A substation equipment fault analysis system based on video fusion technology

Abstract

The application belongs to the field of video processing, and discloses a substation equipment fault analysis system based on a video fusion technology, which comprises a camera module, an illumination adjustment module, a video fusion module and a fault analysis module; the camera module is used for acquiring visible light video and infrared video of substation equipment to be subjected to fault analysis; the illumination adjustment module is used for performing illumination adjustment processing on images contained in the visible light video to obtain adjusted visible light video; the video fusion module is used for performing video fusion calculation on the infrared video and the adjusted visible light video to obtain fusion video; and the fault analysis module is used for judging whether a set type of fault appears in the fusion video or not; the images contained in the visible light video are subjected to illumination adjustment processing before fusion, so that the probability of occurrence of excessively high or excessively low brightness in partial areas in the fused video is effectively reduced, and the accuracy of fault analysis on the substation equipment based on the fused video is improved.

Description

Technical Field

[0001] This invention relates to the field of video processing, and in particular to a substation equipment fault analysis system based on video fusion technology. Background Technology

[0002] Infrared video can distinguish targets from backgrounds based on differences in radiation, which works well in all weather conditions and day / night scenarios. Conversely, visible light video aligns with the human visual system and provides texture details with high spatial resolution and sharpness. Therefore, fusing these two types of video combines the advantages of the thermal radiation information from infrared video and the detailed texture information from visible light video, resulting in more accurate analysis results in the fault analysis of substation equipment.

[0003] In existing technologies, the process of fusing infrared video and visible light video lacks a process for adjusting the illumination of the images in the visible light video. This makes it easy for some areas of the fused video to be too bright or too dim, affecting the accuracy of fault analysis of substation equipment based on the fused video. Summary of the Invention

[0004] The purpose of this invention is to disclose a substation equipment fault analysis system based on video fusion technology, which solves the problem that existing video fusion methods lack a process for adjusting the illumination of images in visible light video during the fusion process, thus affecting the accuracy of fault analysis of substation equipment based on the fused video.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A substation equipment fault analysis system based on video fusion technology includes a camera module, a lighting adjustment module, a video fusion module, and a fault analysis module;

[0007] The camera module is used to acquire visible light and infrared video of the substation equipment to be analyzed for faults;

[0008] The illumination adjustment module is used to perform illumination adjustment processing on the images contained in the visible light video to obtain the adjusted visible light video:

[0009] Step 1: Obtain the first image contained in the visible light video;

[0010] Step two, use the first image as the base image;

[0011] Step 3: Obtain a set of images of the same type based on the base image;

[0012] Step four: Obtain the image with the largest index, fralst, from the set of images of the same type.

[0013] Step 5: If fralst is not the last image in the visible light video, then use fralst as the new base image and proceed to Step 3; if fralst is the last image in the visible light video, then output the set of all images of the same type and proceed to Step 6.

[0014] Step 6: Perform illumination adjustment processing on the images in each set of images of the same type, and combine all the images that have undergone illumination adjustment processing to form the adjusted visible light video;

[0015] The video fusion module is used to perform video fusion calculations on infrared video and adjusted visible light video to obtain fused video;

[0016] The fault analysis module is used to identify and process the fused video to determine whether a fault of a specified type occurs in the fused video.

[0017] Optionally, the camera module includes an infrared camera and a visible light camera;

[0018] Infrared cameras are used to acquire infrared video of substation equipment to be analyzed for faults;

[0019] Visible light cameras are used to acquire visible light video of substation equipment to be analyzed for faults.

[0020] Optionally, a set of images of the same type can be obtained based on the base image, including:

[0021] Let s be the number of the base image, and fra be used to represent it. s Represents the basic image;

[0022] For image fra numbered k in visible light video k , k∈(s,N], calculate fra k and fra s If the classification parameter is less than the set parameter threshold, then fra will be... k Save to fra s In the corresponding set of images of the same type, if the classification parameter is greater than or equal to the set parameter threshold, then the image numbered s-1 in the visible light video is used as fra. s The last image in the set of images of the same type, where N represents the total number of images in the visible light video.

[0023] Optional, calculate fra k and fra s The classification parameters between them include:

[0024] According to the set rules, from fra s Select the calculation window ws ;

[0025] Get fra k In the calculation window w s The corresponding calculation window w k ;

[0026] The function for calculating the classification parameters is:

[0027]

[0028] clapar(fra k ,fra s ) indicates fra k and fra s The classification parameters between them, δ represents the weighting parameters, δ∈(0,1), grad u Indicates w s The gradient of pixel u in the vertical direction. Indicates that pixel u is in w k The vertical gradient of the corresponding pixel in the image, where numw represents w. s The number of pixels in the image, cmpv represents the set average gradient difference, and mid(w s ) indicates w s The median value of the pixels in the matrix, mid(w k ) indicates w k The median of the pixel values ​​of the pixels in the image, where midcp represents the set median pixel value.

[0029] Optionally, illumination adjustment processing is performed on images in each set of images of the same type, including:

[0030] Use fra mi Mi represents the image with the smallest index in a set of images of the same type. mi The number;

[0031] For fra mi Perform image segmentation processing to obtain fra mi The area that belongs to substation equipment (bltequ) and the area that does not belong to substation equipment (nbltequ);

[0032] Perform adjustment unit partitioning on bltequ, dividing bltequ into multiple adjustment units and saving the adjustment units to the set adjset;

[0033] Calculate the adjustment parameters for each adjustment unit in adjset and save the adjustment parameters to the parst collection;

[0034] Use the adjustment parameters in parst to perform lighting adjustment on the corresponding adjustment units in adjset to obtain the desired effect on the fra mi The result of the illumination adjustment process;

[0035] For the image fra numbered j in the set of images of the same type j ,j≠mi, for fra j The process of adjusting light intensity includes:

[0036] In fra j In each pixel, the coordinates of the pixel corresponding to each adjustment unit in the adjset are obtained, thus obtaining the fra. j The adjustment unit in the middle will fra j The adjustment unit in the data is saved to the collection adjset j ;

[0037] Use the adjustment parameters in parser to adjust adjset j The corresponding adjustment unit performs illumination adjustment processing to obtain the desired effect on the fra. j The result of the illumination adjustment process.

[0038] Optionally, video fusion calculations are performed on the infrared video and the adjusted visible light video to obtain a fused video, including:

[0039] The following calculations are performed on images captured at the same time in both infrared video and adjusted visible light video to obtain a fused image:

[0040] Use infr t To represent an image captured at time t in infrared video, use visb. t This represents the image captured at time t in the adjusted visible light video;

[0041] visb t As a reference image, infr t As the image to be registered;

[0042] For infr t Perform image registration operations to obtain the registered image sinfr t ;

[0043] For visb t and sinfr t Perform image fusion calculations to obtain a fused image;

[0044] The fused video is composed of all the fused images.

[0045] Optional, for infr tThe algorithms for image registration include any one of the following: feature-based image registration algorithm, gray-level-based image registration algorithm, and transform domain-based image registration algorithm.

[0046] Optional, for visb t and sinfr t Perform image fusion calculations to obtain a fused image, including:

[0047] Using a weighted fusion method to analyze Visb t and sinfr t The pixels in the image are used to perform image fusion calculations to obtain a fused image.

[0048] In the process of video fusion, this invention first performs illumination adjustment processing on the images contained in the visible light video before fusion, which can effectively reduce the probability of some areas in the fused video being too bright or too dim, thereby improving the accuracy of fault analysis of substation equipment based on the fused video. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a schematic diagram of a substation equipment fault analysis system based on video fusion technology according to the present invention.

[0051] Figure 2 This is a schematic diagram illustrating the modified visible light video obtained by the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] like Figure 1 As shown in one embodiment, the present invention provides a substation equipment fault analysis system based on video fusion technology, including a camera module 101, a lighting adjustment module 102, a video fusion module 103, and a fault analysis module 104.

[0054] Camera module 101 is used to acquire visible light video and infrared video of the substation equipment to be analyzed for faults;

[0055] The illumination adjustment module 102 is used to perform illumination adjustment processing on the images contained in the visible light video to obtain the adjusted visible light video, such as... Figure 2 As shown:

[0056] Step 1: Obtain the first image contained in the visible light video;

[0057] Step two, use the first image as the base image;

[0058] Step 3: Obtain a set of images of the same type based on the base image;

[0059] Step four: Obtain the image with the largest index, fralst, from the set of images of the same type.

[0060] Step 5: If fralst is not the last image in the visible light video, then use fralst as the new base image and proceed to Step 3; if fralst is the last image in the visible light video, then output the set of all images of the same type and proceed to Step 6.

[0061] Step 6: Perform illumination adjustment processing on the images in each set of images of the same type, and combine all the images that have undergone illumination adjustment processing to form the adjusted visible light video;

[0062] The video fusion module 103 is used to perform video fusion calculations on infrared video and adjusted visible light video to obtain fused video;

[0063] The fault analysis module 104 is used to identify and process the fused video to determine whether a fault of a set type occurs in the fused video.

[0064] Furthermore, substation equipment may include transformers, surge arresters, current transformers, etc. For transformers, fault types may include oil leakage, smoke, etc. For surge arresters, fault types may include bushing rupture, fire, etc.

[0065] Optionally, camera module 101 includes an infrared camera and a visible light camera;

[0066] Infrared cameras are used to acquire infrared video of substation equipment to be analyzed for faults;

[0067] Visible light cameras are used to acquire visible light video of substation equipment to be analyzed for faults.

[0068] Infrared cameras can capture relatively clear edges of substation equipment even in poor lighting conditions, providing a foundation for subsequent video fusion and obtaining higher-quality fused video.

[0069] Optionally, a set of images of the same type can be obtained based on the base image, including:

[0070] Let s be the number of the base image, and fra be used to represent it. s Represents the basic image;

[0071] For image fra numbered k in visible light video k , k∈(s,N], calculate fra k and fra s If the classification parameter is less than the set parameter threshold, then fra will be... k Save to fra s In the corresponding set of images of the same type, if the classification parameter is greater than or equal to the set parameter threshold, then the image numbered s-1 in the visible light video is used as fra. s The last image in the set of images of the same type, where N represents the total number of images in the visible light video.

[0072] When multiple consecutive images in a visible light video are captured at the same angle and height, the differences between these images are relatively small, allowing them to be saved into a set of images of the same type. During subsequent illumination adjustment processing, images in the same set can share parameters for adjustment, thus significantly saving time in the illumination adjustment process.

[0073] Optional, calculate fra k and fra s The classification parameters between them include:

[0074] According to the set rules, from fra s Select the calculation window w s ;

[0075] Get fra k In the calculation window w s The corresponding calculation window w k ;

[0076] The function for calculating the classification parameters is:

[0077]

[0078] clapar(fra k ,fra s ) indicates fra k and fra sThe classification parameters between them, δ represents the weighting parameters, δ∈(0,1), grad u Indicates w s The gradient of pixel u in the vertical direction. Indicates that pixel u is in w k The vertical gradient of the corresponding pixel in the image, where numw represents w. s The number of pixels in the image, cmpv represents the set average gradient difference, and mid(w s ) indicates w s The median value of the pixels in the matrix, mid(w k ) indicates w k The median of the pixel values ​​of the pixels in the image, where midcp represents the set median pixel value.

[0079] In calculating fra k and fra s When calculating classification parameters, this invention uses a calculation window obtained from the base image, and all subsequent calculations are performed within this window. This method has two advantages. First, it significantly reduces the number of pixels involved in calculating classification parameters. If the difference between two images is very small, the difference between the calculation windows will also be very small. This means that this invention does not need to calculate classification parameters based on all pixels, thus improving the efficiency of calculating classification parameters. Second, images captured at similar times can share the calculation window in the base image, without having to repeatedly obtain the calculation window for each image, further improving the efficiency of calculating classification parameters.

[0080] When calculating classification parameters, a weighted calculation is performed on both the gradient of each pixel and the median pixel value, which improves the effectiveness of the classification parameters. The larger the average difference of the gradients and the larger the difference of the pixel medians, the greater the difference between the two calculation windows, and the greater the probability that the two images cannot be stored in the same set of images of the same type.

[0081] Optionally, based on the set rules, from fra s Select the calculation window w s ,include:

[0082] Using a segmentation algorithm to process fra s Image segmentation is performed to obtain multiple sub-region images;

[0083] Calculate the window coefficient for each sub-region image separately, and use the region containing the sub-region image with the largest window coefficient as the calculation window w. s ;

[0084] The calculation process for the window coefficient includes:

[0085] For fra sPerform calculations to obtain fra s Edge pixels in the middle;

[0086] Calculate the window coefficient for each sub-region image containing edge pixels:

[0087]

[0088] Where winfac represents the window coefficient, nofal represents the total number of pixels in the cut image, nofeg represents the number of edge pixels in the sub-region image, splig represents the set of pixels in the sub-region image, and gray represents the number of pixels in the cut image. v This represents the grayscale value of a pixel in the sub-region image, and stgray represents the standard value of the set grayscale value variance.

