GIS switch contact identification method and device based on mixed-band light source lighting technology

Abstract

This application relates to a GIS switch contact identification method and device based on hybrid-band light source lighting technology. The method captures images of GIS switch contacts under different-band light sources, processes and fuses the captured images to optimize overall image quality and detail, and then performs image recognition processing on the fused images to obtain a feature map. The switch state can be analyzed and determined based on the number and position of the identification lines on the feature map: if there are no identification lines in the feature map, the switch is in the open state; if there is one identification line in the feature map, the switch is in the fault state; if there are two identification lines in the feature map, the area ratio of the identification lines to the feature image of the identification line area is determined, and the position travel of the identification lines relative to the moving contact is determined. Based on the area ratio and position travel, it is determined whether the GIS switch is fully closed. The method of this application can accurately identify the state of a GIS switch.

Description

Technical Field

[0001] The present application relates to the field of electric power grid technology, and in particular to a GIS switch contact identification method and device based on mixed-band light source lighting technology. Background Art

[0002] As power systems increase their reliability and safety requirements, the performance and stability of GIS equipment, as core equipment in high-voltage power systems, directly impact the stable operation of the entire power system. Accurately identifying the position of switch contacts in real time ensures the safe operation of GIS equipment in various operating states, preventing equipment damage or power system failures caused by misoperation.

[0003] SF6 (sulfur hexafluoride), with a density five times that of air, is widely used in GIS equipment due to its excellent insulation and arc-extinguishing properties. Consequently, GIS equipment is subject to high voltage and requires tight sealing in confined spaces. Therefore, traditional manual inspections primarily rely on visual verification of SF6 density and pressure gauges, checking the temperature of exposed busbars, and inspecting the appearance of the equipment to monitor the overall operation and abnormalities or faults of GIS equipment. However, these inspections cannot automatically identify the open and closed states of GIS switch contacts. With the development of computer vision, image recognition technology is commonly used to detect the open and closed states of switches. This technology can be used to collect, analyze, and process images of switches in different states to ultimately determine the actual open and closed states. However, conventional image recognition technology struggles to accurately observe the status of disconnectors and grounding switches within the high-density SF6 medium through observation windows in dimly lit environments. Therefore, research into GIS switch contact position recognition technology is crucial for ensuring the safe operation of power systems. Summary of the Invention

[0004] In order to overcome at least one deficiency in the prior art, the present application provides a GIS switch contact identification method and device based on mixed-band light source lighting technology.

[0005] In a first aspect, a GIS switch contact identification method based on mixed-band light source lighting technology is provided, comprising:

[0006] Obtain monitoring images of GIS switch contacts under illumination from light sources of different wavelengths;

[0007] The surveillance images illuminated by light sources of different wavelengths are decomposed to obtain the base layer, detail layer, and saliency layer. The base layer, detail layer, and saliency layer corresponding to the surveillance images illuminated by light sources of different wavelengths are fused to obtain a fused image.

[0008] Obtain a standard closed image without identification lines when the switch is fully closed. Based on the standard closed image without identification lines and the fused image, obtain an image of the identification line area on the contact. Binarize and morphologically process the image of the identification line area on the contact to extract a feature image of the identification line area. The standard closed image without identification lines when the switch is fully closed includes the static contact area, the moving contact area, and the connecting rod area.

[0009] Judging whether there is an identification line based on the characteristic image of the identification line area, if not, it is determined that the GIS switch is in the open state, if it exists, the number of identification lines is further judged, if there is one identification line, it is determined that the GIS switch is in the fault state, if there are two identification lines, the area ratio of the identification line occupying the characteristic image of the identification line area is determined, and the position travel of the identification line relative to the moving contact is determined, and whether the GIS switch is in the closed position is determined based on the area ratio and position travel.

[0010] In one embodiment, an image of the area with identification lines on the contacts is obtained based on a standard closed image without identification lines and a fused image of the closed position, including:

[0011] Perform image enhancement on the standard closed image and the fused image without the marking line when the switch is fully closed, to obtain an enhanced standard closed image and an enhanced fused image;

[0012] The enhanced standard composite image and the enhanced fused image are clustered using a superpixel fuzzy clustering algorithm to obtain a clustered standard composite image and a clustered fused image.

[0013] The clustered standard composite image and the clustered fusion image are subjected to difference processing to obtain an image of the identification line area on the contact.

[0014] In one embodiment, determining the position travel of the identification line relative to the moving contact includes:

[0015] Obtain the characteristic diagram of the moving contact in the closed state;

[0016] Calculate the pixel distance between the identification line and the moving contact feature map;

[0017] Based on the dimensional relationship obtained by visual calibration, the pixel point distance is converted into the distance in the actual world coordinate system, that is, the position travel of the identification line relative to the moving contact.

[0018] In one embodiment, obtaining a characteristic diagram of a moving contact in a closed state includes:

[0019] Acquire a contact image and a standard sub-image; the contact image includes the static contact area, and the standard sub-image is an image of the contact in the closed state, including the static contact area and the moving contact area;

[0020] Performing image enhancement on the standard segment image and the contact image to obtain an image-enhanced standard segment image and an image-enhanced contact image;

[0021] The enhanced standard score image and the enhanced contact image are processed by super-pixel fuzzy clustering algorithm to obtain the clustered standard score image and the clustered contact image;

[0022] Perform subtraction processing on the clustered standard score image and the clustered contact image to obtain the image of the moving contact area;

[0023] The image of the moving contact area is binarized and morphologically processed to extract the moving contact feature map.

[0024] In one embodiment, determining whether the GIS switch is fully closed based on the area ratio and the position travel includes:

[0025] The error between the calculated area ratio and the area ratio when the switch is fully closed;

[0026] If the error is greater than the set error threshold and the position travel is less than the set distance threshold, it is determined that the GIS switch is not fully closed; otherwise, it is determined that the GIS switch is fully closed.

[0027] In one embodiment, the light sources of different wavelength bands include white light and yellow light.

[0028] In a second aspect, a GIS switch contact identification device based on mixed-band light source lighting technology is provided, comprising:

[0029] An image acquisition module is used to acquire monitoring images of GIS switch contacts under illumination by light sources of different wavelengths, including white light and yellow light;

[0030] The image decomposition and fusion module is used to decompose the surveillance images illuminated by light sources of different bands to obtain the base layer, detail layer and saliency layer; and fuse the base layer, detail layer and saliency layer corresponding to the surveillance images illuminated by light sources of different bands to obtain a fused image;

[0031] The identification line area feature image acquisition module is used to obtain a standard closing image without identification lines when the switch is fully closed. Based on the standard closing image without identification lines and the fused image, an image of the identification line area on the contact is obtained. The image of the identification line area on the contact is binarized and morphologically processed to extract the feature image of the identification line area. The standard closing image without identification lines when the switch is fully closed includes the static contact area, the moving contact area, and the connecting rod area.

[0032] The judgment module is used to judge whether there is an identification line based on the characteristic image of the identification line area. If not, it is determined that the GIS switch is in the open state. If it exists, it is further determined that the number of identification lines is one. If there is one identification line, it is determined that the GIS switch is in a fault state. If there are two identification lines, it is determined that the area ratio of the identification line to the characteristic image of the identification line area is determined, and the position travel of the identification line relative to the moving contact is determined. According to the area ratio and the position travel, it is determined whether the GIS switch is in the closed state.

[0033] In the third aspect, a GIS switch contact identification system based on hybrid band light source lighting technology is provided, including: a light source, an industrial camera, and an FPGA;

[0034] The light source is used to provide lighting of different wavelength bands to the GIS switch contacts;

[0035] Industrial cameras are used to obtain monitoring images of GIS switch contacts under illumination from light sources of different wavelengths;

[0036] FPGA is used to implement the above-mentioned GIS switch contact identification method based on mixed-band light source lighting technology.

[0037] In one embodiment, the system further includes an LED driver and an MCU. The MCU controls the LED driver through dimming, and the LED driver controls the light source to provide illumination of different wavelength bands to the GIS switch contacts.

[0038] In one embodiment, the system further includes a background monitoring system, and the FPGA sends the obtained GIS switch contact identification results to the background monitoring system, and the background monitoring system displays the GIS switch contact identification results.

[0039] Compared with the existing technology, the present application has the following beneficial effects: The present application's GIS switch contact identification method and device based on mixed-band light source lighting technology solves the problem in the existing technology of being unable to automatically and accurately identify the open and closed status of GIS device switch contacts in high-density media under dim conditions. By collecting GIS switch contact images under different band light sources, processing and fusing the collected images to optimize the overall image quality and detail performance, and then performing image recognition processing on the fused images to obtain a feature map, the switch state can be analyzed and judged based on the number and position of the identification lines on the feature map: if there are no identification lines in the feature map, it means that the switch is in the open state; if there is one identification line in the feature map, it means that the switch is in the fault state; if there are two identification lines in the feature map, the area ratio of the identification lines to the feature image of the identification line area is determined, and the position travel of the identification lines relative to the moving contact is determined. Based on the area ratio and position travel, it is determined whether the GIS switch is in the closed state. The method of the present application can accurately identify the state of the GIS switch. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present application may be better understood by referring to the following description in conjunction with the accompanying drawings, which together with the following detailed description are incorporated into and form a part of this specification. In the drawings:

[0041] Figure 1 A flowchart of a GIS switch contact identification method based on mixed-band light source lighting technology according to an embodiment of the present application is shown;

[0042] Figure 2 A schematic diagram of a GIS switch contact identification method based on mixed-band light source lighting technology according to an embodiment of the present application is shown;

[0043] Figure 3 A flowchart of extracting moving contacts and identification lines is shown;

[0044] Figure 4 A structural block diagram of a GIS switch contact identification device based on mixed-band light source lighting technology according to an embodiment of the present application is shown;

[0045] Figure 5 A structural block diagram of a GIS switch contact identification system based on mixed-band light source lighting technology according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0046] Exemplary embodiments of the present application are described below with reference to the accompanying drawings. For the sake of clarity and conciseness, not all features of actual embodiments are described in this specification. However, it should be understood that in the process of developing any such actual embodiment, many implementation-specific decisions may be made to achieve the developer's specific goals, and these decisions may vary from one implementation to another.

[0047] It is also necessary to explain here that, in order to avoid obscuring the present application due to unnecessary details, the accompanying drawings only show the device structure closely related to the solution according to the present application, while other details that are not closely related to the present application are omitted.

[0048] It should be understood that the present application is not limited to the described embodiments due to the following description with reference to the accompanying drawings. In this document, where feasible, the embodiments may be combined with each other, features between different embodiments may be replaced or borrowed, and one or more features may be omitted in one embodiment.

[0049] The present invention provides a method for identifying GIS switch contacts based on mixed-band light source lighting technology. Figure 1A flowchart of a GIS switch contact identification method based on a mixed-band light source lighting technology according to an embodiment of the present application is shown. Figure 1 , methods include:

[0050] Step S1: Acquire monitoring images of GIS switch contacts under illumination by light sources of different wavelength bands.

[0051] Specifically, when capturing images of GIS switch contacts, a design using time-sharing control of illumination using light sources with different wavelengths is employed. Examples of these light sources include white light and yellow light. The reason for capturing images of GIS switch contacts under both white and yellow light sources is that different wavelengths of light have different reflection and absorption characteristics for objects. Combining images from different wavelengths can improve image detail and contrast. Choosing yellow light as the light source reduces the impact of light scattering in the SF6 environment on image quality. This is because yellow light, with its long wavelength, reduces scattering, improves image contrast, and enhances image detail. It also reduces blurring caused by scattering, thereby improving image sharpness. Reduced scattering reduces background noise, enhancing signal clarity and detail. Using white light for illumination better reproduces the true color of objects, resulting in more natural and accurate images. Therefore, the fusion of images illuminated by white and yellow light facilitates more accurate identification and detection of object features, improving the accuracy of switch contact identification and analysis in the SF6 environment.

[0052] Time-sharing control of LED drivers enables illumination of different wavelengths. The control scheme uses precise time division and pulse-width modulation (PWM) to rapidly switch between different wavelength LEDs. Time is first divided into multiple cycles, each corresponding to a different LED color. During each cycle, the corresponding LED is illuminated, while the other LEDs remain off. Synchronous control ensures that the on-off timing of the different wavelength LEDs does not interfere with each other, and correction is achieved through a light sensor feedback mechanism.

[0053] Light source and image synchronization technology relies primarily on precise triggering mechanisms and coordinated control systems. The core of synchronization technology lies in the generation and transmission of trigger signals. The trigger signal is an electronic pulse generated by a trigger or synchronization module. This signal is sent simultaneously to the light source and image acquisition system to ensure they respond at the same time. Trigger signal transmission requires a high-precision distributor to ensure synchronization of the signals when they arrive at each device. The distributor can replicate and transmit a single trigger signal to multiple devices while maintaining signal stability and synchronization accuracy. At the software level, the control program is responsible for configuring and managing the synchronization process. The software can set trigger modes to precisely control the timing of the light source and camera operation. The control program can also monitor the synchronization status in real time and adjust parameters to suit different shooting conditions. For example, the software can preset the light source brightness level and duration before image acquisition begins, ensuring that the light source brightness reaches the predetermined level at the moment of image exposure.

[0054] Step S2: Decompose the surveillance images illuminated by light sources of different wavelengths to obtain a base layer, a detail layer, and a saliency layer; fuse the base layer, detail layer, and saliency layer corresponding to the surveillance images illuminated by light sources of different wavelengths to obtain a fused image.

[0055] Here, the surveillance image is decomposed into a base layer, a detail layer, and a saliency layer, which are then further fused to obtain a fused image. The specific processing process is conventional, for example, the content of the published patent application number 202211589011.9 can be adopted, and will not be repeated here.

[0056] In other embodiments, decomposing the monitoring image may specifically include: processing the original image through a mean filter to obtain a base layer. The base layer is subtracted from the original image to obtain a detail layer. This step is mainly to decompose the image into a larger area (base layer) and a smaller detail part (detail layer). In order to generate a saliency layer, the image is usually processed using a Laplacian operator and a Gaussian operator. The Laplacian operator is used to extract edge information of the image, while the Gaussian operator is used to calculate a saliency map. Through these operations, a saliency map can be generated, which reflects the saliency of each area in the image.

[0057] Step S3, obtain a standard closed image without identification lines when the switch is in place, obtain an image of the identification line area on the contact based on the standard closed image without identification lines and the fused image, and perform binarization and morphological processing on the image of the identification line area on the contact to extract a feature image of the identification line area; the standard closed image without identification lines when the switch is in place includes the static contact area, the moving contact area and the connecting rod area.

[0058] Step S4: Determine whether there is an identification line based on the characteristic image of the identification line area. If not, determine that the GIS switch is in the open state. If so, further determine the number of identification lines. If there is one identification line, determine that the GIS switch is in a fault state. If there are two identification lines, determine the area ratio of the identification line to the characteristic image of the identification line area, and determine the position travel of the identification line relative to the moving contact. Determine whether the GIS switch is in the closed state based on the area ratio and the position travel. Figure 2 A schematic diagram of a GIS switch contact identification method based on mixed-band light source lighting technology according to an embodiment of the present application is shown.

[0059] This embodiment solves the problem in the prior art that it is impossible to automatically and accurately identify the open and closed status of the switch contacts of GIS equipment in high-density media under dim conditions. By collecting images of GIS switch contacts under light sources of different bands, processing and fusing the collected images to optimize the overall image quality and detail performance, and then performing image recognition processing on the fused images to obtain a feature map, the switch state can be analyzed and judged based on the number and position of the identification lines on the feature map: if there is no identification line in the feature map, it means that the switch is in the open state at this time; if there is one identification line in the feature map, it means that the switch is in the fault state at this time; if there are two identification lines in the feature map, the area ratio of the identification line to the feature image of the identification line area is determined, and the position travel of the identification line relative to the moving contact is determined, and whether the GIS switch is closed is determined based on the area ratio and position travel. The method of this application can accurately identify the state of the GIS switch.

[0060] In one embodiment, Figure 3 The flowchart of extracting the moving contact and the identification line is shown. In step S3, an image of the identification line area on the contact is obtained based on the standard closed image without the identification line and the fused image when the switch is closed, including:

[0061] Step S31 : performing image enhancement on the standard combined image and the fused image without the marking line when the switch is in full swing, to obtain an enhanced standard combined image and an enhanced fused image.

[0062] Here, the image enhancement can be gamma enhancement, which improves the brightness and contrast of the image by adjusting the gamma curve of the image, corrects underexposed or overexposed pictures, improves the detail expression ability of the image, and improves the color balance of the image, thereby improving the accuracy of GIS switch contact position recognition.

[0063] Step S32 , clustering the enhanced standard composite image and the enhanced fused image using a superpixel fuzzy clustering algorithm to obtain a clustered standard composite image and a clustered fused image.

[0064] Here, we use a superpixel fuzzy clustering algorithm to perform clustering. By grouping pixels in the image into superpixels, we simplify the image structure into smaller, more easily analyzed regions. This not only helps improve image recognition accuracy but also provides high-quality input for subsequent image subtraction and other processing.

[0065] Step S33 : performing a difference process on the clustered standard composite image and the clustered fusion image to obtain an image of the identification line area on the contact.

[0066] Here, the image difference is to calculate the difference between the pixels of the two clustered images, so as to extract the different features of the two images, that is, the image of the identification line area on the contact.

[0067] Furthermore, after obtaining the image of the identification line area on the contact, in addition to the identification line information, there are different noises such as point, line or block in the image. By using threshold binarization and morphological processing, the pixel grayscale values ​​of these noise areas can be reduced to zero to eliminate unnecessary interference.

[0068] In one embodiment, in step S4, determining the position travel of the identification line relative to the moving contact includes:

[0069] Step S41, obtaining a characteristic diagram of the moving contact in a closed state;

[0070] Step S42, calculating the pixel distance between the identification line and the moving contact characteristic map;

[0071] Step S43 : Based on the size relationship obtained by visual calibration, the pixel point distance is converted into the distance of the actual world coordinate system, that is, the position stroke of the marking line relative to the moving contact.

[0072] Specifically, step S41, see Figure 3 , obtain the moving contact characteristic diagram in the closed state, including:

[0073] Step S411, obtaining a contact image and a standard sub-image; the contact image includes a static contact area, and the standard sub-image is an image of the contact in a closed state, including a static contact area and a moving contact area;

[0074] Step S412: performing image enhancement on the standard sub-image and the contact image to obtain an image-enhanced standard sub-image and an image-enhanced contact image.

[0075] Here, the image enhancement can be gamma enhancement, which improves the brightness and contrast of the image by adjusting the gamma curve of the image, corrects underexposed or overexposed pictures, improves the detail expression ability of the image, and improves the color balance of the image, thereby improving the accuracy of GIS switch contact position recognition.

[0076] Step S413 : Processing the enhanced standard segment image and the enhanced contact image with a superpixel fuzzy clustering algorithm to obtain a clustered standard segment image and a clustered contact image.

[0077] Here, we use a superpixel fuzzy clustering algorithm to perform clustering. By grouping pixels in the image into superpixels, we simplify the image structure into smaller, more easily analyzed regions. This not only helps improve image recognition accuracy but also provides high-quality input for subsequent image subtraction and other processing.

[0078] Step S414 : performing a difference process on the clustered standard image and the clustered contact image to obtain an image of the moving contact area.

[0079] Here, the image difference is to calculate the difference between the pixels of the two clustered images respectively, so as to extract the different features of the two images, that is, the image of the moving contact area.

[0080] Step S415 , performing binarization and morphological processing on the image of the moving contact area to extract a moving contact feature map.

[0081] Here, after acquiring the image of the moving contact area, in addition to the identification line information, there are also different noises such as point, line or block in the image. By using threshold binarization and morphological processing, the pixel grayscale values ​​of these noise areas can be reduced to zero to eliminate unnecessary interference.

[0082] In one embodiment, in step S4, determining whether the GIS switch is fully closed based on the area ratio and the position travel includes:

[0083] The error between the calculated area ratio and the area ratio when the switch is fully closed; the area ratio when the switch is fully closed can be determined in advance based on actual conditions.

[0084] If the error is greater than the set error threshold and the position travel is less than the set distance threshold, it is determined that the GIS switch is not fully closed; otherwise, it is determined that the GIS switch is fully closed.

[0085] Here, the error threshold and the distance threshold are set according to actual project needs and are not specifically limited.

[0086] Adopting the same inventive concept as the GIS switch contact identification method based on mixed-band light source lighting technology, this embodiment also provides a corresponding GIS switch contact identification device based on mixed-band light source lighting technology. Figure 4 The structural block diagram of the GIS switch contact identification device based on the mixed-band light source lighting technology according to the embodiment of the present application is shown. Figure 4 ,include:

[0087] The image acquisition module 41 is used to acquire monitoring images of the GIS switch contacts under illumination by light sources of different wavelengths, including white light and yellow light;

[0088] The image decomposition and fusion module 42 is used to decompose the surveillance images illuminated by light sources of different wavelengths to obtain a base layer, a detail layer, and a saliency layer; and fuse the base layer, detail layer, and saliency layer corresponding to the surveillance images illuminated by light sources of different wavelengths to obtain a fused image.

[0089] The identification line region feature image acquisition module 43 is used to obtain a standard closing image without identification lines when the switch is fully closed. Based on the standard closing image without identification lines and the fused image, an image of the identification line region on the contact is obtained. The image of the identification line region on the contact is binarized and morphologically processed to extract the feature image of the identification line region. The standard closing image without identification lines when the switch is fully closed includes the static contact region, the moving contact region, and the connecting rod region.

[0090] The judgment module 44 is used to judge whether there is an identification line based on the characteristic image of the identification line area. If not, it is determined that the GIS switch is in the open state. If it exists, it is further determined that the number of identification lines is one. If there is one identification line, it is determined that the GIS switch is in a fault state. If there are two identification lines, it is determined that the area ratio of the identification line to the characteristic image of the identification line area is determined, and the position travel of the identification line relative to the moving contact is determined. Based on the area ratio and the position travel, it is determined whether the GIS switch is in the closed state.

[0091] The GIS switch contact identification device based on mixed-band light source lighting technology of this embodiment has the same inventive concept as the GIS switch contact identification method based on mixed-band light source lighting technology mentioned above. Therefore, the specific implementation method of the device can be seen in the embodiment part of the optimization method of Fourier stack imaging lighting system in the previous text, and its technical effect corresponds to the technical effect of the above method, which will not be repeated here.

[0092] The present application also provides a GIS switch contact identification system based on mixed-band light source lighting technology. Figure 5 The structural block diagram of the GIS switch contact identification system based on the mixed-band light source lighting technology according to the embodiment of the present application is shown. Figure 5 ,The system includes: light source, industrial camera, and FPGA.

[0093] The light source is used to provide illumination of different wavelength bands to the GIS switch contacts; the GIS switch contacts are arranged in the dielectric window.

[0094] Industrial cameras are used to obtain monitoring images of GIS switch contacts under illumination from light sources of different wavelengths;

[0095] The FPGA is used to implement the GIS switch contact identification method based on the mixed-band light source lighting technology of the aforementioned embodiment.

[0096] Furthermore, the system also includes an LED driver and an MCU (Microcontroller Unit); the MCU controls the LED driver through dimming, and the LED driver controls the light source to provide illumination of different wavelength bands to the GIS switch contacts.

[0097] Furthermore, the system also includes a background monitoring system. The FPGA sends the obtained GIS switch contact identification results to the background monitoring system through the Ethernet module. The background monitoring system displays the GIS switch contact identification results to realize automatic identification of the switch contact position and abnormal alarm.

[0098] Furthermore, the system also includes an isolating switching power supply, which provides power for the industrial camera and LED driver to ensure their normal use.

[0099] In summary, this application has the following technical effects:

[0100] This application solves the problem in the prior art that it is impossible to automatically and accurately identify the open and closed status of the switch contacts of GIS equipment in high-density media under dim conditions. By collecting images of GIS switch contacts under light sources of different bands, processing and fusing the collected images to optimize the overall image quality and detail performance, and then performing image recognition processing on the fused images to obtain a feature map, the switch state can be analyzed and judged based on the number and position of the identification lines on the feature map: if there is no identification line in the feature map, it means that the switch is in the open state at this time; if there is one identification line in the feature map, it means that the switch is in the fault state at this time; if there are two identification lines in the feature map, the area ratio of the identification line to the feature image of the identification line area is determined, and the position travel of the identification line relative to the moving contact is determined, and whether the GIS switch is closed is determined based on the area ratio and position travel. The method of this application can accurately identify the state of the GIS switch.

[0101] The above descriptions are merely examples of various embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A GIS switch contact identification method based on mixed-band light source lighting technology, characterized in that: include: Obtain monitoring images of GIS switch contacts under illumination from light sources of different wavelengths; Decompose surveillance images illuminated by light sources of different wavelengths to obtain base layer, detail layer and saliency layer; The base layer, detail layer and saliency layer corresponding to the surveillance images illuminated by light sources of different bands are fused to obtain a fused image; Obtaining a standard closed image without identification lines when the switch is fully closed, obtaining an image of the identification line area on the contact based on the standard closed image without identification lines and the fused image, and performing binarization and morphological processing on the image of the identification line area on the contact to extract a feature image of the identification line area; the standard closed image without identification lines when the switch is fully closed includes a static contact area, a moving contact area, and a connecting rod area; Judging whether there is an identification line based on the characteristic image of the identification line area, if not, it is determined that the GIS switch is in the open state, if it exists, further judging the number of identification lines, if there is one identification line, it is determined that the GIS switch is in the fault state, if there are two identification lines, determining the area ratio of the identification line occupied by the characteristic image of the identification line area, and determining the position travel of the identification line relative to the moving contact, and determining whether the GIS switch is in the closed state based on the area ratio and the position travel.

2. The method according to claim 1, wherein Obtaining an image of the marking line area on the contact according to the standard closing image without the marking line when the switch is fully closed and the fused image, including: Performing image enhancement on the standard closed image without identification lines and the fused image of the closed position to obtain an enhanced standard closed image and an enhanced fused image; Performing clustering processing on the image-enhanced standard composite image and the image-enhanced fused image using a superpixel fuzzy clustering algorithm to obtain a clustered standard composite image and a clustered fused image; The clustered standard composite image and the clustered fusion image are subjected to difference processing to obtain an image of the marking line area on the contact.

3. The method according to claim 1, wherein Determine the position of the identification line relative to the moving contact, including: Obtain the characteristic diagram of the moving contact in the closed state; Calculating the pixel distance between the identification line and the moving contact characteristic map; The pixel point distance is converted into a distance in the actual world coordinate system based on the dimensional relationship obtained by visual calibration, that is, the position stroke of the identification line relative to the moving contact.

4. The method according to claim 3, wherein The obtaining of the moving contact characteristic diagram in the closed state includes: Acquire a contact image and a standard sub-image; the contact image includes a static contact area, and the standard sub-image is a contact image in a closed state, including a static contact area and a moving contact area; Performing image enhancement on the standard sub-image and the contact image to obtain an image-enhanced standard sub-image and an image-enhanced contact image; Processing the image-enhanced standard segment image and the image-enhanced contact image using a superpixel fuzzy clustering algorithm to obtain a clustered standard segment image and a clustered contact image; Performing difference processing on the clustered standard score image and the clustered contact image to obtain an image of the moving contact area; The image of the moving contact area is binarized and morphologically processed to extract a moving contact feature map.

5. The method according to claim 1, wherein Determining whether the GIS switch is fully closed according to the area ratio and the position travel includes: Calculate the error between the area ratio and the area ratio when the switch is fully closed; If the error is greater than the set error threshold and the position travel is less than the set distance threshold, it is determined that the GIS switch is not fully closed; otherwise, it is determined that the GIS switch is fully closed.

6. The method according to claim 1, wherein The light sources of different wavelength bands include white light and yellow light.

7. A GIS switch contact identification device based on mixed-band light source lighting technology, characterized in that: include: An image acquisition module is used to acquire monitoring images of GIS switch contacts under illumination of light sources of different wavelengths; The light sources of different wavelengths include white light and yellow light; The image decomposition and fusion module is used to decompose the surveillance images under different wavelengths to obtain the base layer, detail layer and saliency layer; The base layer, detail layer and saliency layer corresponding to the surveillance images illuminated by light sources of different bands are fused to obtain a fused image; a module for acquiring a characteristic image of the marking line area, configured to acquire a standard closing image without marking lines when the switch is fully closed, obtain an image of the marking line area on the contact based on the standard closing image without marking lines when the switch is fully closed and the fused image, and perform binarization and morphological processing on the image of the marking line area on the contact to extract a characteristic image of the marking line area; the standard closing image without marking lines when the switch is fully closed includes a static contact area, a moving contact area, and a connecting rod area; The judgment module is used to judge whether there is an identification line based on the characteristic image of the identification line area. If not, it is determined that the GIS switch is in the open state. If it exists, the number of identification lines is further judged. If there is one identification line, it is determined that the GIS switch is in a fault state. If there are two identification lines, the area ratio of the identification line occupied by the characteristic image of the identification line area is determined, and the position stroke of the identification line relative to the moving contact is determined. According to the area ratio and the position stroke, it is determined whether the GIS switch is in the closed state.

8. A GIS switch contact identification system based on mixed-band light source lighting technology, characterized in that: include: Light source, industrial camera, FPGA; The light source is used to provide illumination of different wavelength bands to the GIS switch contacts; The industrial camera is used to obtain monitoring images of GIS switch contacts under illumination of light sources of different wavelengths; The FPGA is used to implement the GIS switch contact identification method based on mixed-band light source lighting technology as described in any one of claims 1 to 5.

9. The system according to claim 8, wherein The system further includes an LED driver and an MCU. The MCU controls the LED driver through dimming, and the LED driver controls the light source to provide illumination of different wavelength bands to the GIS switch contacts.

10. The system according to claim 8, further comprising a background monitoring system, wherein the FPGA sends the obtained GIS switch contact identification result to the background monitoring system, and the background monitoring system displays the GIS switch contact identification result.

Images