more obvious, high-pass filtering is applied when removing low-frequency noise, and median filtering is used when there is impulse noise. In this way, the interference of noise in the waveform can be effectively removed. After the preliminary temperature curve L(x) is filtered, the filtered temperature curve obtained is B(x).

[0066] In some embodiments, mean filtering is applied to the preliminary temperature curve L(x). When mean filtering is applied, the calculation formula for each filtered temperature value is as shown in formula (1):

[0067]

[0068] in is the temperature value after mean filtering, x i are the temperature values ​​before filtering, m is the size of the filtering window, k is greater than or equal to 0 and less than or equal to Mm.

[0069] Specifically, the size m of the filter window can be adjusted according to the number of pixels l occupied by the central axis of the insulator in the image, and the adjustment scheme is shown in formula (2).

[0070]

[0071] In formula (2), l means the number of pixels occupied by the central axis of the insulator in the image. If l is less than 200, it is considered that the area occupied by the insulator in the image is too small and it is difficult to extract effective temperature information. When l is higher than 200, the size of the filter window is determined to be 2; when l is higher than 300, the size of the filter window is determined to be 3; when l is higher than 400, the size of the filter window is determined to be 4.

[0072] S400: extracting each local minimum value on the filtered temperature curve B(x), and taking each local minimum value as the composite insulator core rod temperature.

[0073] Referring to the above description, the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position; naturally, along the central axis from the low-voltage end to the high-voltage end of the insulator, the temperature curve B(x) includes the core rod temperature and the shed temperature. Considering that the shed temperature is generally higher than the core rod temperature, the local minimum value is extracted on the temperature curve B(x), and each local minimum value is used as the composite insulator core rod temperature.

[0074] In some embodiments, some points on the central axis may exceed the range of the composite insulator in the image, resulting in the extraction of the temperature value of the image background on the temperature curve, which causes the temperature of a certain part of the temperature curve to be too high or too low. In this case, the temperature points of some backgrounds can be extracted from the image, and the temperature difference between the points on the temperature curve and the background temperature points can be compared. When the difference is less than a certain degree, the corresponding temperature points on the temperature curve are removed.

[0075] S500: performing spline interpolation on each local minimum value extracted to obtain a final core rod temperature curve S(x).

[0076] Specifically, spline interpolation can construct a smooth curve between known data points. Specifically, the original function is approximated by inserting polynomial function segments between data points, so that the interpolation result not only passes through all given data points, but also has a higher smoothness between these data points.

[0077] Specifically, a spline function is a function formed by splicing together polynomial function segments corresponding to each interval within a value range (each polynomial function segment is a spline function), and these polynomial function segments or spline functions meet preset continuity conditions and smoothness conditions at the connection points (or nodes).

[0078] More specifically, the cubic spline interpolation uses a cubic polynomial spline function on each interval, which can ensure the continuity of the first-order derivative and the second-order derivative and has higher smoothness.

[0079] Specifically, when performing cubic spline interpolation, each known data point is used as a node; for a given node (x i ,y i )(where i = 0, 1, ..., n), the spline interpolation function S(x) as shown in formula (3) represents the value of the i-th interval [x i , x i+1 ]The polynomial function segment or spline function is a cubic polynomial:

[0080] S i (x i )=a i +b i (xx i )+c i (xx i ) 2 +d i (xx i ) 3 (3)

[0081] In the above formula, a i , b i , c i , d i The following equations (4) to (7) need to be satisfied:

[0082] Among them, formula (4) represents the interpolation condition:

[0083]

[0084] Among them, formula (5) represents the continuity conditions of the first and second order derivatives:

[0085]

[0086] Wherein, formula (6) indicates that the first-order derivatives of the nodes at both ends of the interpolation curve are known values:

[0087]

[0088] Among them, formula (7) indicates that the second-order derivative of the nodes at both ends of the interpolation curve is 0:

[0089]

[0090] After performing cubic spline interpolation on the extracted local minimum values, a temperature curve can be obtained, which is the final core rod temperature curve S(x). The final core rod temperature curve is more accurate, which is conducive to timely maintenance or repair measures, thereby better ensuring the safety and stable operation of the power system.

[0091] Thus, the composite insulator infrared image core rod temperature curve extraction method proposed in the present invention collects the composite insulator infrared image data set, extracts the center axis position of the composite insulator, extracts the temperature on the center axis to obtain a preliminary temperature curve; filters the temperature curve to remove noise interference; extracts the local minimum point as the core rod temperature; and interpolates the core rod temperature to obtain the final insulator core rod temperature curve.

[0092] In summary, the composite insulator infrared image core rod temperature curve extraction method provided by the present invention combines the temperature curve filtering and interpolation methods, can effectively remove the influence of the shed temperature in the insulator center axis temperature curve in the composite insulator infrared image, can effectively improve the detection accuracy and efficiency of the insulator core rod temperature curve, and can also combine computer vision and image processing technology to improve the accuracy of core rod temperature curve extraction, and realize the automation and intelligence of abnormal temperature rise positioning at the insulator core rod of the composite insulator.

