CN113034501A - Grading ring inclination fault identification method and device based on key point detection - Google Patents
Grading ring inclination fault identification method and device based on key point detection Download PDFInfo
- Publication number
- CN113034501A CN113034501A CN202110574172.XA CN202110574172A CN113034501A CN 113034501 A CN113034501 A CN 113034501A CN 202110574172 A CN202110574172 A CN 202110574172A CN 113034501 A CN113034501 A CN 113034501A
- Authority
- CN
- China
- Prior art keywords
- current
- plane angle
- initial
- angle image
- key point
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Abstract
The invention discloses a method and a device for identifying an inclined fault of a grading ring based on key point detection, wherein the method comprises the following steps: acquiring pre-stored initial position information of the unmanned aerial vehicle, and controlling the unmanned aerial vehicle to fly to the initial position of the unmanned aerial vehicle; controlling an unmanned aerial vehicle to acquire images of an elevation plane angle and an overlook plane angle of the grading ring, and respectively obtaining a current elevation plane angle image and a current overlook plane angle image; inputting the current front-view plane angle image and the current overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the current front-view plane angle image and the current overlook plane angle image; establishing a coordinate system according to the current front view plane angle image and the top view plane angle image, and determining the current position coordinates of key points on the grading ring; and comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points, and identifying the inclination fault of the grading ring according to the comparison result.
Description
Technical Field
The invention relates to the technical field of power inspection, in particular to a grading ring inclination fault identification method and device based on key point detection.
Background
The grading ring is a common electrical appliance in a power transmission line and is often installed at the bottom end of an insulator. The main purpose of the grading ring is to improve the voltage distribution of the insulators in the insulator string and reduce the deterioration rate of the first insulator, and the grading ring is adopted at 110kV and above. In the line construction, because ring type aluminium system equalizer ring is comparatively soft, takes place to warp easily, in case the equalizer ring takes place the electric field distribution that the slope just very big influence composite insulator, makes the equalizer ring can not exert due effect, in addition in case the equalizer ring inclination is too big also will influence transmission line's normal work. Therefore, whether the grading ring is inclined or not needs to be detected in the process of routing inspection of the power transmission line, so that the grading ring can be always in a horizontal state.
Before unmanned aerial vehicle technique does not apply to electric power and patrols and examines, the inspection of whether slope is mostly judged through the manual work by virtue of working experience to the equalizer ring, such a judgment method often can have great artifical error, because the manual work is difficult in order to discover the less slope condition in inclination, and if put the existence of this kind of small-angle slope and increase this inclination very easily at last and influence the normal use of equalizer ring, except there being great error, the manual work still need climb on the power transmission line in the testing process, danger is higher.
With the development of unmanned aerial vehicle technology, unmanned aerial vehicles have been generally adopted for power tower routing inspection. Patent document CN11159889A discloses a method, an apparatus, and a computer device for identifying a tilt fault of a grading ring, and specifically discloses: collecting an electric power tower inspection image, identifying a grading ring area in the electric power tower inspection image, identifying a grading ring profile and an insulator bracket profile where a grading ring is located in the grading ring area through an edge detection algorithm, and calculating an inclination angle of the grading ring according to the grading ring profile and the insulator bracket profile where the grading ring is located. The inclination angle calculated in the scheme is an included angle between a first straight line where a symmetry axis of a minimum fitting graph of the contour of the grading ring is located and a second straight line where a symmetry axis long axis of the minimum fitting graph of the contour of the insulator support is located, the insulator support is used as a reference object, if the position of the insulator support deviates, the included angle cannot be accurately calculated, and misjudgment occurs.
Disclosure of Invention
The invention provides a grading ring inclination fault identification method and device based on key point detection, which can effectively improve the accuracy of grading ring inclination fault identification.
A grading ring inclination fault identification method based on key point detection comprises the following steps:
acquiring pre-stored initial position information of the unmanned aerial vehicle, and controlling the unmanned aerial vehicle to fly to the initial position of the unmanned aerial vehicle;
controlling an unmanned aerial vehicle to acquire images of an elevation plane angle and an overlook plane angle of the grading ring, and respectively obtaining a current elevation plane angle image and a current overlook plane angle image;
inputting the current front-view plane angle image and the current overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the current front-view plane angle image and the current overlook plane angle image;
establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image, and determining the current position coordinates of key points on the grading ring;
and comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points, and identifying the inclination fault of the grading ring according to the comparison result.
Further, training to obtain a key point identification model comprises:
constructing a neural network model;
and carrying out key point marking on the grading ring sample image, and training the neural network model according to the marked grading ring sample image until the model converges to obtain the key point identification model.
Further, before obtaining the initial position information of the pre-stored unmanned aerial vehicle, the method further comprises:
controlling an unmanned aerial vehicle to initially shoot the grading ring, performing image acquisition of an elevation plane angle and an overlook plane angle on the grading ring, respectively obtaining an initial elevation plane angle image and an initial overlook plane angle image, and storing current position information of the unmanned aerial vehicle as initial position information of the unmanned aerial vehicle;
inputting the initial front-view plane angle image and the initial overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the initial front-view plane angle image and the initial overlook plane angle image;
establishing a coordinate system according to the initial front view plane angle image and the initial top view plane angle image, and determining the initial position coordinates of the key points on the grading ring;
and storing the initial position coordinates of the key points.
Further, establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image, and determining the current position coordinates of the key points on the grading ring, including:
establishing an XY coordinate system on the current elevation plane angle image;
establishing a ZX coordinate system on the current overlook plane angle image;
combining the XY coordinate system and the ZX coordinate system to form an XYZ coordinate system;
and determining the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point in the XYZ coordinate system.
Further, the initial position coordinates of the key points comprise initial X-axis coordinates, initial Y-axis coordinates and initial Z-axis coordinates;
comparing the current position coordinates of the key points on the grading ring with the initial position coordinates of the key points stored in advance, and identifying the inclined faults of the grading ring according to the comparison result, wherein the method comprises the following steps:
and comparing the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point with the initial X-axis coordinate, the initial Y-axis coordinate and the initial Z-axis coordinate respectively, and determining the offset direction and the offset of the key point according to the comparison result.
Further, determining the shift direction of the key point according to the comparison result includes:
if the current X-axis coordinate of the key point is larger than the initial X-axis coordinate, determining that the key point deviates to the positive direction of the X axis; if the current X-axis coordinate of the key point is smaller than the initial X-axis coordinate, determining that the key point deviates to the opposite direction of the X axis;
if the current Y-axis coordinate of the key point is larger than the initial Y-axis coordinate, determining that the key point deviates to the positive direction of the Y axis; if the current Y-axis coordinate of the key point is smaller than the initial Y-axis coordinate, determining that the key point deviates to the opposite direction of the Y axis;
if the current Z-axis coordinate of the key point is larger than the initial Z-axis coordinate, determining that the key point deviates to the positive direction of the Z axis; and if the current Z-axis coordinate of the key point is smaller than the initial Z-axis coordinate, determining that the key point deviates to the opposite direction of the Z axis.
Further, determining the offset of the key point according to the comparison result includes:
determining the difference value of the current X-axis coordinate and the initial X-axis coordinate as the offset of the key point in the X-axis direction;
determining the difference value of the current Y-axis coordinate and the initial Y-axis coordinate as the offset of the key point in the Y-axis direction;
and determining the difference value of the current Z-axis coordinate and the initial Z-axis coordinate as the offset of the key point in the Z-axis direction.
Further, the position of the key point comprises at least one of the orientations of true east, true west, true south, true north, south east, south west, north east and north west on the grading ring.
The utility model provides an equalizer ring slope fault recognition device based on key point detects, includes:
the unmanned aerial vehicle control module is used for acquiring pre-stored unmanned aerial vehicle initial position information and controlling the unmanned aerial vehicle to fly to the unmanned aerial vehicle initial position;
the image acquisition control module is used for controlling the unmanned aerial vehicle to acquire an elevation plane angle image and an overlook plane angle image of the grading ring, and respectively acquiring a current elevation plane angle image and a current overlook plane angle image;
the key point identification module is used for inputting the current front view plane angle image and the current overlook plane angle image into a key point identification model obtained by pre-training and identifying key points of an equalizing ring on the current front view plane angle image and the current overlook plane angle image;
the coordinate determination module is used for establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image and determining the current position coordinates of the key points on the grading ring;
and the inclination fault identification module is used for comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points and identifying the inclination faults of the grading ring according to the comparison result.
An electronic device comprises a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is used for reading the instructions and executing the method.
The grading ring inclination fault identification method and device based on key point detection provided by the invention at least have the following beneficial effects:
(1) the unmanned aerial vehicle collects images to identify the inclined faults of the grading rings to replace manual operation, so that the identification accuracy is improved, and the safety problem caused by manual operation is avoided;
(2) the key point recognition of the grading ring is carried out through the key point recognition model, the offset and the offset direction of the grading ring are obtained through the position of the key point, the accuracy is higher, the problem of misjudgment in the prior art is solved, the recognition efficiency is high, the small inclined offset can be recognized, and the recognition precision is high.
Drawings
Fig. 1 is a flowchart of an embodiment of a method for identifying a slanting fault of a grading ring based on key point detection according to the present invention.
Fig. 2 is a schematic structural diagram of an embodiment of the grading ring inclination fault identification device based on key point detection provided by the invention.
Detailed Description
In order to better understand the technical solution, the technical solution will be described in detail with reference to the drawings and the specific embodiments.
Example one
Referring to fig. 1, the present embodiment provides a method for identifying an inclined fault of a grading ring based on key point detection, including:
s1, acquiring pre-stored initial position information of the unmanned aerial vehicle, and controlling the unmanned aerial vehicle to fly to the initial position of the unmanned aerial vehicle;
s2, controlling the unmanned aerial vehicle to acquire images of the front view plane angle and the top view plane angle of the grading ring, and respectively acquiring a current front view plane angle image and a current top view plane angle image;
s3, inputting the current front view plane angle image and the current overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the current front view plane angle image and the current overlook plane angle image;
s4, establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image, and determining the current position coordinates of the key points on the grading ring;
and S5, comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points, and identifying the inclination fault of the grading ring according to the comparison result.
Further, before executing step S1, the method needs to be trained in advance to obtain a keypoint identification model, which specifically includes:
constructing a neural network model;
and carrying out key point marking on the grading ring sample image, and training the neural network model according to the marked grading ring sample image until the model converges to obtain the key point identification model.
As a preferred embodiment, the key point may be selected in multiple numbers, for example, the position of the selected key point includes at least one of the true east, true west, true south, true north, east south, west south, east north, west north.
Further, before step S1 is executed, the initial position coordinates of the grading ring need to be calibrated, which specifically includes:
controlling an unmanned aerial vehicle to initially shoot the grading ring, performing image acquisition of an elevation plane angle and an overlook plane angle on the grading ring, respectively obtaining an initial elevation plane angle image and an initial overlook plane angle image, and storing current position information of the unmanned aerial vehicle as initial position information of the unmanned aerial vehicle;
inputting the initial front-view plane angle image and the initial overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the initial front-view plane angle image and the initial overlook plane angle image;
establishing a coordinate system according to the initial front view plane angle image and the initial top view plane angle image, which specifically comprises the following steps: and establishing an XY coordinate system on the initial front view plane angle image, establishing a ZX coordinate system on the initial top view plane angle image, and combining the XY coordinate system and the ZX coordinate system to form an XYZ coordinate system.
After an XYZ coordinate system is established, the initial position coordinates of the key points on the grading ring can be determined, including the initial X-axis coordinate, the initial Y-axis coordinate and the initial Z-axis coordinate, and the initial position coordinates of the key points are stored.
Further, in step S1, the initial position of the drone is the position of the drone during initial shooting, and it is ensured that the positions of the drone during subsequent shooting and initial shooting are the same, thereby ensuring that the established coordinate system is the same.
Further, in step S2, the drone is controlled to perform photographing at the front view plane angle, that is, photographing at the XY plane angle, and photographing at the top view plane angle, that is, photographing at the ZX plane angle.
Further, in step S4, establishing a coordinate system according to the current elevation plane angle image and the current overhead plane angle image, and determining the current position coordinates of the key points on the grading ring, includes:
establishing an XY coordinate system on the current front view plane angle image, and establishing a ZX coordinate system on the current top view plane angle image; combining the XY coordinate system and the ZX coordinate system to form an XYZ coordinate system; and determining the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point in the XYZ coordinate system.
Further, in step S5, the initial position coordinates of the key points include an initial X-axis coordinate, an initial Y-axis coordinate, and an initial Z-axis coordinate;
comparing the current position coordinates of the key points on the grading ring with the initial position coordinates of the key points stored in advance, and identifying the inclined faults of the grading ring according to the comparison result, wherein the method comprises the following steps:
and comparing the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point with the initial X-axis coordinate, the initial Y-axis coordinate and the initial Z-axis coordinate respectively, and determining the offset direction and the offset of the key point according to the comparison result.
Specifically, if the current X-axis coordinate of the key point is greater than the initial X-axis coordinate, the key point is shifted to the positive direction of the X-axis, otherwise, if the current X-axis coordinate of the key point is less than the initial X-axis coordinate, the key point is shifted to the negative direction of the X-axis, and the difference between the current X-axis coordinate and the initial X-axis coordinate is the shift amount of the key point in the X-axis direction; if the current Y-axis coordinate of the key point is larger than the initial Y-axis coordinate, the key point shifts to the positive direction of the Y-axis, otherwise, if the current Y-axis coordinate of the key point is smaller than the initial Y-axis coordinate, the key point shifts to the negative direction of the Y-axis, and the difference value between the current Y-axis coordinate and the initial Y-axis coordinate is the shift amount of the key point in the Y-axis direction; if the current Z-axis coordinate of the key point is larger than the initial Z-axis coordinate, the key point shifts to the positive direction of the Z axis, otherwise, if the current Z-axis coordinate of the key point is smaller than the initial Z-axis coordinate, the key point shifts to the negative direction of the Z axis, and the difference value between the current Z-axis coordinate and the initial Z-axis coordinate is the shift amount of the key point in the Z-axis direction. The deviation directions and the deviations of the key points in the X-axis direction, the Y-axis direction and the Z-axis direction are obtained, so that the inclination condition of the grading ring can be obtained, and in order to ensure the identification accuracy, comprehensive judgment can be carried out according to the deviation conditions of the key points.
Wherein, the offset X of the key point in the X-axis direction is assumed1Shift of key point in Y-axis direction Y1Amount of shift Z of key point in Z-axis direction1The offset A of the key point in the XYZ coordinate system can be obtained correspondingly according to the offset of the key point on each coordinate axis, and the calculation formula is as follows:
in addition, the offset vector of each coordinate axis is established by taking the initial coordinate of each coordinate axis as a starting point and the coordinate of the key point pointing to each coordinate axis as an end point, and then the offset vector of the key point in the X-axis direction isThe offset vector of the key point in the Y-axis direction isThe offset vector of the key point in the Z-axis direction isThe offset vector of the key point in the XYZ coordinate system can be obtained correspondingly according to the offset vector of the key point on each coordinate axisThe direction of the offset vector is the offset direction of the key point.
The grading ring inclination fault identification method based on key point detection provided by the embodiment at least has the following beneficial effects:
(1) the unmanned aerial vehicle collects images to identify the inclined faults of the grading rings to replace manual operation, so that the identification accuracy is improved, and the safety problem caused by manual operation is avoided;
(2) the key point recognition of the grading ring is carried out through the key point recognition model, the offset and the offset direction of the grading ring are obtained through the position of the key point, the accuracy is higher, the problem of misjudgment in the prior art is solved, the recognition efficiency is high, the small inclined offset can be recognized, and the recognition precision is high.
Example two
Referring to fig. 2, the present embodiment provides an equalizing ring inclination fault identifying device based on key point detection, including:
the unmanned aerial vehicle control module 201 is used for acquiring pre-stored unmanned aerial vehicle initial position information and controlling the unmanned aerial vehicle to fly to the unmanned aerial vehicle initial position;
the image acquisition control module 202 is used for controlling the unmanned aerial vehicle to acquire an elevation plane angle image and an overlook plane angle image of the grading ring, and respectively acquiring a current elevation plane angle image and a current overlook plane angle image;
the key point identification module 203 is configured to input the current front view plane angle image and the current top view plane angle image to a key point identification model obtained through pre-training, and identify key points of a grading ring on the current front view plane angle image and the current top view plane angle image;
the coordinate determination module 204 is configured to establish a coordinate system according to the current elevation plane angle image and the current overlook plane angle image, and determine current position coordinates of key points on the grading ring;
and the inclination fault identification module 205 is configured to compare the current position coordinates of the key points on the grading ring with pre-stored initial position coordinates of the key points, and identify an inclination fault of the grading ring according to a comparison result.
Further, a model building module 206 is included for building a neural network model; and carrying out key point marking on the grading ring sample image, and training the neural network model according to the marked grading ring sample image until the model converges to obtain the key point identification model.
Further, the system also comprises an initial control module 207, which is used for controlling the unmanned aerial vehicle to initially shoot the grading ring, performing image acquisition of an elevation plane angle and an overlook plane angle on the grading ring, respectively obtaining an initial elevation plane angle image and an initial overlook plane angle image, and storing the current position information of the unmanned aerial vehicle as initial position information; inputting the initial front-view plane angle image and the initial overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the initial front-view plane angle image and the initial overlook plane angle image; establishing a coordinate system according to the initial front view plane angle image and the initial top view plane angle image, and determining the initial position coordinates of the key points on the grading ring; and storing the initial position coordinates of the key points.
Further, the coordinate determination module 204 is further configured to establish an XY coordinate system on the current elevation plane angle image; establishing a ZX coordinate system on the current overlook plane angle image; combining the XY coordinate system and the ZX coordinate system to form an XYZ coordinate system; and determining the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point in the XYZ coordinate system.
Further, the initial position coordinates of the key points comprise initial X-axis coordinates, initial Y-axis coordinates and initial Z-axis coordinates;
the tilt fault identifying module 205 is further configured to compare the current X-axis coordinate, the current Y-axis coordinate, and the current Z-axis coordinate of the key point with the initial X-axis coordinate, the initial Y-axis coordinate, and the initial Z-axis coordinate, respectively, and determine an offset direction and an offset amount of the key point according to a comparison result; specifically, if the current X-axis coordinate of the key point is larger than the initial X-axis coordinate, determining that the key point deviates to the positive direction of the X axis; if the current X-axis coordinate of the key point is smaller than the initial X-axis coordinate, determining that the key point deviates to the opposite direction of the X axis; determining the difference value of the current X-axis coordinate and the initial X-axis coordinate as the offset of the key point in the X-axis direction; if the current Y-axis coordinate of the key point is larger than the initial Y-axis coordinate, determining that the key point deviates to the positive direction of the Y axis; if the current Y-axis coordinate of the key point is smaller than the initial Y-axis coordinate, determining that the key point deviates to the opposite direction of the Y axis; determining the difference value of the current Y-axis coordinate and the initial Y-axis coordinate as the offset of the key point in the Y-axis direction; if the current Z-axis coordinate of the key point is larger than the initial Z-axis coordinate, determining that the key point deviates to the positive direction of the Z axis; if the current Z-axis coordinate of the key point is smaller than the initial Z-axis coordinate, determining that the key point deviates to the opposite direction of the Z axis; and determining the difference value of the current Z-axis coordinate and the initial Z-axis coordinate as the offset of the key point in the Z-axis direction.
Further, the position of the key point comprises at least one of the orientations of true east, true west, true south, true north, south east, south west, north east and north west on the grading ring.
The grading ring inclination fault identification method based on key point detection provided by the embodiment at least has the following beneficial effects:
(1) the unmanned aerial vehicle collects images to identify the inclined faults of the grading rings to replace manual operation, so that the identification accuracy is improved, and the safety problem caused by manual operation is avoided;
(2) the key point recognition of the grading ring is carried out through the key point recognition model, the offset and the offset direction of the grading ring are obtained through the position of the key point, the accuracy is higher, the problem of misjudgment in the prior art is solved, the recognition efficiency is high, the small inclined offset can be recognized, and the recognition precision is high.
The embodiment also provides an electronic device, which includes a processor and a memory, where the memory stores a plurality of instructions, and the processor is configured to read the instructions and execute the method according to the first embodiment.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention. It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A grading ring inclination fault identification method based on key point detection is characterized by comprising the following steps:
acquiring pre-stored initial position information of the unmanned aerial vehicle, and controlling the unmanned aerial vehicle to fly to the initial position of the unmanned aerial vehicle;
controlling an unmanned aerial vehicle to acquire images of an elevation plane angle and an overlook plane angle of the grading ring, and respectively obtaining a current elevation plane angle image and a current overlook plane angle image;
inputting the current front-view plane angle image and the current overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the current front-view plane angle image and the current overlook plane angle image;
establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image, and determining the current position coordinates of key points on the grading ring;
and comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points, and identifying the inclination fault of the grading ring according to the comparison result.
2. The method of claim 1, wherein training the derived keypoint recognition model comprises:
constructing a neural network model;
and carrying out key point marking on the grading ring sample image, and training the neural network model according to the marked grading ring sample image until the model converges to obtain the key point identification model.
3. The method of claim 1, wherein before obtaining the pre-stored initial position information of the drone, the method further comprises:
controlling an unmanned aerial vehicle to initially shoot the grading ring, performing image acquisition of an elevation plane angle and an overlook plane angle on the grading ring, respectively obtaining an initial elevation plane angle image and an initial overlook plane angle image, and storing current position information of the unmanned aerial vehicle as initial position information of the unmanned aerial vehicle;
inputting the initial front-view plane angle image and the initial overlook plane angle image into a key point recognition model obtained by pre-training, and recognizing key points of a grading ring on the initial front-view plane angle image and the initial overlook plane angle image;
establishing a coordinate system according to the initial front view plane angle image and the initial top view plane angle image, and determining the initial position coordinates of the key points on the grading ring;
and storing the initial position coordinates of the key points.
4. The method according to claim 1, wherein establishing a coordinate system according to the current elevation plane angle image and the current top view plane angle image, and determining current position coordinates of key points on the grading ring comprises:
establishing an XY coordinate system on the current elevation plane angle image;
establishing a ZX coordinate system on the current overlook plane angle image;
combining the XY coordinate system and the ZX coordinate system to form an XYZ coordinate system;
and determining the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point in the XYZ coordinate system.
5. The method of claim 4, wherein the keypoint initial position coordinates comprise an initial X-axis coordinate, an initial Y-axis coordinate, and an initial Z-axis coordinate;
comparing the current position coordinates of the key points on the grading ring with the initial position coordinates of the key points stored in advance, and identifying the inclined faults of the grading ring according to the comparison result, wherein the method comprises the following steps:
and comparing the current X-axis coordinate, the current Y-axis coordinate and the current Z-axis coordinate of the key point with the initial X-axis coordinate, the initial Y-axis coordinate and the initial Z-axis coordinate respectively, and determining the offset direction and the offset of the key point according to the comparison result.
6. The method of claim 5, wherein determining the direction of shift of the keypoints from the comparison comprises:
if the current X-axis coordinate of the key point is larger than the initial X-axis coordinate, determining that the key point deviates to the positive direction of the X axis; if the current X-axis coordinate of the key point is smaller than the initial X-axis coordinate, determining that the key point deviates to the opposite direction of the X axis;
if the current Y-axis coordinate of the key point is larger than the initial Y-axis coordinate, determining that the key point deviates to the positive direction of the Y axis; if the current Y-axis coordinate of the key point is smaller than the initial Y-axis coordinate, determining that the key point deviates to the opposite direction of the Y axis;
if the current Z-axis coordinate of the key point is larger than the initial Z-axis coordinate, determining that the key point deviates to the positive direction of the Z axis; and if the current Z-axis coordinate of the key point is smaller than the initial Z-axis coordinate, determining that the key point deviates to the opposite direction of the Z axis.
7. The method of claim 5, wherein determining the amount of keypoint shift based on the comparison comprises:
determining the difference value of the current X-axis coordinate and the initial X-axis coordinate as the offset of the key point in the X-axis direction;
determining the difference value of the current Y-axis coordinate and the initial Y-axis coordinate as the offset of the key point in the Y-axis direction;
and determining the difference value of the current Z-axis coordinate and the initial Z-axis coordinate as the offset of the key point in the Z-axis direction.
8. The method of any of claims 1-7, wherein the location of the keypoint comprises at least one of a true east, a true west, a true south, a true north, a south east, a south west, a north east, and a north west orientation on the grading ring.
9. The utility model provides an equalizer ring slope fault recognition device based on key point detects which characterized in that includes:
the unmanned aerial vehicle control module is used for acquiring pre-stored unmanned aerial vehicle initial position information and controlling the unmanned aerial vehicle to fly to the unmanned aerial vehicle initial position;
the image acquisition control module is used for controlling the unmanned aerial vehicle to acquire an elevation plane angle image and an overlook plane angle image of the grading ring, and respectively acquiring a current elevation plane angle image and a current overlook plane angle image;
the key point identification module is used for inputting the current front view plane angle image and the current overlook plane angle image into a key point identification model obtained by pre-training and identifying key points of an equalizing ring on the current front view plane angle image and the current overlook plane angle image;
the coordinate determination module is used for establishing a coordinate system according to the current front view plane angle image and the current top view plane angle image and determining the current position coordinates of the key points on the grading ring;
and the inclination fault identification module is used for comparing the current position coordinates of the key points on the grading ring with the prestored initial position coordinates of the key points and identifying the inclination faults of the grading ring according to the comparison result.
10. An electronic device comprising a processor and a memory, the memory storing a plurality of instructions, the processor configured to read the instructions and perform the method of any of claims 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110574172.XA CN113034501B (en) | 2021-05-25 | 2021-05-25 | Grading ring inclination fault identification method and device based on key point detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110574172.XA CN113034501B (en) | 2021-05-25 | 2021-05-25 | Grading ring inclination fault identification method and device based on key point detection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113034501A true CN113034501A (en) | 2021-06-25 |
CN113034501B CN113034501B (en) | 2021-10-15 |
Family
ID=76456199
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110574172.XA Active CN113034501B (en) | 2021-05-25 | 2021-05-25 | Grading ring inclination fault identification method and device based on key point detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113034501B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107729808A (en) * | 2017-09-08 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | A kind of image intelligent acquisition system and method for power transmission line unmanned machine inspection |
US20180357788A1 (en) * | 2016-08-11 | 2018-12-13 | Changzhou Campus of Hohai University | UAV Inspection Method for Power Line Based on Human Visual System |
CN111598889A (en) * | 2020-05-26 | 2020-08-28 | 南方电网数字电网研究院有限公司 | Grading ring inclination fault identification method and device and computer equipment |
CN112396582A (en) * | 2020-11-16 | 2021-02-23 | 南京工程学院 | Mask RCNN-based equalizing ring skew detection method |
-
2021
- 2021-05-25 CN CN202110574172.XA patent/CN113034501B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180357788A1 (en) * | 2016-08-11 | 2018-12-13 | Changzhou Campus of Hohai University | UAV Inspection Method for Power Line Based on Human Visual System |
CN107729808A (en) * | 2017-09-08 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | A kind of image intelligent acquisition system and method for power transmission line unmanned machine inspection |
CN111598889A (en) * | 2020-05-26 | 2020-08-28 | 南方电网数字电网研究院有限公司 | Grading ring inclination fault identification method and device and computer equipment |
CN112396582A (en) * | 2020-11-16 | 2021-02-23 | 南京工程学院 | Mask RCNN-based equalizing ring skew detection method |
Also Published As
Publication number | Publication date |
---|---|
CN113034501B (en) | 2021-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112199993B (en) | Method for identifying transformer substation insulator infrared image detection model in any direction based on artificial intelligence | |
CN108846418B (en) | Cable equipment temperature abnormity positioning and identifying method | |
CN108124487B (en) | Cloud meter reading method and device | |
CN110827251A (en) | Power transmission line locking pin defect detection method based on aerial image | |
CN110703800A (en) | Unmanned aerial vehicle-based intelligent identification method and system for electric power facilities | |
CN108229587B (en) | Autonomous transmission tower scanning method based on hovering state of aircraft | |
CN111220159B (en) | Path optimization method for multi-unmanned aerial vehicle cooperative inspection task | |
CN111814850A (en) | Defect detection model training method, defect detection method and related device | |
CN111428748A (en) | Infrared image insulator recognition and detection method based on HOG characteristics and SVM | |
CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
CN109813276B (en) | Base station antenna downward inclination angle measuring method and system | |
CN109345513A (en) | A kind of cigarette package defect inspection method with cigarette packet Attitude Calculation | |
CN113255590A (en) | Defect detection model training method, defect detection method, device and system | |
CN111598889A (en) | Grading ring inclination fault identification method and device and computer equipment | |
CN111539924A (en) | Defect detection method, device and equipment for suspension clamp and storage medium | |
CN110490261B (en) | Positioning method for power transmission line inspection image insulator | |
CN115359295A (en) | Decoupling knowledge distillation hardware target detection method and system | |
CN112241739A (en) | Method, device, equipment and computer readable medium for identifying text errors | |
CN113947714B (en) | Multi-mode collaborative optimization method and system for video monitoring and remote sensing | |
CN113034501B (en) | Grading ring inclination fault identification method and device based on key point detection | |
CN115908295A (en) | Power grid insulator defect detection method and system based on deep learning | |
EP4150662A1 (en) | Semiconductor overlay measurements using machine learning | |
KR20180082166A (en) | The Apparatus For Detecting Solar Panel | |
CN114022763A (en) | Foreign matter detection method and device for high-voltage overhead line and readable storage medium | |
CN113128247A (en) | Image positioning identification verification method and server |
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 | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 225300 No. 2 Fenghuang West Road, Jiangsu, Taizhou Applicant after: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. TAIZHOU POWER SUPPLY BRANCH Applicant after: ZHONGXIN HANCHUANG (BEIJING) TECHNOLOGY Co.,Ltd. Address before: 100094 room 129, 1st floor, building 5, 8 Dongbeiwang West Road, Haidian District, Beijing Applicant before: ZHONGXIN HANCHUANG (BEIJING) TECHNOLOGY Co.,Ltd. Applicant before: STATE GRID JIANGSU ELECTRIC POWER Co.,Ltd. TAIZHOU POWER SUPPLY BRANCH |
|
GR01 | Patent grant | ||
GR01 | Patent grant |