Method and system for identifying abnormality of electrical components in a meter box

Abstract

The present invention discloses a method and system for identifying abnormalities in electrical components within a meter box. The method comprises: determining whether each infrared image in a set of infrared images is associated with each other; if an infrared image does not have an associated relationship, segmenting the infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the infrared image; searching for other infrared images corresponding to the at least one infrared sub-image in the set of infrared images; splicing other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the infrared image; and fusing the target visible light image with the infrared image to obtain a target mixed image. This method effectively expands the application scenarios for identifying abnormalities in meter boxes based on dual spectrums and can improve the accuracy of identifying abnormalities in electrical components.

Description

Technical Field

[0001] The present invention belongs to the technical field of meter box abnormality identification, and in particular relates to a method and system for identifying abnormalities of electrical components in a meter box. Background Art

[0002] The meter box is used to detect the power data of electrical appliances. When the meter box is operating normally, the detected power data is consistent with the actual power data, so it can convey accurate power data to people. When the meter box is operating abnormally, the collected power data deviates from the actual power data and cannot convey accurate power data to people. Therefore, it is necessary to monitor the electrical components in the meter box to detect any abnormalities in a timely manner to avoid large losses.

[0003] In the existing technology, by acquiring visible light images and infrared images, fusing the two, and then identifying abnormal temperatures on the fused images, high-precision temperature recognition can be achieved compared to the traditional single infrared image recognition method. For example, a dual-spectral camera is installed in the meter box, and the visible light image and infrared image in the meter box are respectively collected by the dual-spectral camera, and subsequent identification is performed. It has the advantages of non-contact, high precision, real-time monitoring, simple installation and maintenance, etc., and can effectively improve the accuracy and real-time performance of temperature monitoring of electrical components in the meter box.

[0004] However, when monitoring multiple meter boxes simultaneously, if the visible light images of one or several meter boxes are missing due to obstruction or partial damage to the equipment, the visible light image and infrared image cannot be fused, resulting in insufficient subsequent recognition accuracy. Summary of the Invention

[0005] The present invention provides a method and system for identifying abnormalities in electrical components in a meter box, which are used to solve the technical problem that the fusion of visible light images and infrared images cannot be performed due to the loss of visible light images, resulting in insufficient subsequent identification accuracy.

[0006] In a first aspect, the present invention provides a method for identifying abnormalities in electrical components in a meter box, comprising:

[0007] Obtaining visible light images and infrared images of multiple meter boxes, and correlating the visible light image and the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object;

[0008] Preprocessing each infrared image, and clustering each preprocessed infrared image according to a preset division strategy to obtain at least one infrared image set;

[0009] Determine whether all infrared images in a certain infrared image set have a correlation relationship;

[0010] If there is no association relationship for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set;

[0011] splicing other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fusing the target visible light image with the certain infrared image to obtain a target mixed image;

[0012] The target mixed image is input into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

[0013] In a second aspect, the present invention provides a system for identifying abnormalities of electrical components in a meter box, comprising:

[0014] an acquisition module configured to acquire visible light images and infrared images of a plurality of meter boxes, and associate the visible light image with the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object;

[0015] a clustering module configured to pre-process each infrared image and cluster the pre-processed infrared images according to a preset division strategy to obtain at least one infrared image set;

[0016] A judgment module configured to judge whether all infrared images in a certain infrared image set have a correlation relationship;

[0017] a segmentation module configured to, if no association relationship exists for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set;

[0018] a splicing module configured to splice other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and to fuse the target visible light image with the certain infrared image to obtain a target mixed image;

[0019] The output module is configured to input the target mixed image into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

[0020] In a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for identifying abnormalities of electrical components in a meter box of any embodiment of the present invention.

[0021] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the method for identifying abnormalities of electrical components in a meter box of any embodiment of the present invention.