[0093] The following combination Figure 2 , further illustrating the composite insulator infrared image core rod temperature curve extraction method of the embodiment of the present invention, comprising the following steps:

[0094] 1) Composite insulator infrared image collection and preprocessing.

[0095] Specifically, infrared images of composite insulators on transmission lines are collected and photographed. The infrared images of insulators are as follows: Figure 3A The infrared images of composite insulators taken are screened to remove the images with poor quality such as incomplete, severe occlusion, and unclear.

[0096] Specifically, the screened infrared image is preprocessed. The infrared image is processed using a denoising algorithm, such as Gaussian filtering and bilateral filtering, to reduce noise in the image and improve image quality. Specifically, Gaussian filtering can blur the image by weighted averaging the neighborhood pixels around each pixel. Specifically, bilateral filtering can consider the spatial distance and grayscale difference between pixels to remove noise while retaining edge information.

[0097] The nearest neighbor interpolation method is further used to uniformly scale the image to a size of 640 × 512. This can improve the computational efficiency of the algorithm while retaining the temperature information of the infrared image.

[0098] 2) Extraction of the centerline of the insulator.

[0099] Specifically, the position of the composite insulator is extracted from the composite insulator infrared image, and the center axis of the insulator is marked.

[0100] 3) Extraction of temperature curve of central axis;

[0101] Specifically, the temperature information of the center axis of the insulator in the infrared image is extracted, and the extracted temperature curve L(x) is as follows: Figure 3B shown.

[0102] 4) Curve filtering reduces the impact of noise.

[0103] The composite insulator temperature curve is filtered to remove the noise in the temperature curve L(x). Specifically, the filtering scheme selects mean filtering. Since the number of waveform pixels is 200<l<300, the filtering window is set to m=2, and the mean filtering represented by equation (1) is used to Figure 3B The temperature curve shown is filtered to obtain Figure 3C The filtered temperature curve B(x) is shown in Figure 2. Figure 3B , Figure 3C The noise content is significantly reduced.

[0104] 5) The local minimum of the curve is extracted as the core rod temperature.

[0105] exist Figure 3C The local minimum is extracted from the temperature curve of the composite insulator shown. Figure 3D The solid dots in the figure indicate the location of the minimum value on the current temperature curve. These local minimum points are used to represent the temperature at the location of the composite insulator core rod.

[0106] 6) Spline interpolation fits the final temperature curve.

[0107] Specifically, the local minimum points are screened, the parts overlapping with the image background are excluded, and then the local minimum points are spline interpolated to obtain the final temperature curve S(x) as shown in Figure 3E shown.

[0108] It can be found that compared with Figure 3B The temperature curve shown, Figure 3E The maximum temperature difference of the temperature curve shown in the figure is reduced from 1.8℃ to 0.4℃, which means that the current scheme can successfully extract the core rod temperature and avoid the influence of the shed temperature on the temperature curve of the composite insulator core rod.

[0109] like Figure 5 As shown, the composite insulator infrared image core rod temperature curve extraction device of the embodiment of the present application includes:

[0110] An acquisition unit 10 is used to collect infrared images of composite insulators of transmission lines and pre-process the infrared images to obtain pre-processed infrared images;

[0111] A labeling unit 20 is used to label the central axis position in the preprocessed infrared image using a deep learning-based target detection model;

[0112] The central axis temperature extraction unit 30 is used to extract a preliminary temperature curve using the marked central axis position; filter the preliminary temperature curve to obtain a filtered temperature curve;

[0113] A core rod temperature extraction unit 40 is used to extract each local minimum value on the filtered temperature curve and use each local minimum value as the core rod temperature of the composite insulator;

[0114] The temperature curve generating unit 50 is used to perform spline interpolation on each extracted local minimum value to obtain a final core rod temperature curve.

[0115] Furthermore, the acquisition unit is used to pre-process the infrared image, including:

[0116] Image denoising and image size unification processing;

[0117] The image denoising includes: using Gaussian filtering and / or bilateral filtering.

[0118] Furthermore, the labeling unit is used to label the central axis position in the preprocessed infrared image using a deep learning-based target detection model, including:

[0119] On the composite insulator infrared image with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; the Yolo series target detection model based on deep learning is used to mark the central axis position in the composite insulator area.

[0120] Furthermore, the central axis temperature extraction unit is used to extract a preliminary temperature curve using the marked central axis position, including:

[0121] Using the marked central axis position, the temperature value of each pixel on the central axis of the infrared image is extracted, and the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position to obtain a preliminary temperature curve.

[0122] The present application also provides a terminal to execute the method. Figure 5It shows a schematic diagram of a terminal provided by some embodiments of the present application. Figure 5 As shown, the terminal 8 includes: a processor 800, a memory 801, a bus 802 and a communication interface 803, and the processor 800, the communication interface 803 and the memory 801 are connected via the bus 802; the memory 801 stores a computer program that can be run on the processor 800, and the processor 800 executes the method provided by any embodiment described in the present application when running the computer program.

[0123] The memory 801 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the device network element and at least one other network element is realized through at least one communication interface 803 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0124] The bus 802 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 801 is used to store programs, and the processor 800 executes the program after receiving an execution instruction. The method disclosed in any implementation of the embodiment of the present application may be applied to the processor 800, or implemented by the processor 800.

[0125] The processor 800 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the method can be completed by the hardware integrated logic circuit or software instructions in the processor 800. The processor 800 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor are combined and executed. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 801, and the processor 800 reads the information in the memory 801 and completes the steps of the method in combination with its hardware.

[0126] The terminal provided in the embodiment of the present application and the method in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the method adopted, operated or implemented by them.

[0127] like Figure 6 As shown, an embodiment of the present application also provides a computer-readable storage medium corresponding to the method provided by the embodiment, and the computer-readable storage medium is a CD, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the method provided by any of the embodiments.

[0128] It should be noted that examples of the computer-readable storage medium may also 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 optical or magnetic storage media, which are not listed here one by one.

[0129] The computer-readable storage medium provided by the embodiments of the present application is based on the same inventive concept as the method of the embodiments of the present application, and has the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0130] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these adjustments and modifications of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these changes and modifications.

Claims

1. A composite insulator infrared image core rod temperature curve extraction method, characterized in that: include: Collect infrared images of composite insulators of transmission lines and pre-process the infrared images to obtain pre-processed infrared images; Use a deep learning-based object detection model to mark the central axis position in the preprocessed infrared image; Extract a preliminary temperature curve using the marked central axis position; filter the preliminary temperature curve to obtain a filtered temperature curve; Extracting each local minimum value on the filtered temperature curve, and taking each local minimum value as the core rod temperature of the composite insulator; The extracted local minimum values ​​are subjected to spline interpolation to obtain the final core rod temperature curve.

2. The composite insulator infrared image core rod temperature curve extraction method according to claim 1, characterized in that: The preprocessing of infrared images comprises: Image denoising and image size unification processing; The image denoising includes: using Gaussian filtering and / or bilateral filtering.

3. The composite insulator infrared image core rod temperature curve extraction method according to claim 1, characterized in that: The method of using a deep learning-based target detection model to mark the central axis position in the preprocessed infrared image includes: On the composite insulator infrared image with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; the Yolo series target detection model based on deep learning is used to mark the central axis position in the composite insulator area.

4. The composite insulator infrared image core rod temperature curve extraction method according to claim 1, characterized in that: The method of extracting a preliminary temperature curve by using the marked central axis position includes: Using the marked central axis position, the temperature value of each pixel on the central axis of the infrared image is extracted, and the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position to obtain a preliminary temperature curve.

5. A composite insulator infrared image core rod temperature curve extraction device, characterized in that: include: An acquisition unit, used for collecting infrared images of composite insulators of transmission lines and preprocessing the infrared images to obtain preprocessed infrared images; A labeling unit, used for labeling the central axis position in the preprocessed infrared image using a deep learning-based target detection model; The central axis temperature extraction unit is used to extract a preliminary temperature curve using the marked central axis position; filter the preliminary temperature curve to obtain a filtered temperature curve; A core rod temperature extraction unit is used to extract each local minimum value on the filtered temperature curve, and use each local minimum value as the core rod temperature of the composite insulator; Temperature curve generation The temperature curve generation unit is used to perform spline interpolation on each local minimum value extracted to obtain the final core rod temperature curve.

6. The composite insulator infrared image core rod temperature curve extraction device according to claim 5, characterized in that: The acquisition unit is used to pre-process the infrared image, including: Image denoising and image size unification processing; The image denoising includes: using Gaussian filtering and / or bilateral filtering.

7. The composite insulator infrared image core rod temperature curve extraction device according to claim 5, characterized in that: The labeling unit is used to label the central axis position in the preprocessed infrared image using a deep learning-based target detection model, and includes: On the composite insulator infrared image with uniform image size, the composite insulator part is cut out to obtain the composite insulator area; the Yolo series target detection model based on deep learning is used to mark the central axis position in the composite insulator area.

8. The composite insulator infrared image core rod temperature curve extraction device according to claim 5, characterized in that: The central axis temperature extraction unit is used to extract a preliminary temperature curve using the marked central axis position, including: Using the marked central axis position, the temperature value of each pixel on the central axis of the infrared image is extracted, and the temperature values ​​are arranged from the low-voltage end to the high-voltage end of the insulator according to the pixel position to obtain a preliminary temperature curve.

9. A terminal, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any one of claims 1 to 4.

10. A computer storage medium, characterized in that: Computer executable instructions are stored, and the computer executable instructions are used to execute the method according to any one of claims 1 to 5.

Images