Method, device and system for monitoring operation state of power equipment

Abstract

The invention provides a power equipment operation state monitoring method, device and monitoring system, and relates to the technical field of power equipment inspection. The method comprises the following steps: acquiring a video stream of the indicating lamp of the power equipment, and extracting an indicating lamp area image of each frame of picture from the video stream according to a pre-acquired calibration picture of the indicating lamp; calculating the gray difference of the adjacent frames of indicating lamp area images, generating a plurality of frame difference maps, and judging whether the state of the indicating lamp is flickering according to the plurality of frame difference maps; if the state of the indicating lamp is not flickering, extracting a preset number of indicating lamp area images, and determining that the state of the indicating lamp is normally on or normally off according to the brightness of the extracted indicating lamp area images; and determining the operation state of the power equipment based on the state of the indicator lamp. According to the invention, the problems of low efficiency and error proneness caused by manual identification of the power equipment indicating lamp in power inspection can be solved.

Description

Technical Field

[0001] This invention relates to the field of power equipment inspection technology, and in particular to a method, device and monitoring system for monitoring the operating status of power equipment. Background Technology

[0002] In the operation and maintenance management system of power equipment rooms, core equipment is typically equipped with multi-color indicator lights. These indicator lights, through different on / off rhythms and color combinations (such as constant on, constant off, high-frequency flashing, low-frequency flashing, etc.), convey key information in real time, such as normal equipment operation, fault alarms, and standby status. Therefore, accurate identification and real-time monitoring of the indicator light's working status is a crucial technical aspect for ensuring the safe and stable operation of power equipment and achieving automated operation and maintenance as well as rapid fault location.

[0003] Currently, monitoring the operational status of power equipment mainly relies on two technological approaches: Firstly, there is the traditional manual inspection mode. Maintenance personnel periodically conduct on-site checks, visually identifying the status of indicator lights. However, this method is limited by human efficiency and is difficult to adapt to high-density, large-scale equipment cluster scenarios. At the same time, the identification results heavily depend on the experience, fatigue level, and concentration of the inspectors, and are prone to missed or incorrect judgments due to subjective judgment biases, failing to meet the needs of modern data centers for 24 / 7, high-reliability monitoring.

[0004] Secondly, there is automated monitoring technology based on image processing. Although some technical solutions use camera acquisition combined with optical recognition algorithms to replace manual work, there are still obvious technical bottlenecks in practical applications: insufficient robustness of status recognition, poor adaptability to small targets and low brightness, and weak anti-interference ability in complex environments, which directly affect the accuracy of equipment status recognition and the stability of long-term operation. Summary of the Invention

[0005] This invention provides a method, device, and system for monitoring the operating status of power equipment, in order to solve the problems of low accuracy and poor adaptability of existing power equipment operating status monitoring methods.

[0006] In a first aspect, embodiments of the present invention provide a method for monitoring the operating status of power equipment, including: Acquire video streams of power equipment indicator lights, and extract indicator light area images from each frame of the video stream based on pre-collected calibration images of the indicator lights; Calculate the grayscale difference between adjacent frame indicator light regions to generate multiple frame difference maps, and determine whether the indicator light is flashing based on the multiple frame difference maps; If the indicator light is not flashing, a preset number of indicator light area images are extracted, and the state of the indicator light is determined to be either constantly on or constantly off based on the brightness of the extracted indicator light area images. The operating status of the power equipment is determined based on the status of the indicator lights.

[0007] In one possible implementation, determining whether the indicator light is flashing based on the plurality of frame difference maps includes: Each frame difference map is binarized. The binarization process involves setting the values ​​of pixels in the frame difference map whose grayscale difference is greater than or equal to a first preset threshold as first values, and setting the values ​​of pixels whose grayscale difference is less than the first preset threshold as second values. Morphological optimization is performed on the binarized frame difference map; Determine the area percentage of the pixel with the first value in the frame difference map; Based on the area ratio corresponding to each frame difference map, determine whether the indicator light is flashing.

[0008] In one possible implementation, determining whether the indicator light is flashing based on the area proportion corresponding to each frame difference map includes: If the number of frame difference images with an area ratio greater than the second preset threshold reaches the third preset threshold, then the state of the indicator light is determined to be flashing. Otherwise, the indicator light is determined to be not flashing.

[0009] In one possible implementation, determining whether the indicator light is constantly on or constantly off based on the brightness of the extracted indicator light area image includes: The extracted indicator light area image is subjected to brightness binarization processing. The brightness binarization processing is to set the values ​​of pixels in the indicator light area image whose brightness is greater than or equal to a fourth preset threshold as a third value, and set the values ​​of pixels whose brightness is less than the fourth preset threshold as a fourth value. Morphological optimization is performed on the indicator area image after brightness binarization. Calculate the area percentage of the pixel with the third value in the indicator light area image; Based on the area ratio of each indicator light region image, the state of the indicator light is determined to be either constantly on or constantly off.

[0010] In one possible implementation, determining whether the indicator light is constantly on or constantly off based on the area proportion corresponding to each indicator light region image includes: If the area ratio of all indicator light regions is greater than or equal to the fifth preset threshold, then the indicator light is determined to be constantly lit. If the area ratio of all indicator light regions is less than the fifth preset threshold, then the indicator light is determined to be in a state of constant off.

[0011] In one possible implementation, the area percentage of the pixel whose determined value is a first value in the frame difference map includes: The Suzuki contour detection algorithm is used to detect connected regions of pixels whose values ​​are the first value. The area ratio of the connected region to the frame difference map is calculated to obtain the area percentage.

[0012] In one possible implementation, extracting the indicator light area image of each frame from the video stream based on the pre-acquired calibration image of the indicator light includes: Feature points are extracted from the first frame of the video stream and the calibration image, and feature points of the two images are matched by calculating feature descriptors to generate a homography matrix; Based on the homography matrix, the positions of the indicator lights in the calibration image are mapped to each frame of the video stream, and the indicator light area image of each frame is extracted based on the indicator light positions.

[0013] In one possible implementation, after extracting the indicator area image of each frame from the video stream, the method further includes: The extracted indicator light area image is preprocessed; the preprocessing includes: Gaussian filtering for noise reduction, histogram equalization, and color space conversion.

[0014] Secondly, embodiments of the present invention provide a power equipment operation status monitoring device, comprising: The acquisition module is used to acquire the video stream of the indicator lights of the power equipment, and extract the indicator light area image of each frame from the video stream based on the pre-collected calibration image of the indicator lights; The first judgment module is used to calculate the grayscale difference between adjacent frame indicator light area images, generate multiple frame difference maps, and determine whether the state of the indicator light is flashing based on the multiple frame difference maps; The second judgment module is used to extract a preset number of indicator light area images if the indicator light is not flashing, and determine whether the indicator light is always on or always off based on the brightness of the extracted indicator light area images. The analysis module is used to determine the operating status of the power equipment based on the status of the indicator lights.

[0015] Thirdly, embodiments of the present invention provide a monitoring system, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation of the first aspect.

[0016] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: This invention acquires video streams of power equipment indicator lights and extracts indicator light area images from each frame of the video stream based on pre-collected calibration images of the indicator lights. This corrects camera position deviations and ensures accurate indicator light area extraction. By calculating the grayscale difference between adjacent indicator light area images, multiple frame difference maps are generated. Based on these multiple frame difference maps, it is possible to accurately determine whether the indicator light is flashing. If the indicator light is not flashing, a preset number of indicator light area images are extracted. Based on the brightness of the extracted indicator light area images, it is possible to accurately determine whether the indicator light is constantly on or constantly off. Finally, based on the indicator light status, the operating status of the power equipment can be accurately determined. This invention has low computational complexity, allows for long-term inspection and identification tasks with only one calibration, saving significant manpower and costs. Furthermore, this method exhibits good accuracy and adaptability in various environments. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of the power equipment operation status monitoring method provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the structure of the power equipment operation status monitoring device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the monitoring system provided in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] This application is applicable to the monitoring of the operating status of various power equipment in industrial and maintenance scenarios, and is especially suitable for scenarios with dense equipment deployment and low efficiency of manual inspection.

