CN114863311A - Automatic tracking method and system for inspection target of transformer substation robot - Google Patents
Automatic tracking method and system for inspection target of transformer substation robot Download PDFInfo
- Publication number
- CN114863311A CN114863311A CN202210284103.XA CN202210284103A CN114863311A CN 114863311 A CN114863311 A CN 114863311A CN 202210284103 A CN202210284103 A CN 202210284103A CN 114863311 A CN114863311 A CN 114863311A
- Authority
- CN
- China
- Prior art keywords
- target
- inspection
- robot
- image
- automatic tracking
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for automatically tracking a patrol target of a transformer substation robot, wherein the method comprises the following steps: acquiring and preprocessing the video data of the robot polling; identifying and obtaining the target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model; and determining the distance and relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot so that the target image is always in the center area of the image. The invention creatively provides an automatic tracking method of the inspection target, which can accurately and automatically track the inspection equipment in real time, so that the target image is always in the central area of the image, the problem of inaccurate inspection target identification in the prior art is solved, and the accuracy of target identification and positioning is improved.
Description
Technical Field
The invention relates to the technical field of robot inspection, in particular to a method and a system for automatically tracking inspection targets of a transformer substation robot.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The transformer substation inspection robot system is a complex system integrating mechatronics, multi-sensor fusion, navigation and behavior planning, robot vision and wireless transmission. The inspection robot carries a CCD camera, an infrared thermal imager, a pickup and ultrasonic waves, autonomously or remotely inspects outdoor high-voltage equipment by navigation positioning and planning an optimal path, acquires information such as an infrared chart, an image and audio of the equipment, automatically identifies thermal defects, switch or disconnecting link states of the equipment, and generates uniform and standard alarm items and inspection reports.
Due to the positioning error of the robot and the error of the freedom degree of the robot body, the polling target can deviate or even be lost when the robot gets an image from a fixed point.
In the prior art, the traditional machine learning algorithm is adopted to position the target in the inspection image, the recognition error rate is high, the positioning is not accurate, and the camera of the robot can not be aligned to the inspection target all the time, so that the extraction of the inspection target is not facilitated.
Disclosure of Invention
In order to solve the problems, the invention provides a method and a system for automatically tracking a patrol target of a transformer substation robot, wherein the patrol target is positioned by utilizing a multi-target detection algorithm YOLO based on deep learning; and calculating the deflection direction of the camera by combining the position coordinates of the inspection target in the image through a visual servo method, so that the inspection target is always positioned in the middle area of the video.
In some embodiments, the following technical scheme is adopted:
a method for automatically tracking a patrol target of a transformer substation robot comprises the following steps:
acquiring and preprocessing the video data of the robot polling;
identifying and obtaining the target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
and determining the distance and relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot so that the target image is always in the center area of the image.
As an optional scheme, acquiring the inspection video data of the robot, and preprocessing the inspection video data specifically as follows:
and decomposing the video frame into a plurality of images, analyzing the background characteristics of the inspection image, and filtering the inspection image by adopting a histogram equalization technology.
As an optional scheme, the trained target detection model specifically includes: improved YOLO target detection model.
As an optional scheme, the improved YOLO target detection model takes a residual error network as a feature extraction network.
Optionally, the improved YOLO target detection model is provided with a data enhancement module before the feature extraction network.
As an optional scheme, the improved YOLO target detection model removes a full connection layer on the basis of an original YOLO algorithm.
As an optional scheme, a hole convolution network is finally set in the improved YOLO target detection model, and semantic features of the feature map are increased.
In other embodiments, the following technical solutions are adopted:
the utility model provides a transformer substation robot patrols and examines target automatic tracking system, includes:
the data acquisition module is used for acquiring the inspection video data of the robot and carrying out pretreatment;
the inspection target identification module is used for identifying and obtaining a target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
and the target tracking module is used for determining the distance and the relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot, so that the target image is always in the center area of the image.
In other embodiments, the following technical solutions are adopted:
a terminal device comprising a processor and a memory, the processor being arranged to implement instructions; the storage is used for storing a plurality of instructions, and the instructions are suitable for being loaded by the processor and executing the automatic tracking method for the inspection target of the substation robot.
In other embodiments, the following technical solutions are adopted:
a computer readable storage medium, wherein a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the substation robot inspection target automatic tracking method.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention creatively provides an automatic tracking method of the inspection target, which can accurately and real-timely detect and automatically track the target of the inspection equipment, so that the target image is always in the central area of the image, the problem of inaccurate inspection target identification in the prior art is solved, and the accuracy of target identification and positioning is improved.
(2) The invention creatively provides an improved YOLO target detection model, and a data enhancement module is added on the basis of the original YOLO algorithm, so that training data can be expanded, and the diversity of training samples is increased; adding a hole convolution network to highlight semantic features of the feature map; the improved YOLO target detection model can reduce the occupancy rate of video memory, reduce the false detection rate of routing inspection target identification and improve the accuracy of target detection.
Additional features and advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
Fig. 1 is a schematic diagram of an automatic tracking method for a patrol target of a transformer substation robot in an embodiment of the invention;
fig. 2 is a schematic diagram of an improved YOLO target detection model training process in the embodiment of the present invention.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
In one or more embodiments, a method for automatically tracking a patrol target of a transformer substation robot is disclosed, and the method comprises the steps of firstly, collecting a patrol video by the transformer substation patrol robot through a carried visible light camera, decomposing a video frame into a plurality of images, and processing the images by using an image preprocessing technology, so as to filter the interference of background and other veiling glare on equipment identification; then, positioning the routing inspection target by utilizing a multi-target detection algorithm YOLO based on deep learning; and finally, calculating the deflection direction of the camera by combining the position coordinates of the inspection target in the image through a visual servo method, so that the inspection target is always positioned in the middle area of the video.
As a specific implementation manner, the method for automatically tracking the inspection target of the substation robot in this embodiment specifically includes, with reference to fig. 1, the following steps:
s101: the method comprises the steps of obtaining robot inspection video data, decomposing a video frame into a plurality of images, and processing the images by using an image preprocessing technology, so that interference of a background and other stray light on equipment identification is filtered.
Specifically, the transformer substation robot shoots an inspection video by carrying a visible light camera, analyzes the background characteristics of the inspection video, and performs filtering processing on the inspection video by adopting a histogram equalization technology, so that the contrast and the definition of a video picture are improved, the quality of the acquired video is ensured, and a foundation is laid for the following intelligent identification;
the video frame is decomposed into a plurality of images, and the images are processed by using an image preprocessing technology, so that the interference of background and other stray light on equipment identification is filtered.
S102: identifying and obtaining the target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
in the embodiment, in order to realize the front-end real-time detection and identification of the substation robot, the original YOLOv1 algorithm is improved, a YOLO target detection model based on a residual error network structure is designed, and the residual error network model is selected as a feature extraction module of a target detector. The improvement measures are as follows:
1) adding data enhancement module (including adjusting image saturation, brightness, translation, random cropping, etc.);
2) the fully connected layer is removed so that the model can accept multiple sizes of input (necessarily multiples of 32) and thus can be multi-scale trained and predicted. (default input is 448 × 448 × 3, the number of grids is 14 × 14, one object is detected per grid, and up to 256 objects are detected);
3) replacing the feature extraction network with a resnet series (selecting resnet 18);
4) and adding a hole convolution module at the last of the model, and increasing the semantic features of the feature map under the condition of keeping the resolution of the feature map.
The routing inspection image recognition is carried out through the YOLO algorithm, the complexity of model operation can be reduced, the reasoning time of a deep learning model is shortened, and the method has great significance for improving the real-time performance of intelligent recognition.
With reference to fig. 2, the training process of the present embodiment for the improved YOLO target detection model is as follows:
(1) collecting a transformer substation inspection image sample, calibrating a target position in the image, and dividing a training set and a test set after the sample is manufactured.
(2) Extracting image features using resnet18 as the underlying network of the YOLO framework;
(3) and classifying and positioning the target by utilizing a subsequent classification module and a frame regression module.
(4) And calculating training loss according to the predicted value and the true value of the model, and continuously optimizing the parameters of the network.
S103: and determining the distance and relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot so that the target image is always in the center area of the image.
In the embodiment, the inspection target of the transformer substation robot is automatically tracked through visual servo; firstly, obtaining the coordinate position of the center point of the inspection equipment in the image by using a YOLO target detection algorithm, and then calculating the distance between the center point of the inspection equipment and the center point of the image and the position of the inspection equipment relative to the center of the image. And finally, calculating the deflection direction of the transformer substation robot carrying the camera according to the distance between the central points and the relative position relation, and continuously moving the camera according to a certain direction until the inspection target is in the central area of the image.
According to the method, the inspection target can be always in the central area of the inspection image, and the target detection and automatic tracking of the inspection equipment can be accurately and timely carried out.
Example two
In one or more embodiments, disclosed is a substation robot inspection target automatic tracking system, including:
the data acquisition module is used for acquiring the inspection video data of the robot and carrying out pretreatment;
the inspection target identification module is used for identifying and obtaining a target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
and the target tracking module is used for determining the distance and the relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot, so that the target image is always in the center area of the image.
It should be noted that, a specific implementation of the above module has been described in the first embodiment, and is not described in detail here.
EXAMPLE III
In one or more embodiments, a terminal device is disclosed, which includes a server, where the server includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the substation robot inspection target automatic tracking method in the first embodiment. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software.
Example four
In one or more embodiments, a computer-readable storage medium is disclosed, in which a plurality of instructions are stored, and the instructions are adapted to be loaded by a processor of a terminal device and execute the substation robot inspection target automatic tracking method described in the first embodiment.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A method for automatically tracking a patrol target of a transformer substation robot is characterized by comprising the following steps:
acquiring and preprocessing the video data of the robot polling;
identifying and obtaining the target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
and determining the distance and relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot so that the target image is always in the center area of the image.
2. The automatic tracking method for the inspection target of the substation robot according to claim 1, characterized by acquiring inspection video data of the robot and preprocessing the inspection video data, specifically:
and decomposing the video frame into a plurality of images, analyzing the background characteristics of the inspection image, and filtering the inspection image by adopting a histogram equalization technology.
3. The automatic tracking method for the inspection tour target of the substation robot according to claim 1, wherein the trained target detection model specifically comprises: improved YOLO target detection model.
4. The substation robot inspection target automatic tracking method according to claim 3, wherein the improved YOLO target detection model takes a residual error network as a feature extraction network.
5. The substation robot inspection target automatic tracking method according to claim 3, wherein the improved YOLO target detection model is provided with a data enhancement module before a feature extraction network.
6. The automatic tracking method for the inspection tour target of the transformer substation robot as claimed in claim 3, wherein the improved YOLO target detection model removes a full connection layer on the basis of an original YOLO algorithm.
7. The automatic tracking method for the inspection tour target of the substation robot as claimed in claim 3, wherein a hole convolution network is arranged at the end of the improved YOLO target detection model, and semantic features of a feature map are increased.
8. The utility model provides a transformer substation robot patrols and examines target automatic tracking system which characterized in that includes:
the data acquisition module is used for acquiring the inspection video data of the robot and carrying out pretreatment;
the inspection target identification module is used for identifying and obtaining a target image position of the robot inspection equipment based on the preprocessed video image data and the trained target detection model;
and the target tracking module is used for determining the distance and the relative position relationship between the center point of the inspection equipment and the center point of the image, and calculating the deflection direction of a camera carried by the inspection robot, so that the target image is always in the center area of the image.
9. A terminal device comprising a processor and a memory, the processor being arranged to implement instructions; the memory is used for storing a plurality of instructions, wherein the instructions are suitable for being loaded by the processor and executing the substation robot inspection target automatic tracking method according to any one of claims 1-7.
10. A computer readable storage medium having stored therein a plurality of instructions, wherein the instructions are adapted to be loaded by a processor of a terminal device and to perform the substation robot inspection target automatic tracking method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210284103.XA CN114863311A (en) | 2022-03-22 | 2022-03-22 | Automatic tracking method and system for inspection target of transformer substation robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210284103.XA CN114863311A (en) | 2022-03-22 | 2022-03-22 | Automatic tracking method and system for inspection target of transformer substation robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114863311A true CN114863311A (en) | 2022-08-05 |
Family
ID=82627915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210284103.XA Pending CN114863311A (en) | 2022-03-22 | 2022-03-22 | Automatic tracking method and system for inspection target of transformer substation robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114863311A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN116258466A (en) * | 2023-05-15 | 2023-06-13 | 国网山东省电力公司菏泽供电公司 | Multi-mode power scene operation specification detection method, system, equipment and medium |
-
2022
- 2022-03-22 CN CN202210284103.XA patent/CN114863311A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN116258466A (en) * | 2023-05-15 | 2023-06-13 | 国网山东省电力公司菏泽供电公司 | Multi-mode power scene operation specification detection method, system, equipment and medium |
| CN116258466B (en) * | 2023-05-15 | 2023-10-27 | 国网山东省电力公司菏泽供电公司 | Multi-mode power scene operation specification detection method, system, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110059558B (en) | Orchard obstacle real-time detection method based on improved SSD network | |
| CN109977813B (en) | Inspection robot target positioning method based on deep learning framework | |
| CN102175701B (en) | System and method for online flaw detection of industrial X-ray machine | |
| CN108009515B (en) | Power transmission line positioning and identifying method of unmanned aerial vehicle aerial image based on FCN | |
| CN114863311A (en) | Automatic tracking method and system for inspection target of transformer substation robot | |
| CN102915432B (en) | A kind of vehicle-mounted microcomputer image/video data extraction method and device | |
| CN111754498A (en) | Conveyor belt carrier roller detection method based on YOLOv3 | |
| CN110443212B (en) | Positive sample acquisition method, device, equipment and storage medium for target detection | |
| CN111695373B (en) | Zebra stripes positioning method, system, medium and equipment | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| CN106815576B (en) | Target tracking method based on continuous space-time confidence map and semi-supervised extreme learning machine | |
| CN113052295B (en) | Training method of neural network, object detection method, device and equipment | |
| CN115272204A (en) | Bearing surface scratch detection method based on machine vision | |
| CN115439458A (en) | Industrial image defect target detection algorithm based on depth map attention | |
| CN110610123A (en) | Multi-target vehicle detection method and device, electronic equipment and storage medium | |
| CN114781514A (en) | Floater target detection method and system integrating attention mechanism | |
| CN112017172A (en) | System and method for detecting defects of deep learning product based on raspberry group | |
| CN112419261A (en) | Visual acquisition method and device with abnormal point removing function | |
| CN116052222A (en) | Cattle face recognition method for naturally collecting cattle face image | |
| CN116416613A (en) | Citrus fruit identification method and system based on improved YOLO v7 | |
| CN114359552A (en) | Instrument image identification method based on inspection robot | |
| CN114331961A (en) | Method for defect detection of an object | |
| CN112784675B (en) | Target detection method and device, storage medium and terminal | |
| CN115984780B (en) | Industrial solid waste warehouse-in and warehouse-out judging method and device, electronic equipment and medium | |
| CN113723841B (en) | On-line detection method for tool missing in assembled prefabricated part |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |