Electric spark monitoring method and system of power distribution cabinet and readable storage medium

Abstract

The invention discloses an electric spark monitoring method and system for a power distribution cabinet and a readable storage medium, and the method comprises the steps: intercepting a to-be-recognized image from a video stream, carrying out the brightness comparison of the to-be-recognized image and a preset standard image, if the brightness of the to-be-recognized image changes, obtaining a plurality of continuous frames of images of the to-be-recognized image in a preset time period before and after the to-be-recognized image, storing the image into a spark identification image set; performing light-emitting area identification on the to-be-identified image with the changed brightness to obtain an identification frame of the light-emitting area, and intercepting each frame of image in the spark identification image set according to the identification frame; and carrying out brightness identification on the intercepted local image, fitting a brightness change curve according to the brightness identification result of each local image, judging the cause of brightness change according to the brightness change curve, and determining whether to carry out electric spark early warning or not according to the judgment result. Whether electric sparks are generated or not is helped to be judged according to the change trend of the brightness change curve, so that the influence of an interference source on electric spark monitoring is eliminated.

Description

Technical Field

[0001] The present invention relates to the technical field of computer data processing, and particularly to a method, a system and a readable storage medium for monitoring electric sparks in a power distribution cabinet. Background Art

[0002] Since electronic devices such as circuit breakers and leakage protection switches are installed inside the power distribution cabinet. Especially the circuit breaker, as the core component of the power distribution cabinet, is used to connect and disconnect the circuit, and quickly cut off the circuit in case of faults such as short circuit and overload to provide reliable protection. As time goes by, switching devices such as circuit breakers are prone to generate electric sparks during the switching process, which may cause a fire risk. For the current monitoring of electric sparks, the mainstream methods currently include the electric spark detection method based on the single-chip microcomputer algorithm, the electric spark detection method based on electromagnetic induction, and the electric spark monitoring method through image modeling, etc.

[0003] When monitoring electric sparks in the above manner, it only preliminarily judges whether electric sparks are generated, but for the brightness changes inside the power distribution cabinet caused by some external interference sources such as opening the door and flashlight light, etc., the brightness of the picture will change instantaneously, and it is easy to misjudge electric sparks. Summary of the Invention

[0004] The purpose of the present invention is to provide a method, a system and a readable storage medium for monitoring electric sparks in a power distribution cabinet. By summarizing the images with brightness changes and drawing a brightness change curve, the change trend of the brightness change curve is used to help assist in judging whether electric sparks are generated, so as to eliminate the influence of interference sources on the monitoring of electric sparks.

[0005] To achieve the above purpose, the present application provides the following solutions: On the one hand, the present invention provides a method for monitoring electric sparks in a power distribution cabinet, which specifically includes the following steps: S1. Obtain the video stream collected in real time by the monitoring device inside the power distribution cabinet, and intercept the images to be recognized from the video stream at a preset time interval; S2. Compare the brightness of the image to be recognized with the preset standard image. If the brightness of the image to be recognized changes, obtain a continuous number of frames of images within a preset time period before and after the image to be recognized with the changed brightness as the starting point, and store them in the spark recognition image set; S3. Identify the light-emitting area of the image to be recognized with the changed brightness to obtain the recognition frame of the light-emitting area, and intercept each frame of the image in the spark recognition image set according to the recognition frame to obtain the local image marked by the recognition frame in each frame of the image; S4. Perform brightness recognition on the local images, fit a brightness change curve according to the brightness recognition results of each local image, judge the reason for the brightness change according to the brightness change curve, and decide whether to give an electric spark warning according to the judgment result.

[0006] In some specific embodiments, in step S3, the image to be recognized is evenly divided, each frame of the image is segmented into n recognition regions of the same size, and the n recognition regions are arranged in an i×j matrix.

[0007] In some specific embodiments, the specific process of step S3 includes: The image to be recognized is segmented into n recognition regions, and each row is traversed in a window moving and traversing manner. For the j recognition regions in each row, the following steps are performed: Obtain a number of consecutive recognition regions according to the set window length; Calculate the brightness values of a number of consecutive recognition regions respectively, draw a curve according to the brightness values of the number of consecutive recognition regions. If the curve conforms to the trend of changing from small to large and then to small, the number of consecutive recognition regions is used as the first light-emitting region; Move the window according to the set window moving step value to obtain the next number of consecutive recognition regions until all the j recognition regions in the current row are traversed; Summarize the first light-emitting regions obtained from each row to obtain the light-emitting region.

