CN112749656B - ORB feature matching and yolo-based air switch state detection method and device - Google Patents
ORB feature matching and yolo-based air switch state detection method and device Download PDFInfo
- Publication number
- CN112749656B CN112749656B CN202110017776.4A CN202110017776A CN112749656B CN 112749656 B CN112749656 B CN 112749656B CN 202110017776 A CN202110017776 A CN 202110017776A CN 112749656 B CN112749656 B CN 112749656B
- Authority
- CN
- China
- Prior art keywords
- image
- air switch
- yolo
- switch
- data set
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- 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)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Manipulator (AREA)
Abstract
The invention provides an ORB feature matching and yolo-based air switch state detection method, which comprises the following steps: A. shooting a device image sample containing an air switch to establish a dataset; B. training and exporting a yolo model; C. manufacturing a standard template; D. shooting an equipment image containing an air switch to be detected, and extracting ORB characteristic points of the equipment image and generating corresponding characteristic vectors; E. performing feature matching to locate an air switch to be detected, and correcting the equipment image in the step D; F. obtaining a switch image to be detected which is basically consistent with a standard template; G. b, inputting the switch image to be detected into the yolo model processed in the step B, and obtaining the number of the switches and the state of each switch; F. judging whether the air switch state is abnormal or not, and alarming. The invention has strong anti-interference performance, low false detection rate and high accuracy, and can reduce the workload.
Description
Technical Field
The invention relates to an ORB feature matching and yolo-based air switch state detection method and device.
Background
The stability of the power communication system directly affects the reliability and stability of the operation of the power grid, so that the maintenance of the power communication equipment is important, and the air switch is a very important electric appliance, and integrates control and various protection functions. The air switch not only can complete contact and breaking of a circuit, but also can protect short circuits, serious overload, undervoltage and the like of the circuit or electrical equipment, so that detection and maintenance of the air switch are indispensable links of an electric power communication machine room. At present, the detection mode of manual detection or fixed camera monitoring is generally adopted for the detection of the equipment in the electric power communication machine room, and the two modes have defects: the manual detection mode is easy to cause false detection, false detection or missing detection due to subjective factors of personnel; the detection mode of fixed camera control has limited detection range, when the equipment to be detected is increased, the detection cost is increased, and the detection mode of fixed camera control mostly adopts a template matching algorithm to detect the air switch state, and the method is easily influenced by external environment and has poor robustness.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides an ORB feature matching and yolo-based air switch state detection method and device, which have the advantages of strong interference resistance, low false detection rate, high accuracy and capability of reducing workload.
The invention is realized by the following technical scheme:
the air switch state detection method based on ORB feature matching and yolo comprises the following steps:
A. controlling the inspection robot to shoot equipment image samples containing the air switch at a specified place in a specified gesture to establish a data set, and recording the position and gesture when shooting each image sample;
B. training and deriving a yolo model according to the equipment image sample of the data set;
C. manufacturing standard templates corresponding to all the air switches according to the equipment image samples of the data set, and extracting feature points of all the standard templates and generating corresponding feature vectors;
D. c, controlling the inspection robot to shoot an equipment image containing the air switch to be detected according to the position and the gesture recorded in the step A, and extracting ORB characteristic points of the equipment image and generating corresponding characteristic vectors;
E. carrying out RANSAC feature matching on the feature points and the feature vectors in the step D and the feature points and the feature vectors of the corresponding standard templates in the step C respectively so as to locate an air switch to be detected in the equipment image, and acquiring a perspective transformation matrix so as to correct the equipment image in the step D;
F. c, according to the feature matching result, performing ROI buckling on the equipment image obtained in the step E to obtain a switch image to be detected, wherein the switch image to be detected is basically consistent with the standard template;
G. b, inputting the switch image to be detected into the yolo model processed in the step B, obtaining the position and state of each switch, and determining the number of air switches according to the number of the positions;
H. and C, comparing the number and the state of the air switches obtained in the step G with the power company management background data, judging whether the state of the air switches is abnormal or not, and if so, giving an alarm if the state of the air switches is abnormal or if the air switches are missed.
Further, the step a further includes: after image processing operations of mirror image inversion, translation transformation, affine transformation, random noise addition and random brightness change are performed on the device image samples, a data set is built.