[0020] Taking data center operation and maintenance as a typical application scenario as an example, the system can automatically identify and monitor the status of indicator lights on power equipment within the data center. Specifically, various indicator lights configured on power equipment (such as power indicator lights, port data transmission lights, and fault alarm lights) are key visual indicators reflecting the operating status of the equipment. The working modes of different types of indicator lights (always on, always off, flashing at a specific frequency) directly correspond to core information such as the stability of the equipment's power supply, the validity of port connections, data transmission status, and fault alarms. This embodiment can complete the real-time collection, analysis, and judgment of indicator light status, not only replacing traditional manual inspections but also enabling timely detection of abnormal states of fault alarm lights through automated monitoring, providing technical support for the intelligent and efficient operation and maintenance of data centers.

[0021] Similarly, the technical logic of this invention can also be applied to indicator light recognition scenarios for other fixed equipment, such as indicator lights for computer room servers and warning lights for industrial control cabinets.

[0022] See Figure 1 The document illustrates a flowchart of the power equipment operation status monitoring method provided in an embodiment of the present invention, which is described in detail below: Step S101: Acquire the video stream of the power equipment indicator light, and extract the indicator light area image of each frame from the video stream based on the pre-collected calibration image of the indicator light.

[0023] In this embodiment, to ensure the accuracy of subsequent monitoring of the operating status of power equipment, the inspection robot needs to be pre-controlled to perform a calibration task during the system deployment phase. The core objective is to obtain the reference position information of the indicator lights of the equipment to be identified. The specific operation is as follows: At the preset point corresponding to the equipment to be identified, the inspection robot is controlled to move to the optimal imaging distance of 1-3 meters directly in front of the equipment to collect calibration images. This distance range is determined based on the size of common indicator lights in industrial scenarios and the focal length parameters of the camera mounted on the robot. This ensures that the indicator lights in the image are clear and have a reasonable proportion. At the same time, it is necessary to ensure that the equipment is unobstructed and the lighting conditions are stable, avoiding dust, obstacles, or direct strong light from interfering with the calibration accuracy. After the data collection is completed, the system automatically records and saves the geographical coordinates and attitude information of the equipment point.

[0024] During the calibration phase, the inspection robot utilizes its YOLO-based target detection algorithm to automatically detect indicator lights in the calibration image, outputting initial detection bounding boxes (including position coordinates and size information) for each indicator light. Subsequently, through the robot's system interface, staff can manually verify and fine-tune the detection bounding boxes. This includes manually correcting the boundaries of the bounding boxes to address potential detection offsets or missed detections, ensuring complete and accurate selection of the indicator light area. Finally, the calibration results—including the calibration image, corrected indicator light detection box coordinates, device location information, and device ID—are linked and stored in the system database to form a standardized calibration dataset for use in subsequent recognition stages.

[0025] After entering the recognition phase, the inspection robot executes its tasks according to the preset inspection path. Once the positioning module confirms that it has arrived at the calibration point of the device to be identified, it activates the image acquisition module to acquire a video stream of the target area. To balance recognition accuracy and data transmission efficiency, the video acquisition parameters are set as follows: acquisition duration is approximately 5 seconds, frame rate is 25fps, and video resolution is strictly consistent with the calibration image to ensure image quality matching. After acquisition, the video stream is transmitted in real time to the remote monitoring platform via an industrial Ethernet or 5G wireless communication module, where the platform executes subsequent image processing and status recognition logic.

[0026] In one possible implementation, the monitoring platform achieves precise localization of the indicator light area through image feature matching and spatial mapping. The specific process is as follows: First, image feature points are extracted from the first frame of the video stream and the corresponding calibration images pre-stored in the database. Then, feature descriptors are calculated for the extracted feature points, and precise matching of feature points between the two images is achieved through similarity comparison of the feature descriptors. Based on the effective feature point pairs after matching, a homography matrix representing the image coordinate mapping relationship under the two perspectives is constructed. Finally, the homography matrix is ​​used to map the precisely labeled indicator light position coordinates in the calibration images to each frame of the video stream. Based on the mapped coordinate information, an image cropping algorithm is used to extract the indicator light area image frame by frame, providing a focused target area for subsequent status recognition.

[0027] As a preferred example, the monitoring platform can employ the Scale-Invariant Feature Transform (SIFT) algorithm to perform feature point extraction and matching operations. The specific implementation details are as follows: The SIFT algorithm is used to detect feature points in the first frame of the video stream and the calibration image. This algorithm possesses scale invariance and rotation invariance, effectively resisting feature distortion caused by slight pose deviations of the inspection robot and fluctuations in industrial lighting. For each detected feature point, a SIFT feature descriptor is calculated. This descriptor can fully characterize the gray-level gradient distribution information around the feature point, providing a reliable basis for feature matching. The Fast Library for Approximate Nearest Neighbors (FLANN) feature matcher can be used to perform feature point matching. To eliminate mismatches caused by similar background textures and noise interference during the matching process, random sampling consistency (RandomSampling Consensus) can be introduced. The SampleConsensus (RANSAC) algorithm filters the matching results, retaining valid feature point pairs with geometric consistency. Based on the valid feature point pairs, a robust homography matrix is ​​obtained. The homography matrix can describe the coordinate mapping relationship of points in the same plane in images from different viewpoints. Since there is a spatial transformation relationship between the first frame of the video stream and the calibration image (described by the homography matrix), the position coordinates of the indicator light in the calibration image are substituted into the homography matrix for calculation to obtain the accurate corresponding position of the indicator light in the first frame of the video stream. Finally, the indicator light area image is cropped out.

[0028] In one possible implementation, to improve the accuracy of subsequent monitoring of the operating status of power equipment, a series of preprocessing operations can be performed on the image of the indicator light area extracted frame by frame, such as Gaussian filtering for noise reduction, histogram equalization, and color space conversion.

[0029] Step S102: Calculate the grayscale difference between adjacent frame indicator light area images, generate multiple frame difference maps, and determine whether the indicator light is flashing based on the multiple frame difference maps.

[0030] In this embodiment, each frame difference image is first binarized. Binarization involves setting the values ​​of pixels with grayscale differences greater than or equal to a first preset threshold as a first value, and setting the values ​​of pixels with grayscale differences less than the first preset threshold as a second value. The first preset threshold can be determined by considering the noise level of the video captured by the inspection robot, the brightness variation of the indicator lights, and the lighting conditions in the industrial environment. For example, in a stable indoor industrial setting, it can be set to 20-30 (grayscale value range 0-255). If the first preset threshold is set to 30, the first value can be set to 255, and the second value can be set to 0, then pixels with grayscale differences ≥30 in the frame difference image will be marked as white, and pixels with grayscale differences <30 will be marked as black.

[0031] Then, morphological optimization is performed on the binarized frame difference map. Morphological optimization eliminates noise interference in the frame difference map through dilation and erosion operations.

[0032] Then, the area ratio of pixels with the first value in the frame difference image is determined. If the number of frame difference images with an area ratio greater than the second preset threshold reaches the third preset threshold, it indicates that the indicator light has a large grayscale change between adjacent frames, and the state is determined to be flashing; otherwise, the state of the indicator light is determined not to be flashing. The setting of the second preset threshold is related to the proportion of the indicator light in the indicator light area image. For example, if the indicator light occupies 30% in the switch indicator light image, the second preset threshold can be set to 25%. The third preset threshold can be set according to actual needs, such as 5, 6, 7, etc.

[0033] Regarding the calculation of the area ratio: the Suzuki contour detection algorithm can be used to detect connected regions of pixels with the first value (white); the area ratio of the connected regions to the frame difference map is calculated to obtain the area ratio. Other methods can also be used, which are not limited in this embodiment.

[0034] Step S103: If the indicator light is not flashing, extract a preset number of indicator light area images, and determine whether the indicator light is always on or always off based on the brightness of the extracted indicator light area images.

[0035] In this embodiment, if the indicator light is not flashing, a preset number of indicator light area images are extracted, such as uniformly extracting 3 frames. Similar to step S102, the extracted indicator light area images are subjected to brightness binarization processing, that is, the values ​​of pixels with brightness greater than or equal to a fourth preset threshold in the indicator light area image are set as a third value, and the values ​​of pixels with brightness less than the fourth preset threshold are set as a fourth value. The fourth preset threshold is set according to requirements, scenarios, etc., and the third value can be set to 255 and the fourth value can be set to 0.

[0036] Then, morphological optimization is performed on the indicator light area image after brightness binarization. Next, the area percentage of the indicator light area image corresponding to the third value (white, corresponding to higher brightness pixels) is calculated. If the area percentage of each frame of the indicator light area image is greater than or equal to a fifth preset threshold, it indicates the existence of a continuous high-brightness area within the indicator light area, and the indicator light state is determined to be constantly on. If the area percentage of each frame of the indicator light area image is less than the fifth preset threshold, it indicates the absence of a continuous high-brightness area within the indicator light area, and its state is determined to be constantly off. The fifth preset threshold can be set based on the standard brightness characteristics of the target indicator light in an industrial scenario. It should be noted that since the indicator light has been determined not to be flashing, the possible states of the indicator light are only constantly on (area percentages all greater than or equal to the fifth preset threshold) or constantly off (area percentages all less than the fifth preset threshold). If some area percentages are detected to be greater than or equal to the fifth preset threshold, and some area percentages are detected to be less than the fifth preset threshold, it indicates an error, and an error message can be displayed, such as indicating that no indicator light state was detected.

[0037] Step S104: Determine the operating status of the power equipment based on the status of the indicator lights.

[0038] This invention acquires video streams of power equipment indicator lights and extracts indicator light area images from each frame of the video stream based on pre-collected calibration images of the indicator lights. This corrects camera position deviations and ensures accurate indicator light area extraction. By calculating the grayscale difference between adjacent indicator light area images, multiple frame difference maps are generated. Based on these multiple frame difference maps, it is possible to accurately determine whether the indicator light is flashing. If the indicator light is not flashing, a preset number of indicator light area images are extracted. Based on the brightness of the extracted indicator light area images, it is possible to accurately determine whether the indicator light is constantly on or constantly off. Finally, based on the indicator light status, the operating status of the power equipment can be accurately determined. This invention has low computational complexity, allows for long-term inspection and identification tasks with only one calibration, saving significant manpower and costs. Furthermore, this method exhibits good accuracy and adaptability in various environments.

[0039] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0040] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0041] Figure 2 A schematic diagram of the power equipment operation status monitoring device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the power equipment operation status monitoring device 2 includes: The acquisition module 21 is used to acquire the video stream of the power equipment indicator light and extract the indicator light area image of each frame from the video stream based on the pre-collected calibration image of the indicator light. The first judgment module 22 is used to calculate the grayscale difference of adjacent frame indicator light area images, generate multiple frame difference maps, and determine whether the state of the indicator light is flashing based on the multiple frame difference maps; The second judgment module 23 is used to extract a preset number of indicator light area images if the indicator light is not flashing, and determine whether the indicator light is always on or always off based on the brightness of the extracted indicator light area images. Analysis module 24 is used to determine the operating status of power equipment based on the status of indicator lights.

[0042] In one possible implementation, the first determining module 22 is used to: Each frame difference map is binarized. The binarization process involves setting the values ​​of pixels in the frame difference map whose grayscale difference is greater than or equal to a first preset threshold as first values, and setting the values ​​of pixels whose grayscale difference is less than the first preset threshold as second values. Morphological optimization is performed on the binarized frame difference map; Determine the area percentage of the pixel with the first value in the frame difference map; Based on the area ratio corresponding to each frame difference map, determine whether the indicator light is flashing.

[0043] In one possible implementation, the first determining module 22 is used to: If the number of frame difference images with an area ratio greater than the second preset threshold reaches the third preset threshold, then the state of the indicator light is determined to be flashing. Otherwise, the indicator light is determined to be not flashing.

[0044] In one possible implementation, the second judgment module 23 is used for: The extracted indicator light area image is subjected to brightness binarization processing. The brightness binarization processing is to set the values ​​of pixels in the indicator light area image whose brightness is greater than or equal to a fourth preset threshold as a third value, and set the values ​​of pixels whose brightness is less than the fourth preset threshold as a fourth value. Morphological optimization is performed on the indicator area image after brightness binarization. Calculate the area percentage of the pixel with the third value in the indicator light area image; Based on the area ratio of each indicator light region image, the state of the indicator light is determined to be either constantly on or constantly off.

[0045] In one possible implementation, the second judgment module 23 is used for: If the area ratio of all indicator light regions is greater than or equal to the fifth preset threshold, then the indicator light is determined to be constantly lit. If the area ratio of all indicator light regions is less than the fifth preset threshold, then the indicator light is determined to be in a state of constant off.

[0046] In one possible implementation, the first determining module 22 is used to: The Suzuki contour detection algorithm is used to detect connected regions of pixels whose values ​​are the first value. The area ratio of the connected region to the frame difference map is calculated to obtain the area percentage.

[0047] In one possible implementation, the acquisition module 21 is used for: Feature points are extracted from the first frame of the video stream and the calibration image, and feature points of the two images are matched by calculating feature descriptors to generate a homography matrix; Based on the homography matrix, the positions of the indicator lights in the calibration image are mapped to each frame of the video stream, and the indicator light area image of each frame is extracted based on the indicator light positions.

[0048] In one possible implementation, after extracting the indicator area image of each frame from the video stream, the acquisition module 21 is further configured to: The extracted indicator light area image is preprocessed; the preprocessing includes: Gaussian filtering for noise reduction, histogram equalization, and color space conversion.

[0049] This invention acquires video streams of power equipment indicator lights and extracts indicator light area images from each frame of the video stream based on pre-collected calibration images of the indicator lights. This corrects camera position deviations and ensures accurate indicator light area extraction. By calculating the grayscale difference between adjacent indicator light area images, multiple frame difference maps are generated. Based on these multiple frame difference maps, it is possible to accurately determine whether the indicator light is flashing. If the indicator light is not flashing, a preset number of indicator light area images are extracted. Based on the brightness of the extracted indicator light area images, it is possible to accurately determine whether the indicator light is constantly on or constantly off. Finally, based on the indicator light status, the operating status of the power equipment can be accurately determined. This invention has low computational complexity, allows for long-term inspection and identification tasks with only one calibration, saving significant manpower and costs. Furthermore, this method exhibits good accuracy and adaptability in various environments.

[0050] Figure 3 This is a schematic diagram of the monitoring system provided in an embodiment of the present invention. Figure 3 As shown, the monitoring system 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the above-described method embodiments. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module in the above-described device embodiments.

[0051] For example, computer program 32 can be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of computer program 32 in monitoring system 3.

[0052] The monitoring system 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of monitoring system 3 and does not constitute a limitation on monitoring system 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, monitoring system 3 may also include input / output devices, network access devices, buses, etc.

[0053] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0054] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0055] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for monitoring the operating status of power equipment, characterized in that, include: Acquire video streams of power equipment indicator lights, and extract indicator light area images from each frame of the video stream based on pre-collected calibration images of the indicator lights; Calculate the grayscale difference between adjacent frame indicator light regions to generate multiple frame difference maps, and determine whether the indicator light is flashing based on the multiple frame difference maps; If the indicator light is not flashing, a preset number of indicator light area images are extracted, and the state of the indicator light is determined to be either constantly on or constantly off based on the brightness of the extracted indicator light area images. The operating status of the power equipment is determined based on the status of the indicator lights.

2. The method for monitoring the operating status of power equipment according to claim 1, characterized in that, The step of determining whether the indicator light is flashing based on the multiple frame difference maps includes: Each frame difference map is binarized. The binarization process involves setting the values ​​of pixels in the frame difference map whose grayscale difference is greater than or equal to a first preset threshold as first values, and setting the values ​​of pixels whose grayscale difference is less than the first preset threshold as second values. Morphological optimization is performed on the binarized frame difference map; Determine the area percentage of the pixel with the first value in the frame difference map; Based on the area ratio corresponding to each frame difference map, determine whether the indicator light is flashing.

3. The method for monitoring the operating status of power equipment according to claim 2, characterized in that, The step of determining whether the indicator light is flashing based on the area proportion corresponding to each frame difference map includes: If the number of frame difference images with an area ratio greater than the second preset threshold reaches the third preset threshold, then the state of the indicator light is determined to be flashing. Otherwise, the indicator light is determined to be not flashing.

4. The method for monitoring the operating status of power equipment according to claim 1, characterized in that, The step of determining whether the indicator light is constantly on or constantly off based on the brightness of the extracted indicator light area image includes: The extracted indicator light area image is subjected to brightness binarization processing. The brightness binarization processing is to set the values ​​of pixels in the indicator light area image whose brightness is greater than or equal to a fourth preset threshold as a third value, and set the values ​​of pixels whose brightness is less than the fourth preset threshold as a fourth value. Morphological optimization is performed on the indicator area image after brightness binarization. Calculate the area percentage of the pixel with the third value in the indicator light area image; Based on the area ratio of each indicator light region image, the state of the indicator light is determined to be either constantly on or constantly off.

5. The method for monitoring the operating status of power equipment according to claim 4, characterized in that, The step of determining whether the indicator light is constantly on or constantly off based on the area proportion corresponding to each indicator light area image includes: If the area ratio of all indicator light regions is greater than or equal to the fifth preset threshold, then the indicator light is determined to be constantly lit. If the area ratio of all indicator light regions is less than the fifth preset threshold, then the indicator light is determined to be in a state of constant off.

6. The method for monitoring the operating status of power equipment according to claim 2, characterized in that, The area percentage of the pixels whose determined value is the first value in the frame difference map includes: The Suzuki contour detection algorithm is used to detect connected regions of pixels whose values ​​are the first value. The area ratio of the connected region to the frame difference map is calculated to obtain the area percentage.

7. The method for monitoring the operating status of power equipment according to any one of claims 1 to 6, characterized in that, The step of extracting the indicator light area image of each frame from the video stream based on the pre-acquired calibration image of the indicator light includes: Feature points are extracted from the first frame of the video stream and the calibration image, and feature points of the two images are matched by calculating feature descriptors to generate a homography matrix; Based on the homography matrix, the positions of the indicator lights in the calibration image are mapped to each frame of the video stream, and the indicator light area image of each frame is extracted based on the indicator light positions.

8. The method for monitoring the operating status of power equipment according to any one of claims 1 to 6, characterized in that, After extracting the indicator light area image of each frame from the video stream, the method further includes: The extracted indicator light area image is preprocessed; the preprocessing includes: Gaussian filtering for noise reduction, histogram equalization, and color space conversion.

9. A power equipment operation status monitoring device, characterized in that, include: The acquisition module is used to acquire the video stream of the indicator lights of the power equipment, and extract the indicator light area image of each frame from the video stream based on the pre-collected calibration image of the indicator lights; The first judgment module is used to calculate the grayscale difference between adjacent frame indicator light area images, generate multiple frame difference maps, and determine whether the state of the indicator light is flashing based on the multiple frame difference maps; The second judgment module is used to extract a preset number of indicator light area images if the indicator light is not flashing, and determine whether the indicator light is always on or always off based on the brightness of the extracted indicator light area images. The analysis module is used to determine the operating status of the power equipment based on the status of the indicator lights.

10. A monitoring system, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images