Switch cabinet multi-mode multi-channel on-line monitoring equipment

Abstract

The invention relates to the technical field of online monitoring, in particular to switch cabinet multi-mode multi-channel online monitoring equipment which comprises a video data acquisition module, a switch cabinet part recognition module, a multi-channel image fusion module, an equipment state monitoring module and an abnormal picture early warning module. According to the invention, the ambient light sensor is combined with visible light and infrared videos to analyze brightness balance and thermal signal coverage, improve imaging integrity, guarantee quality evaluation comprehensiveness, extract component contours and analyze temperature distribution, and is combined with a multi-source data analysis structure to improve identification precision, dynamically adjust video fusion proportion and improve component definition. Visual consistency of a complex environment is guaranteed, images, temperature and form data are fused, abnormal heating and loosening are accurately detected, suspicious areas are marked in real time, abnormal visualization and evaluation are enhanced through automatic marking and data analysis, and timely early warning is achieved. A multi-channel fusion combination form and temperature monitoring form closed-loop optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of online monitoring, and in particular to a multi-mode and multi-channel online monitoring device for a switch cabinet. Background Art

[0002] The field of online monitoring technology includes real-time monitoring and anomaly detection of the operating status of power equipment. Its core content is to ensure the safe and stable operation of the power system through remote data collection and analysis. This technical field involves dynamic monitoring of internal environmental parameters of switch cabinets, including key indicators such as mechanical component displacement, contact temperature, arc activity and insulation status. It is necessary to combine video image acquisition and sensor data fusion technology to build a complete visual information chain of equipment operating status, systematically covering data synchronization transmission, abnormal feature extraction and risk warning functions.

[0003] Among them, a multi-mode multi-channel online monitoring device for switch cabinets refers to a device based on a combination of a fixed camera and an infrared thermal imager. The device continuously captures the circuit breaker movement trajectory and contact contact status in the switch cabinet through a video stream, and simultaneously superimposes the real-time values ​​collected by the temperature sensor on the video screen. Specific technical means include presetting a multi-angle camera unit inside the cabinet to cover the operating mechanism and busbar area, using image grayscale value comparison to analyze the degree of contact oxidation, dynamically matching temperature data with video frames through a timestamp alignment mechanism, and triggering local screen magnification and data marking functions based on preset thresholds.

[0004] The existing technology has the problem of insufficient adaptability to changes in light intensity in video data acquisition and processing, resulting in the image quality being greatly disturbed by ambient light, and it is difficult to ensure the stability of the video signal in complex lighting environments. In the process of component recognition, only a single edge detection method is relied on, ignoring the role of the thermal signal distribution pattern, which affects the accuracy of component recognition and makes it difficult to accurately match the internal structure of the equipment. The fusion of visible light and infrared video fails to dynamically adjust the fusion ratio according to the visibility of the component, resulting in distortion of image information under different lighting environments, affecting the visibility of key components, and thus reducing the reliability of equipment status monitoring. Status monitoring mainly relies on the numerical judgment of temperature sensors, lacks a comprehensive analysis of morphological changes, and is prone to missed detection or misjudgment of abnormal situations. The abnormal picture warning is triggered only based on the set threshold, and no in-depth analysis of the change trend of the abnormal area is conducted, resulting in a lack of flexibility in the warning mechanism and difficulty in responding to potential faults in a timely manner. Overall, the existing technology has obvious deficiencies in terms of light adaptability, component recognition accuracy, image fusion strategy, comprehensiveness of abnormal detection, and flexibility of the warning mechanism, which affects the accuracy and reliability of equipment monitoring. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a multi-mode multi-channel online monitoring device for a switch cabinet.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a switch cabinet multi-mode multi-channel online monitoring device, comprising: The video data acquisition module obtains the switch cabinet video, calls the ambient light sensor to measure the light intensity, calculates the brightness balance of the visible light image, analyzes the thermal signal coverage of the infrared video, evaluates the imaging integrity of the two types of videos, and forms a video signal quality evaluation result; The switch cabinet component identification module calls the video signal quality evaluation result, detects the edge features in the visible light video, extracts the contour information of the internal components of the switch cabinet, and analyzes the temperature distribution pattern in the infrared video to identify the heat source characteristics of key components. Combined with the internal structures of the two types of video analysis equipment, a component identification result is formed; The multi-channel image fusion module calls the component recognition result, dynamically fuses the visible light and infrared videos, adjusts the fusion weight of the difference area based on the visibility of the component, and generates a fused image data stream; The equipment status monitoring module obtains the fused image data stream, analyzes the temperature change trend of the infrared signal during equipment operation, combines the component morphology changes in the visible light image, detects abnormal status, and highlights suspicious areas in real time on the monitoring interface to output equipment status monitoring results.

[0007] As a further solution of the present invention, the video signal quality assessment results include brightness balance, thermal signal coverage, and imaging integrity; the component recognition results include component contour information, temperature distribution pattern, heat source characteristics, and internal structure of the equipment; the fused image data stream includes image fusion ratio and imaging effect; the equipment status monitoring results include temperature change trends, component morphology changes, status abnormalities, and suspicious areas.

[0008] As a further solution of the present invention, the video data acquisition module includes: The video signal acquisition submodule obtains the visible light video and infrared video data of the switch cabinet, calls the ambient light sensor to measure the ambient light intensity value, and generates visible light video data, infrared video data and ambient light intensity value; The illumination analysis submodule calculates the mean and variance of the brightness of the pixels on the screen based on the visible light video data, the infrared video data and the ambient light intensity value, analyzes the distribution range and continuity of the thermal signal intensity in the infrared video data, and generates a brightness balance value and a thermal signal coverage status; The integrity assessment submodule calls the brightness balance value and thermal signal coverage status, compares the preset brightness deviation standard with the thermal signal coverage standard, evaluates the integrity of visible light and infrared video imaging, and generates a video signal quality assessment result.

[0009] As a further solution of the present invention, the specific calculation formula for analyzing the distribution range and continuity of the thermal signal intensity in the infrared video data is: ; Calculate the continuity characteristic value of thermal signal distribution, generate brightness balance value and thermal signal coverage status; in, Representative Continuity characteristic value of thermal signal distribution in frame infrared image, Representative The number of pixels in the frame that are identified as heat signal areas, Representative The pixel in the heat signal area is The heat intensity value in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The standard deviation of the brightness of all pixels in the frame, Representative The average brightness of all pixels in the frame, A very small constant to prevent the denominator from being zero.

[0010] As a further solution of the present invention, the switch cabinet component identification module includes: The edge contour extraction submodule calls the visible light video frame brightness data and signal-to-noise ratio data in the video signal quality assessment result, detects the amplitude of the edge gradient change of the video frame, selects edge points according to the gradient direction and preset standards, connects adjacent points to form a closed boundary, extracts the geometric shape information of the switch cabinet components and generates a component contour feature set; The heat source association analysis submodule calls the geometric boundary range of the component contour feature set to limit the detection range of the abnormal temperature difference area, calculates the temperature difference between the target area and the surrounding area by analyzing the temperature distribution data of the infrared video frame, screens the abnormal temperature rise area according to the preset threshold, records the coordinates of the highest temperature point and the change trend, and generates a heat source distribution map in combination with the component contour; The structural analysis submodule calls the temperature distribution data of the heat source distribution map and the geometric boundary information of the component contour feature set, spatially aligns the center point of the component boundary with the highest temperature point of the heat source, calculates whether the position deviation between the two is less than the preset spatial tolerance range, screens the component heat source association groups that meet the conditions, and generates component recognition results based on the contour morphology and heat source distribution law in the association group.

[0011] As a further solution of the present invention, the multi-channel image fusion module includes: The component recognition submodule calls the edge contour features and texture features in the component recognition result, obtains the input frame sequence of the visible light image and the infrared image, extracts the regional coordinates of the same component in the visible light image and the infrared image, compares the edge sharpness value and texture complexity value of the corresponding component in the visible light image and the infrared image, and screens to obtain a component difference region coordinate set; The difference allocation submodule calculates the grayscale mean and standard deviation of the corresponding area of ​​the visible light image based on the component difference area coordinate set, and simultaneously calculates the thermal radiation intensity mean and contrast value of the corresponding area of ​​the infrared image, performs a difference operation on the visible light grayscale mean and the infrared thermal radiation intensity mean, determines the fusion ratio by combining the ratio of the standard deviation to the contrast value, and generates a difference area fusion ratio set; The dynamic fusion submodule calls the component difference area coordinate set and the difference area fusion ratio set, performs pixel-by-pixel superposition operation on the difference area of ​​the visible light image and the infrared image, fuses the non-difference area according to a preset ratio, integrates all area results and splices them into a continuous frame sequence to generate a fused image data stream.

[0012] As a further solution of the present invention, the specific calculation formula for comparing the edge sharpness value and texture complexity value of the components corresponding to the visible light and infrared images is: ; Calculate the composite index value of feature comparison and filter out the coordinate set of component difference areas; Represents the composite index value of feature comparison, Represents the visible light image The edge sharpness value of pixels, Represents the infrared image The edge sharpness value of pixels, Represents the visible light image The texture complexity value of each pixel, Represents the infrared image The texture complexity value of each pixel, Represents the total number of pixels involved in the comparison of edge features, Represents the total number of pixels involved in the calculation of texture features.

[0013] As a further solution of the present invention, the equipment status monitoring module includes: The image fusion submodule synchronously collects the infrared image temperature sequence and the visible light image frame sequence during the operation of the monitoring device according to the fused image data stream, aligns the temperature value of each pixel point of the infrared image with the spatial coordinates of the corresponding area of ​​the visible light image on the time axis, and superimposes the temperature data on the visible light image coordinate system to generate multi-source image fusion data; The temperature trend analysis submodule calls the temperature sequence in the multi-source image fusion data, extracts the temperature extreme difference value and the temperature change amplitude between adjacent frames in the continuous time window of the equipment area, compares the extreme difference value with the temperature fluctuation range corresponding to the preset equipment type, screens the extreme difference exceeding limit area, calculates the statistical fluctuation range area whose change amplitude exceeds the benchmark change amplitude of similar equipment, and generates the temperature anomaly mark area; The morphological association determination submodule extracts the component contour deformation amount in the temperature anomaly marked area, calculates the mean Euclidean distance of the contour point set between adjacent frames based on the visible light image frame sequence, matches the spatial coordinates of the area where the deformation amount exceeds the assembly tolerance range with the temperature anomaly area, marks the boundaries of the area where both the deformation amount exceeds the standard and the temperature anomaly exists, and maps the coordinates of the temperature anomaly area to the monitoring interface to generate the equipment status monitoring results.

[0014] As a further solution of the present invention, the specific calculation formula for calculating the mean value of the Euclidean distance of the contour point sets between adjacent frames is: ; in, Represents the mean Euclidean distance of contour point sets between adjacent frames, represents the Euclidean distance of the contour point set between the i-th frame and the j-th frame, represents the average Euclidean distance between all contour point sets, and Q represents the total number of calculated contour point sets.

[0015] As a further solution of the present invention, it also includes: The abnormal screen warning module calls the equipment status monitoring results, automatically marks the abnormal area on the monitoring screen and adjusts the display mode, and at the same time compares and analyzes the data of the abnormal area, evaluates the severity of the problem, and triggers abnormal screen warning information; The abnormal screen warning information includes abnormal area marking, adjustment of display mode, data comparison and analysis, and assessment of problem severity; The abnormal picture warning module includes: The abnormal area marking submodule obtains the equipment operation data set in the equipment status monitoring result, calls the preset benchmark comparison rule, calculates the difference between the partition values ​​in the data set and the corresponding benchmark item by item, selects the partitions whose differences exceed the set deviation threshold, maps the partition coordinates to the monitoring screen, and generates the abnormal area coordinates; The display adjustment and comparison submodule calls the coordinates of the abnormal area, extracts the real-time monitoring data stream in the abnormal area, adjusts the color saturation and flickering frequency in the screen display settings, synchronously obtains the normal data mean value within the operation cycle of the same area, compares the real-time data stream with the data mean in segmented time series, and generates an abnormal data comparison set; The abnormal assessment trigger submodule extracts the maximum deviation amplitude and the mean fluctuation frequency based on the abnormal data comparison set, combines the load degree and the proportion of long operation time in the equipment operation data set, calculates the abnormal deviation value, matches the deviation value with the preset warning level threshold range, triggers the corresponding level of warning instructions, and generates abnormal screen warning information.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, the ambient light sensor is combined with visible light and infrared video to analyze the brightness balance and thermal signal coverage, improve the imaging integrity, ensure the comprehensiveness of quality assessment, extract the component outline and analyze the temperature distribution, combine the multi-source data analysis structure, improve the recognition accuracy, dynamically adjust the video fusion ratio, improve the component clarity, ensure the visual consistency of complex environments, integrate images with temperature and morphological data, accurately detect abnormal heating and looseness, mark suspicious areas in real time, and automatically mark and analyze abnormal visualization and evaluation through data analysis to achieve timely warning. Multi-channel fusion combines morphology and temperature monitoring to form a closed-loop optimization to improve monitoring accuracy and detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system flow chart of the present invention; Figure 2 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0020] See also Figure 1 , a switch cabinet multi-mode multi-channel online monitoring device includes: The video data acquisition module obtains the switch cabinet video, calls the ambient light sensor to measure the light intensity, calculates the brightness balance of the visible light image, analyzes the thermal signal coverage of the infrared video, evaluates the imaging integrity of the two types of videos, and forms a video signal quality evaluation result; The switch cabinet component recognition module calls the video signal quality assessment results, detects the edge features in the visible light video, extracts the contour information of the internal components of the switch cabinet, and analyzes the temperature distribution pattern in the infrared video to identify the heat source characteristics of key components. It combines the internal structures of the two types of video analysis equipment to form component recognition results; The multi-channel image fusion module calls the component recognition results to dynamically fuse the visible light and infrared videos. Based on the visibility of the components, it adjusts the fusion weight of the difference area, optimizes the imaging effect at night or in different lighting environments, and generates a fused image data stream. The equipment status monitoring module obtains the fused image data stream, analyzes the temperature change trend of the infrared signal during the operation of the equipment, combines the component morphology changes in the visible light image, detects abnormal conditions, and highlights suspicious areas in real time on the monitoring interface, outputting the equipment status monitoring results; The abnormal screen warning module calls the equipment status monitoring results, automatically marks the abnormal area on the monitoring screen, and adjusts the display mode. At the same time, it compares and analyzes the data in the abnormal area, evaluates the severity of the problem, and triggers abnormal screen warning information.