Further, the step B includes:
b1, marking the air switch area in each equipment image sample in the data set by a rectangular frame, wherein the data marking records the position (x, y, w, h) of the air switch area in the image and the state (on, off) of the air switch, wherein (x, y) represents the upper left corner coordinate of the image, and w and h represent the width and height of the air switch area respectively;
b2, dividing the data set processed in the step B1 into a training set and a testing set;
and B3, setting yolo model parameters, training by using a training set, and exporting the yolo model when the accuracy of the testing set reaches 99%.
Further, the step C includes:
c1, respectively buckling each air switch in each equipment image sample in the data set to serve as a standard template;
c2, extracting OEB characteristic points of each standard template respectively, and generating corresponding characteristic vectors according to the characteristic points;
and C3, numbering and recording each standard template according to the position information corresponding to the standard template.
Further, in the step E, a corresponding standard template is selected according to the position in the step D.
Further, step H further includes: when the air state of the switch is abnormal, background alarming is carried out; and when the detection omission exists, a verification signal is sent to a worker.
Further, the invention is realized by the following technical scheme:
ORB feature matching and yolo-based air switch state detection device comprises:
a data set establishing module: for controlling the inspection robot to take the image samples of the equipment containing the air switch at the designated location in the designated pose to create a dataset and to record the position and pose of each image sample taken;
and a pretreatment module: training and deriving a yolo model from the device image samples of the dataset; the standard templates corresponding to the air switches are manufactured according to the equipment image samples of the data set, and feature point extraction and corresponding feature vector generation are carried out on the standard templates;
the feature point acquisition module is used for: the device image processing device is used for controlling the inspection robot to shoot a device image containing the air switch to be detected according to the recorded position and posture, extracting characteristic points of the device image and generating corresponding characteristic vectors;
and a feature matching module: the method comprises the steps of performing RANSAC feature matching on feature points and feature vectors obtained by a feature point obtaining module and feature points and feature vectors of a corresponding standard template respectively to locate an air switch to be detected in an equipment image, obtaining a perspective transformation matrix to correct the equipment image, and performing ROI buckling on the corrected equipment image according to a feature matching result to obtain an image of the switch to be detected, wherein the image is basically consistent with the standard template;
and an identification and alarm module: the method comprises the steps of inputting a switch image to be detected into a pre-processed yolo model to obtain the position and state of each switch, and determining the number of air switches according to the number of the positions; and comparing the number and the state of the switches with the power company management background data, judging whether the air switch state is abnormal or not, and if the air switch state is missed, alarming if the air switch state is abnormal or missed.
Furthermore, the data set establishment module establishes the data set after performing image processing operations of mirror image inversion, translation transformation, affine transformation, random noise addition and random brightness change on the device image sample before establishing the data set.
Further, an industrial camera capable of moving up and down is arranged on the inspection robot.
Further, the feature point obtaining module specifically includes:
and (5) buckling and taking the module: the air switches are used for respectively buckling all the equipment image samples in the data set and serve as standard templates;
and an extraction module: the method comprises the steps of extracting OEB characteristic points of standard templates respectively, and generating corresponding characteristic vectors according to the characteristic points;
numbering module: and numbering and recording the standard templates according to the position information corresponding to the standard templates.
The invention has the following beneficial effects:
1. according to the method, the ORB characteristic matching and the yolo model are combined to detect the states of the air switches on the power equipment, the states of the air switches can be rapidly and accurately judged, in practical application, a plurality of groups of air switches are usually arranged on one equipment panel, if the state identification is directly carried out by using the yolo model, the relative positions of the air switches cannot be determined due to the movement error of the inspection robot, whether the detected target is the target required to be detected or not cannot be judged, and false detection is extremely easy to occur, therefore, before the yolo model is used, the air switch groups required to be detected at this time are positioned through ORB characteristic matching, the images are corrected through perspective transformation, the air switch groups required to be detected are basically consistent with the standard templates, and then the yolo model is used, so that the interference of factors such as distance, angle and illumination caused by the movement error of the robot can be avoided, the false detection rate is reduced, the accuracy can reach 98.8%, the detection accuracy of the yolo model is about 94% when the method is used, and the accuracy of detection of the yolo model is lower when the template is used singly; furthermore, if ORB feature matching is not used first, if interference of factors such as distance, angle and the like is avoided, more data are needed to train the yolo model, the preparation of training data is not small in workload, satisfactory effects cannot be ensured, and if feature matching is performed first, interference factors are not needed to be considered, data needed for training are greatly reduced, and workload is effectively reduced.