[0089] In selecting the calculation window, this invention does not select it randomly, but rather obtains fra s The window coefficients of the sub-region image are obtained. Before calculating the window coefficients, this invention first obtains the fra value through edge calculation. s The algorithm first identifies edge pixels in the image, and then calculates the window coefficient only for the sub-region containing these edge pixels. This avoids selecting sub-regions containing only background pixels as the calculation window. Since this invention calculates classification parameters based on a calculation window, if the calculation window only includes background pixels, then when the shooting angle changes (i.e., when the difference between two images is significant), analyzing the calculation windows of the two images will still yield smaller classification parameters, which is clearly not realistic.

[0090] When calculating the window coefficient, this invention calculates from two directions: the number of edge pixels and the variance of the gray values ​​of the pixels. It can select a sub-region image with a high proportion of edge pixels and a large variance between pixels, that is, a high probability that the pixels are not all background pixels, and then use the region where the sub-region image is located as the calculation window.

[0091] Optionally, a quadtree partitioning algorithm can be used to partition fra. s Perform image segmentation processing.

[0092] Optionally, illumination adjustment processing is performed on images in each set of images of the same type, including:

[0093] Use fra mi Mi represents the image with the smallest index in a set of images of the same type. mi The number;

[0094] For fra mi Perform image segmentation processing to obtain fra miThe area that belongs to substation equipment (bltequ) and the area that does not belong to substation equipment (nbltequ);

[0095] Perform adjustment unit partitioning on bltequ, dividing bltequ into multiple adjustment units and saving the adjustment units to the set adjset;

[0096] Calculate the adjustment parameters for each adjustment unit in adjset and save the adjustment parameters to the parst collection;

[0097] Use the adjustment parameters in parst to perform lighting adjustment on the corresponding adjustment units in adjset to obtain the desired effect on the fra mi The result of the illumination adjustment process;

[0098] For the image fra numbered j in the set of images of the same type j ,j≠mi, for fra j The process of adjusting light intensity includes:

[0099] In fra j In each pixel, the coordinates of the pixel corresponding to each adjustment unit in the adjset are obtained, thus obtaining the fra. j The adjustment unit in the middle will fra j The adjustment unit in the data is saved to the collection adjset j ;

[0100] Use the adjustment parameters in parser to adjust adjset j The corresponding adjustment unit performs illumination adjustment processing to obtain the desired effect on the fra. j The result of the illumination adjustment process.

[0101] In the illumination adjustment process, this invention first acquires the adjustment parameters of each adjustment unit in the image with the smallest number, i.e., the earliest captured image in the set of images of the same type. Then, based on the adjustment parameters, illumination adjustment processing is performed on the image with the smallest number and other images in the set of images of the same type. Since only one image needs to be processed to acquire the adjustment units and adjustment parameters during the adjustment process, and since the pixels in the bltequ are the pixels used for subsequent video recognition, the above embodiments of this invention do not perform adjustment processing on the pixels in the nbltequ. Therefore, this invention can significantly improve the speed of illumination adjustment processing.

[0102] The main advantage is that the differences between images in a set of similar images are relatively small. That is, multiple images may be taken from the same angle and at the same height. Therefore, these multiple images can actually be regarded as the same image. The same set of adjustment parameters and the results of adjustment unit division are used for adjustment, thereby improving the speed of illumination adjustment processing.

[0103] Optional, for fra mi Perform image segmentation processing to obtain fra mi The area bltequ that belongs to substation equipment and the area nbltequ that does not belong to substation equipment include:

[0104] Image segmentation algorithm is used to process fra mi Perform image segmentation processing to obtain fra mi The area that belongs to the substation equipment is bltequ, and the area that does not belong to the substation equipment is nbltequ.

[0105] Specifically, threshold-based image segmentation algorithms such as the OTSU algorithm can be used to segment fra. mi Perform image segmentation processing.

[0106] Optionally, the bltequ can be divided into multiple regulation units, including:

[0107] A non-overlapping sliding window of size K×K is used on bltequ, with each pixel in the sliding window acting as a pixel in an adjustment unit.

[0108] Specifically, for areas in BLTEE that cannot completely fill the sliding window, an adjustment unit is formed by only the pixels belonging to BLTEE within that area. Dividing the adjustment into multiple units avoids the inaccuracy issues caused by traditional global adjustments. Because the illumination distribution varies in different areas of an image, applying the same adjustment parameter to all pixels is clearly inappropriate.

[0109] Optionally, calculate the adjustment parameters for each adjustment unit in adjset separately, including:

[0110] Obtain the grayscale histogram S = {s1, ..., s2} of the adjustment unit. k ,…,s 255};s k The grayscale histogram value represents the grayscale value k.

[0111] Use the following function to calculate the adjustment parameters:

[0112]

[0113] sd This represents the grayscale histogram value of grayscale value d, Θ represents the set first enhancement coefficient, Ω represents the set second enhancement coefficient, and adjpar(g) represents the adjustment parameter of pixel value g.

[0114] Specifically, the grayscale histograms of each adjustment unit are not consistent. Therefore, each adjustment unit adopts an adaptive approach to obtain adjustment parameters, thereby improving the applicability of the adjustment parameters.

[0115] Optionally, the lighting can be adjusted using the adjustment parameters in parst on the corresponding adjustment units in adjset, including:

[0116] For the adjustment parameter adjpar(g) in parst, the adjustment unit corresponding to adjpar(g) in adjset is denoted as adjunt;

[0117] Use setau to save the pixels in adjunt.

[0118] For pixel value g in setau, the lighting adjustment process is as follows:

[0119]

[0120] agry(g) represents the pixel value of a pixel after applying lighting adjustment to the pixel with pixel value g.

[0121] Specifically, use the adjustment parameters in parser to adjust adjset j The process of performing illumination adjustment on the corresponding adjustment unit in the parst is the same as the process of performing illumination adjustment on the corresponding adjustment unit in adjset using the adjustment parameters in parst, except that the adjustment units involved in the calculation are different.

[0122] Optionally, video fusion calculations are performed on the infrared video and the adjusted visible light video to obtain a fused video, including:

[0123] The following calculations are performed on images captured at the same time in both infrared video and adjusted visible light video to obtain a fused image:

[0124] Use infr t To represent an image captured at time t in infrared video, use visb. t This represents the image captured at time t in the adjusted visible light video;

[0125] visb t As a reference image, infr t As the image to be registered;

[0126] For infrt Perform image registration operations to obtain the registered image sinfr t ;

[0127] For visb t and sinfr t Perform image fusion calculations to obtain a fused image;

[0128] The fused video is composed of all the fused images.

[0129] Optional, for infr t The algorithms for image registration include any one of the following: feature-based image registration algorithm, gray-level-based image registration algorithm, and transform domain-based image registration algorithm.

[0130] Feature-based image registration algorithms include BFMatcher, FLANN, and SANSAC.

[0131] Gray-scale-based image registration algorithms directly utilize the similarity of gray-scale indices between two images, determine the transformation parameters between the reference image and the image to be registered based on information within the image, and use a search method to find the maximum or minimum similarity point.

[0132] Optional, for visb t and sinfr t Perform image fusion calculations to obtain a fused image, including:

[0133] Using a weighted fusion method to analyze Visb t and sinfr t The pixels in the image are used to perform image fusion calculations to obtain a fused image.

[0134] In the process of video fusion, this invention first performs illumination adjustment processing on the images contained in the visible light video before fusion, which can effectively reduce the probability of some areas in the fused video being too bright or too dim, thereby improving the accuracy of fault analysis of substation equipment based on the fused video.

[0135] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A substation equipment fault analysis system based on video fusion technology, characterized in that, It includes a camera module, a lighting adjustment module, a video fusion module, and a fault analysis module; The camera module is used to acquire visible light and infrared video of the substation equipment to be analyzed for faults; The illumination adjustment module is used to perform illumination adjustment processing on the images contained in the visible light video to obtain the adjusted visible light video: Step 1: Obtain the first image contained in the visible light video; Step two, use the first image as the base image; Step 3: Obtain a set of images of the same type based on the base image; Step four: Obtain the image with the largest index, fralst, from the set of images of the same type. Step 5: If fralst is not the last image in the visible light video, then use fralst as the new base image and proceed to Step 3; if fralst is the last image in the visible light video, then output the set of all images of the same type and proceed to Step 6. Step 6: Perform illumination adjustment processing on the images in each set of images of the same type, and combine all the images that have undergone illumination adjustment processing to form the adjusted visible light video; The video fusion module is used to perform video fusion calculations on infrared video and adjusted visible light video to obtain fused video; The fault analysis module is used to identify and process the fused video to determine whether a fault of a specified type occurs in the fused video.

2. The substation equipment fault analysis system based on video fusion technology according to claim 1, characterized in that, The camera module includes infrared cameras and visible light cameras; Infrared cameras are used to acquire infrared video of substation equipment to be analyzed for faults; Visible light cameras are used to acquire visible light video of substation equipment to be analyzed for faults.

3. The substation equipment fault analysis system based on video fusion technology according to claim 1, characterized in that, A collection of images of the same type is obtained based on a base image, including: Let s be the number of the base image, and fra be used to represent it. s Represents the basic image; For image fra numbered k in visible light video k , k∈(s,N], calculate fra k and fra s If the classification parameter is less than the set parameter threshold, then fra will be... k Save to fra s In the corresponding set of images of the same type, if the classification parameter is greater than or equal to the set parameter threshold, then the image numbered s-1 in the visible light video is used as fra. s The last image in the set of images of the same type, where N represents the total number of images in the visible light video.

4. The substation equipment fault analysis system based on video fusion technology according to claim 3, characterized in that, Calculate fra k and fra s The classification parameters between them include: According to the set rules, from fra s Select the calculation window w s ; Get fra k In the calculation window w s The corresponding calculation window w k ; The function for calculating the classification parameters is: clapar(fra k ,fra s ) indicates fra k and fra s The classification parameters between them, δ represents the weighting parameters, δ∈(0,1), grad u Indicates w s The gradient of pixel u in the vertical direction, grad u,wk Indicates that pixel u is in w k The vertical gradient of the corresponding pixel in the image, where numw represents w. s The number of pixels in the image, cmpv represents the set average gradient difference, and mid(w s ) indicates w s The median value of the pixels in the matrix, mid(w k ) indicates w k The median of the pixel values ​​of the pixels in the image, where midcp represents the set median pixel value.

5. A substation equipment fault analysis system based on video fusion technology according to claim 1, characterized in that, The illumination adjustment process is performed on the images in each set of images of the same type, including: Use fra mi Mi represents the image with the smallest index in a set of images of the same type. mi The number; For fra mi Perform image segmentation processing to obtain fra mi The area that belongs to substation equipment (bltequ) and the area that does not belong to substation equipment (nbltequ); Perform adjustment unit partitioning on bltequ, dividing bltequ into multiple adjustment units and saving the adjustment units to the set adjset; Calculate the adjustment parameters for each adjustment unit in adjset and save the adjustment parameters to the parst collection; Use the adjustment parameters in parst to perform lighting adjustment on the corresponding adjustment units in adjset to obtain the desired effect on the fra mi The result of the illumination adjustment process; For the image fra numbered j in the set of images of the same type j ,j≠mi, for fra j The process of adjusting light intensity includes: In fra j In each pixel, the coordinates of the pixel corresponding to each adjustment unit in the adjset are obtained, thus obtaining the fra. j The adjustment unit in the middle will fra j The adjustment unit in the data is saved to the collection adjset j ; Use the adjustment parameters in parser to adjust adjset j The corresponding adjustment unit performs illumination adjustment processing to obtain the desired effect on the fra. j The result of the illumination adjustment process.

6. The substation equipment fault analysis system based on video fusion technology according to claim 1, characterized in that, Video fusion calculations are performed on the infrared video and the adjusted visible light video to obtain the fused video, including: The following calculations are performed on images captured at the same time in both infrared video and adjusted visible light video to obtain a fused image: Use infr t To represent an image captured at time t in infrared video, use visb. t This represents the image captured at time t in the adjusted visible light video; visb t As a reference image, infr t As the image to be registered; For infr t Perform image registration operations to obtain the registered image sinfr t ; For visb t and sinfr t Perform image fusion calculations to obtain a fused image; The fused video is composed of all the fused images.

7. A substation equipment fault analysis system based on video fusion technology according to claim 6, characterized in that, For infr t The algorithms for image registration include any one of the following: feature-based image registration algorithm, gray-scale-based image registration algorithm, and transform domain-based image registration algorithm.

8. A substation equipment fault analysis system based on video fusion technology according to claim 6, characterized in that, For visb t and sinfr t Perform image fusion calculations to obtain a fused image, including: Using a weighted fusion method to analyze Visb t and sinfr t The pixels in the image are used to perform image fusion calculations to obtain a fused image.

Images