[0022] The method and system for identifying abnormalities of electrical components in a meter box of the present application, if there is no association relationship for a certain infrared image, then the certain infrared image is segmented based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and other infrared images corresponding to the at least one infrared sub-image are searched in a certain infrared image set, and other visible light images associated with other infrared images are spliced ​​to obtain a target visible light image corresponding to the certain infrared image, and the target visible light image is fused with a certain infrared image to obtain a target mixed image. The target mixed image can be obtained even in the absence of a single visible light image or several visible light images, thereby effectively increasing the application scenarios of meter box abnormality identification based on dual spectrum, and can improve the accuracy of electrical component abnormality identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 A flowchart of a method for identifying abnormalities in electrical components in a meter box provided by one embodiment of the present invention;

[0025] Figure 2 This is a structural block diagram of a system for identifying abnormalities in electrical components in a meter box provided by one embodiment of the present invention;

[0026] Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0028] See also Figure 1 , which shows a flow chart of a method for identifying abnormalities of electrical components in a meter box of the present application.

[0029] like Figure 1 As shown, the method for identifying abnormalities of electrical components in a meter box specifically includes the following steps:

[0030] Step S101: Obtain visible light images and infrared images in multiple meter boxes, and associate the visible light image and infrared image of the same meter box to obtain an association relationship, wherein both the visible light image and the infrared image contain at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object.

[0031] In this step, a first marking frame is assigned to each component object in a certain visible light image to obtain a target visible light image, and a second marking frame is assigned to each component object in a certain infrared image after grayscale processing to obtain a target grayscale image; a two-dimensional coordinate system is constructed, and the target visible light image and the target grayscale image are respectively imported into the two-dimensional coordinate system to obtain the first coordinate information of the center point of each first marking frame and the second coordinate information of the center point of each second marking frame; it is determined whether the difference between a certain first coordinate and a certain second coordinate is greater than a preset threshold; if it is greater than the preset threshold, the certain first marking frame is not associated with the certain second marking frame; if it is not greater than the preset threshold, the certain first marking frame is associated with the certain second marking frame to obtain a certain association sub-relationship, and all the first coordinate information and all the second coordinate information are traversed to obtain the association relationship between the certain visible light image and the certain infrared image.

[0032] It should be noted that when a visible light image is input into a trained convolutional neural network model, the convolutional neural network model can label each component object in a visible light image, thereby outputting a target visible light image, wherein the target visible light image contains at least one first marking box. Furthermore, when a grayscale infrared image is input into a trained VGGNet network model, the VGGNet network model can label each component object in a certain infrared image, thereby outputting a target grayscale image, wherein the target grayscale image contains at least one second marking box. Both the convolutional neural network model and the VGGNet network model are obtained through machine learning using multiple sets of training images and detection images.

[0033] Specifically, the target visible light image and target grayscale image are imported into a two-dimensional coordinate system, and the lower left corner vertices of the target visible light image and target grayscale image are aligned with the coordinate origin. This allows the coordinates of the center points of each first marked frame in the target visible light image and the coordinates of the center points of each second marked frame in the target grayscale image to be obtained. If the difference between a first coordinate and a second coordinate is greater than a preset threshold, it indicates that the component objects corresponding to the first marked frame and the second marked frame are most likely different, and therefore the first marked frame is not associated with the second marked frame.

[0034] Step S102 : pre-processing each infrared image, and clustering each pre-processed infrared image according to a preset division strategy to obtain at least one infrared image set.

[0035] In this step, image enhancement processing is performed on each infrared image, and the processed infrared images are classified according to the type of meter box. All infrared images of a certain category are sorted based on the location information of each meter box on a preset electronic map to obtain a certain infrared image sequence; a preset first sliding window is used to slide on a certain infrared image sequence, and the infrared images obtained by each sliding are clustered to obtain at least one infrared image set.

[0036] In this embodiment, since the internal electrical component structure of the same meter box type may be the same, the infrared images of the same meter box type are clustered, and based on the installation distance of the meter box, the infrared images of the meter boxes with the same internal electrical component structure can be clustered as accurately as possible, thereby facilitating the subsequent acquisition of visible light images.

[0037] Step S103 : determining whether all infrared images in a certain infrared image set have an associated relationship.

[0038] In a specific embodiment, after determining whether all infrared images in a certain infrared image set are associated, if an infrared image is associated, the infrared image is directly fused with a corresponding visible light image to obtain a target mixed image.

[0039] Step S104: if there is no association relationship for a certain infrared image, the infrared image is segmented based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and other infrared images corresponding to the at least one infrared sub-image are searched in the certain infrared image set.

