CN114463270A

CN114463270A - Disconnecting link state identification method and device, electronic device and storage medium

Info

Publication number: CN114463270A
Application number: CN202111666725.0A
Authority: CN
Inventors: 杨剑波; 孙海涛; 李宁钏; 赵蕾; 熊剑平
Original assignee: Zhejiang Dahua Technology Co Ltd
Current assignee: Zhejiang Dahua Technology Co Ltd
Priority date: 2021-12-30
Filing date: 2021-12-30
Publication date: 2022-05-10

Abstract

The application relates to a method, a device, an electronic device and a storage medium for identifying the state of a disconnecting link, wherein the method comprises the following steps: detecting the image to be recognized according to a preset first target detection model to obtain a disconnecting link target boundary, a disconnecting link type and a disconnecting link basic opening and closing state of a disconnecting link in the image to be recognized; based on the knife switch target boundary, obtaining a knife switch target image only containing a knife switch from the image to be identified; detecting the position relation between a knife arm in a knife-switch target image and a preset knife-switch structural member corresponding to the knife arm; and determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relation. Through the method and the device, the final state of the disconnecting link is determined by utilizing the type of the disconnecting link, the preliminary closing state of the disconnecting link and the position relation, and the accuracy of disconnecting link state identification is improved.

Description

Disconnecting link state identification method and device, electronic device and storage medium

Technical Field

The present disclosure relates to the field of image processing technologies, and in particular, to a method and an apparatus for identifying a status of a disconnecting link, an electronic apparatus, and a storage medium.

Background

In the routing inspection task of the intelligent substation, the state identification of the disconnecting link is an important link. Traditional manual inspection wastes time and energy, and is low in efficiency. The existing image processing method for identifying the state of the disconnecting link is mostly the traditional image processing technology or needs to manually preset a calibration line to obtain the parameters of a field deployment camera and the like. The traditional image processing technology is used for identifying that the swing arm of the disconnecting link has no robustness under the conditions of complex background and large light change. Links such as manually presetting a calibration line, acquiring on-site deployment camera parameters and the like reduce the intelligent degree, the process is complicated, and the professional experience of professionals is required. Therefore, the problem that the accuracy of the identification result is not high exists in the knife switch state identification in the prior art.

Aiming at the problem of low accuracy of the identification result in the related technology, no effective solution is provided at present.

Disclosure of Invention

The embodiment provides a method, a device, an electronic device and a storage medium for identifying a disconnecting link state, so as to solve the problem that the accuracy of an identification result is not high in the related art.

In a first aspect, in this embodiment, there is provided a method for identifying a status of a knife switch, the method comprising,

detecting an image to be identified containing a disconnecting link according to a preset first target detection model to obtain the disconnecting link type and the disconnecting link basic opening and closing state of the disconnecting link in the image to be identified;

determining the position relation between the knife arm of the knife switch and a knife switch structural member based on the image to be identified; the knife switch structural part is preset aiming at the knife arm;

and determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relation.

In some embodiments, the determining, based on the image to be identified, a positional relationship between a knife arm of the knife switch and a knife switch structural member includes:

obtaining a disconnecting link target image containing the disconnecting link from the image to be identified; the area proportion of the disconnecting link in the disconnecting link target image is larger than an area proportion threshold value;

the determining the position relationship between the knife arm of the knife switch and the knife switch structural member based on the image to be identified comprises the following steps:

and detecting the knife arm in the knife-switch target image and the knife-switch structural member, and determining the position relation between the knife arm of the knife-switch and the preset knife-switch structural member based on the detection result.

In some embodiments, when detecting an image to be identified including a disconnecting link and obtaining a disconnecting link type and a disconnecting link basic on-off state of the disconnecting link in the image to be identified, the method further includes:

detecting an image to be identified containing a disconnecting link according to a preset first target detection model to obtain a disconnecting link target boundary of the disconnecting link in the image to be identified;

the obtaining of the knife switch target image containing the knife switch from the image to be identified comprises:

and obtaining the disconnecting link target image only containing the disconnecting link from the image to be identified based on the disconnecting link target boundary.

In some of these embodiments, where the knife switch is of the non-three-post horizontal rotation type, the knife switch comprises two knife arms; the detecting of the position relationship between the knife arm in the knife-switch target image and the preset knife-switch structural member corresponding to the knife arm comprises detecting the position relationship between one knife arm and the other knife arm in the two knife arms in the knife-switch target image.

In some embodiments, the detecting the position relationship between one of the two knife arms and the other knife arm in the knife gate target image includes detecting an included angle between the two knife arms in the knife gate target image.

In some of these embodiments, when the knife switch is of the three-post horizontal rotation type, the knife switch comprises a knife arm and two connecting parts; the detecting of the position relationship between the knife arm in the knife-switch target image and the preset knife-switch structural member corresponding to the knife arm comprises detecting the position relationship between the knife arm in the knife-switch target image and the two connecting parts.

In some embodiments, the detecting the position relationship between the knife arm in the knife-switch target image and the two connecting components includes detecting an included angle between the knife arm in the knife-switch target image and a connecting line of the two connecting components.

In some embodiments, when the type of the disconnecting link is non-three-column horizontal rotation type, the determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relationship includes,

if the on-off state of the knife switch base is a closing state and meets the condition gamma belongs to [ 180-alpha, 180-alpha + alpha ], the state of the knife switch is in place for closing;

if the on-off state of the disconnecting link base is a closing state and does not meet the condition gamma which belongs to [ 180-alpha, 180-alpha + alpha ], the state of the disconnecting link is an abnormal state;

if the on-off state of the disconnecting link base is an opening state and the condition gamma is in the range of [0 DEG, beta ], the disconnecting link is in place;

if the disconnecting link basic opening and closing state is an opening state and the condition gamma is not met, belonging to [0 DEG, beta ], the state of the disconnecting link is an abnormal state;

wherein gamma is an included angle between the two knife arms in the knife switch target image, alpha is a closing in-place threshold value, alpha belongs to [0 DEG and 90 DEG ], beta is a breaking in-place threshold value, and beta belongs to [0 DEG and 90 DEG ].

In some embodiments, the determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relationship includes,

if the on-off state of the disconnecting link base is an opening state and meets the condition gamma epsilon [ beta, 360 degrees ], the disconnecting link is in place;

if the on-off state of the disconnecting link base is an off state and does not meet the condition gamma epsilon [ beta, 360 degrees ], the state of the disconnecting link is an abnormal state;

wherein gamma is an included angle between a knife arm in the knife switch target image and a connecting line of the two connecting parts, alpha is a closing in-place threshold value, alpha belongs to [0 DEG, 90 DEG ], beta is a breaking in-place threshold value, and beta belongs to [0 DEG, 90 DEG ].

In some of these embodiments, the method further comprises,

according to a preset semantic segmentation model, segmenting the disconnecting link target image to obtain a disconnecting link arm mask image of the disconnecting link; the tool arm mask graph comprises a tool arm target area and tool arm type information; the knife arm type information is used for distinguishing two knife arms in the knife switch target image;

and respectively calculating included angles between the two cutter arms and a preset axis according to the cutter arm mask image, and calculating the included angle between the two cutter arms based on the included angles between the two cutter arms and the preset axis.

In some embodiments, the calculating the included angles between the two tool arms and the preset axis according to the tool arm mask map includes,

respectively obtaining corresponding first cutter arm target areas of the two cutter arms by using a preset algorithm based on the cutter arm target area and the cutter arm type information, and calculating a circumscribed rectangle with the smallest area of the first cutter arm target areas;

and calculating the included angle between the straight line of the preset side of the external rectangle and the preset axis.

In a second aspect, in this embodiment, a device for identifying a disconnecting link state is provided, where the device includes a detection module, configured to detect, according to a preset first target detection model, an image to be identified that includes a disconnecting link, and obtain a disconnecting link type and a disconnecting link basic opening and closing state of the disconnecting link in the image to be identified;

the processing module is used for determining the position relation between the knife arm of the knife switch and the knife switch structural member based on the image to be identified; the knife switch structural part is preset aiming at the knife arm;

and the identification module is used for determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relation.

In a third aspect, in the present embodiment, there is provided an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the computer program to execute the method for identifying a state of a disconnecting link according to the first aspect.

In a fourth aspect, in the present embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method for identifying a state of a disconnecting link according to the first aspect.

Compared with the related art, the method, the device, the electronic device and the storage medium for identifying the state of the disconnecting link provided in the embodiment determine the final state of the disconnecting link by using the type of the disconnecting link, the preliminary closing state of the disconnecting link and the position relationship, so that the accuracy of identifying the state of the disconnecting link is improved.

The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the application.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is a block diagram of a hardware configuration of a terminal that executes the method of identifying a state of a switch of the present embodiment;

FIG. 2 is a flow chart of a method for identifying the status of a disconnecting link according to the present embodiment;

FIG. 3 is a flow chart of another method for identifying the status of the knife switch according to the present embodiment;

FIG. 4 is a flow chart of another method for identifying the status of the knife switch according to the present embodiment;

FIG. 5 is a flow chart of the method for identifying the status of the knife switch in the preferred embodiment;

FIG. 6 is a tool arm mask illustration of the preferred embodiment;

FIG. 7 is a tool arm mask illustration of the preferred embodiment detecting a tool arm circumscribing rectangle;

FIG. 8 is a block diagram of a three-post horizontal rotary knife switch according to the preferred embodiment;

fig. 9 is a block diagram of a disconnecting link state identifying device according to the present embodiment.

Detailed Description

For a clearer understanding of the objects, aspects and advantages of the present application, reference is made to the following description and accompanying drawings.

Unless defined otherwise, technical or scientific terms used herein shall have the same general meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of the terms "a" and "an" and "the" and similar referents in the context of this application do not denote a limitation of quantity, either in the singular or the plural. The terms "comprises," "comprising," "has," "having," and any variations thereof, as referred to in this application, are intended to cover non-exclusive inclusions; for example, a process, method, and system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or modules, but may include other steps or modules (elements) not listed or inherent to such process, method, article, or apparatus. Reference throughout this application to "connected," "coupled," and the like is not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Reference to "a plurality" in this application means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. In general, the character "/" indicates a relationship in which the objects associated before and after are an "or". The terms "first," "second," "third," and the like in this application are used for distinguishing between similar items and not necessarily for describing a particular sequential or chronological order.

The method embodiments provided in the present embodiment may be executed in a terminal, a computer, a server, or a similar computing device. Hereinafter, description will be given by taking an example of operation on a terminal, and fig. 1 is a hardware configuration block diagram of a terminal that executes the disconnecting link state identification method of the present embodiment. As shown in fig. 1, the terminal may include one or more processors 102 (only one shown in fig. 1) and a memory 104 for storing data, wherein the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA. The terminal may also include a transmission device 106 for communication functions and an input-output device 108. It will be understood by those of ordinary skill in the art that the structure shown in fig. 1 is merely an illustration and is not intended to limit the structure of the terminal described above. For example, the terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

The memory 104 may be used to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the knife switch state identification method in the present embodiment, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the above-mentioned method. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used to receive or transmit data via a network. The network described above includes a wireless network provided by a communication provider of the terminal. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.

Fig. 2 is a flowchart of a method for identifying a switch status according to this embodiment, and as shown in fig. 2, the flowchart includes the following steps:

step S210, detecting the image to be identified containing the disconnecting link according to a preset first target detection model to obtain the disconnecting link type and the disconnecting link basic opening and closing state of the disconnecting link in the image to be identified. Specifically, after the image to be recognized is obtained, the image to be recognized includes a main body of the disconnecting link, of which the current state of the disconnecting link needs to be recognized, and the image to be recognized is detected according to a preset first target detection model. For example, a YOLOV5S target detection model can be used to detect an image to be recognized, the image to be recognized is input into a YOLOV5S target detection model trained in advance, and the model outputs a disconnecting link type and a disconnecting link basic opening and closing state. The types of the knife switch include a double-column horizontal rotary knife switch, a three-column horizontal rotary knife switch, a single-arm horizontal telescopic knife switch, a single-arm vertical telescopic knife switch, a double-arm vertical telescopic knife switch and the like, and are not particularly limited herein; the basic disconnecting and connecting state of the disconnecting switch is the disconnecting and connecting state of the disconnecting switch which is preliminarily identified according to the first target detection model, and the basic disconnecting and connecting state of the disconnecting switch is a switching-on state or a switching-off state.

Step S220, determining the position relation between the knife arm of the knife switch and the knife switch structural part based on the image to be identified; the knife switch structural part is preset aiming at the knife arm.

Specifically, from the image to be recognized, knife arm, connector and other knife structure members of the knife are detected. Here, the structure of the knife switch corresponding to the knife arm may be preset according to the type of the knife switch detected in step S210. For different types of disconnecting links, the relative position between the disconnecting link and the disconnecting link structural member can be represented by the position relation between the disconnecting link and the disconnecting link structural member, and the state of the disconnecting link can be more accurately determined according to the relative position of the disconnecting link. For example, when the type of the knife switch is non-three-column horizontal rotation type, the knife switch includes two knife arms, and the detection of the position relationship between the knife arm in the image to be recognized and the knife switch structural member corresponding to the knife arm may be the detection of the position relationship between one knife arm and the other knife arm in the two knife arms in the image to be recognized. For a non-three-column horizontal rotary type disconnecting link, when the disconnecting link base is in a switching-on state, if two disconnecting link arms are basically parallel, the state of the disconnecting link is judged to be in place; when the basic on-off state of the disconnecting link is disconnecting, if the two disconnecting links are basically parallel, judging that the state of the disconnecting link is in place; otherwise, the state of the knife switch is judged to be an abnormal state. When the type of the disconnecting link is three-column horizontal rotation type, the disconnecting link comprises a disconnecting link arm and two connecting parts, and the position relation between the disconnecting link arm in the image to be recognized and a preset disconnecting link structural part corresponding to the disconnecting link arm is detected, wherein the position relation between the disconnecting link arm in the image to be recognized and the two connecting parts can be detected. For the three-column horizontal rotary type disconnecting link, when the disconnecting link base is switched on, if the connecting line between the disconnecting link and the two connecting parts is basically parallel, the state of the disconnecting link is judged to be in place; when the basic on-off state of the disconnecting link is disconnecting, if the maximum distance of a connecting line between the disconnecting link and the two connecting parts is greater than a preset distance or the included angle of the connecting line between the disconnecting link and the two connecting parts is greater than a preset angle, judging that the disconnecting link is in place; otherwise, the state of the knife switch is judged to be an abnormal state.

And step S230, determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relation.

Specifically, the state of the disconnecting link is determined according to the disconnecting link basic opening and closing state and the type of the disconnecting link detected in step S210 and the position relationship determined in step S220. The states of the knife switch comprise a closing in-place state, a breaking in-place state and an abnormal state. In this embodiment, the type of the knife switch and the preliminary closing state of the knife switch are obtained by detecting the image to be recognized through a preset target detection model, the position relationship between the knife arm and the knife switch structural member is detected according to the image to be recognized, and the final state of the knife switch is determined according to the type of the knife switch, the preliminary closing state of the knife switch and the position relationship. The final state of the disconnecting link is determined by the type of the disconnecting link and the preliminary closing state and position relation of the disconnecting link, and the accuracy of disconnecting link state identification is improved.

Further, in some embodiments, based on the step S220, the determining the position relationship between the knife arm of the knife switch and the knife switch structural member based on the image to be identified specifically includes the following steps:

step S222, obtaining a disconnecting link target image containing a disconnecting link from the image to be identified; the area ratio of the disconnecting link in the disconnecting link target image is larger than the area ratio threshold value.

Specifically, the image to be identified is detected, and a disconnecting link target image containing a disconnecting link is obtained. For example, after the switch body is detected, a circumscribed polygon with the smallest area corresponding to the switch body region is used as a switch target boundary, where the circumscribed polygon may be a circumscribed rectangle. And determining the image area framed by the knife switch target boundary as a knife switch target image, or determining the image area framed after the knife switch target boundary is expanded to a certain extent as a target image. The area ratio of the disconnecting link in the disconnecting link target image is larger than the area ratio threshold value, so that the subsequent disconnecting link structural member detection is facilitated.

Step S224, detecting the knife arm and the knife switch structural member in the knife switch target image, and determining the position relation between the knife arm of the knife switch and the preset knife switch structural member based on the detection result.

Specifically, knife gate structural members such as knife arms and connecting pieces in knife gate target images are detected. Here, the knife switch structure corresponding to the knife arm may be preset according to the knife switch type detected in step S210. For different types of disconnecting links, the relative position between the disconnecting link and the disconnecting link structural member can be represented by the position relation between the disconnecting link and the disconnecting link structural member, and the state of the disconnecting link can be more accurately determined according to the relative position of the disconnecting link. Further, in some embodiments, based on the step S210, when detecting the to-be-identified image including the disconnecting link and obtaining the disconnecting link type and the disconnecting link basic opening and closing state of the disconnecting link in the to-be-identified image, the method further includes the following steps:

step S212, detecting the image to be identified containing the disconnecting link according to a preset first target detection model to obtain the disconnecting link target boundary of the disconnecting link in the image to be identified.

Specifically, the image to be recognized is detected according to a preset first target detection model, and a disconnecting link target boundary, a disconnecting link type and a disconnecting link basic opening and closing state are obtained. For example, a YOLOV5S target detection model can be used to detect an image to be recognized, the image to be recognized is input into a YOLOV5S target detection model trained in advance, and the model outputs a disconnecting link target boundary, a disconnecting link type and a disconnecting link basic on-off state. The knife switch target boundary is the boundary of the region of the knife switch main body in the image to be identified; the types of the knife switch include a double-column horizontal rotary knife switch, a three-column horizontal rotary knife switch, a single-arm horizontal telescopic knife switch, a single-arm vertical telescopic knife switch, a double-arm vertical telescopic knife switch and the like, and are not particularly limited herein; the basic disconnecting and connecting state of the disconnecting switch is the disconnecting and connecting state of the disconnecting switch which is preliminarily identified according to the first target detection model, and the basic disconnecting and connecting state of the disconnecting switch is a switching-on state or a switching-off state.

On the basis of obtaining the edge of the disconnecting link target, based on the step S222, obtaining the disconnecting link target image including the disconnecting link from the image to be identified, which specifically includes the following steps:

step S223, based on the edge of the knife-switch target, obtaining a knife-switch target image containing the knife-switch from the image to be identified.

Specifically, based on the edge of the disconnecting link target obtained in step S212, the image to be recognized is processed to obtain a disconnecting link target image. The image area framed by the knife switch target boundary can be determined as a knife switch target image, and the image area framed after the knife switch target boundary is expanded to a certain extent can also be determined as a target image.

In some embodiments, the detecting the position relationship between the knife arm in the knife gate target image and the preset knife gate structural member corresponding to the knife arm may be detecting a distance between the knife arm in the knife gate target image and the preset knife gate structural member corresponding to the knife arm.

In some embodiments, the detecting of the position relationship between the knife arm in the knife gate target image and the preset knife gate structural member corresponding to the knife arm may be detecting an included angle between the knife arm in the knife gate target image and the preset knife gate structural member corresponding to the knife arm, or detecting an included angle between the knife arm in the knife gate target image and a preset knife gate structural member connecting line corresponding to the knife arm.

Fig. 3 is a flowchart of another method for identifying a state of a knife switch in this embodiment, and as shown in fig. 3, the flowchart includes the following steps:

and S310, detecting the image to be recognized according to a preset first target detection model to obtain a disconnecting link target boundary, a disconnecting link type and a disconnecting link basic opening and closing state of a disconnecting link in the image to be recognized.

Step S320, based on the edge of the knife switch target, obtaining a knife switch target image only containing the knife switch from the image to be identified.

Step S330, when the type of the disconnecting link is non-three-column horizontal rotation type, detecting an included angle between two disconnecting link arms in the disconnecting link target image.

Specifically, when the knife switch type is a non-three-column horizontal rotation type, the knife switch comprises two knife arms. And detecting an included angle between two knife arms in the knife switch target image by using a preset semantic segmentation model.

And step S340, determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the included angle.

Specifically, if the on-off state of the knife switch base is a closing state and the condition gamma is met, wherein gamma belongs to [180 ° -alpha, 180 ° + alpha ], the state of the knife switch is judged to be in place; if the basic on-off state of the disconnecting link is a closing state and the condition gamma is not met, wherein the gamma belongs to [ 180-alpha, 180-alpha + alpha ], judging that the state of the disconnecting link is an abnormal state; if the basic opening and closing state of the disconnecting link is the opening state and the condition gamma is met, wherein gamma is an element [0 DEG, beta ], judging that the state of the disconnecting link is in place; and if the disconnecting link basic opening and closing state is the disconnecting link state and the condition gamma is not met, determining that the state of the disconnecting link is the abnormal state. Wherein gamma is an included angle between two knife arms in the knife switch target image, alpha is a switch-on in-place threshold value, alpha belongs to [0 DEG and 90 DEG ], beta is a switch-off in-place threshold value, and beta belongs to [0 DEG and 90 DEG ].

This embodiment is through detecting the contained angle between two sword arms, and the final state of switch is judged according to the contained angle between two sword arms, has further improved the degree of accuracy that the switch state detected.

Further, in some embodiments, based on the step S330, detecting an included angle between two knife arms in the knife gate target image specifically includes the following steps:

and S332, segmenting the knife gate target image according to a preset semantic segmentation model to obtain a knife arm mask image of the knife gate. The tool arm mask graph comprises a tool arm target area and tool arm type information, and the tool arm type information is used for distinguishing two tool arms in the knife switch target image.

Specifically, the knife gate target image is input into a pre-trained semantic segmentation model, and the model outputs a knife arm mask image of the knife gate, wherein the knife arm mask image comprises a knife arm target area and knife arm type information. The knife arm types of different knife switches comprise a left knife arm, a right knife arm, an upper knife arm and a lower knife arm. For example, the tool arm type information may be background information of tool arm target areas of different tool arm types in the tool arm mask map.

In some embodiments, a lightweight HrNet semantic segmentation model may be used to segment a knife gate target image to obtain a knife arm mask map of the knife gate. The number of input channels from the second stage to the fourth stage of the main network of the lightweight HrNet semantic segmentation model is 18, and the number of BASIC blocks in the first stage to the fourth stage is 2. The lightweight HrNet semantic segmentation model reduces the model parameters by reducing the number of input channels and the number of blocks, thereby reducing the amount of model calculation.

Step 334, respectively calculating included angles between the two tool arms and a preset axis according to the tool arm mask map, and calculating an included angle between the two tool arms based on the included angles between the two tool arms and the preset axis.

Specifically, based on the tool arm target region and the tool arm type information obtained in step S332, a preset algorithm is used to obtain first tool arm target regions corresponding to the two tool arms, respectively, a circumscribed rectangle with the smallest area of the first tool arm target region is calculated, and an included angle between a straight line where a preset side of the circumscribed rectangle is located and a preset axis is calculated. The maximum connected domain algorithm can be adopted to find the region with the largest area in the tool arm target region of each tool arm, and the circumscribed rectangle with the smallest area of the region with the largest area is calculated.

In the embodiment, the semantic segmentation model is utilized to define two knife arms of the knife switch into different types, and the model outputs knife arm type information and a knife arm target area, so that the problem that an included angle cannot be calculated due to the fact that the two knife arms are overlapped in an image is solved.

Fig. 4 is a flowchart of another method for identifying a state of a knife switch in this embodiment, and as shown in fig. 4, the flowchart includes the following steps:

and S410, detecting the image to be recognized according to a preset first target detection model to obtain the disconnecting link target boundary, the disconnecting link type and the disconnecting link basic on-off state of the disconnecting link in the image to be recognized.

Step S420, based on the edge of the knife switch target, obtaining a knife switch target image only containing the knife switch from the image to be identified.

And step S430, when the type of the disconnecting link is three-column horizontal rotation type, detecting an included angle between the disconnecting link and a connecting line of the two connecting parts in the disconnecting link target image.

Specifically, a pre-trained second target detection model is used for detecting a knife arm and two connecting parts in a knife switch target image, and an included angle between the knife arm and a connecting line of the two connecting parts is calculated according to the detected knife arm and the connecting parts. The second object detection model may be the same as or different from the first object detection model, and is not limited in particular.

And step S440, determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the included angle.

Specifically, if the on-off state of the knife switch base is a closing state and the condition gamma is met, wherein gamma belongs to [180 ° -alpha, 180 ° + alpha ], the state of the knife switch is in place for closing; if the basic on-off state of the disconnecting link is a closing state and the condition gamma is not met, wherein the gamma belongs to [ 180-alpha, 180-alpha + alpha ], the state of the disconnecting link is an abnormal state; if the on-off state of the disconnecting link base is the opening state and the condition gamma is met, wherein gamma belongs to [ beta, 360 degrees ], the state of the disconnecting link is in place; if the on-off state of the disconnecting link base is an off state and the condition gamma is not met, the state of the disconnecting link is an abnormal state; wherein gamma is an included angle between a knife arm in a knife switch target image and a connecting line of two connecting parts, alpha is a switch-on in-place threshold value, alpha belongs to [0 DEG and 90 DEG ], beta is a switch-off in-place threshold value, and beta belongs to [0 DEG and 90 DEG ].

This embodiment judges the final state of switch according to the contained angle between the connecting wire of sword arm and two adapting unit through detecting the contained angle between the connecting wire of sword arm and two adapting unit, has further improved the degree of accuracy that the switch state detected.

The present embodiment is described and illustrated below by means of preferred embodiments.

Fig. 5 is a flowchart of the method for identifying the status of the disconnecting link according to the preferred embodiment. As shown in fig. 5, the method for identifying the state of the disconnecting link includes the following steps:

step S501, a closing in-place threshold value alpha and a separating in-place threshold value beta are configured.

Specifically, α is a closing in-place threshold value, α ∈ [0 °, 90 ° ], β is an opening in-place threshold value, β ∈ [0 °, 90 ° ], α is preferably 10 °, and β is preferably 20 °.

And step S502, detecting the range of the knife switch body, the type of the knife switch and the on-off state of the knife switch base. If the knife switch type is the three-column horizontal rotation type, step S506 is executed, otherwise step S503 is executed.

Firstly, data acquisition is carried out by a camera, scene images containing the disconnecting link equipment in the transformer substation are acquired, the acquired data are divided into a training set and a testing set, and the main body range, the type and the on-off state of the disconnecting link are marked. Subsequently, the YOLOV5S target detection model was trained using the data of the training set, and the model was tested using the data of the test set. And after the model training and testing are finished, the image to be tested is used as the model input, and the model outputs the main body range, the type and the on-off state of the disconnecting link. And cutting the image to be detected according to the main body range of the disconnecting link to obtain a disconnecting link target image.

And step S503, cutting the tool arm in the knife switch target image.

The method comprises the steps of marking the range and the category of a tool arm for an acquired image, wherein the category comprises a left tool arm, a right tool arm, an upper tool arm and a lower tool arm, manufacturing a training set and a testing set, then training a light-weight HrNet semantic segmentation model by using data of the training set, and testing the model by using data of the testing set. Finally, test set data is input, and the model divides the range of the cutter arm. After the model training and testing are completed, the knife gate target image obtained in the step S502 is used as a model input, and the model outputs a knife arm mask image, which includes a knife arm area and a knife arm category. For example, for a knife switch with the knife arm type of left knife arm and right knife arm, the knife arm mask diagram is shown in fig. 6, and fig. 6 is a knife arm mask diagram illustration of the preferred embodiment. The tool arm mask image comprises a left tool arm target area 601, a right tool arm target area 602 and tool arm type information, wherein the left tool arm target area 601 is an area where a left tool arm is located, and the right tool arm target area 602 is an area where a right tool arm is located. As in the blade mask of fig. 6, the left and right blades have different backgrounds to distinguish the different types of blades.

Step S504, the tool arm profile is detected.

And (4) finding the region with the largest area in each tool arm by using the tool arm mask map obtained in the step (S503) and adopting a maximum connected domain algorithm, and calculating the circumscribed rectangle with the smallest area in the region. Fig. 7 is a tool arm mask diagram illustrating the detection of the circumscribed rectangle of the tool arm according to the preferred embodiment. As shown in fig. 7, the maximum connected domain algorithm is applied to the left tool arm target region 601 to find the maximum region 701 of the left tool arm, the maximum connected domain algorithm is applied to the right tool arm target region 602 to find the maximum region 702 of the right tool arm, the left tool arm circumscribed rectangle 703 with the minimum area of the maximum region 701 of the left tool arm is calculated, and the right tool arm circumscribed rectangle 704 with the minimum area of the maximum region 702 of the right tool arm is calculated. The left tool arm circumscribed rectangle 703 is a rectangle with the smallest inner product of the circumscribed rectangle of the maximum area 701 of the left tool arm, and the right tool arm circumscribed rectangle 704 is a rectangle with the smallest inner product of the circumscribed rectangle of the maximum area 702 of the right tool arm. For example, when the left tool arm target area 601 and the right tool arm target area 602 both have only one connected area, the maximum area 701 of the left tool arm is the left tool arm target area 601, and the maximum area 702 of the right tool arm is the right tool arm target area 602; when a knife arm region in a knife switch target image is blocked, one knife arm may be divided into a plurality of regions, a left knife arm target region 601 or a right knife arm target region 602 may have a plurality of connected regions, at this time, a region with the largest area in the left knife arm target region 601 is detected as a left knife arm maximum region 701, and a region with the largest area in the right knife arm target region 602 is detected as a right knife arm maximum region 702.

Step S505, a tool arm angle is calculated.

And respectively calculating included angles between the straight line of the long edges of the two cutter arms and the horizontal direction according to the cutter arm circumscribed rectangle obtained in the step S504, and respectively recording the included angles as a and b. First, the slope k, theta (arctan) (k) 180/pi is calculated according to two end points of the tool arm line, and if theta is less than 0, theta (theta) 180+ theta is calculated.

Step S506, the rotating member and the connecting member are detected.

When the type of the switch detected in step S502 is the three-column horizontal rotation type, the two arms of the three-column horizontal rotation type switch are integrated, and the included angle is always 0, so that the rotating member and the connecting member need to be detected. Fig. 8 is a block diagram of a three-post horizontally-rotating knife switch according to the preferred embodiment, and as shown in fig. 8, the knife switch includes a knife arm 801, a rotating member 802, a first connecting member 803, and a second connecting member 804. At this time, the included angle between the knife arm and the connecting part is calculated. The blade arm, the rotating part and the connecting part are detected on the blade target image obtained in step S502 by using a YOLOV5S target detection model. Specifically, the knife gate target image is divided into a training set and a testing set, and a knife arm, a rotating part and a connecting part are marked. Subsequently, the YOLOV5S target detection model was trained using the data of the training set, and the model was tested using the data of the test set. And after the model training and testing are finished, the disconnecting link target image is used as a model input, and the model outputs the knife arm, the rotating part and the connecting part. The angle c of the blade arm 801 to the horizontal and the angle d of the connecting line of the first connecting part 803 and the second connecting part 804 to the horizontal are calculated.

Step S507, an included angle γ is calculated.

Calculating the included angle gamma according to the angles of the two arm lines obtained in the step S505 or the angle obtained in the step S506₁，γ₁| a-b | or γ₁I c-d i. According to the state of the knife arm obtained in the step S502, if the on-off state of the knife switch base is the on state, and gamma is set₁Less than 90 °, then γ is 180 ° - γ₁(ii) a If the basic on-off state of the knife switch is off and gamma₁Greater than 90 °, then γ is 180 ° - γ₁(ii) a In other cases, γ ═ γ₁。

In step S508, the state of the knife switch is identified.

If the knife switch type is not a three-column horizontal rotation type, then,

(1) the disconnecting link base switching-on/off state detected in the step S502 is switched on, 180 degrees-alpha-gamma-180 degrees + alpha, and the disconnecting link state is switched on in place;

(2) the state of the disconnecting link detected in step S502 is open-off, and 0 ° < ═ γ < ═ β, the state of the disconnecting link is open-off in place;

(3) otherwise, the state of the knife switch is abnormal opening and closing.

If the knife switch type is a three-column horizontal rotation type, then,

(1) the disconnecting link base switching-on/off state detected in the step S502 is switching-on, 180 ° - α ═ γ ═ 180 ° + α, and the disconnecting link state is switching-on in place;

(2) the on-off state of the disconnecting link base detected in the step S502 is an open switching state, and γ > is β, and the disconnecting link state is an open switching state;

(3) otherwise, the state of the knife switch is abnormal opening and closing.

In the embodiment, the types and the opening and closing states of the disconnecting links are obtained by using the target detection model, so that accurate prior information is provided for calculating included angles of the disconnecting links of different types, and the universality is improved; and defining the two cutter arms into different categories by utilizing a semantic segmentation model, so that the problem that the included angle cannot be calculated due to the overlapping of the two cutter arms is solved. The method is based on the traditional image processing method, is greatly influenced by light change, has poor robustness, and improves the accuracy of knife switch state identification based on the knife switch state identification of a deep learning model of target detection and semantic segmentation.

In this embodiment, a device for identifying a state of a disconnecting link is further provided, and the device is used to implement the above embodiments and preferred embodiments, which have already been described and will not be described again. The terms "module," "unit," "subunit," and the like as used below may implement a combination of software and/or hardware for a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

Fig. 9 is a block diagram of a disconnecting link status recognition device according to the present embodiment, and as shown in fig. 9, the device includes:

the detection module 910 is configured to detect an image to be identified including a switch according to a preset first target detection model, so as to obtain a switch type and a switch basic on-off state of the switch in the image to be identified;

the processing module 920 is configured to determine a position relationship between a knife arm of the knife switch and a knife switch structural member based on the image to be identified; the knife switch structural part is preset aiming at the knife arm;

and the identification module 930 is configured to determine the state of the disconnecting link according to the basic switching-on/off state of the disconnecting link, the type of the disconnecting link, and the position relationship.

There is also provided in this embodiment an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.

Optionally, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.

Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:

s1, detecting the image to be recognized according to a preset first target detection model to obtain a disconnecting link target boundary, a disconnecting link type and a disconnecting link basic opening and closing state of a disconnecting link in the image to be recognized;

s2, obtaining a disconnecting link target image only containing a disconnecting link from the image to be identified based on the disconnecting link target boundary;

s3, detecting the position relation between the knife arm in the knife-switch target image and a knife-switch structural member corresponding to the knife arm;

and S4, determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relation.

It should be noted that, for specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and optional implementations, and details are not described again in this embodiment.

In addition, in combination with the method for identifying the state of the disconnecting link provided in the above embodiment, a storage medium may also be provided in this embodiment. The storage medium having stored thereon a computer program; the computer program, when executed by a processor, implements the steps of any one of the above-described method for identifying a state of a knife switch.

It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to be limiting. All other embodiments, which can be derived by a person skilled in the art from the examples provided herein without any inventive step, shall fall within the scope of protection of the present application.

It is obvious that the drawings are only examples or embodiments of the present application, and it is obvious to those skilled in the art that the present application can be applied to other similar cases according to the drawings without creative efforts. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.

The term "embodiment" is used herein to mean that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly or implicitly understood by one of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the patent protection. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for identifying the state of a disconnecting link is characterized by comprising the following steps,

2. The method for identifying the state of the disconnecting link according to claim 1, wherein the determining the position relationship between the disconnecting link arm and the disconnecting link structural member based on the image to be identified comprises:

3. The method for identifying the disconnecting link state according to claim 2, wherein when detecting an image to be identified including a disconnecting link to obtain the disconnecting link type and the disconnecting link basic on-off state of the disconnecting link in the image to be identified, the method further comprises:

and obtaining the knife switch target image containing the knife switch from the image to be identified based on the knife switch target boundary.

4. The method for identifying the state of the disconnecting link according to claim 1, wherein when the type of the disconnecting link is a non-three-column horizontal rotation type, the disconnecting link comprises two disconnecting link arms; the detecting of the position relationship between the knife arm in the knife-switch target image and the preset knife-switch structural member corresponding to the knife arm comprises detecting the position relationship between one knife arm and the other knife arm in the two knife arms in the knife-switch target image.

5. The method for identifying the state of the knife switch according to claim 4, wherein the detecting the position relationship between one of the two knife arms and the other knife arm in the knife switch target image comprises detecting an included angle between the two knife arms in the knife switch target image.

6. The method for identifying the state of the disconnecting link according to claim 1, wherein when the type of the disconnecting link is a three-column horizontal rotation type, the disconnecting link comprises a disconnecting link arm and two connecting parts; the detecting of the position relationship between the knife arm in the knife-switch target image and the preset knife-switch structural member corresponding to the knife arm comprises detecting the position relationship between the knife arm in the knife-switch target image and the two connecting parts.

7. The method for identifying the state of the disconnecting link according to claim 6, wherein the detecting the position relationship between the disconnecting link arm in the disconnecting link target image and the two connecting components comprises detecting an included angle between the disconnecting link arm in the disconnecting link target image and a connecting line of the two connecting components.

8. The method for identifying the state of the disconnecting link according to claim 5, wherein when the type of the disconnecting link is a non-three-post horizontal rotation type, the determining the state of the disconnecting link according to the basic switching-on/off state of the disconnecting link, the type of the disconnecting link and the positional relationship comprises,

if the on-off state of the disconnecting link base is an off state and meets the condition gamma belongs to [0 DEG, beta ], the state of the disconnecting link is in place;

9. The method for identifying the state of the disconnecting link according to claim 7, wherein the determining the state of the disconnecting link according to the basic on-off state of the disconnecting link, the type of the disconnecting link and the position relationship comprises,

10. The method of claim 5 or 8, further comprising,

and respectively calculating included angles between the two tool arms and a preset axis according to the tool arm mask image, and calculating the included angle between the two tool arms based on the included angles between the two tool arms and the preset axis.

11. The method for identifying the status of the knife switch according to claim 8, wherein the calculating the included angles between the two knife arms and the preset axis respectively according to the knife arm mask map comprises,

12. A device for identifying the state of a disconnecting link is characterized by comprising,

the detection module is used for detecting an image to be identified containing a disconnecting link according to a preset first target detection model to obtain the disconnecting link type and the disconnecting link basic opening and closing state of the disconnecting link in the image to be identified;

13. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform the method for identifying a status of a disconnecting link according to any one of claims 1 to 11.

14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for identification of the state of a knife switch of any one of claims 1 to 11.