Methods, devices, equipment, and storage media for identifying the open / closed state of a disconnector.

Abstract

This invention relates to the field of power monitoring technology and discloses a method, device, equipment, and storage medium for identifying the open / closed state of a disconnector. The method determines the real-time coordinate spatial position of a target disconnector using a disconnector point cloud image, and adjusts the position of the target disconnector in the point cloud image using the real-time coordinate spatial position and a preset coordinate spatial position of the target disconnector to obtain an image of the target disconnector to be identified. The method then identifies the disconnector arm in the image to be identified and calculates the proportion of the disconnector arm in the image, determining the open / closed state of the disconnector based on the proportion. This method detects the open / closed state of a disconnector by acquiring a point cloud image of the disconnector. Point cloud images are not affected by factors such as ambient lighting or the shooting angle of the monitoring equipment, which can lead to unclear images. Furthermore, the algorithm's identification process is simple, solving the problem of low accuracy in identifying the open / closed state of disconnectors in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of power monitoring technology, and in particular to a method, device, equipment, and storage medium for identifying the open / closed state of a disconnector. Background Technology

[0002] After each power outage or restoration operation at a substation, the open / closed status of the disconnect switches needs to be confirmed. To improve the accuracy of disconnect switch status confirmation, this confirmation mainly includes auxiliary equipment confirmation and manual confirmation.

[0003] Manually confirming whether the disconnector is in place increases the workload of on-site personnel and introduces subjectivity into the operation, potentially leading to misjudgments.

[0004] The methods of confirming the status of disconnectors using auxiliary equipment mainly include installing magnetic induction devices on the disconnectors and using surveillance cameras. Installing magnetic induction devices generally requires a power outage, which can conflict with maintenance work due to limited downtime. For surveillance cameras, images of the disconnectors are first acquired, and then image processing technology is used to determine their status. The biggest problem with this method is that the final result is affected by the camera's installation angle, ambient lighting, and subsequent image processing algorithms. Considering the complexity and diversity of on-site environments, the accuracy of this method is difficult to guarantee. Therefore, considering the importance of confirming the disconnector's open / closed status and the on-site conditions, it is necessary to propose a highly applicable and accurate disconnector open / closed status identification technology. Summary of the Invention

[0005] The main objective of this invention is to solve the problem that existing disconnector opening / closing status detection schemes have low accuracy in identifying the opening / closing status of disconnectors.

[0006] The first aspect of the present invention provides a method for identifying the open / closed state of a disconnector, the method comprising:

[0007] Acquire point cloud data of disconnectors within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of disconnectors;

[0008] The real-time coordinate spatial position of the target disconnector is determined based on the disconnector point cloud image, and whether the target disconnector is parallel to the preset detection plane is determined based on the real-time coordinate spatial position.

[0009] If not, the point cloud image of the switch is adjusted to obtain the image to be identified of the target switch;

[0010] Identify the disconnector arm in the image to be identified, calculate the proportion of the disconnector arm in the image to be identified, and determine the open / closed state of the disconnector based on the proportion.

[0011] In a first implementation of the first aspect of the present invention, the step of acquiring point cloud data of disconnectors within a set location area and extracting a point cloud image of a target disconnector from the point cloud data includes:

[0012] Point cloud data of the knife switch is collected by a depth camera located at the position of the knife switch within a set location area, wherein the set location area includes the position of the knife switch;

[0013] The target switch in the switch point cloud data is determined, and the corresponding installation information is determined based on the target switch, wherein the installation information includes at least the initial coordinate space position of the target switch;

[0014] Based on the initial coordinate space position, the position of the target disconnector in the disconnector point cloud data is determined, and the point cloud data at the position is extracted to obtain the disconnector point cloud image.

[0015] In a second implementation of the first aspect of the present invention, the step of determining the position of the target disconnector in the disconnector point cloud data based on the initial coordinate space position, and extracting the point cloud data at the position to obtain a disconnector point cloud image, includes:

[0016] Based on the initial coordinate space position and the shooting parameters of the depth camera, the depth information of the target knife switch relative to the depth camera is calculated;

[0017] Calculate the depth value of each object in the knife switch point cloud data relative to the depth camera, and remove the data of objects whose depth values ​​do not conform to the depth information from the knife switch point cloud data to obtain the knife switch point cloud image.

[0018] In a third implementation of the first aspect of the present invention, before determining whether the target switch is parallel to a preset detection plane based on the real-time coordinate spatial position, the method further includes:

[0019] A visible light camera is used to capture images of disconnect switches within a designated area, and the image location information of the target disconnect switch is determined based on the disconnect switch images.

[0020] The real-time coordinate spatial position is corrected based on the image position information of the target disconnector.

[0021] In a fourth implementation of the first aspect of the present invention, adjusting the point cloud image of the disconnector to obtain the image to be identified of the target disconnector includes:

[0022] The position of the target disconnector in the point cloud image is adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image of the target disconnector to be identified.

[0023] In a fifth implementation of the first aspect of the present invention, adjusting the position of the target disconnector in the disconnector point cloud image using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image to be identified of the target disconnector includes:

[0024] Using the angle information between the target switch and the horizontal plane in the real-time coordinate space position and the preset coordinate space position of the target switch, and the position information of the intersection point of the perpendicular line from the midpoint of the target switch and the z-axis of the depth camera coordinate system, the position of the target switch in the switch point cloud image is transformed to obtain the image of the target switch to be identified.

[0025] In a sixth implementation of the first aspect of the present invention, the step of identifying the knife switch arm in the image to be identified and calculating the proportion of the knife switch arm in the image to be identified includes:

[0026] Extract the height information of the disconnector in the preset coordinate space;

[0027] Based on the height information, the region to be detected in the image to be identified is determined, wherein the region to be detected is the identification region of the switch arm;

[0028] The knife switch arm image in the area to be detected is extracted using a knife switch arm recognition algorithm;

[0029] Calculate the ratio of the image of the switch arm to the image corresponding to the region to be detected.

[0030] In a seventh implementation of the first aspect of the present invention, the step of extracting the knife switch arm image in the area to be detected using a knife switch arm recognition algorithm includes:

[0031] The detection area image of the target switch is extracted by using a Gaussian filtering algorithm to filter the image of the area to be detected.

[0032] The image and location of the switch arm are identified from the detection area image using a switch arm recognition algorithm, and the location is marked.

[0033] Based on the aforementioned markers, an image segmentation algorithm is used to segment the image of the switch arm.

[0034] In an eighth implementation of the first aspect of the present invention, calculating the ratio of the image of the switch arm to the image corresponding to the region to be detected includes:

[0035] Using the erosion operation in image processing technology, the image within the marked area is eroded to obtain the contour information of the knife switch arm;

[0036] The proportion of each of the knife switch arms relative to the marked area is calculated based on the contour information of each knife switch arm.

[0037] In a ninth implementation of the first aspect of the present invention, the step of using an erosion operation in image processing technology to erode the image within the marked area to obtain the contour information of the knife switch arm includes:

[0038] The outline of the knife gate arm is obtained by performing an erosion operation on the image within the marked area using OpenCV's built-in functions.

[0039] The boundary of the contour is determined, and the area of ​​the contour is calculated based on the boundary to obtain the contour information of the knife switch arm.

[0040] In a tenth implementation of the first aspect of the present invention, determining the state of the disconnector based on the proportion includes:

[0041] The ratio is matched with the ratio of the disconnector arm in each state in the preset disconnector state table, and the open / closed state of the disconnector is determined based on the matching result.

[0042] or,

[0043] Determine whether the percentage is greater than a preset threshold for the minimum outline area ratio of the disconnector in the closed position; if yes, determine that the disconnector is in the closed position; if no, determine that the disconnector is in the open position.

[0044] A second aspect of the present invention provides a device for identifying the open / closed state of a disconnector switch, the device comprising:

[0045] The image acquisition module is used to acquire point cloud data of the disconnector within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of the disconnector;

[0046] The image extraction module is used to determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and to determine whether the target disconnector is parallel to a preset detection plane based on the real-time coordinate spatial position; if not, the disconnector point cloud image is adjusted to obtain the image of the target disconnector to be identified.

[0047] The status recognition module is used to identify the disconnector arm in the image to be recognized, calculate the proportion of the disconnector arm in the image to be recognized, and determine the open / closed state of the disconnector based on the proportion.

[0048] In a first implementation of the second aspect of the present invention, the image acquisition module includes:

[0049] A point cloud acquisition unit is used to acquire point cloud data of the knife switch within a set location area by a depth camera located at the knife switch location, wherein the set location area includes the position of the knife switch;

[0050] The first determining unit is used to determine the target disconnector in the disconnector point cloud data and determine the corresponding installation information based on the target disconnector, wherein the installation information includes at least the initial coordinate space position of the target disconnector;

[0051] The first extraction unit is used to determine the position of the target switch in the switch point cloud data based on the initial coordinate space position, and extract the point cloud data at the position to obtain the switch point cloud image.

[0052] In a second implementation of the second aspect of the present invention, the image extraction module is specifically used for:

[0053] Based on the initial coordinate space position and the shooting parameters of the depth camera, the depth information of the target knife switch relative to the depth camera is calculated;

[0054] Calculate the depth value of each object in the knife switch point cloud data relative to the depth camera, and remove the data of objects whose depth values ​​do not conform to the depth information from the knife switch point cloud data to obtain the knife switch point cloud image.

[0055] In a third implementation of the second aspect of the present invention, the image extraction module is further configured to:

[0056] A visible light camera is used to capture images of disconnect switches within a designated area, and the image location information of the target disconnect switch is determined based on the disconnect switch images.

[0057] The real-time coordinate spatial position is corrected based on the image position information of the target disconnector.

[0058] In a fourth implementation of the second aspect of the present invention, the image extraction module is specifically used for:

[0059] The position of the target disconnector in the point cloud image is adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image of the target disconnector to be identified.

[0060] In a fifth implementation of the second aspect of the present invention, the image extraction module is specifically used for:

[0061] Using the corrected real-time coordinate space position and the angle information between the target switch and the horizontal plane in the preset coordinate space position of the target switch, and the position information of the intersection point of the perpendicular line from the midpoint of the target switch and the z-axis of the depth camera coordinate system, the position of the target switch in the switch point cloud image is transformed to obtain the image of the target switch to be identified.

[0062] In a sixth implementation of the second aspect of the present invention, the state recognition module includes:

[0063] The second extraction unit is used to extract the height information of the knife switch in the preset coordinate space position;

[0064] The second determining unit is used to determine the region to be detected in the image to be identified based on the height information, wherein the region to be detected is the identification region of the knife switch arm;

[0065] The third extraction unit is used to extract the knife switch arm image in the area to be detected using a knife switch arm recognition algorithm;

[0066] The calculation unit is used to calculate the ratio of the image of the switch arm to the image corresponding to the area to be detected.

[0067] In a seventh implementation of the second aspect of the present invention, the third extraction unit is specifically used for:

[0068] The detection area image of the target switch is extracted by using a Gaussian filtering algorithm to filter the image of the area to be detected.

[0069] The image and location of the switch arm are identified from the detection area image using a switch arm recognition algorithm, and the location is marked.

[0070] Based on the aforementioned markers, an image segmentation algorithm is used to segment the image of the switch arm.

[0071] In an eighth implementation of the second aspect of the present invention, the computing unit is specifically used for:

[0072] Using the erosion operation in image processing technology, the image within the marked area is eroded to obtain the contour information of the knife switch arm;

[0073] The proportion of each of the knife switch arms relative to the marked area is calculated based on the contour information of each knife switch arm.

[0074] In a ninth implementation of the second aspect of the present invention, the identification unit is specifically used for:

[0075] The outline of the knife gate arm is obtained by performing an erosion operation on the image within the marked area using OpenCV's built-in functions.

[0076] The boundary of the contour is determined, and the area of ​​the contour is calculated based on the boundary to obtain the contour information of the knife switch arm.

[0077] In a tenth implementation of the second aspect of the present invention, the state recognition module further includes a recognition unit, specifically used for:

[0078] The ratio is matched with the ratio of the disconnector arm in each state in the preset disconnector state table, and the open / closed state of the disconnector is determined based on the matching result.

[0079] or,

[0080] Determine whether the percentage is greater than a preset threshold for the minimum outline area ratio of the disconnector in the closed position; if yes, determine that the disconnector is in the closed position; if no, determine that the disconnector is in the open position.

[0081] A third aspect of the present invention provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a circuit; the at least one processor invokes the instructions in the memory to cause the electronic device to perform the various steps of the above-described method for identifying the open / closed state of a disconnector.

[0082] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the steps of the above-described method for identifying the open / closed state of a disconnector.

[0083] Beneficial effects:

[0084] In the technical solution of this invention, the real-time coordinate spatial position of the target disconnector is determined by the disconnector point cloud image, and the position of the target disconnector in the point cloud image is adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image to be identified of the target disconnector; the disconnector arm in the image to be identified is identified, and the proportion of the disconnector arm in the image to be identified is calculated, and the disconnector's open / closed state is determined based on the proportion. This method detects the disconnector's open / closed state by acquiring the disconnector's point cloud image, and the point cloud image is not affected by factors such as ambient lighting or the shooting angle of the monitoring equipment, which can lead to unclear images. Furthermore, the algorithm recognition process is simple, solving the problem of low accuracy in identifying the disconnector's open / closed state in existing technologies. Attached Figure Description

[0085] Figure 1 A hardware framework diagram for the identification of the open / closed state of a disconnector based on 3D image segmentation provided in an embodiment of the present invention;

[0086] Figure 2This is a schematic diagram of the first embodiment of the method for identifying the open / closed state of a disconnector in this invention.

[0087] Figure 3 This is a schematic diagram of a second embodiment of the method for identifying the open / closed state of a disconnector in this invention.

[0088] Figure 4 This is a schematic diagram of a third embodiment of the method for identifying the open / closed state of a disconnector in this invention.

[0089] Figure 5 This is a schematic diagram of the coordinate transformation of the disconnector in the point cloud image of the disconnector in an embodiment of the present invention;

[0090] Figure 6 This is a schematic diagram of a disconnector after coordinate changes in an embodiment of the present invention;

[0091] Figure 7 This is another schematic diagram of the disconnector after coordinate changes in an embodiment of the present invention;

[0092] Figure 8 This is a schematic diagram illustrating the determination of the open and closed states of the disconnector in an embodiment of the present invention;

[0093] Figure 9 This is a schematic diagram of one embodiment of the disconnection / opening status identification device of the present invention;

[0094] Figure 10 This is a schematic diagram of another embodiment of the disconnection / opening status identification device of the present invention;

[0095] Figure 11 This is a schematic diagram of one embodiment of the electronic device in this invention. Detailed Implementation

[0096] Existing technologies for determining the status of disconnectors rely heavily on traditional 2D vision solutions, as changes in the surrounding environment significantly impact the judgment results. While some technologies utilize deep learning to convert 2D to 3D space, this process is complex, requires sophisticated hardware, and is cumbersome to deploy. This invention addresses the specific problem of determining the open / closed position of disconnectors by providing a point cloud image-based detection method. Point cloud images significantly improve the environmental adaptability of disconnector status recognition, ensuring high accuracy while having low hardware requirements and facilitating simple and easy-to-implement field deployment.

[0097] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0098] For ease of understanding, the specific process of the embodiments of the present invention is described below. The method proposed in the embodiments of the present invention can be based on... Figure 1 The provided hardware framework Figure 1 The present invention provides a hardware framework for identifying the open / closed status of a disconnector based on 3D image segmentation. This hardware framework comprises a disconnector, a main control unit, an image acquisition module, a target image extraction module, a target image processing module, a disconnector segmentation and extraction module, and a status recognition module. The modules are connected via a communication bus, wireless communication, or other means. The image acquisition module can be a depth camera, such as a LiDAR. The image acquisition module is installed near the disconnector to monitor it and acquire point cloud images of the disconnector. These images are then sent to the target image extraction module, the target image processing module, the disconnector segmentation and extraction module, and the status recognition module for information extraction and recognition, thereby obtaining the detection result of the disconnector's open / closed status.

[0099] Please see Figure 1 and 2 The first embodiment of the method for identifying the open / closed state of a disconnector in this invention includes the following steps:

[0100] 101. Obtain the point cloud data of the disconnect switch within the set location area, and extract the point cloud image of the target disconnect switch from the point cloud data;

[0101] In this embodiment, the point cloud data of the disconnect switch can be collected by LiDAR. Specifically, the LiDAR is set up near the disconnect switch and communicates with the main control unit. The LiDAR monitors electrical equipment within a certain range, including the disconnect switch, and sets a monitoring area in the monitoring screen, which is the location area of ​​the disconnect switch. When the LiDAR collects point cloud data, the main control unit extracts the point cloud data of the disconnect switch from the set location area after receiving the point cloud data, and extracts the point cloud image of the disconnect switch.

[0102] Specifically, this step involves acquiring point cloud data of the knife switch within a designated location area using a depth camera positioned at the knife switch location. The designated location area includes the position of the knife switch.

[0103] The target switch in the switch point cloud data is determined, and the corresponding installation information is determined based on the target switch, wherein the installation information includes at least the initial coordinate space position of the target switch;

[0104] Based on the initial coordinate space position, the position of the target disconnector in the disconnector point cloud data is determined, and the point cloud data at the position is extracted to obtain the disconnector point cloud image.

[0105] In practical applications, the main control unit includes a storage area and a controller. The storage area stores the installation information of the disconnect switch within the monitoring range of the depth camera, which is the spatial coordinate information of the disconnect switch. The spatial coordinate position information is constructed with the depth camera as the origin. The storage area also stores the angle information between the disconnect switch and the horizontal plane within the monitoring range of the depth camera, and the position information of the intersection point of the perpendicular line from the midpoint of the disconnect switch and the z-axis of the depth camera coordinate system.

[0106] 102. Determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and determine whether the target disconnector is parallel to the preset detection plane based on the real-time coordinate spatial position;

[0107] 103. If not, adjust the point cloud image of the switch to obtain the image of the target switch to be identified;

[0108] In this embodiment, the adjustment of the knife switch point cloud image is specifically achieved by using the real-time coordinate spatial position and the preset coordinate spatial position of the target knife switch to adjust the position of the target knife switch in the knife switch point cloud image, thereby obtaining the image of the target knife switch to be identified.

[0109] In this embodiment, based on the relationship between the image coordinate system and the spatial coordinate system of the disconnector in the point cloud image, the real-time coordinate spatial position of the target disconnector in the point cloud image is calculated. This real-time coordinate spatial position will have a certain offset from the standard position when the disconnector is installed. After obtaining the disconnector point cloud image, the specific position of the target disconnector in the point cloud image is identified. Then, the point cloud image coordinates of the target disconnector are calculated in combination with the coordinate system of the image. Based on the relationship between the coordinate changes and the point cloud image coordinates, the corresponding spatial coordinates are calculated, thereby obtaining the real-time coordinate spatial position of the target disconnector.

[0110] In practical applications, in order to improve the accuracy of identifying the open and closed state of the disconnector, it is necessary to maintain the front view of the disconnector. Specifically, the position difference is calculated based on the real-time coordinate space position and the preset coordinate space position of the disconnector. Based on the position difference, the target disconnector is adjusted to the front view to obtain the image to be identified.

[0111] In this embodiment, the position of the target knife switch in the point cloud image is transformed by using the angle information between the target knife switch and the horizontal plane in the real-time coordinate space position and the preset coordinate space position of the target knife switch, and the position information of the intersection point of the perpendicular line from the midpoint of the target knife switch and the z-axis of the depth camera coordinate system, to obtain the image of the target knife switch to be identified.

[0112] Specifically, the target image extraction module extracts the knife gate by removing objects that do not belong to the coordinate space position of the knife gate from the perspective of the depth camera. For example, in order to detect the state of a knife gate, the point cloud image in the 3D space is extracted based on the coordinate information of each vertex or edge of the knife gate stored in the memory, and the point cloud image of the knife gate is obtained.

[0113] The target image processing module performs coordinate transformation on the target knife switch image based on the angle information between the knife switch and the horizontal plane stored in the memory and the position information of the intersection point of the perpendicular line from the midpoint of the knife switch and the z-axis of the depth camera coordinate system, to obtain the image of the target knife switch to be identified.

[0114] 104. Identify the disconnector arms in the image to be identified, calculate the proportion of the disconnector arms in the image to be identified, and determine the open / closed state of the disconnector based on the proportion.

[0115] In this embodiment, the identification of the knife switch arm can be achieved through a feature model, or by matching the knife switch model to determine the knife switch contour in the image to be identified, and then determining the knife switch arm based on the contour. Here, the knife switch model refers to the learning of the contour features of the knife switch arm and the contour features outside the knife switch arm through a neural network learning algorithm to construct an identification model that can identify the knife switch arm.

[0116] In this implementation, after identifying the coordinate information or pixel information of the switch arm in the image to be identified, the area of ​​the switch arm is calculated based on the coordinate information or pixel information, thereby calculating the proportion of the switch arm in the image to be identified, and thus determining the open / closed state of the switch.

[0117] In practical applications, the open / closed state of the switch is determined based on the proportion of the switch arms in the image to be identified. Specifically, the total length of each switch arm is calculated, and the distance between the switch arms is calculated based on the total length of each switch arm and the total length of the image to be identified. The switch is then used to determine whether it is in the open or closed state.

[0118] In summary, the point cloud image of the disconnector is used to identify its open / closed state. The point cloud image can be accurately acquired in any environment and is not affected by factors such as ambient lighting or the shooting angle of the monitoring equipment, which would result in unclear images. Furthermore, the algorithm recognition process is simple, solving the problem of low accuracy in identifying the open / closed state of disconnectors in existing technologies.

[0119] Meanwhile, identifying the open / closed state of the switch through point cloud images has low requirements for the recognition algorithm and low requirements for hardware, making it easy to deploy. The algorithm only involves simple coordinate conversion, image slicing, and comparison, and has a fast processing speed and low computing power requirements.

[0120] Please see Figure 3 This is another embodiment of the method for identifying the open / closed state of a disconnector in this invention. The identification method includes:

[0121] 201. Obtain the point cloud data of the disconnect switch within the set location area, and extract the point cloud image of the target disconnect switch from the point cloud data;

[0122] In this step, a depth camera positioned at the location of the disconnector acquires point cloud data of the disconnector within a designated location area, wherein the designated location area includes the location of the disconnector; a target disconnector is identified in the disconnector point cloud data, and corresponding installation information is determined based on the target disconnector, wherein the installation information includes at least the initial coordinate space position of the target disconnector; the position of the target disconnector in the disconnector point cloud data is determined based on the initial coordinate space position, and point cloud data at the position is extracted to obtain a disconnector point cloud image.

[0123] The step of determining the position of the target knife switch in the knife switch point cloud data based on the initial coordinate space position, and extracting the point cloud data at the position to obtain a knife switch point cloud image, includes: calculating the depth information of the target knife switch relative to the depth camera based on the initial coordinate space position and the shooting parameters of the depth camera; calculating the depth value of each object in the knife switch point cloud data relative to the depth camera, and removing the data of objects whose depth values ​​do not conform to the depth information from the knife switch point cloud data to obtain a knife switch point cloud image.

[0124] In practical applications, LiDAR is used to collect point cloud data of the knife switch and extract the knife switch point cloud image. Specifically, under the control command of the main controller, the depth camera collects the knife switch point cloud image information within a set location area. Based on the coordinate space position of the knife switch in the depth camera's view, objects that do not belong to that coordinate space position are removed, and the knife switch is extracted. For example, in order to detect the state of a certain knife switch, the point cloud image in the 3D space is extracted based on the coordinate information of each vertex or edge of the knife switch stored in the memory, and the knife switch point cloud image is obtained.

[0125] 202. Determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and determine whether the target disconnector is parallel to the preset detection plane based on the real-time coordinate spatial position;

[0126] 203. If so, adjust the position of the target switch in the point cloud image of the switch using the real-time coordinate space position and the preset coordinate space position of the target switch to obtain the image of the target switch to be identified;

[0127] In this step, the position of the target switch in the point cloud image is transformed by using the angle information between the target switch and the horizontal plane in the real-time coordinate space position and the preset coordinate space position of the target switch, as well as the position information of the intersection point of the perpendicular line from the midpoint of the target switch and the z-axis of the depth camera coordinate system, to obtain the image of the target switch to be identified.

[0128] In this embodiment, to ensure the accuracy of the obtained real-time coordinate spatial position, the method further includes: using a visible light camera to acquire images of the disconnector within a set location area, and determining the image position information of the target disconnector based on the disconnector images; and correcting the real-time coordinate spatial position based on the image position information of the target disconnector.

[0129] In practical applications, the standard parameters of the disconnect switch are recorded and stored in the device. Specifically, the spatial coordinate information of the disconnect switch within the monitoring range of the depth camera is stored. The spatial coordinate position information is constructed with the depth camera as the origin. The memory also stores the angle information between the disconnect switch and the horizontal plane within the monitoring range of the depth camera, and the position information of the intersection point of the perpendicular line from the midpoint of the disconnect switch and the z-axis of the depth camera coordinate system. Here, the angle information between the disconnect switch and the horizontal plane and the position information of the intersection point of the perpendicular line from the midpoint of the disconnect switch and the z-axis of the depth camera coordinate system are recorded as preset information.

[0130] Based on the preset information of the stored disconnect switch, and according to the angle information between the disconnect switch and the horizontal plane and the position information of the intersection point of the perpendicular line from the midpoint of the disconnect switch and the z-axis of the depth camera coordinate system stored in the memory, the target disconnect switch image is transformed to obtain the image of the target disconnect switch to be identified.

[0131] 204. Extract the height information of the disconnector in the preset coordinate space;

[0132] In this step, the height information is preset. After the knife switch is installed, its Z-axis coordinate is fixed. What will change is that the knife switch has a certain rotational offset. Therefore, when extracting the specific image of the knife switch, the height position in the image can be calculated based on the height information, and then step 205 can be executed based on the height position.

[0133] 205. Determine the detection area in the image to be identified based on the height information. The detection area is the identification area of ​​the switch arm.

[0134] In this step, the region where the disconnector is located in the image to be identified is determined based on the aforementioned height position, and then this region is marked as the identification region. Subsequent identification of the disconnector's open / closed status can directly extract and identify the identification region.

[0135] 206. Extract the image of the switch arm in the area to be detected using a switch arm recognition algorithm;

[0136] In this embodiment, a Gaussian filtering algorithm is used to filter the image of the area to be detected to extract the detection area image of the target disconnector; a disconnector arm recognition algorithm is used to identify the image and position of the disconnector arm from the detection area image, and the position is marked; based on the marking, an image segmentation algorithm is used to segment the disconnector arm image.

[0137] In practical applications, the switch arm recognition algorithm can be a recognition module trained on a switch image. This recognition module extracts switch features in the area to be detected, identifies the switch arm based on the switch features, and then combines the switch features to form a switch arm image.

[0138] 207. Calculate the ratio of the disconnector arm image to the image corresponding to the area to be detected, and determine the disconnector's open / closed state based on the ratio.

[0139] In this step, the image within the marked area is eroded using an image processing technique to obtain the contour information of the switch arm; the proportion of each switch arm relative to the marked area is calculated based on the contour information of each switch arm; and the open / closed state of the switch is determined based on the proportion.

[0140] The step of determining the open / closed state of the disconnector based on the ratio includes: matching the ratio with the disconnector arm ratio of each state in a preset disconnector state table, and determining the state of the disconnector based on the matching result; or, determining whether the ratio is greater than a preset minimum contour area ratio threshold for the disconnector in the closed state; if yes, then determining that the disconnector is in the closed state; if no, then determining that the disconnector is in the open state.

[0141] Specifically, the erosion operation is implemented using erosion functions in image processing technology. For example, OpenCV's built-in functions are used to perform erosion operations on the image within the marked area to obtain the outline of the knife switch arm; the boundary of the outline is determined, and the area of ​​the outline is calculated based on the boundary to obtain the outline information of the knife switch arm.

[0142] In summary, the method involves acquiring point cloud data of disconnectors within a designated area and extracting a point cloud image of the target disconnector from this data. Based on the point cloud image, the real-time coordinate spatial position of the target disconnector is determined. The position of the target disconnector in the point cloud image is then adjusted using the real-time coordinate spatial position and a preset coordinate spatial position, resulting in an image to be identified. The disconnector arm in the image to be identified is then identified, and its proportion within the image is calculated. Based on this proportion, the open / closed state of the disconnector is determined. Through this method, the point cloud image can adapt to various environments, ensuring high accuracy while having low hardware requirements and simple, easy-to-implement field deployment. This solves the problem of low accuracy in identifying the open / closed state of disconnectors in existing disconnector open / closed state detection schemes.

[0143] Please see Figure 1 and 4 This is another embodiment of the method for identifying the open / closed state of a disconnector in this invention. The solution provided in this embodiment is based on the following method combined with Figure 1The hardware implementation in this system, specifically the status recognition of disconnectors in substations, is described in detail below. This hardware is based on a LiDAR-based hardware and software system, comprising: a main control unit (storage area and controller), an image acquisition module, a target image extraction module, a target image processing module, a disconnector segmentation and extraction module, and a status recognition module. The main control unit stores the fixed status information of the disconnectors and the point cloud images acquired by the LiDAR. The image acquisition module receives controller commands and acquires the current point cloud image of the disconnector. The target image extraction module filters out distant objects from the image based on the known location of the disconnector to improve the saliency of the target disconnector. The target image processing module considers the positional relationship between each disconnector and the camera, using spatial coordinate transformation to convert each disconnector plane into a plane facing the camera. The disconnector segmentation and extraction module extracts the disconnectors from the point cloud image based on their known positional relationships, filters out some noise, and then segments the image. The status recognition module determines the status of the disconnector based on the segmentation of the disconnector plane. The target image extraction module filters out excessively distant objects from the image based on the known position of the knife gate, thereby improving the saliency of the target knife gate. This is simply a simple criterion for filtering objects in the point cloud space using known distance information, without involving complex calculations. The spatial coordinate transformation in the target image processing module is a simple coordinate matrix transformation, which accurately obtains the rotated spatial position of the knife gate. The knife gate segmentation and extraction module extracts the knife gate from the captured point cloud image by transforming its known spatial position to the rotated coordinate system. Considering the potential presence of noise, Gaussian filtering is used to filter this portion of the image. The state recognition module first extracts 1 / 4 of the area at the middle of both sides of the knife gate based on its position information. Then, it performs contour detection on the knife gate within this area and calculates the contour area. If the contour area is greater than half of the area of ​​the region, it is determined to be in a closed position; otherwise, it is determined to be in a separated position. This recognition method includes:

[0144] 301. Obtain the point cloud image of the switch;

[0145] By installing a depth camera (LiDAR) near the switch in the substation to acquire point cloud images of the switch, and extracting the switch from the point cloud images based on information such as the distance, height, and angle of the switch, the point cloud image of the target switch is obtained.

[0146] 302. Filter out irrelevant information in the point cloud image of the disconnect switch;

[0147] The irrelevant information in this step can be understood as images of all electrical devices or wiring information other than the disconnect switch. Specifically, the disconnect switch region in the disconnect switch point cloud image is marked using a disconnect switch recognition model, and then the point cloud data outside the disconnect switch region is deleted to obtain the image of the target disconnect switch.

[0148] 303. Identify the image of the disconnector from the point cloud image of the disconnector after filtering out irrelevant information;

[0149] 304. Using the stored angle information between the switch and the horizontal plane and the position information of the intersection point of the perpendicular line from the midpoint of the switch and the z-axis of the depth camera coordinate system, the image of the identified switch is transformed to obtain the front view of the switch.

[0150] 305. Segment the front view of the disconnect switch from the point cloud image of the disconnect switch to obtain the disconnect switch image;

[0151] The target image extraction module extracts the knife gate by removing objects that do not belong to the coordinate space position of the knife gate from the perspective of the depth camera. For example, in order to detect the state of a knife gate, the module extracts the point cloud image in the 3D space based on the coordinate information of each vertex or edge of the knife gate stored in the memory, and obtains the point cloud image of the knife gate.

[0152] The target image processing module performs coordinate transformation on the target knife switch image based on the angle information between the knife switch and the horizontal plane stored in the memory and the position information of the intersection point of the perpendicular line from the midpoint of the knife switch and the z-axis of the depth camera coordinate system, to obtain the image of the target knife switch to be identified.

[0153] The specific transformation method is to first rotate the device by θ around the y-axis, and then translate the origin to position o' in the xz-axis plane. Here, θ is the angle between the plane containing the switch and the horizontal line, and position o' is the intersection of the perpendicular line from the midpoint of the switch and the z-axis of the depth camera coordinate system, where y = 0. For example... Figure 5 As shown in θ, the specific changes are described in the following equation:

[0154]

[0155] Based on the height of the disconnect switch, the location of the disconnect switch is extracted, and then the area is filtered by a Gaussian filtering algorithm to obtain the image of the area to be detected by the disconnect switch.

[0156] A schematic diagram of a single disconnect switch after coordinate system transformation is shown below. Figure 6 As shown, based on the height of the disconnect switch, the cut-off point can be... Figure 6 The dashed box region shown is analyzed. The imaging of the plane containing the dashed box region at different distances is detected, and then a Gaussian filtering algorithm is applied to filter the dashed box region. A planar image of the disconnector in the closed position is obtained at the origin o' of the new coordinate system (this plane is determined based on the distance from the disconnector plane to the origin o'), and the disconnector position is extracted based on the height. Figure 6 The area marked by the dashed box is the image on this plane when the disconnect switch is in the closed position; the extraction effect when the disconnect switch is in the open position is as follows. Figure 7 As shown.

[0157] The status recognition module extracts a certain proportion of each side of the disconnector position image. It identifies the disconnector's open / closed state by calculating the contour area within the set area. If the contour area of ​​the set area is greater than half the total area of ​​the selected area, the disconnector is considered to be in the closed position; otherwise, it is considered to be in the open position. Specifically, the status recognition module first extracts 1 / 4 of each side of the disconnector's center position based on the known disconnector position information. Figure 8 As shown in the solid line frame, the region within the solid line frame is then subjected to contour detection, and the contour area is calculated. If the contour area is greater than 1 / 2 of the region area, it is determined to be in a merging position; otherwise, it is determined to be in a split position.

[0158] To eliminate the influence of noise and impurities within the bounding box, OpenCV's built-in functions can be used to perform erosion operations on the region within the bounding box before contour detection.

[0159] 306. Identify the open / closed state of the disconnector based on the disconnector image.

[0160] By implementing the above method, the real-time coordinate spatial position of the target disconnector is determined through the disconnector point cloud image. The position of the target disconnector in the point cloud image is then adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector, resulting in an image of the target disconnector to be identified. The disconnector arms in the image to be identified are then identified, and the proportion of each arm in the image is calculated. Based on this proportion, the open / closed state of the disconnector is determined. This method can accurately determine the open / closed state of a disconnector with simple hardware configuration.

[0161] Furthermore, the location of the knife switch is determined by using LiDAR to acquire point cloud images of the knife switch and visible light cameras to acquire images. The location of the knife switch determined by the visible light camera is then verified against the location obtained by the LiDAR to ensure the accuracy of the identified knife switch location. This improves the adaptability to changes in external light and enables accurate judgment of the knife switch status under different locations, scenes, and lighting conditions.

[0162] The above describes the method for identifying the open / closed state of the disconnector in an embodiment of the present invention. The following describes the device for identifying the open / closed state of the disconnector in an embodiment of the present invention. Please refer to [link to relevant documentation]. Figure 9 One embodiment of the device for identifying the open / closed state of a disconnector in this invention includes:

[0163] Image acquisition module 910 is used to acquire point cloud data of disconnectors within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of disconnectors;

[0164] The image extraction module 920 is used to determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and to determine whether the target disconnector is parallel to a preset detection plane based on the real-time coordinate spatial position; if not, the disconnector point cloud image is adjusted to obtain the image of the target disconnector to be identified.

[0165] The status recognition module 930 is used to identify the disconnector arm in the image to be recognized, calculate the proportion of the disconnector arm in the image to be recognized, and determine the open / closed state of the disconnector based on the proportion.

[0166] In this embodiment of the invention, the open / closed state is identified based on the point cloud image of the disconnector. The point cloud image can be accurately acquired in any environment and is not affected by factors such as ambient lighting or the shooting angle of the monitoring equipment, which would result in unclear images. Furthermore, the algorithm recognition process is simple, solving the problem of low accuracy in identifying the open / closed state of disconnectors in the prior art.

[0167] Please see Figure 10 A second embodiment of the device for identifying the open / closed state of a disconnector in this invention includes:

[0168] Image acquisition module 910 is used to acquire point cloud data of disconnectors within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of disconnectors;

[0169] The image extraction module 920 is used to determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and to determine whether the target disconnector is parallel to a preset detection plane based on the real-time coordinate spatial position; if not, the disconnector point cloud image is adjusted to obtain the image of the target disconnector to be identified.

[0170] The status recognition module 930 is used to identify the disconnector arm in the image to be recognized, calculate the proportion of the disconnector arm in the image to be recognized, and determine the open / closed state of the disconnector based on the proportion.

[0171] In this embodiment, the image acquisition module 910 includes:

[0172] The point cloud acquisition unit 911 is used to acquire point cloud data of the knife switch within a set location area by a depth camera located at the knife switch location, wherein the set location area includes the position of the knife switch;

[0173] The first determining unit 912 is used to determine the target switch in the switch point cloud data and determine the corresponding installation information based on the target switch, wherein the installation information includes at least the initial coordinate space position of the target switch;

[0174] The first extraction unit 913 is used to determine the position of the target switch in the switch point cloud data based on the initial coordinate space position, and extract the point cloud data at the position to obtain the switch point cloud image.

[0175] In this embodiment, the image extraction module 920 is specifically used for:

[0176] Based on the initial coordinate space position and the shooting parameters of the depth camera, the depth information of the target knife switch relative to the depth camera is calculated;

[0177] Calculate the depth value of each object in the knife switch point cloud data relative to the depth camera, and remove the data of objects whose depth values ​​do not conform to the depth information from the knife switch point cloud data to obtain the knife switch point cloud image.

[0178] In this embodiment, the image extraction module 920 is further used for:

[0179] A visible light camera is used to capture images of disconnect switches within a designated area, and the image location information of the target disconnect switch is determined based on the disconnect switch images.

[0180] The real-time coordinate spatial position is corrected based on the image position information of the target disconnector.

[0181] In this embodiment, the image extraction module 920 is specifically used for:

[0182] The position of the target disconnector in the point cloud image is adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image of the target disconnector to be identified.

[0183] In this embodiment, the image extraction module 920 is specifically used for:

[0184] Using the angle information between the target switch and the horizontal plane in the real-time coordinate space position and the preset coordinate space position of the target switch, and the position information of the intersection point of the perpendicular line from the midpoint of the target switch and the z-axis of the depth camera coordinate system, the position of the target switch in the switch point cloud image is transformed to obtain the image of the target switch to be identified.

[0185] In this embodiment, the state recognition module 930 includes:

[0186] The second extraction unit 931 is used to extract the height information of the knife switch in the preset coordinate space position;

[0187] The second determining unit 932 is used to determine the region to be detected in the image to be identified based on the height information, wherein the region to be detected is the identification region of the knife switch arm;

[0188] The third extraction unit 933 is used to extract the image of the knife switch arm in the area to be detected using a knife switch arm recognition algorithm;

[0189] The calculation unit 934 is used to calculate the ratio of the image of the switch arm to the image corresponding to the area to be detected.

[0190] In this embodiment, the third extraction unit 933 is specifically used for:

[0191] The detection area image of the target switch is extracted by using a Gaussian filtering algorithm to filter the image of the area to be detected.

[0192] The image and location of the switch arm are identified from the detection area image using a switch arm recognition algorithm, and the location is marked.

[0193] Based on the aforementioned markers, an image segmentation algorithm is used to segment the image of the switch arm.

[0194] In this embodiment, the computing unit 934 is specifically used for:

[0195] Using the erosion operation in image processing technology, the image within the marked area is eroded to obtain the contour information of the knife switch arm;

[0196] The proportion of each of the knife switch arms relative to the marked area is calculated based on the contour information of each knife switch arm.

[0197] In this embodiment, the identification unit 934 is specifically used for:

[0198] The outline of the knife gate arm is obtained by performing an erosion operation on the image within the marked area using OpenCV's built-in functions.

[0199] The boundary of the contour is determined, and the area of ​​the contour is calculated based on the boundary to obtain the contour information of the knife switch arm.

[0200] In this embodiment, the state recognition module 930 further includes a recognition unit 935, specifically used for:

[0201] The ratio is matched with the ratio of the disconnector arm in each state in the preset disconnector state table, and the open / closed state of the disconnector is determined based on the matching result.

[0202] or,

[0203] Determine whether the percentage is greater than a preset threshold for the minimum outline area ratio of the disconnector in the closed position; if yes, determine that the disconnector is in the closed position; if no, determine that the disconnector is in the open position.

[0204] This embodiment determines the real-time coordinate spatial position of the target disconnector using a disconnector point cloud image. Then, using this real-time coordinate spatial position and a preset coordinate spatial position of the target disconnector, the position of the target disconnector in the point cloud image is adjusted to obtain an image of the target disconnector to be identified. The disconnector arms in the image to be identified are then identified, and the proportion of each arm in the image is calculated. Based on this proportion, the open / closed state of the disconnector is determined. This method can achieve accurate determination of the open / closed state of the disconnector through simple hardware configuration.

[0205] Furthermore, the location of the knife switch is determined by using LiDAR to acquire point cloud images of the knife switch and visible light cameras to acquire images. The location of the knife switch determined by the visible light camera is then verified against the location obtained by the LiDAR to ensure the accuracy of the identified knife switch location. This improves the adaptability to changes in external light and enables accurate judgment of the knife switch status under different locations, scenes, and lighting conditions.

[0206] above Figure 9-10 The identification device for the open / closed state of the switch in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The electronic equipment in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0207] Figure 11 This is a schematic diagram of the structure of an electronic device 1000 provided in an embodiment of the present invention. The electronic device 1000 can vary significantly due to differences in configuration or performance. It may include one or more central processing units (CPUs) 1010 (e.g., one or more processors) and a memory 1020, and one or more storage media 1030 (e.g., one or more mass storage devices) for storing application programs 1033 or data 1032. The memory 1020 and storage media 1030 can be temporary or persistent storage. The program stored in the storage media 1030 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the electronic device 1000. Furthermore, the processor 1010 may be configured to communicate with the storage media 1030 and execute the series of instruction operations in the storage media 1030 on the electronic device 1000 to implement the various steps of the method provided in the above embodiment.

[0208] Electronic device 1000 may also include one or more power supplies 1040, one or more wired or wireless network interfaces 1050, one or more input / output interfaces 1060, and / or one or more operating systems 1031, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 11The illustrated electronic device structure does not constitute a limitation on the computer device provided by the present invention, and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0209] The electronic device provided in this embodiment of the invention includes: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; the at least one processor invokes the instructions in the memory to cause the electronic device to execute the various steps of the above-described method for identifying the open / closed state of a disconnector.

[0210] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform each step of the method for identifying the open / closed state of the switch.

[0211] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0212] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0213] 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.

Claims

1. A method for identifying the open / closed state of a disconnector, characterized in that, The method for identifying the open / closed state of the disconnector includes: Acquire point cloud data of disconnectors within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of disconnectors; The real-time coordinate spatial position of the target disconnector is determined based on the disconnector point cloud image, and whether the target disconnector is parallel to the preset detection plane is determined based on the real-time coordinate spatial position. If not, the point cloud image of the switch is adjusted to obtain the image to be identified of the target switch; The detection area in the image to be identified is determined, the contour information of the switch arm is extracted from the detection area using a switch arm identification algorithm, the proportion of the switch arm in the image to be identified is calculated based on the contour information, and the open / closed state of the switch is determined based on the proportion.

2. The method for identifying the open / closed state of a disconnector according to claim 1, characterized in that, The step of acquiring point cloud data of disconnectors within a set location area and extracting the point cloud image of the target disconnector from the point cloud data includes: Point cloud data of the knife switch is collected by a depth camera located at the position of the knife switch within a set location area, wherein the set location area includes the position of the knife switch; The target switch in the switch point cloud data is determined, and the corresponding installation information is determined based on the target switch, wherein the installation information includes at least the initial coordinate space position of the target switch; Based on the initial coordinate space position, the position of the target disconnector in the disconnector point cloud data is determined, and the point cloud data at the position is extracted to obtain the disconnector point cloud image.

3. The method for identifying the open / closed state of a disconnector according to claim 2, characterized in that, The step of determining the position of the target disconnector in the disconnector point cloud data based on the initial coordinate space position, and extracting the point cloud data at the position to obtain the disconnector point cloud image, includes: Based on the initial coordinate space position and the shooting parameters of the depth camera, the depth information of the target knife switch relative to the depth camera is calculated; Calculate the depth value of each object in the knife switch point cloud data relative to the depth camera, and remove the data of objects whose depth values ​​do not conform to the depth information from the knife switch point cloud data to obtain the knife switch point cloud image.

4. The method for identifying the open / closed state of a disconnector according to claim 1, characterized in that, Before determining whether the target switch is parallel to a preset detection plane based on the real-time coordinate spatial position, the method further includes: A visible light camera is used to capture images of disconnect switches within a designated area, and the image location information of the target disconnect switch is determined based on the disconnect switch images. The real-time coordinate spatial position is corrected based on the image position information of the target disconnector.

5. The method for identifying the open / closed state of a disconnector according to claim 1, characterized in that, The step of adjusting the point cloud image of the disconnect switch to obtain the image to be identified of the target disconnect switch includes: The position of the target disconnector in the point cloud image is adjusted using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image of the target disconnector to be identified.

6. The method for identifying the open / closed state of a disconnector according to claim 5, characterized in that, The step of adjusting the position of the target disconnector in the point cloud image of the disconnector using the real-time coordinate spatial position and the preset coordinate spatial position of the target disconnector to obtain the image of the target disconnector to be identified includes: Using the angle information between the target switch and the horizontal plane in the real-time coordinate space position and the preset coordinate space position of the target switch, and the position information of the intersection point of the perpendicular line from the midpoint of the target switch and the z-axis of the depth camera coordinate system, the position of the target switch in the switch point cloud image is transformed to obtain the image of the target switch to be identified.

7. The method for identifying the open / closed state of a disconnector according to any one of claims 1-6, characterized in that, The process of determining the region to be detected in the image to be identified, extracting the contour information of the switch arm from the region to be detected using a switch arm recognition algorithm, and calculating the proportion of the switch arm in the image to be identified based on the contour information includes: Extract the height information of the disconnector from the preset coordinate space location; Based on the height information, the region to be detected in the image to be identified is determined, wherein the region to be detected is the identification region of the switch arm; The contour information of the knife switch arm in the area to be detected is extracted using a knife switch arm recognition algorithm; Calculate the proportion of the contour information to the image corresponding to the region to be detected.

8. The method for identifying the open / closed state of a disconnector according to claim 7, characterized in that, The step of extracting the contour information of the knife switch arm in the area to be detected using the knife switch arm recognition algorithm includes: The detection area image of the target switch is extracted by using a Gaussian filtering algorithm to filter the image of the area to be detected. The image and location of the switch arm are identified from the detection area image using a switch arm recognition algorithm, and the location is marked. Based on the markers, an image segmentation algorithm is used to segment the switch arm image, and the contour information of the switch arm is determined based on the switch arm image.

9. The method for identifying the open / closed state of a disconnector according to claim 8, characterized in that, The calculation of the proportion of the contour information to the image corresponding to the region to be detected includes: Using the erosion operation in image processing technology, the image within the marked area is eroded to obtain the contour information of the knife switch arm; The proportion of each of the knife switch arms relative to the marked area is calculated based on the contour information of each knife switch arm.

10. The method for identifying the open / closed state of a disconnector according to claim 9, characterized in that, The step of using the erosion operation in image processing technology to erode the image within the marked area to obtain the contour information of the knife switch arm includes: The outline of the knife gate arm is obtained by performing an erosion operation on the image within the marked area using OpenCV's built-in functions. The boundary of the contour is determined, and the area of ​​the contour is calculated based on the boundary to obtain the contour information of the knife switch arm.

11. The method for identifying the open / closed state of a disconnector according to any one of claims 1-6, characterized in that, Determining the open / closed state of the disconnector based on the ratio includes: The ratio is matched with the ratio of the disconnector arm in each state in the preset disconnector state table, and the open / closed state of the disconnector is determined based on the matching result. or, Determine whether the percentage is greater than a preset threshold for the minimum outline area ratio of the disconnector in the closed position; if yes, determine that the disconnector is in the closed position; if no, determine that the disconnector is in the open position.

12. A device for identifying the open / closed state of a disconnector switch, characterized in that, The device for identifying the open / closed state of the disconnector includes: The image acquisition module is used to acquire point cloud data of the disconnector within a set location area, and extract the point cloud image of the target disconnector from the point cloud data of the disconnector; The image extraction module is used to determine the real-time coordinate spatial position of the target disconnector based on the disconnector point cloud image, and to determine whether the target disconnector is parallel to a preset detection plane based on the real-time coordinate spatial position; if not, the disconnector point cloud image is adjusted to obtain the image of the target disconnector to be identified. The state recognition module is used to determine the detection area in the image to be recognized, extract the contour information of the switch arm from the detection area using the switch arm recognition algorithm, calculate the proportion of the switch arm in the image to be recognized based on the contour information, and determine the open / closed state of the switch based on the proportion.

13. The device for identifying the open / closed state of a disconnector according to claim 12, characterized in that, The status recognition module includes: The second extraction unit is used to extract the height information of the disconnector in a preset coordinate space position; The second determining unit is used to determine the region to be detected in the image to be identified based on the height information, wherein the region to be detected is the identification region of the knife switch arm; The third extraction unit is used to extract the contour information of the knife switch arm in the area to be detected using a knife switch arm recognition algorithm; The calculation unit is used to calculate the proportion of the contour information to the image corresponding to the region to be detected.

14. An electronic device, characterized in that, The electronic device includes: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; The at least one processor invokes the instructions in the memory to cause the electronic device to perform the steps of the method for identifying the open / closed state of a disconnector as described in any one of claims 1-11.

15. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the method for identifying the open / closed state of the disconnector as described in any one of claims 1-11.

Images