[0008] In some specific embodiments, the specific process of step S3 includes: S31. Take the recognition region at the center position of the matrix as the central region; S32. Starting from the central region, compare the brightness value of the central region with the brightness values of its adjacent recognition regions. When there is a recognition region whose brightness value is greater than or equal to the brightness value of the central region, the recognition region is used as the central region to execute step S33; S33. Repeat step S32 until all the recognition regions are traversed, gather all the recognition regions used as the central region, and select the recognition region with the highest brightness value in the central region and its surrounding recognition regions as the light-emitting region.

[0009] In some specific embodiments, taking the recognition region at the center position of the matrix as the central region, starting from the central region, compare the brightness value of the central region with the brightness values of its surrounding recognition regions. When the brightness value of the central region is higher than the brightness values of all its surrounding recognition regions, continue to judge whether the brightness value of each recognition region is greater than the brightness value at the corresponding position in the standard image. If so, the recognition region and the central region are used as the light-emitting region together.

[0010] In some specific embodiments, the specific process of step S4 is: Find the peak value of the brightness change curve. When the change trend of the brightness change curve is a jump and there are multiple peak values within a preset time period, judge that the reason for the brightness change is an electric spark; When the change trend of the brightness change curve shows a slow increase first, then remains constant for a certain preset time and then gradually decreases within the preset time period, or when the change trend of the brightness change curve within the preset time period is to jump to the brightness peak and then remain unchanged, it is determined that the cause of the brightness change is an interference source.

[0011] In some specific implementation manners, when it is determined that the cause of the brightness change is an electric spark, the following steps are executed: Count the number of occurrences of the brightness peak of the brightness change curve. If the number of occurrences of the brightness peak within the preset time period is less than the preset threshold, an early warning of electric spark generation is output; When the number of occurrences of the brightness peak within the preset time period is greater than the preset threshold and the time between two adjacent occurrences of the brightness peak is less than the set value, an emergency maintenance warning is given.

[0012] In some specific implementation manners, the preset time period is the continuous lighting time of the electric spark interference source.

[0013] In a second aspect, the present application provides an electric spark monitoring system for a power distribution cabinet, which applies the electric spark monitoring method for a power distribution cabinet described in the first aspect, and includes: A data acquisition module, configured to obtain the video stream collected in real time by the monitoring device in the power distribution cabinet, and intercept a plurality of frames of images to be recognized from the video stream at preset time intervals; A trigger module, configured to compare the brightness of the image to be recognized with a preset standard image. If the brightness of the image to be recognized changes, a continuous plurality of frames of images within a preset time period before and after the image to be recognized with the changed brightness are obtained and stored in the spark recognition image set; An image intercepting module, which recognizes the light-emitting area of the image to be recognized with the changed brightness to obtain the recognition frame of the light-emitting area, and intercepts each frame of the image in the spark recognition image set according to the recognition frame to obtain the local image marked by the recognition frame in each frame of the image; A spark judgment module, configured to perform brightness recognition on the local images, fit the brightness change curve according to the brightness recognition results of the local images, determine the cause of the brightness change according to the brightness change curve, and decide whether to give an electric spark warning according to the judgment result.

[0014] In a third aspect, the present application provides a computer-readable storage medium, including: One or more processors; A storage unit, configured to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors can implement an electric spark monitoring method for a power distribution cabinet as described in the first aspect.

[0015] The beneficial effects of the present invention are: The present invention quickly identifies the changed image, obtains the image set composed of the images before and after the changed image, performs brightness recognition on each image in the image set, fits a brightness change curve based on the brightness recognition results, determines the scene that the change trend of the brightness change curve conforms to, thereby differentiates the reasons for the change in image brightness, and thus differentiates the electric spark and other interference sources to avoid false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a method for monitoring electric sparks in a power distribution cabinet provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the brightness change curve of the electric spark provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the brightness change curve of the interference source provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0018] Unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and values set forth in these embodiments do not limit the scope of the present invention.