[0021] The video signal quality assessment results include brightness balance, thermal signal coverage, and imaging integrity. The component recognition results include component contour information, temperature distribution pattern, heat source characteristics, and internal structure of the equipment. The fused image data stream includes image fusion ratio and imaging effect. The equipment status monitoring results include temperature change trends, component morphology changes, abnormal status, and suspicious areas. Abnormal image warning information includes abnormal area marking, display adjustment, data comparison and analysis, and assessment of problem severity.

[0022] See also Figure 2 , the video data acquisition module includes: The video signal acquisition submodule obtains the visible light video and infrared video data of the switch cabinet, calls the ambient light sensor to measure the ambient light intensity value, and generates visible light video data, infrared video data and ambient light intensity value; First, start the video acquisition terminal and locate it to the target area of ​​the switch cabinet. Set the acquisition angle, acquisition frame rate and focal length parameters through the control program to obtain clear visible light image data. At the same time, the infrared video acquisition channel runs in parallel to record thermal imaging of the same area. The ambient light sensor module is started synchronously. Its installation position must ensure that it is unobstructed and as close to the surface of the device being photographed as possible. It is used to detect the actual illumination level in the environment. In an environment where the light changes frequently or the light source is complex (such as strong light or backlight conditions in the factory), the sensor sampling frequency should be increased to above 10Hz to ensure that the measurement data is continuous and valid. The collected data includes the current light level. Intensity value (in lux). If the illuminance value detected at a certain moment is 350 lux, the value is associated with the timestamp of the synchronously acquired video frame. During the whole process, each set of visible light and infrared video data is labeled with the acquisition time and light value to form a complete data pair. For example, in a switch cabinet inspection, an infrared video and a visible light video are acquired at 10:30 in the morning. The synchronously recorded ambient light intensity is 520 lux. The two videos will carry the light value for subsequent processing by the light analysis submodule. The final video data can be matched with the ambient illumination to form multi-dimensional visual data.

[0023] The illumination analysis submodule calculates the mean and variance of the brightness of the pixels on the screen based on the visible light video data, infrared video data and the ambient light intensity value, analyzes the distribution range and continuity of the thermal signal intensity in the infrared video data, and generates the brightness balance value and thermal signal coverage status; The specific calculation formula for analyzing the distribution range and continuity of thermal signal intensity in infrared video data is: ; Calculate the continuity characteristic value of thermal signal distribution, generate brightness balance value and thermal signal coverage status; in, Representative Continuity characteristic value of thermal signal distribution in frame infrared image, Representative The number of pixels in the frame that are identified as heat signal areas, Representative The pixel in the heat signal area is The heat intensity value in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The standard deviation of the brightness of all pixels in the frame, Representative The average brightness of all pixels in the frame, To prevent a very small constant with a denominator of zero; Detailed explanation of the formula and the process of formula calculation and derivation: This formula is used to calculate the Distribution continuity eigenvalue of thermal signal in infrared image frame The following are the definitions and derivations of the parameters in the formula, including how each parameter is calculated: Usually, image processing algorithms are used to automatically identify and count hot spots in infrared images. Assume that 500 pixels are detected.

[0024] From the infrared camera data output, the heat intensity of each pixel is recorded. For example, select some specific pixel heat intensity values ​​[200,220,230,210,205].

[0025] The calculation method is to sum the thermal intensity values ​​of all thermal signal pixels and divide by Using the previous pixel value, the average heat intensity .

[0026] The calculation method is the same as . Assume that the average heat intensity of the previous frame is 211.

[0027] and It is obtained by analyzing the visible light image of the corresponding frame. , .

[0028] The specific calculation process is as follows: First, calculate the absolute value of the difference between each thermal signal pixel and the average thermal intensity, sum it up and divide it by : ; Then calculate the ratio of the brightness standard deviation to the brightness mean, and add a very small constant: ; Finally, calculate the square root of the difference in average heat intensity between the previous and next frames: ; Substitute the values ​​of these three parts into In the formula: ; This result shows that the Frame compared to The frames show remarkable stability in the continuity feature of the thermal signal distribution. The calculation of this characteristic value reflects the variability of the thermal signal and is directly related to the generation of the thermal signal coverage state.

[0029] The integrity assessment submodule calls the brightness balance value and thermal signal coverage status, compares the preset brightness deviation standard with the thermal signal coverage standard, evaluates the integrity of visible light and infrared video imaging, and generates a video signal quality assessment result; First, the preset brightness deviation allowable range in the system is called, for example, an error of ±20% is allowed. During the processing, the brightness mean and deviation ratio in the visible light analysis results are read and compared with the set standard value. If the deviation ratio is within 15%, it is marked as qualified brightness, otherwise it is marked as abnormal brightness. The judgment of the thermal signal coverage status is based on the preset coverage continuity standard, which is usually set as more than 70% of the area in the picture needs to be covered by the thermal signal, and the temperature difference between adjacent areas shall not be greater than 5°C. If the system detection results show that the thermal signal area only occupies 45% of the picture and there is an obvious temperature difference fault, its thermal coverage status is judged to be substandard. In an actual scene, for example The average brightness analyzed from the video taken during daily equipment inspections deviates from the benchmark by 17%, and the thermal signal coverage is 82%. However, there are multiple areas of sudden temperature changes. The video signal is judged to have poor thermal continuity. The system ultimately generates a video signal quality assessment result of "incomplete infrared and uneven brightness at the edge of visible light". The assessment result is recorded in the inspection log for manual review and system tuning. All values ​​involved in the comparison process, such as "deviation standard" and "coverage standard", can be configured by operation and maintenance personnel according to different equipment scenarios. For example, the lower limit of thermal signal coverage allowed for some high-voltage switchgear can be increased to 60% to adapt to thermal reflection interference under complex wiring structures.

[0030] See also Figure 2 , the switchgear component identification module includes: The edge contour extraction submodule calls the visible light video frame brightness data and signal-to-noise ratio data in the video signal quality assessment results, detects the amplitude of the edge gradient change of the video frame, selects edge points according to the gradient direction and preset standards, connects adjacent points to form a closed boundary, extracts the geometric shape information of the switch cabinet components and generates a component contour feature set; First, the overall brightness level of each frame of the image needs to be analyzed. In actual operation, the frame brightness can be extracted by averaging the grayscale value of each pixel in the image. For example, when processing a 640×480 resolution image, the current frame brightness value can be obtained by traversing the pixel grayscale values ​​row by row and column by column, and then the sum is counted and divided by the total number of pixels. If the average grayscale of the frame is 125, this is recorded as the current frame brightness. Then the signal-to-noise ratio data corresponding to the frame is extracted. The image denoising can be performed using the difference between the frame image and the previous and next frames, and then the image noise and signal intensity ratio is counted. Combined with historical settings, if the signal-to-noise ratio is lower than 20dB, the frame will be marked as a poor quality frame. After completing the image quality assessment, the edge detection stage is entered. Common algorithms such as Canny edge detection are used. First, the image is Gaussian blurred, and then the gradient change values ​​of the image in the X and Y directions are extracted to determine the edge strength of each pixel. By setting the edge strength threshold such as 80, the edge points with drastic changes are retained, and the gradient directions of these points are extracted. Here, the direction is the direction angle of the grayscale change in the image. For example, if the direction angle of a point is detected to be 90 degrees, if the system preset standard direction is horizontal or vertical, then the point can be retained as an edge candidate. Continue to process all candidate edge points, and by judging the pixel adjacency relationship between the points, connect the adjacent and directional continuous edge points into a closed curve, and finally form several closed area boundaries. These closed contours are used to calculate geometric features, such as the circumscribed rectangle of the contour, contour area, center coordinates, etc. For example, a contour with a width of 120 pixels, a height of 220 pixels, and a center coordinate of (210,210) is identified. This feature is recorded and organized as the geometric contour feature information of the circuit breaker component, stored in the feature set, and used for subsequent structural analysis and comparison operations.

[0031] The heat source association analysis submodule calls the geometric boundary range of the component contour feature set to limit the detection range of the abnormal temperature difference area. By analyzing the temperature distribution data of the infrared video frame, the temperature difference between the target area and the surrounding area is calculated, and the abnormal temperature rise area is screened according to the preset threshold. The coordinates of the highest temperature point and the change trend are recorded, and the heat source distribution map is generated in combination with the component contour; First, read the boundary coordinate values ​​of each target component from the contour feature set. For example, if the coordinates of the upper left corner of a component contour are (100, 150) and the coordinates of the lower right corner are (320, 270), then this rectangular area is the detection area range of the component. The system usually expands the detection range to prevent the edge heat source from being missed. After setting the extension range to 10 pixels on each side, the analysis area becomes (90, 140)-(330, 280). The system extracts the temperature value of each pixel in the above area from the infrared video frame, and calculates the average temperature of the pixels in the target area. Assume that the average temperature of the area is 65°C. At the same time, sample the 20 pixels outside the boundary of the area as the background area, extract its average temperature, and assume that it is 46°C. The temperature difference between the two can be calculated as 19°C. The system presets the temperature difference threshold based on the material and load level of the switch cabinet components. For example, 15°C is set as the abnormal threshold for a high-power circuit breaker, and this time 19°C has exceeded the threshold. Then the area is confirmed to be an abnormal temperature rise area, and the highest temperature point is located in the target area, such as the coordinates (210,190) and the temperature is 72.4°C. The system continues to track the temperature change of this point in continuous video frames, extracts the average temperature change trend within 10 frames, such as from 68°C in the first frame to 72.4°C in the tenth frame. The system determines that the temperature rise continues, integrates the coordinates, maximum temperature and change trend information of the point into the heat source characteristic information of the current component, and generates the component heat source distribution map in combination with the geometric contour position relationship.

[0032] The structural analysis submodule calls the temperature distribution data of the heat source distribution map and the geometric boundary information of the component contour feature set, spatially aligns the component boundary center point with the heat source highest temperature point, calculates whether the position deviation between the two is less than the preset spatial tolerance range, selects the component heat source association group that meets the conditions, and generates the component recognition result according to the contour morphology and heat source distribution law in the association group; First, determine the center point position of the component contour area, and obtain it by finding the median of the left and right and upper and lower coordinates of the boundary box. For example, the upper left of the component boundary is (100,150) and the lower right is (320,270), then the center point is (210,210). The system then reads the coordinates of the highest temperature point recorded in the heat source map corresponding to the component area, for example (215,205), to determine whether the spatial distance between the two points is within the allowable tolerance range. Set the tolerance range to 15 pixels, and calculate the distance through the center point and the heat source point coordinates. If the result is 7 pixels, it meets the matching conditions and the heat source point is bound to the contour. Subsequently, the system performs a structural matching analysis on the bound heat source and the component boundary to check whether the heat source is located in a specific structural area of ​​the contour. For example, if the component is a rectangular structure and the heat source point is located in the central area of ​​its long side, it can be classified as an "edge heating" type. By counting the spatial distribution characteristics of the heat sources, such as when multiple heat source points are concentrated within 10% of the upper edge of the boundary, the system further determines that the heat source characteristic type is "upper edge heat accumulation". Finally, the system integrates each heat source that meets the spatial tolerance requirements with the component boundary information to generate component identification results, including the associated heat source position, number of heat source points, maximum temperature, boundary matching deviation and other information, to form the final heat source-structure matching identification table.

[0033] See also Figure 2 , the multi-channel image fusion module includes: The component recognition submodule calls the edge contour features and texture features in the component recognition results, obtains the input frame sequence of the visible light image and the infrared image, extracts the regional coordinates of the same component in the visible light and infrared images, compares the edge sharpness values ​​and texture complexity values ​​of the corresponding components in the visible light and infrared images, and screens out the component difference region coordinate set; The specific calculation formula for comparing the edge sharpness value and texture complexity value of the corresponding parts of the visible light and infrared images is: ; Calculate the composite index value of feature comparison and filter out the coordinate set of component difference areas; Represents the composite index value of feature comparison, Represents the visible light image The edge sharpness value of pixels, Represents the infrared image The edge sharpness value of pixels, Represents the visible light image The texture complexity value of each pixel, Represents the infrared image The texture complexity value of each pixel, Represents the total number of pixels involved in the comparison of edge features, Represents the total number of pixels involved in the calculation of texture features; This formula is used to evaluate the difference in features of a part between visible and infrared images. The formula consists of two parts: one part calculates the overall difference in edge sharpness, and the other part calculates the average difference in texture complexity.

[0034] Edge sharpness difference calculation: ; Texture complexity difference calculation: ; Parameter setting and data acquisition: and It is extracted from the actual image data through image processing algorithms. Usually, edge sharpness can be calculated by methods such as Sobel operator or Canny edge detection.

[0035] and Texture complexity can be calculated using texture analysis methods such as gray-level co-occurrence matrix (GLCM).

[0036] and Determined by image resolution and selected area.

[0037] Numerical examples and calculation process: Assume a specific scenario. Pixels, Pixels. Measured by image processing software, and The average values ​​are 15 and 10 respectively. and The average values ​​are 5 and 3 respectively.

[0038] Edge sharpness difference calculation: ; Texture complexity difference calculation: ; Result interpretation: The calculated edge sharpness difference value is about 180.28, and the texture complexity difference value is 8. These two values ​​combined indicate that there is a significant difference in the feature performance of the analyzed parts between visible light and infrared images. Feature comparison composite index value The value is 180.28+8=188.28, indicating that the image has obvious differences in the characteristics of the visual and infrared spectrum. By analyzing this value, we can determine which areas are inconsistent in the two image modes, thus effectively guiding subsequent image processing and analysis.

[0039] The actual monitoring and image processing algorithm extract data to ensure the reality of the parameters and the accuracy of the calculation. This result provides a scientific basis for subsequent steps, such as the selection and analysis of component difference areas.

[0040] The difference allocation submodule calculates the grayscale mean and standard deviation of the corresponding area of ​​the visible light image based on the component difference area coordinate set, and simultaneously calculates the thermal radiation intensity mean and contrast value of the corresponding area of ​​the infrared image. It performs a difference operation on the visible light grayscale mean and the infrared thermal radiation intensity mean, and determines the fusion ratio based on the ratio of the standard deviation to the contrast value to generate a difference area fusion ratio set. The corresponding area is extracted from the visible light image, and the pixel grayscale values ​​in the area are statistically processed. First, the average grayscale value is calculated to reflect the brightness benchmark of the area, and then the discrete degree of the grayscale value distribution in the area is counted as an indicator of the grayscale uniformity of the area. For example, in an area of ​​100×100 pixels, the system will read and summarize the grayscale values ​​of all pixels. If the average grayscale value is 123, it means that the overall brightness is medium, and its standard deviation is 26, indicating that the brightness change in the area is relatively obvious. After that, the thermal radiation intensity value of the corresponding area of ​​the infrared image is synchronously read. Because the pixel value of the infrared image is essentially the encoding of the thermal intensity, the system directly uses it as the thermal value input for statistics, and calculates the average thermal value of the area. For example, if it is 143, it means that there is a certain heat source in the area. At the same time, the thermal intensity contrast is calculated according to the difference between the maximum and minimum values ​​in the pixel. For example, if the maximum thermal value is 180 and the minimum thermal value is 100, the difference is 80, and the overall thermal contrast is medium. By comparing the brightness mean of the visible light area with the infrared thermal value mean, the numerical difference is calculated. For example, if the difference is 20 units, the fusion ratio is calculated by combining the ratio of the grayscale distribution standard deviation to the infrared contrast. For example, if the standard deviation of the area is 26 and the infrared contrast is 92 after conversion, the fusion ratio is set to 0.16. This ratio means that the infrared image has a higher weight in this area. The system records the ratio for each difference area and generates a fusion ratio set for the difference area. The fusion ratios of all areas are associated and stored with the coordinates as indexes to ensure that subsequent operations can accurately extract the weight settings for each area.

[0041] The dynamic fusion submodule calls the component difference area coordinate set and the difference area fusion ratio set, performs pixel-by-pixel superposition operation on the difference area of ​​the visible light image and the infrared image, fuses the non-difference area according to the preset ratio, integrates all area results and splices them into a continuous frame sequence to generate a fused image data stream; The dynamic fusion submodule calls the difference area coordinate set and the fusion ratio set to perform pixel-by-pixel fusion operations on the difference areas of the input visible light image and the infrared image. During the fusion execution process, the system first extracts the coordinate range of the difference area identified in each frame image, and reads the visible light pixel value and infrared pixel value in the area pixel by pixel. Then, according to the fusion ratio corresponding to the area in the ratio set, the final fusion value of each pixel is calculated. For example, if the grayscale of a pixel point in the visible light image is 120, the grayscale in the infrared image is 140, and the fusion ratio is 0.16, the value synthesized by the system according to the ratio is 123. The non-difference area does not use the difference fusion ratio, but uses a unified preset The ratio is preset to 0.5, for example, infrared and visible light each account for half. If the infrared of a pixel is 110 and the visible light is 100, it will be 105 after fusion. The system processes all pixels in the image in sequence and calculates the fused image pixel by pixel. After completing the single-frame processing, all image frames are connected in sequence to form a continuous fused image sequence. The output of this sequence is the final fused image data stream. The fusion operation in the whole process is based on the identified difference area and the preset fusion strategy to ensure that the fusion logic of different areas in each frame of the image is consistent. All operations keep the image size and pixel structure consistent with the original image to avoid displacement or distortion. The final result can be directly used in subsequent target detection or monitoring systems.

[0042] See also Figure 2 , the equipment status monitoring module includes: The image fusion submodule collects infrared image temperature sequence and visible light image frame sequence synchronously during the operation of the monitoring equipment according to the fused image data stream, aligns the temperature value of each pixel of the infrared image with the spatial coordinates of the corresponding area of ​​the visible light image on the time axis, and superimposes the temperature data on the visible light image coordinate system to generate multi-source image fusion data; After receiving the fused image data stream, the image fusion submodule needs to synchronize the time axis of the collected infrared image and visible light image. Taking 30 frames per second for infrared images and 25 frames per second for visible light images as an example, the infrared image can be time-aligned according to the visible light frame rate through key frame interpolation, that is, matching the corresponding infrared temperature frame in the visible light frame with a similar time stamp. If the time points do not completely overlap, the infrared image can be reconstructed by time interpolation to make up for the missing frames or multi-frame interference and complete the synchronous acquisition. In terms of spatial alignment, the feature point matching method is used to extract edges or corner points with high stability in the image, such as using the Harris method to identify the edge of the device contour, and then the infrared image is converted and aligned to the visible light image coordinate system through affine transformation. In actual device scenarios, such as electrical The infrared image taken from the front of the device needs to be superimposed on the 1080P high-definition image. The infrared image can be interpolated and enlarged and then projected according to the positioning mark. After mapping, for the spatial coordinates of each pixel point in the visible light image, find its corresponding area in the infrared image, extract the temperature data of the area and assign it to the coordinate position. For example, the temperature of the infrared image area corresponding to the heat source concentration area of ​​the equipment at the center of the image is 52.1 degrees Celsius, that is, the same temperature is assigned to the coordinate. Subsequently, the temperature value is converted into a displayable layer, for example, different temperature ranges are assigned different color codes in a pseudo-color manner, and the image is superimposed pixel by pixel with the original visible light image to complete the image layer fusion and generate a fused image sequence containing infrared temperature, visible light color and spatial coordinate information for subsequent processing and analysis.

[0043] The temperature trend analysis submodule calls the temperature sequence in the multi-source image fusion data, extracts the temperature extreme value in the continuous time window of the equipment area and the temperature change amplitude between adjacent frames, compares the extreme value with the temperature fluctuation range corresponding to the preset equipment type, screens the extreme difference area, calculates the statistical fluctuation range area where the change amplitude exceeds the benchmark change amplitude of similar equipment, and generates the temperature anomaly mark area; After the temperature trend analysis submodule obtains the fused image sequence, it needs to continuously extract the temperature data in several frames of images for a specific device area. Usually, a certain number of continuous frames are selected in a time window manner, for example, 10 frames of data are extracted with a window length of 0.4 seconds to form a temperature change sequence for each pixel. In actual operation, the temperature sensitive parts inside the device are selected, such as the radiator area of ​​the power module, and the maximum and minimum temperature differences in this period are recorded through the temperature sequence. The temperature extremes are calculated and compared with the preset allowable fluctuation range of the device. For example, the temperature fluctuation of a certain type of equipment should not exceed 0.7°C when it is working normally. If the actual fluctuation exceeds this range, the temperature is calculated. Threshold, this area is marked as extreme difference abnormality; then, the temperature change values ​​between adjacent frames are further extracted, and the average fluctuation amplitude is statistically compared with the historical average temperature difference benchmark when similar equipment is running. The historical benchmark value is usually obtained through large sample long-term operation data. For example, the normal variation range of the fan module is between 0.1℃ and 0.2℃. If the current observation data is significantly higher than this interval, it is marked as a fluctuation abnormality area; finally, the distribution of these abnormal areas is statistically analyzed, and the degree of abnormality of the entire area is evaluated by the area occupied by the abnormal area and the abnormal pixel density, and the temperature abnormality result is output to provide a labeling basis for subsequent analysis.

[0044] The morphological association judgment submodule extracts the contour deformation of the component in the temperature anomaly mark area, calculates the mean Euclidean distance of the contour point set between adjacent frames based on the visible light image frame sequence, matches the spatial coordinates of the area where the deformation exceeds the assembly tolerance range with the temperature anomaly area, marks the boundaries of the area where both the deformation exceeds the standard and the temperature anomaly exists, and maps the coordinates of the temperature anomaly area to the monitoring interface to generate the equipment status monitoring results; The specific calculation formula for calculating the mean Euclidean distance of contour point sets between adjacent frames is: ; in, Represents the mean Euclidean distance of contour point sets between adjacent frames, represents the Euclidean distance of the contour point set between the i-th frame and the j-th frame, represents the average Euclidean distance between all contour point sets, and Q represents the total number of calculated contour point sets; Euclidean distance of contour point set

[0045] The Euclidean distance of a contour point set is the straight-line distance between two contour points in the image. , we need to first extract the contour point set in the image. For example, suppose in the image, the coordinates of the contour point in the i-th frame are , the coordinates of the contour point in the jth frame are According to the three-dimensional coordinates of the contour points, the Euclidean distance It can be calculated by the following formula: ; Assume that the coordinates of the contour points between the i-th frame and the j-th frame are and ,but: ; Calculated is the Euclidean distance of contour points between two frames.

[0046] Calculate the average Euclidean distance of all contour points

[0047] The average Euclidean distance of all contour points is calculated for all Assume that in a certain calculation, there are 5 pairs of contour points, and the Euclidean distance of each contour point set is calculated as: ; The average Euclidean distance for: ; so, .

[0048] Calculate the mean Euclidean distance : After calculating the Euclidean distance between each pair of contour points, in order to measure the fluctuation of the Euclidean distance of all contour point sets, the formula needs to be calculated by taking each pair of distances and the mean The absolute value of the difference between the two, and then the average value of all absolute differences. During the calculation process, it is necessary to traverse all , and calculate the average Euclidean distance between each pair of contour points and The difference between. Assuming N is 5, then: ; The calculations show that: ; so, .

[0049] Weight parameters and adjustment coefficients are set based on: When calculating the mean Euclidean distance, the selection of each parameter is related to the actual image data acquisition method and image quality. Through the actual image monitored, it may be necessary to consider the effects of illumination changes, camera position errors, etc., and then set different weight parameters or adjustment coefficients to optimize the calculation results. In this formula, there is no explicit weight parameter, but in actual applications, different distance values ​​may be given different weights based on the reliability of contour points between different frames, especially when the image acquisition process may be affected by noise.

[0050] See also Figure 2 , the abnormal screen warning module includes: The abnormal area marking submodule obtains the equipment operation data set in the equipment status monitoring results, calls the preset benchmark comparison rules, calculates the difference between the partition values ​​in the data set and the corresponding benchmarks one by one, selects the partitions whose differences exceed the set deviation threshold, maps the partition coordinates to the monitoring screen, and generates the abnormal area coordinates; First, it is necessary to preset benchmark rules based on the equipment type, application scenario, and previous stable operation cycle data. For example, in a certain power inspection system, a standard model of operation data is established for 10kV outdoor switchgear equipment, covering key indicators such as temperature, voltage, current, and vibration. Each indicator has a corresponding benchmark reference value under normal conditions. For example, the switchgear contact temperature benchmark is 75°C, the equipment vibration amplitude is 0.3g, and the current is 180A. After collecting the current cycle monitoring data, the real-time values ​​of each indicator are compared in turn, such as the current temperature is 78°C, the vibration is 0.45g, and the current is 195A, and then compared with the benchmark values ​​in turn to obtain The offset values ​​are filtered through the preset deviation judgment thresholds. For example, the offset allowable thresholds are set to 2°C, 0.1g, and 10A respectively. These three data items all exceed the standard range and are recorded as abnormal items. The system further combines the binding relationship between the monitoring point and the partition coordinates to quickly map the partition corresponding to the abnormal indicator to the image monitoring interface. The partition-coordinate conversion relationship table is referenced during the mapping process. For example, the temperature anomaly is located in the area where the main switch contacts are located (the numbered block is #A3, corresponding to the coordinates 12,8). The system aggregates these abnormal items into an abnormal coordinate set as the basis for subsequent screen highlighting and alarming.

[0051] The display adjustment and comparison submodule calls the coordinates of the abnormal area, extracts the real-time monitoring data stream in the abnormal area, adjusts the color saturation and flicker frequency in the screen display settings, synchronously obtains the normal data mean within the operation cycle of the same area, compares the real-time data stream with the data mean in segmented time series, and generates an abnormal data comparison set; After receiving the coordinates of the abnormal area, the display adjustment and comparison submodule immediately retrieves the real-time monitoring data stream in the area for processing. For example, when the system receives the temperature abnormality coordinates (12, 8), it retrieves the temperature data of the area in the past 10 seconds from the database or cache: [77.5, 78.0, 78.2, 78.4, 78.7] ℃, and changes the visual display parameters of the area where the coordinates are located, such as increasing the saturation of the original layer to 180%, and setting the flashing frequency of the area to once every 2 seconds to enhance its recognition. On this basis, the system synchronously calls the The coordinate point corresponds to the normal mean data in the historical stable operation cycle. For example, the historical mean temperature of the area is 75.3°C. The system processes the current temperature flow in segments (such as one segment every 10 seconds), and calculates the average temperature in each segment. For example, if the current segment mean is 78.16°C, the deviation from the normal value is calculated. Through the processing of multiple consecutive time periods, the system forms an abnormal comparison data set containing time tags and deviation values, such as a deviation of 2.86°C in the T1 segment and 3.12°C in the T2 segment, etc., to provide basic data support for subsequent fluctuation trends and alarm levels.

[0052] The abnormal assessment trigger submodule extracts the maximum deviation amplitude and the average fluctuation frequency based on the abnormal data comparison set, and calculates the abnormal deviation value by combining the load degree and the long-time operation ratio in the equipment operation data set. The deviation value is matched with the preset warning level threshold interval, triggering the corresponding level of warning instructions and generating abnormal screen warning information; First, extract the maximum deviation amplitude in each time period. For example, the maximum temperature deviation value in multiple time periods is 3.20℃. The system records this as the extreme deviation of this cycle and analyzes the overall fluctuation frequency. For example, if the sampling frequency is once per second, and the temperature change exceeds ±0.2℃ 10 times in the continuous 30-second data, the fluctuation frequency can be qualitatively 0.33Hz. Then the system retrieves the equipment operation data of the abnormal point and extracts the current cycle load level. For example, if the equipment current is monitored to be 195A and the rated current is 180A, the load level is calculated to be 1.08 times, which is classified as the "high load" area (>1.00 times is defined as high). Combined with the equipment's running time, for example, the equipment has been running for 16 hours that day, and the total plan is 24 hours, then the running time is 24 hours. The ratio is 0.667, which is classified as "medium to high" (divided into <0.4 low, 0.4-0.7 medium, >0.7 high according to the set interval). The system will weight the extreme deviation, load intensity and operation ratio to generate a comprehensive deviation value. For example, with a weight ratio of 4:3:3, the calculated comprehensive deviation value is 2.30℃, which is then determined by the system matching the set warning level interval. For example, the first level warning is 2.5℃ and above, the second level is 1.5℃ to 2.5℃, and the third level is below 1.5℃. The current result falls into the second level warning interval, and the system generates a "Level-2" warning instruction, and based on the abnormal coordinate information (such as coordinates 12,8), it is superimposed and displayed on the monitoring main interface, forming a corresponding warning mark and detailed information for subsequent disposal personnel to identify and confirm.

[0053] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A switch cabinet multi-mode multi-channel online monitoring device, characterized in that: include: The video data acquisition module obtains the switch cabinet video, calls the ambient light sensor to measure the light intensity, calculates the brightness balance of the visible light image, analyzes the thermal signal coverage of the infrared video, evaluates the imaging integrity of the two types of videos, and forms a video signal quality evaluation result; The switch cabinet component identification module calls the video signal quality evaluation result, detects the edge features in the visible light video, extracts the contour information of the internal components of the switch cabinet, and analyzes the temperature distribution pattern in the infrared video to identify the heat source characteristics of key components. Combined with the internal structures of the two types of video analysis equipment, a component identification result is formed; The multi-channel image fusion module calls the component recognition result, dynamically fuses the visible light and infrared videos, adjusts the fusion weight of the difference area based on the visibility of the component, and generates a fused image data stream; The equipment status monitoring module obtains the fused image data stream, analyzes the temperature change trend of the infrared signal during equipment operation, combines the component morphology changes in the visible light image, detects abnormal status, and highlights suspicious areas in real time on the monitoring interface to output equipment status monitoring results.

2. The switch cabinet multi-mode multi-channel online monitoring device according to claim 1 is characterized in that: The video signal quality assessment results include brightness balance, thermal signal coverage, and imaging integrity; the component recognition results include component contour information, temperature distribution pattern, heat source characteristics, and internal structure of the equipment; the fused image data stream includes image fusion ratio and imaging effect; the equipment status monitoring results include temperature change trends, component morphology changes, status abnormalities, and suspicious areas.

3. The switch cabinet multi-mode multi-channel online monitoring device according to claim 1 is characterized in that: The video data acquisition module comprises: The video signal acquisition submodule obtains the visible light video and infrared video data of the switch cabinet, calls the ambient light sensor to measure the ambient light intensity value, and generates visible light video data, infrared video data and ambient light intensity value; The illumination analysis submodule calculates the mean and variance of the brightness of the pixels on the screen based on the visible light video data, the infrared video data and the ambient light intensity value, analyzes the distribution range and continuity of the thermal signal intensity in the infrared video data, and generates a brightness balance value and a thermal signal coverage status; The integrity assessment submodule calls the brightness balance value and thermal signal coverage status, compares the preset brightness deviation standard with the thermal signal coverage standard, evaluates the integrity of visible light and infrared video imaging, and generates a video signal quality assessment result.

4. The switch cabinet multi-mode multi-channel online monitoring device according to claim 3 is characterized in that: The specific calculation formula for analyzing the distribution range and continuity of the thermal signal intensity in the infrared video data is: ； Calculate the continuity characteristic value of thermal signal distribution, generate brightness balance value and thermal signal coverage status; in, Representative Continuity characteristic value of thermal signal distribution in frame infrared image, Representative The number of pixels in the frame that are identified as heat signal areas, Representative The pixel in the heat signal area is The heat intensity value in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The average thermal intensity of all pixels in the thermal signal area in the frame, Representative The standard deviation of the brightness of all pixels in the frame, Representative The average brightness of all pixels in the frame, A very small constant to prevent the denominator from being zero.

5. The switch cabinet multi-mode multi-channel online monitoring device according to claim 3 is characterized in that: The switch cabinet component identification module includes: The edge contour extraction submodule calls the visible light video frame brightness data and signal-to-noise ratio data in the video signal quality assessment result, detects the amplitude of the edge gradient change of the video frame, selects edge points according to the gradient direction and preset standards, connects adjacent points to form a closed boundary, extracts the geometric shape information of the switch cabinet components and generates a component contour feature set; The heat source association analysis submodule calls the geometric boundary range of the component contour feature set to limit the detection range of the abnormal temperature difference area, calculates the temperature difference between the target area and the surrounding area by analyzing the temperature distribution data of the infrared video frame, screens the abnormal temperature rise area according to the preset threshold, records the coordinates of the highest temperature point and the change trend, and generates a heat source distribution map in combination with the component contour; The structural analysis submodule calls the temperature distribution data of the heat source distribution map and the geometric boundary information of the component contour feature set, spatially aligns the center point of the component boundary with the highest temperature point of the heat source, calculates whether the position deviation between the two is less than the preset spatial tolerance range, screens the component heat source association groups that meet the conditions, and generates component recognition results based on the contour morphology and heat source distribution law in the association group.

6. The switch cabinet multi-mode multi-channel online monitoring device according to claim 5 is characterized in that: The multi-channel image fusion module comprises: The component recognition submodule calls the edge contour features and texture features in the component recognition result, obtains the input frame sequence of the visible light image and the infrared image, extracts the regional coordinates of the same component in the visible light image and the infrared image, compares the edge sharpness value and texture complexity value of the corresponding component in the visible light image and the infrared image, and screens to obtain a component difference region coordinate set; The difference allocation submodule calculates the grayscale mean and standard deviation of the corresponding area of ​​the visible light image based on the component difference area coordinate set, and simultaneously calculates the thermal radiation intensity mean and contrast value of the corresponding area of ​​the infrared image, performs a difference operation on the visible light grayscale mean and the infrared thermal radiation intensity mean, determines the fusion ratio by combining the ratio of the standard deviation to the contrast value, and generates a difference area fusion ratio set; The dynamic fusion submodule calls the component difference area coordinate set and the difference area fusion ratio set, performs pixel-by-pixel superposition operation on the difference area of ​​the visible light image and the infrared image, fuses the non-difference area according to a preset ratio, integrates all area results and splices them into a continuous frame sequence to generate a fused image data stream.

7. The switch cabinet multi-mode multi-channel online monitoring device according to claim 6 is characterized in that: The specific calculation formula for comparing the edge sharpness value and texture complexity value of the components corresponding to the visible light and infrared images is: ； Calculate the composite index value of feature comparison and filter out the coordinate set of component difference areas; Represents the composite index value of feature comparison, Represents the visible light image The edge sharpness value of pixels, Represents the infrared image The edge sharpness value of pixels, Represents the visible light image The texture complexity value of each pixel, Represents the infrared image The texture complexity value of each pixel, Represents the total number of pixels involved in the comparison of edge features, Represents the total number of pixels involved in the calculation of texture features.

8. The switch cabinet multi-mode multi-channel online monitoring device according to claim 6, characterized in that: The equipment status monitoring module comprises: The image fusion submodule synchronously collects the infrared image temperature sequence and the visible light image frame sequence during the operation of the monitoring device according to the fused image data stream, aligns the temperature value of each pixel point of the infrared image with the spatial coordinates of the corresponding area of ​​the visible light image on the time axis, and superimposes the temperature data on the visible light image coordinate system to generate multi-source image fusion data; The temperature trend analysis submodule calls the temperature sequence in the multi-source image fusion data, extracts the temperature extreme difference value and the temperature change amplitude between adjacent frames in the continuous time window of the equipment area, compares the extreme difference value with the temperature fluctuation range corresponding to the preset equipment type, screens the extreme difference exceeding limit area, calculates the statistical fluctuation range area whose change amplitude exceeds the benchmark change amplitude of similar equipment, and generates the temperature anomaly mark area; The morphological association determination submodule extracts the component contour deformation amount in the temperature anomaly marked area, calculates the mean Euclidean distance of the contour point set between adjacent frames based on the visible light image frame sequence, matches the spatial coordinates of the area where the deformation amount exceeds the assembly tolerance range with the temperature anomaly area, marks the boundaries of the area where both the deformation amount exceeds the standard and the temperature anomaly exists, and maps the coordinates of the temperature anomaly area to the monitoring interface to generate the equipment status monitoring results.

9. The switch cabinet multi-mode multi-channel online monitoring device according to claim 8, characterized in that: The specific calculation formula for calculating the mean value of the Euclidean distance of the contour point set between adjacent frames is: ； in, Represents the mean Euclidean distance of contour point sets between adjacent frames, represents the Euclidean distance of the contour point set between the i-th frame and the j-th frame, represents the average Euclidean distance between all contour point sets, and Q represents the total number of calculated contour point sets.

10. The switch cabinet multi-mode multi-channel online monitoring device according to claim 8, characterized in that: Also includes: The abnormal screen warning module calls the equipment status monitoring results, automatically marks the abnormal area on the monitoring screen and adjusts the display mode, and at the same time compares and analyzes the data of the abnormal area, evaluates the severity of the problem, and triggers abnormal screen warning information; The abnormal screen warning information includes abnormal area marking, adjustment of display mode, data comparison and analysis, and assessment of problem severity; The abnormal picture warning module includes: The abnormal area marking submodule obtains the equipment operation data set in the equipment status monitoring result, calls the preset benchmark comparison rule, calculates the difference between the partition values ​​in the data set and the corresponding benchmark item by item, selects the partitions whose differences exceed the set deviation threshold, maps the partition coordinates to the monitoring screen, and generates the abnormal area coordinates; The display adjustment and comparison submodule calls the coordinates of the abnormal area, extracts the real-time monitoring data stream in the abnormal area, adjusts the color saturation and flickering frequency in the screen display settings, synchronously obtains the normal data mean value within the operation cycle of the same area, compares the real-time data stream with the data mean in segmented time series, and generates an abnormal data comparison set; The abnormal assessment trigger submodule extracts the maximum deviation amplitude and the mean fluctuation frequency based on the abnormal data comparison set, combines the load degree and the proportion of long operation time in the equipment operation data set, calculates the abnormal deviation value, matches the deviation value with the preset warning level threshold range, triggers the corresponding level of warning instructions, and generates abnormal screen warning information.

Images