[0040] In this step, an infrared image is converted into a grayscale image, and a second marking frame is assigned to each component object in the grayscale image; a preset second sliding window is slid along the edge of the second marking frame, and the area covered by the second sliding window is fused with the second marking frame to obtain a target area; the target area is segmented to obtain an infrared sub-image corresponding to the infrared image, wherein the infrared sub-image only contains the target area.

[0041] In this embodiment, the second sliding window is slid along the edge of a second marking frame, and the area covered by the second sliding window is merged with the second marking frame to obtain the target area, so that more pixel features can be obtained, which is convenient for reflecting the difference in subsequent similarity.

[0042] Furthermore, the similarity between a certain infrared sub-image and other infrared images in a certain infrared image set is calculated; and another infrared image with the greatest similarity is selected as the infrared image corresponding to the certain infrared sub-image.

[0043] Step S105 , stitching other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fusing the target visible light image with the certain infrared image to obtain a target mixed image.

[0044] In this step, the visible light image corresponding to each infrared sub-image is obtained according to the associated sub-relationship, and it is determined whether the difference between the number of the same visible light images and the total number of visible light images corresponding to each infrared sub-image is greater than a preset number threshold; if it is greater than the preset number threshold, a certain visible light image is directly defined as the target visible light image; if it is not greater than the preset number threshold, the other visible light images are cropped based on the associated sub-relationship, and the cropped first marking frames are fused into the visible light image to be processed based on the two-dimensional coordinate system, wherein the visible light image to be processed is: the visible light image with the largest number of the same visible light images corresponding to each infrared sub-image.

[0045] Step S106: input the target mixed image into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

[0046] In this step, a convolutional neural network is iteratively trained based on the target mixed image to produce an image recognition model. The convolutional neural network architecture consists of components such as convolutional layers, activation layers, pooling layers, and fully connected layers. During training, deep features are extracted from both visible and infrared images. The features are then fused in the pooling layer, and the fully connected layer outputs the object recognition results.

[0047] In summary, the method of the present application, if there is no association relationship for a certain infrared image, then the certain infrared image is segmented based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and other infrared images corresponding to the at least one infrared sub-image are searched in a certain infrared image set, and other visible light images associated with other infrared images are spliced ​​to obtain a target visible light image corresponding to a certain infrared image, and the target visible light image is fused with a certain infrared image to obtain a target mixed image. The target mixed image can be obtained even in the absence of a single visible light image or several visible light images, thereby effectively increasing the application scenarios of meter box anomaly identification based on dual spectrum, and can improve the accuracy of electrical component anomaly identification.

[0048] See also Figure 2 , which shows a structural block diagram of a system for identifying abnormalities of electrical components in a meter box of the present application.

[0049] like Figure 2 As shown, the electrical component abnormality identification system 200 in the meter box includes an acquisition module 210, a clustering module 220, a judgment module 230, a segmentation module 240, a splicing module 250 and an output module 260.

[0050] Among them, the acquisition module 210 is configured to acquire visible light images and infrared images in multiple meter boxes, and associate the visible light image and infrared image of the same meter box to obtain an association relationship, wherein the visible light image and the infrared image each contain at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object; the clustering module 220 is configured to pre-process each infrared image, and cluster each pre-processed infrared image according to a preset division strategy to obtain at least one infrared image set; the judgment module 230 is configured to judge whether each infrared image in a certain infrared image set has an association relationship; the segmentation module 240 is configured to determine whether a certain infrared image If there is no association relationship, the infrared image is segmented based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the infrared image, and other infrared images corresponding to the at least one infrared sub-image are searched in the infrared image set; the splicing module 250 is configured to splice other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the infrared image, and fuse the target visible light image with the infrared image to obtain a target mixed image; the output module 260 is configured to input the target mixed image into a preset image recognition model, and the image recognition model outputs the abnormal component object in the target mixed image.

[0051] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects also apply to Figure 2 The modules in it will not be described in detail here.

[0052] In other embodiments, embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the method for identifying abnormalities of electrical components in a meter box in any of the above method embodiments;

[0053] As an embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0054] Obtaining visible light images and infrared images of multiple meter boxes, and correlating the visible light image and the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object;

[0055] Preprocessing each infrared image, and clustering each preprocessed infrared image according to a preset division strategy to obtain at least one infrared image set;

[0056] Determine whether all infrared images in a certain infrared image set have a correlation relationship;

[0057] If there is no association relationship for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set;

[0058] splicing other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fusing the target visible light image with the certain infrared image to obtain a target mixed image;

[0059] The target mixed image is input into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

[0060] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the electrical component anomaly identification system in the meter box, etc. In addition, the computer-readable storage medium may include a high-speed random access memory and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the electrical component anomaly identification system in the meter box via a network. Examples of the aforementioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0061] Figure 3 Schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3The example of a bus connection is shown. Memory 320 is the aforementioned computer-readable storage medium. Processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in memory 320, thereby implementing the above-mentioned method embodiment for identifying abnormalities in electrical components within a meter box. Input device 330 can receive input digital or character information and generate key signal input related to user settings and function control of the system for identifying abnormalities in electrical components within a meter box. Output device 340 may include a display device such as a display screen.

[0062] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0063] As an embodiment, the electronic device is applied to a system for identifying abnormalities in electrical components in a meter box, and is used for a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0064] Obtaining visible light images and infrared images of multiple meter boxes, and correlating the visible light image and the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object;

[0065] Preprocessing each infrared image, and clustering each preprocessed infrared image according to a preset division strategy to obtain at least one infrared image set;

[0066] Determine whether all infrared images in a certain infrared image set have a correlation relationship;

[0067] If there is no association relationship for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set;

[0068] splicing other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fusing the target visible light image with the certain infrared image to obtain a target mixed image;

[0069] The target mixed image is input into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

[0070] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods of each embodiment or certain portions of the embodiments.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for identifying abnormalities in electrical components in a meter box, characterized in that: include: Obtaining visible light images and infrared images of multiple meter boxes, and associating the visible light image and the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object, and associating the visible light image and the infrared image of the same meter box to obtain the association relationship includes: Assigning a first marking frame to each component object in a visible light image to obtain a target visible light image, and assigning a second marking frame to each component object in a grayscale processed infrared image to obtain a target grayscale image; Constructing a two-dimensional coordinate system, and importing the target visible light image and the target grayscale image into the two-dimensional coordinate system respectively, to obtain first coordinate information of the center point of each first marking frame, and second coordinate information of the center point of each second marking frame; Determine whether the difference between a first coordinate and a second coordinate is greater than a preset threshold; If it is greater than a preset threshold, the first marking box is not associated with the second marking box; If it is not greater than a preset threshold, a first marked frame is associated with a second marked frame to obtain a certain associated sub-relationship, and all first coordinate information and all second coordinate information are traversed to obtain the associated relationship between the certain visible light image and the certain infrared image; Preprocessing each infrared image, and clustering each preprocessed infrared image according to a preset division strategy to obtain at least one infrared image set; Determine whether all infrared images in a certain infrared image set have a correlation relationship; If there is no association relationship for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set, wherein searching for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set includes: Calculating similarity between a certain infrared sub-image and other infrared images in the certain infrared image set; Selecting another infrared image with the greatest similarity as the infrared image corresponding to the certain infrared sub-image; Stitching the other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fusing the target visible light image with the certain infrared image to obtain a target mixed image, wherein the stitching of the other visible light images associated with the other infrared images to obtain the target visible light image corresponding to the certain infrared image includes: Obtaining visible light images corresponding to the respective infrared sub-images according to the associated sub-relationships, and determining whether a difference between the number of identical visible light images and the total number of visible light images corresponding to the respective infrared sub-images is greater than a preset number threshold; If the number is greater than a preset threshold, the certain visible light image is directly defined as a target visible light image; If the number is not greater than a preset threshold, the other visible light images are cropped based on the associated sub-relationship, and each first marking frame obtained by cropping is fused into the visible light image to be processed based on a two-dimensional coordinate system, wherein the visible light image to be processed is: the visible light image with the largest number of the same number among the visible light images corresponding to the infrared sub-images; The target mixed image is input into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

2. A method for identifying abnormalities of electrical components in a meter box according to claim 1, characterized in that: The method of clustering the pre-processed infrared images according to the preset division strategy to obtain at least one infrared image set includes: Classify each infrared image according to the type of meter box, and sort all infrared images of a certain category based on the location information of each meter box on a preset electronic map to obtain a certain infrared image sequence; A preset first sliding window is used to slide on the certain infrared image sequence, and each infrared image obtained by each sliding is clustered to obtain at least one infrared image set.

3. The method for identifying abnormalities of electrical components in a meter box according to claim 1, characterized in that: After determining whether all infrared images in a certain infrared image set are associated with each other, the method further includes: If a certain infrared image has a correlation relationship, the infrared image is directly fused with a corresponding visible light image to obtain a target mixed image.

4. The method for identifying abnormalities of electrical components in a meter box according to claim 1, characterized in that: The step of segmenting the infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the infrared image includes: converting the infrared image into a grayscale image, and assigning a second marking frame to each component object in the grayscale image; Sliding a preset second sliding window along an edge of a second marking frame, and fusing an area covered by the second sliding window with the second marking frame to obtain a target area; The target area is segmented to obtain an infrared sub-image corresponding to the infrared image, wherein the infrared sub-image only includes the target area.

5. A system for identifying abnormalities of electrical components in a meter box, characterized in that: include: The acquisition module is configured to acquire visible light images and infrared images of multiple meter boxes, and associate the visible light image and the infrared image of the same meter box to obtain an association relationship, wherein each of the visible light image and the infrared image contains at least one component object, and the association relationship contains an association sub-relationship corresponding to the at least one component object, and the association of the visible light image and the infrared image of the same meter box to obtain the association relationship includes: Assigning a first marking frame to each component object in a visible light image to obtain a target visible light image, and assigning a second marking frame to each component object in a grayscale processed infrared image to obtain a target grayscale image; Constructing a two-dimensional coordinate system, and importing the target visible light image and the target grayscale image into the two-dimensional coordinate system respectively, to obtain first coordinate information of the center point of each first marking frame, and second coordinate information of the center point of each second marking frame; Determine whether the difference between a first coordinate and a second coordinate is greater than a preset threshold; If it is greater than a preset threshold, the first marking box is not associated with the second marking box; If it is not greater than a preset threshold, a first marked frame is associated with a second marked frame to obtain a certain associated sub-relationship, and all first coordinate information and all second coordinate information are traversed to obtain the associated relationship between the certain visible light image and the certain infrared image; a clustering module configured to pre-process each infrared image and cluster the pre-processed infrared images according to a preset division strategy to obtain at least one infrared image set; A judgment module configured to judge whether all infrared images in a certain infrared image set have a correlation relationship; The segmentation module is configured to, if no association relationship exists for a certain infrared image, segment the certain infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the certain infrared image, and search for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set, wherein searching for other infrared images corresponding to the at least one infrared sub-image in the certain infrared image set includes: Calculating similarity between a certain infrared sub-image and other infrared images in the certain infrared image set; Selecting another infrared image with the greatest similarity as the infrared image corresponding to the certain infrared sub-image; a stitching module configured to stitch the other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the certain infrared image, and fuse the target visible light image with the certain infrared image to obtain a target mixed image, wherein stitching the other visible light images associated with the other infrared images to obtain the target visible light image corresponding to the certain infrared image includes: Obtaining visible light images corresponding to the respective infrared sub-images according to the associated sub-relationships, and determining whether a difference between the number of identical visible light images and the total number of visible light images corresponding to the respective infrared sub-images is greater than a preset number threshold; If the number is greater than a preset threshold, the certain visible light image is directly defined as a target visible light image; If the number is not greater than a preset threshold, the other visible light images are cropped based on the associated sub-relationship, and each first marking frame obtained by cropping is fused into the visible light image to be processed based on a two-dimensional coordinate system, wherein the visible light image to be processed is: the visible light image with the largest number of the same number among the visible light images corresponding to the infrared sub-images; The output module is configured to input the target mixed image into a preset image recognition model, and the image recognition model outputs abnormal component objects in the target mixed image.

6. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images