[0019] At the same time, it should be understood that for the sake of description, the dimensions of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0020] In addition, for the sake of clarity and conciseness, the descriptions of well-known structures, functions and configurations may be omitted. Those of ordinary skill in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0021] The techniques, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods and devices should be regarded as part of the authorization specification.

[0022] In all the examples shown and discussed here, any specific value should be construed as merely exemplary, not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0023] Example 1 As Figure 1 shown, this embodiment provides a method for monitoring electric sparks in a distribution cabinet, which specifically includes the following steps: S1. Obtain the video stream collected in real time by the monitoring device in the distribution cabinet, and intercept the images to be recognized from the video stream at preset time intervals; Specifically, since the distribution cabinet is generally sealed, which is equivalent to a darkroom and is very sensitive to light, therefore, the monitoring video of the monitoring device in the distribution cabinet is obtained for image frame analysis to monitor the change in brightness in the distribution cabinet. When it is detected that the continuous image frames in the distribution cabinet change, it is determined that an electric spark may have occurred. The video segments within the preset time periods before and after the occurrence of the continuous frames are obtained for further analysis to exclude the interference of other interference sources. Here, other interference sources may be situations such as suddenly opening the cabinet door, the indicator light turning on, and the light penetrating through the observation window of the distribution cabinet, which may all cause the change in brightness in the distribution cabinet. The difference between an electric spark and an interference source is that an electric spark will instantaneously release strong light when it is generated. The release cycle of an electric spark can be roughly divided into four stages: dielectric breakdown, spark generation, spark maintenance, and spark extinction; the characteristic presented in the video stream is that the brightness of a certain area changes within an extremely short time, while the brightness of an interference source lasts for a relatively long time, generally greater than 1 s, and the brightness change is relatively uniform, showing a rule of slowly rising, maintaining for a period of time after reaching the maximum brightness and then going out. Therefore, the images with brightness changes can be found first by the method of skipping frames. Skipping frames can be every other frame or multiple frames, and the time interval of skipping frames can be set according to the longest time of the electric spark release cycle to avoid missing and quickly finding the images with brightness changes.

[0024] S2. Compare the brightness of the image to be recognized with the preset standard image. If the brightness of the image to be recognized changes, obtain a continuous number of frames of images within the preset time periods before and after the image to be recognized with the changed brightness as the starting point, and store them in the spark recognition image set; The standard image referred to in this embodiment can be an image taken in the distribution cabinet under normal circumstances. For example, it can be compared with the images in the video stream for a long period of time. If the images do not change for a long period of time, the images within this period can be used as the standard image. Or the standard image can be the previous frame of the current image to be recognized as the comparison image. When comparing the brightness of the current image to be recognized with the standard image, it can be to compare the average brightness of the two images. If the change range of the average brightness is within the set range, it indicates that the brightness has changed. And in order to reflect the brightness change of the interference source to the greatest extent, the preset time period is set as the continuous lighting time of the electric spark interference source (which can be measured or set artificially). In this way, by statistically analyzing the brightness change situations of the images before and after the image to be recognized with the changed brightness, the brightness change curve within a period of time can be fitted.

[0025] S3, identifying the luminous area of ​​the image to be identified with changed brightness, obtaining an identification frame of the luminous area, intercepting each frame of the spark identification image set according to the identification frame, and obtaining a local image marked by the identification frame in each frame of the image; Since the changes caused by sparks or other interference sources are changes within a certain area, the luminous area can be divided first, so as to obtain the brightness change value of the luminous area in each frame of the image and obtain a more accurate brightness change curve.

[0026] Specifically, if it is necessary to locate the luminous area, since the distribution cabinet is a closed and dark environment, when a light source appears, it will illuminate a certain area. Therefore, the image to be recognized is evenly divided, and each frame of the image is divided into n recognition areas of the same size. The n recognition areas are arranged into an i×j matrix, so as to quickly locate the luminous area. The i×j matrix can be expressed as: , the specific process can be: S01, dividing the image to be identified into n identification areas, traversing each row by window moving traversal, and for the j identification areas in each row, taking the first row as an example, performing the following steps: S011. Obtain several continuous recognition areas A according to the set window length w 1m， …, A 1(m +w) ; S012, respectively calculate the brightness values ​​of several continuous recognition areas, draw a curve according to the brightness values ​​of several continuous recognition areas, if the curve conforms to the trend change from small to large and then to small, then the several continuous recognition areas are used as the first luminous area; if the curve does not conform to the trend change from small to large and then to small, for example, A 1m , …, A 1(m +w) If the brightness values ​​are all equal, it means that there is no brightness change at present. Then you can set the window moving step value to w, that is, move the window by w; or when it changes from small to large, you can move the window moving step value forward by two; S013, moving the window according to the set window movement step value to obtain the next several continuous recognition areas until the j recognition areas of the current row are traversed; S02. Summarize the first light-emitting areas obtained in each row to obtain a light-emitting area.

[0027] When summarizing, when the identification areas of the first luminous areas in two adjacent rows intersect (that is, there are identification areas in the same column), the first luminous areas in the two rows are merged. When the identification areas of the first luminous areas in two adjacent rows do not intersect and the values ​​of the column numbers of the closest identification areas are also greatly different, it means that there are multiple luminous areas. In this case, it is only necessary to perform statistics according to the luminous areas separately, that is, each luminous area is identified and judged after curve fitting.

[0028] In some embodiments, the general brightness change area usually appears in the middle of the screen. Therefore, the search can start from the central area first to improve the processing speed. The specific process of step S3 may further include: S31. Take the recognition area at the center position of the matrix as the central area; the center position of the matrix is , where represents rounding up. That is, when i is even, , and when i is odd, ; S32. Obtain all adjacent recognition areas of the central area (for example, for the central area at the center position of the matrix area, eight adjacent recognition areas need to be obtained. When other recognition areas are used as the central area, the already compared recognition areas can be filtered out, and only seven adjacent recognition areas need to be compared). Compare the brightness value of the central area with the brightness values of its adjacent recognition areas respectively. When there is a recognition area whose brightness value is greater than or equal to the brightness value of the central area, take the recognition area as the central area and repeat the above steps; S33. Until all recognition areas are traversed, gather all the recognition areas that serve as the central area, and select the recognition area with the highest brightness value in the central area and all its surrounding recognition areas as the light-emitting area.

[0029] Alternatively, when taking the recognition area at the center position of the matrix as the central area, starting from the central area, compare the brightness value of the central area with the brightness values of the recognition areas around it. When the brightness value of the central area is higher than the brightness values of all the surrounding recognition areas, continue to determine whether the brightness value of each recognition area is greater than the brightness value at the corresponding position in the standard image. If so, take the recognition area and the central area together as the light-emitting area.

[0030] S4. Perform brightness recognition on the local images, fit the brightness change curve according to the brightness recognition results of each local image, determine the cause of the brightness change based on the brightness change curve, and decide whether to issue an electric spark warning according to the determination result.

[0031] Specifically, the specific process of step S4 is as follows: Find the peak of the brightness change curve. When the change trend of the brightness change curve is a jump within a preset time period and there are multiple peaks, determine that the cause of the brightness change is an electric spark; When the change trend of the brightness change curve shows a slow increase first, then remains for a certain preset time and then gradually decreases within a preset time period, or when the change trend of the brightness change curve is a jump to the brightness peak and then remains unchanged, determine that the cause of the brightness change is an interference source.

[0032] To a certain extent, the brightness change curve can reflect the change of the light source that causes the brightness change inside the power distribution cabinet. For example, the fitted brightness change curve uses the brightness value as the vertical coordinate and time as the horizontal coordinate, and plots the brightness values of all frame images within a preset time period before and after the image to be recognized on the brightness change curve. As Figure 2 shown, the brightness change curve of the electric spark generally jumps from a brightness value of 0 to the brightness peak in a very short time and then quickly changes back to 0. And when the electric spark is generated, there are generally multiple brightness peak change trends similar to pulses. While the brightness change curve of suddenly opening the door is generally from a brightness value of 0, gradually rising to the brightness peak as the opening amplitude increases, and then remaining for a period of time until it becomes 0, and there is a certain slope. And the brightness change curve of the indicator light as Figure 3 shown, is generally that the brightness value jumps from 0 to the brightness peak in a very short time and remains unchanged for a long time.

[0033] In order to distinguish the types of electric sparks generated, when it is determined that the cause of the brightness change is an electric spark, the following steps are executed: Count the number of times the brightness peak appears in the brightness change curve. If the number of times the brightness peak appears within the preset time period is less than the preset threshold, an early warning of electric spark generation is output; When the number of times the brightness peak appears within the preset time period is greater than the preset threshold and the time between two adjacent brightness peak appearances is less than the set value, an emergency maintenance warning is given.

[0034] For the short - term sparks caused by the aging and switching of relay switches or the generation of static electricity due to dry air, the duration is short and the energy is not high. It can be observed continuously. If the frequency is high, for example, if it appears continuously for a certain period of time, it indicates that there may be a fault that requires maintenance. When identifying short - term sparks, the type of short - term sparks can be further identified. For example, select the local image with the largest brightness value from all local images as the spark frame image, and perform spark shape recognition on the spark frame image; when the shape is radial, it is a fault spark generated during the aging and switching of the relay switch, and when the shape is rectangular, it is a static electricity spark caused by dry air generating static electricity.

[0035] Embodiment 2 The present application provides an electric spark monitoring system for a power distribution cabinet, which applies the electric spark monitoring method for a power distribution cabinet described in the first aspect, including: A data acquisition module, configured to obtain the video stream collected in real time by the monitoring device inside the power distribution cabinet, and intercept a plurality of frames of images to be recognized from the video stream at a preset time interval; A trigger module is used to compare the brightness of the image to be recognized with a preset standard image. If the brightness of the image to be recognized changes, a continuous number of frames of images within a preset time period before and after the image to be recognized with the changed brightness as the starting point are obtained and stored in the spark recognition image set. An image intercepting module recognizes the light-emitting area of the image to be recognized with the changed brightness to obtain the recognition frame of the light-emitting area, and intercepts each frame of the image in the spark recognition image set according to the recognition frame to obtain the partial image marked by the recognition frame in each frame of the image. A spark judgment module is used to recognize the brightness of the partial images, fit the brightness change curve according to the brightness recognition results of the partial images, determine the reason for the change in brightness according to the brightness change curve, and decide whether to issue an electric spark warning based on the judgment result.

[0036] Embodiment 3 Provided is a computer-readable storage medium, including: One or more processors; A storage unit for storing one or more programs, which, when executed by the one or more processors, can enable the one or more processors to implement the electric spark monitoring method of a power distribution cabinet described in Embodiment 1.

[0037] As mentioned above, it is only a preferred embodiment of the present invention, and does not impose any form of limitation on the present invention. According to the technical essence of the present invention, any simple modification, equivalent replacement, and improvement made to the above embodiments within the spirit and principle of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for monitoring electric sparks in a power distribution cabinet, characterized in that: The specific steps include: S1. Obtain the video stream collected in real time by the monitoring equipment in the power distribution cabinet, and extract several frames of images to be identified from the video stream at preset time intervals; S2, compare the brightness of the image to be identified with a preset standard image. If the brightness of the image to be identified changes, obtain a number of consecutive frames of images within a preset time period before and after the image to be identified with the changed brightness as the starting point, and store them in a spark recognition image set; S3, identifying the luminous area of ​​the image to be identified with changed brightness, obtaining an identification frame of the luminous area, intercepting each frame of the spark identification image set according to the identification frame, and obtaining a local image marked by the identification frame in each frame of the image; S4. Perform brightness recognition on the local images, fit the brightness change curve according to the brightness recognition results of each local image, determine the cause of the brightness change according to the brightness change curve, and decide whether to issue an electric spark warning based on the determination result.

2. The method for monitoring sparks in a power distribution cabinet according to claim 1, characterized in that: In step S3, the image to be recognized is evenly divided, each frame of the image is divided into n recognition areas of the same size, and the n recognition areas are arranged into an i×j matrix.

3. The method for monitoring sparks in a power distribution cabinet according to claim 2, characterized in that: The specific process of step S3 includes: The image to be identified is divided into n identification areas, and each row is traversed by window moving traversal. For each row of j identification areas, the following steps are performed: Obtain several continuous recognition areas according to the set window length; Calculate the brightness values ​​of several continuous identification areas respectively, draw a curve according to the brightness values ​​of several continuous identification areas, and if the curve conforms to the trend of changing from small to large and then to small, then take the several continuous identification areas as the first light-emitting areas; Move the window according to the set window moving step value to obtain the next several consecutive recognition areas until all j recognition areas in the current row are traversed; The first light-emitting areas obtained in each row are summarized to obtain a light-emitting area.

4. The method for monitoring sparks in a power distribution cabinet according to claim 2, characterized in that: The specific process of step S3 includes: S31, taking the recognition area at the center of the matrix as the central area; S32, starting from the central area, comparing the brightness value of the central area with the brightness values ​​of the adjacent recognition areas, when there is a recognition area whose brightness value is greater than or equal to the brightness value of the central area, the recognition area is taken as the central area to execute step S33; S33, repeat step S32 until all the recognition areas are traversed, collect all the recognition areas as the central areas, and select the recognition area with the highest brightness value in the central area and the recognition areas around it as the luminous areas.

5. The method for monitoring sparks in a power distribution cabinet according to claim 3, characterized in that: Taking the recognition area at the center of the matrix as the central area, starting from the central area, compare the brightness value of the central area with the brightness values ​​of the recognition areas around it. When the brightness value of the central area is higher than the brightness values ​​of all the surrounding recognition areas, continue to judge whether the brightness value of each recognition area is greater than the brightness value of the corresponding position in the standard image. If so, the recognition area and the central area are taken as the luminous area together.

6. The method for monitoring sparks in a power distribution cabinet according to claim 1, characterized in that: The specific process of step S4 is: Find the peak value of the brightness change curve. When the change trend of the brightness change curve in a preset time period is a jump and there are multiple peak values, it is determined that the cause of the brightness change is electric spark; When the brightness change curve in the preset time period shows a trend of first slowly increasing, then maintaining for a preset time and then gradually decreasing, or when the brightness change curve in the preset time period shows a trend of jumping to the brightness peak and then remaining unchanged, it is determined that the cause of the brightness change is an interference source.

7. The method for monitoring electric sparks in a power distribution cabinet according to claim 6, characterized in that: When it is determined that the cause of the brightness change is electric spark, perform the following steps: The number of times the brightness peak value of the brightness change curve appears is counted. If the number of times the brightness peak value appears within a preset time period is less than a preset threshold, an electric spark warning is output; When the number of brightness peaks within a preset time period is greater than a preset threshold and the time between two adjacent brightness peaks is less than a set value, an emergency maintenance warning is issued.

8. The method for monitoring electric sparks in a power distribution cabinet according to claim 1, characterized in that: The preset time interval is greater than the continuous lighting time when the electric spark is generated, and the preset time period is the continuous lighting time of the electric spark interference source.

9. A spark monitoring system for a power distribution cabinet, using the spark monitoring method for a power distribution cabinet as claimed in claim 1, characterized in that: include: The data acquisition module is used to obtain the video stream collected in real time by the monitoring equipment in the power distribution cabinet, and to extract several frames of images to be identified from the video stream at preset time intervals; A trigger module is used to compare the brightness of the image to be identified with a preset standard image. If the brightness of the image to be identified changes, a number of consecutive frames of images within a preset time period before and after the image to be identified with the changed brightness are obtained and stored in a spark recognition image set; The image capture module recognizes the luminous area of ​​the image to be recognized with changed brightness, obtains the recognition frame of the luminous area, and captures each frame of the spark recognition image set according to the recognition frame to obtain the local image marked by the recognition frame in each frame of the image; The spark judgment module is used to identify the brightness of local images, fit the brightness change curve according to the brightness recognition results of each local image, determine the cause of the brightness change according to the brightness change curve, and decide whether to issue an electric spark warning based on the judgment result.

10. A computer-readable storage medium, characterized in that: include: one or more processors; A storage unit, used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement an electric spark monitoring method for a distribution cabinet as described in any one of claims 1-9.

Images