Drawings
The invention is described in further detail below with reference to the accompanying drawings.
FIG. 1 is a flow chart of the present invention.
Fig. 2 is a standard template corresponding to an air switch to be detected.
Fig. 3 is an image of an apparatus including an air switch to be detected.
Fig. 4 is a feature point matching and perspective transformation process image.
Fig. 5 is an image of the corrected air switch to be detected after the buckle is removed.
Detailed Description
As shown in fig. 1, the method for detecting the air switch state based on ORB feature matching and yolo comprises the following steps:
A. controlling the inspection robot to shoot equipment image samples containing the air switch at a specified place in a specified gesture, performing image processing operations of mirror image overturning, translation transformation, affine transformation, random noise adding and random brightness change on the equipment image samples, establishing a data set, and recording the position and gesture when shooting each image sample; an industrial camera capable of moving up and down is arranged on the inspection robot, so that the gesture when an image sample is shot refers to the specific position of the industrial camera on the inspection robot; the inspection robot structure, the industrial camera setting structure, the inspection robot control and the specific operation of the image processing are all in the prior art;
B. training and deriving a yolo model (in this embodiment, a yolo3 model is specifically performed) according to the device image samples of the data set; the method specifically comprises the following steps:
b1, marking the air switch area in each equipment image sample in the data set by a rectangular frame, wherein the data marking records the position (x, y, w, h) of the air switch area in the image and the state (on, off) of the air switch, wherein (x, y) represents the upper left corner coordinate of the image, and w and h represent the width and height of the air switch area respectively;
b2, randomly dividing the data set processed in the step B1 according to the proportion of 2:8 to form a training set and a testing set, wherein the training set is used for training a yolo3 model, and the testing set is used for testing the performance of the yolo3 model;
b3, setting yolo3 model parameters, training by using a training set, and exporting a yolo3 model when the accuracy of a test set reaches 99%;
the parameter setting is specifically as follows: scaling the picture size to 416×416×3, setting the batch training sample number batch to 64, learning rate learning_rate to 0.001, maximum iteration number to 100000, and using loss (object) as a loss function, the calculation formula is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,
indicating whether the ith grid, jth Anchor box, is responsible for this object, if so
Otherwise, 0, the loss function is the prior art;
C. as shown in fig. 2, a standard template corresponding to each air switch is manufactured according to the equipment image sample of the data set, and feature point extraction and corresponding feature vector generation are carried out on each standard template; the method specifically comprises the following steps:
c1, manually buckling each air switch in each equipment image sample in the data set respectively, and storing the air switches as a standard template;
c2, extracting OEB characteristic points of each standard template respectively, and generating corresponding characteristic vectors according to the characteristic points;
c3, numbering and recording each standard template according to the position information corresponding to the standard template;
D. c, controlling the inspection robot to shoot an equipment image containing the air switch to be detected according to the position and the gesture recorded in the step A, and extracting ORB characteristic points of the equipment image and generating corresponding characteristic vectors, as shown in figure 3;
E. as shown in fig. 4, selecting a corresponding standard template according to the position in the step D, and performing RANSAC feature matching on the feature points and the feature vectors in the step D and the feature points and the feature vectors of the corresponding standard template in the step C respectively to locate an air switch to be detected in the equipment image, and acquiring a perspective transformation matrix to correct the equipment image in the step D; although the inspection robot is controlled to shoot equipment images according to the recorded positions and postures, because the robot moves to have larger errors, factors such as shooting angles, distances, illumination and the like are complex and changeable, the shot equipment images and the equipment images in the data set also have larger errors, and therefore, characteristic point matching and perspective transformation are needed to be performed first; wherein, RANSAC feature matching and perspective transformation matrix are the prior art;
F. performing ROI buckling on the equipment image subjected to the step E according to the feature matching result to obtain a switch image to be detected, wherein the switch image to be detected is basically consistent with the standard template, as shown in FIG. 5;
G. b, inputting the switch image to be detected into the yolo model processed in the step B, obtaining the position and state of each switch, and determining the number of air switches according to the number of the positions;
H. comparing the number and the state of the air switches obtained in the step G with the power company management background data, judging whether the state of the air switches is abnormal or not, and if so, giving an alarm, wherein the alarm specifically comprises the following steps:
when the air state of the switch is abnormal, background alarming is carried out; and when the detection omission exists, a verification signal is sent to a worker.
ORB feature matching and yolo-based air switch state detection device comprises:
a data set establishing module: the method comprises the steps of controlling a patrol robot to shoot equipment image samples containing an air switch at a specified place in a specified gesture, performing image inversion, translation transformation, affine transformation, random noise addition and random brightness change image processing operation on the equipment image samples, establishing a data set for the processed equipment image samples, and recording positions and gestures when shooting the image samples;
and a pretreatment module: training and deriving a yolo model from the device image samples of the dataset; the standard templates corresponding to the air switches are manufactured according to the equipment image samples of the data set, and feature point extraction and corresponding feature vector generation are carried out on the standard templates;
the feature point acquisition module is used for: the device image processing device is used for controlling the inspection robot to shoot a device image containing the air switch to be detected according to the recorded position and posture, extracting characteristic points of the device image and generating corresponding characteristic vectors; the module specifically comprises: and (5) buckling and taking the module: the air switches are used for respectively buckling all the equipment image samples in the data set and serve as standard templates;
and an extraction module: the method comprises the steps of extracting OEB characteristic points of standard templates respectively, and generating corresponding characteristic vectors according to the characteristic points;
numbering module: and numbering and recording the standard templates according to the position information corresponding to the standard templates.
And a feature matching module: the method comprises the steps of performing RANSAC feature matching on feature points and feature vectors obtained by a feature point obtaining module and feature points and feature vectors of a corresponding standard template respectively to locate an air switch to be detected in an equipment image, obtaining a perspective transformation matrix to correct the equipment image, and performing ROI buckling on the corrected equipment image according to a feature matching result to obtain an image of the switch to be detected, wherein the image is basically consistent with the standard template;
and an identification and alarm module: the method comprises the steps of inputting a switch image to be detected into a pre-processed yolo model to obtain the position and state of each switch, and determining the number of air switches according to the number of the positions; and comparing the number and the state of the switches with the power company management background data, judging whether the air switch state is abnormal or not, and if the air switch state is missed, alarming if the air switch state is abnormal or missed.
The foregoing description is only illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the scope of the invention, i.e., the invention is not to be limited to the details of the claims and the description, but rather is to cover all modifications which are within the scope of the invention.
Claims (6)
1. The air switch state detection method based on ORB feature matching and yolo is characterized by comprising the following steps of: the method comprises the following steps:
A. controlling the inspection robot to shoot equipment image samples containing the air switch at a specified place in a specified gesture to establish a data set, and recording the position and gesture when shooting each image sample;
B. training and deriving a yolo model according to the equipment image sample of the data set;
C. manufacturing standard templates corresponding to all the air switches according to the equipment image samples of the data set, and extracting feature points of all the standard templates and generating corresponding feature vectors;
D. c, controlling the inspection robot to shoot an equipment image containing the air switch to be detected according to the position and the gesture recorded in the step A, and extracting ORB characteristic points of the equipment image and generating corresponding characteristic vectors;
E. carrying out RANSAC feature matching on the feature points and the feature vectors in the step D and the feature points and the feature vectors of the corresponding standard templates in the step C respectively so as to locate an air switch to be detected in the equipment image, and acquiring a perspective transformation matrix so as to correct the equipment image in the step D;
F. c, according to the feature matching result, performing ROI buckling on the equipment image obtained in the step E to obtain a switch image to be detected, wherein the switch image to be detected is basically consistent with the standard template;
G. b, inputting the switch image to be detected into the yolo model processed in the step B, obtaining the position and state of each switch, and determining the number of air switches according to the number of the positions;
H. comparing the number and the state of the air switches obtained in the step G with the power company management background data, judging whether the state of the air switches is abnormal or not, and if so, giving an alarm;
the step B comprises the following steps:
b1, marking the air switch area in each equipment image sample in the data set by a rectangular frame, wherein the data marking records the position x, y, w, h of the air switch area in the image and the on and off states of the air switch, wherein x and y represent the horizontal and vertical coordinates of the upper left corner of the image, and w and h represent the width and height of the air switch area respectively;
b2, dividing the data set processed in the step B1 into a training set and a testing set;
b3, setting yolo model parameters, training by using a training set, and exporting the yolo model when the accuracy of the testing set reaches 99%;
the step C comprises the following steps:
c1, respectively buckling each air switch in each equipment image sample in the data set to serve as a standard template;
c2, extracting OEB characteristic points of each standard template respectively, and generating corresponding characteristic vectors according to the characteristic points;
c3, numbering and recording each standard template according to the position information corresponding to the standard template;
in the step E, a corresponding standard template is selected according to the position in the step D.
2. The ORB feature matching and yolo based air switch state detection method of claim 1 wherein: the step A further comprises the following steps: after image processing operations of mirror image inversion, translation transformation, affine transformation, random noise addition and random brightness change are performed on the device image samples, a data set is built.
3. The ORB feature matching and yolo based air switch state detection method of claim 1 wherein: step H further comprises: when the air state of the switch is abnormal, background alarming is carried out; and when the detection omission exists, a verification signal is sent to a worker.
4. ORB feature matching and yolo-based air switch state detection device is characterized in that: comprising the following steps:
a data set establishing module: for controlling the inspection robot to take the image samples of the equipment containing the air switch at the designated location in the designated pose to create a dataset and to record the position and pose of each image sample taken;
and a pretreatment module: training and deriving a yolo model from the device image samples of the dataset; the standard templates corresponding to the air switches are manufactured according to the equipment image samples of the data set, and feature point extraction and corresponding feature vector generation are carried out on the standard templates; training and deriving a yolo model comprises the steps of marking the air switch area in each equipment image sample in a data set by a rectangular frame, and recording the position x, y, w, h of the air switch area in the image and the on and off states of the air switch by the data marking; dividing the processed data set into a training set and a testing set, setting yolo model parameters, training by using the training set, and exporting the yolo model when the accuracy of the testing set reaches 99%; the feature point acquisition module specifically comprises: and (5) buckling and taking the module: the air switches are used for respectively buckling all the equipment image samples in the data set and serve as standard templates; and an extraction module: the method comprises the steps of extracting OEB characteristic points of standard templates respectively, and generating corresponding characteristic vectors according to the characteristic points; numbering module: numbering and recording each standard template according to the position information corresponding to the standard template; wherein x and y represent the left upper corner horizontal and vertical coordinates of the image, and w and h represent the width and height of the air switch area respectively;
the feature point acquisition module is used for: the device image processing device is used for controlling the inspection robot to shoot a device image containing the air switch to be detected according to the recorded position and posture, extracting characteristic points of the device image and generating corresponding characteristic vectors;
and a feature matching module: the method comprises the steps of performing RANSAC feature matching on feature points and feature vectors obtained by a feature point obtaining module and feature points and feature vectors of corresponding standard templates respectively to locate an air switch to be detected in an equipment image, obtaining a perspective transformation matrix to correct the equipment image, and performing ROI buckling on the corrected equipment image according to a feature matching result to obtain an image of the switch to be detected, wherein the image is basically consistent with the standard templates, and the corresponding standard templates are selected according to recorded positions;
and an identification and alarm module: the method comprises the steps of inputting a switch image to be detected into a pre-processed yolo model to obtain the position and state of each switch, and determining the number of air switches according to the number of the positions; and comparing the number and the state of the switches with the power company management background data, judging whether the air switch state is abnormal or not, and if the air switch state is missed, alarming if the air switch state is abnormal or missed.
5. The ORB feature matching and yolo based air switch state detection device of claim 4 wherein: the data set establishment module establishes the data set after performing image processing operations of mirror image inversion, translation transformation, affine transformation, random noise addition and random brightness change on the equipment image sample before establishing the data set.
6. The ORB feature matching and yolo based air switch state detection device of claim 4 wherein: the inspection robot is provided with an industrial camera capable of moving up and down.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110017776.4A CN112749656B (en) | 2021-01-07 | 2021-01-07 | ORB feature matching and yolo-based air switch state detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110017776.4A CN112749656B (en) | 2021-01-07 | 2021-01-07 | ORB feature matching and yolo-based air switch state detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112749656A CN112749656A (en) | 2021-05-04 |
| CN112749656B true CN112749656B (en) | 2023-06-20 |
Family
ID=75650252
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110017776.4A Active CN112749656B (en) | 2021-01-07 | 2021-01-07 | ORB feature matching and yolo-based air switch state detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112749656B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113191350B (en) * | 2021-06-03 | 2023-04-18 | 河南科技大学 | Method and equipment for detecting state of switch knob of aircraft cockpit |
| CN113674352A (en) * | 2021-07-28 | 2021-11-19 | 浙江大华技术股份有限公司 | Switch state detection method, electronic device and storage medium |
| CN115578630B (en) * | 2022-11-09 | 2023-03-21 | 广东科凯达智能机器人有限公司 | Method and system for judging working gear of air switch |
| CN116935079B (en) * | 2023-09-07 | 2024-02-20 | 深圳金三立视频科技股份有限公司 | Linear switch state monitoring method and terminal based on vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957325A (en) * | 2010-10-14 | 2011-01-26 | 山东鲁能智能技术有限公司 | Substation equipment appearance abnormality recognition method based on substation inspection robot |
| CN110059556A (en) * | 2019-03-14 | 2019-07-26 | 天津大学 | A kind of transformer substation switch division condition detection method based on deep learning |
| CN110570392A (en) * | 2019-07-26 | 2019-12-13 | 深圳供电局有限公司 | method, device, system, equipment and medium for detecting on-off state of substation equipment |
| CN111738109A (en) * | 2020-06-09 | 2020-10-02 | 杭州古德微机器人有限公司 | Van-type cargo vehicle carriage door state identification method based on deep learning |
-
2021
- 2021-01-07 CN CN202110017776.4A patent/CN112749656B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957325A (en) * | 2010-10-14 | 2011-01-26 | 山东鲁能智能技术有限公司 | Substation equipment appearance abnormality recognition method based on substation inspection robot |
| CN110059556A (en) * | 2019-03-14 | 2019-07-26 | 天津大学 | A kind of transformer substation switch division condition detection method based on deep learning |
| CN110570392A (en) * | 2019-07-26 | 2019-12-13 | 深圳供电局有限公司 | method, device, system, equipment and medium for detecting on-off state of substation equipment |
| CN111738109A (en) * | 2020-06-09 | 2020-10-02 | 杭州古德微机器人有限公司 | Van-type cargo vehicle carriage door state identification method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 一种变电站智能巡检机器人的仪表图像空间变换算法;许志瑜;章海兵;;电工技术(第13期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112749656A (en) | 2021-05-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112749656B (en) | ORB feature matching and yolo-based air switch state detection method and device | |
| CN112766103B (en) | Machine room inspection method and device | |
| CN102837406B (en) | Mold monitoring method based on FAST-9 image characteristic rapid registration algorithm | |
| CN110555889A (en) | CALTag and point cloud information-based depth camera hand-eye calibration method | |
| CN110400315A (en) | A kind of defect inspection method, apparatus and system | |
| CN110850723A (en) | Fault diagnosis and positioning method based on transformer substation inspection robot system | |
| CN109638959B (en) | Power equipment remote signaling function debugging method and system based on AR and deep learning | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN111740699A (en) | Photovoltaic panel fault detection and identification method and device and unmanned aerial vehicle | |
| CN111144401A (en) | Touch screen control operation method for deep learning and visual servo of power plant centralized control room | |
| CN116678337A (en) | Image recognition-based bridge girder erection machine girder front and rear pivot point position height difference and girder deformation monitoring and early warning system and method | |
| CN109086643B (en) | Color box label detection method and system based on machine vision | |
| CN111967323B (en) | Electric power live working safety detection method based on deep learning algorithm | |
| CN110962128B (en) | Substation inspection and stationing method and inspection robot control method | |
| CN117119170A (en) | Camera holder testing method | |
| CN111951161A (en) | Target identification method and system and inspection robot | |
| CN116563391A (en) | Automatic laser structure calibration method based on machine vision | |
| CN115861407A (en) | Safe distance detection method and system based on deep learning | |
| CN110708519A (en) | Interactive monitoring method for industrial field environment | |
| CN112396633B (en) | Target tracking and track three-dimensional reproduction method and device based on single camera | |
| CN114266822A (en) | Workpiece quality inspection method and device based on binocular robot, robot and medium | |
| CN106846302B (en) | Detection method for correctly taking tool and examination table based on method | |
| CN111923042A (en) | Virtualization processing method and system for cabinet grid and inspection robot | |
| CN109034097B (en) | Image-based switch equipment inspection positioning method | |
| CN117435998B (en) | Fault detection method of intelligent robot |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |