CN117168411A - Electronic fence line levelness detection device of transformer substation - Google Patents
Electronic fence line levelness detection device of transformer substation Download PDFInfo
- Publication number
- CN117168411A CN117168411A CN202311140283.5A CN202311140283A CN117168411A CN 117168411 A CN117168411 A CN 117168411A CN 202311140283 A CN202311140283 A CN 202311140283A CN 117168411 A CN117168411 A CN 117168411A
- Authority
- CN
- China
- Prior art keywords
- electronic fence
- fence line
- aerial vehicle
- unmanned aerial
- image
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 239000002184 metal Substances 0.000 claims abstract description 70
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 32
- 238000000034 method Methods 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 22
- 238000012360 testing method Methods 0.000 claims description 20
- 238000003062 neural network model Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 11
- 238000003708 edge detection Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 230000005611 electricity Effects 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000003672 processing method Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000011478 gradient descent method Methods 0.000 claims description 4
- 230000001678 irradiating effect Effects 0.000 claims description 4
- 238000005096 rolling process Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 230000000087 stabilizing effect Effects 0.000 claims description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 4
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 3
- 238000007665 sagging Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003197 gene knockdown Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a transformer substation electronic fence line levelness detection device, which belongs to the technical field of electronic fences, and comprises an unmanned aerial vehicle, a cradle head and a camera, wherein the cradle head is arranged below the middle part of the unmanned aerial vehicle and is fixedly connected with the unmanned aerial vehicle, the camera is arranged on the cradle head, a flexible metal wire which is imaged on the camera is hung on the top of a lens barrel of the camera through a heavy hammer, a memory, a controller and a wireless communicator are arranged in the unmanned aerial vehicle, program instructions are stored in the memory, and the controller executes the program instructions to realize the following steps: the electronic fence line identification step is used for identifying the electronic fence line and the image of the flexible metal wire from the shooting image obtained by the camera; the step of detecting the electronic fence line is used for judging whether the levelness of the fence line of the detected electronic fence meets the standard according to the included angle between the flexible metal wire and the electronic fence line in the image; a step of data communication; the device can be used for rapidly and accurately detecting whether the electronic fence line is straight or damaged.
Description
Technical Field
The invention belongs to the technical field of electronic fences, and particularly relates to a transformer substation electronic fence line levelness detection device.
Background
The electronic fence is used as the most advanced perimeter anti-theft alarm system, is an active intrusion anti-crossing fence, can make counterattack to intrusion attempts, can knock down an intruder, delay intrusion time, does not threaten the life of people, and can send intrusion signals to monitoring equipment so as to ensure that substation management personnel can know the condition of an alarm area in time and can quickly make processing.
At present, transformer substation electronic fence detection mainly utilizes insulating tools such as insulating rods and shovels and comprising metal heads to overlap or pull up two adjacent metal wires of the electronic fence. Such test methods can result in scribe damage to the metal wire or result in the inability of the two metal wires to recover so far that the offset is too close or too far apart, and the test speed is uneven due to repeated lifting of the lowering tool. The electronic fence for the transformer substation is mainly used for the fence with the height not lower than 2m, and is generally poor in observation field of an operator and inconvenient to operate due to the change of the terrain around the fence, so that a testing tool cannot be smoothly lapped, testing is difficult, and testing efficiency is affected. In addition, the distance between metal wires of the electronic fence is 15-20cm, the alarm outputs a single pulse with a high voltage peak value of 5-10kV, a low voltage peak value of 0.7-1kV, a current peak value of <10A, a pulse duration of less than or equal to 0.1S and a maximum output energy of less than or equal to 4J, the interval time of each pulse is 1S, a person can not accurately estimate whether the distance between the metal wires of each layer of the electronic fence meets the regulation through naked eye observation, and a common test tool can not measure whether the signal parameters of each pulse of the electronic fence meet the regulation.
Disclosure of Invention
In view of the defects of the existing electronic fence testing method, the invention provides a transformer substation electronic fence parameter testing device which can rapidly and accurately test whether an electronic fence line is straight or not and whether the distance is relatively short or not, and is easy to cause discharge interference.
The invention is realized in the following way:
the invention provides a transformer substation electronic fence line levelness detection device, which comprises an unmanned aerial vehicle, a cradle head and a camera, wherein the cradle head is arranged below the middle part of the unmanned aerial vehicle and is fixedly connected with the unmanned aerial vehicle, the camera is arranged on the cradle head, a flexible metal wire which is imaged on the camera is hung on the top of a lens barrel of the camera through a heavy hammer, a memory, a controller and a wireless communicator are arranged in the unmanned aerial vehicle, the memory, the controller and the wireless communicator are electrically connected with a power supply of the unmanned aerial vehicle, the controller is electrically connected with the memory, the wireless communicator and a control system of the cradle head and the unmanned aerial vehicle, and a program instruction is stored in the memory, and when the controller executes the program instruction, the controller is used for realizing the following steps:
the electronic fence line identification step is used for identifying the electronic fence line and the image of the flexible metal wire used for assisting calculation from the shooting image acquired by the camera;
The electronic fence line detection step is used for judging whether the levelness of the fence line of the electronic fence meets the standard or not according to the included angle between the flexible metal wire and the electronic fence line in the image based on the shooting image acquired by the camera;
and the step of data communication is used for communicating the detection result with the outside and controlling the flight command of the unmanned aerial vehicle.
The technical effect of the levelness detection device for the transformer substation electronic fence line provided by the invention is as follows: through the arrangement of the unmanned aerial vehicle, the cradle head and the camera are suspended in the air and are stabilized at fixed positions; capturing the position of the electronic fence line and the image of the flexible metal wire through the camera; a vertical line appears in a picture shot by the camera through the arrangement of the flexible metal wire, and whether the electronic fence line meets the levelness standard is judged according to the included angle at the intersection of the flexible metal wire and the electronic fence line; the weight of the flexible metal wire can be given by arranging the heavy hammer so as to ensure that the flexible metal wire is vertical to the ground.
On the basis of the technical scheme, the levelness detection device for the transformer substation electronic fence line can be further improved as follows:
the electronic fence line identification step specifically comprises the steps of utilizing a pre-trained fence line identification model to conduct electronic fence line identification on the shooting image, and marking the identified electronic fence line in the shooting image.
Further, the steps of establishing and training the fence line identification model specifically comprise:
step 1, acquiring a history fence image and preprocessing the history fence image, wherein the preprocessing comprises the following steps: denoising, gray level processing and enhancement;
step 2, taking the preprocessed historical fence image as a training sample, and dividing the training sample into a training set, a verification set and a test set, wherein the ratio of the training sample to the verification set to the test set is 7:2:1;
step 3, selecting a feedforward neural network model as a basic model, selecting an activation function, and defining a loss function and an optimization algorithm, wherein the activation function selects a ReLU activation function, the loss function selects a cross entropy loss function, and the optimization algorithm selects a random gradient descent method;
step 4, selecting an Xavier initialization mode to initialize weights and bias parameters;
and 5, training the defined feedforward neural network model by using a training sample, performing forward propagation by inputting the training sample into the defined feedforward neural network model, calculating a gradient by using backward propagation, and updating network parameters by using an optimization algorithm to obtain a trained fence line identification model.
And denoising the history fence image by adopting a median filtering method. The median filtering method is an important treatment method in the signal processing process, is an important means in the pretreatment stage, and is used for eliminating abnormal points and noise; is a nonlinear signal processing method. Under certain conditions, the problem of image blurring caused by a linear filter, such as a neighborhood averaging method, is solved, and the method has obvious effects of eliminating impulse interference and image scanning noise. No statistical properties are required for processing the image.
And then carrying out gray scale processing by adopting an average value method, and carrying out enhancement processing on the image subjected to the gray scale processing by using an edge enhancement filter so as to improve the quality, enhancement characteristics or definition of the image, wherein the commonly used edge enhancement filter comprises a MAX-MIN filter, a differential filter and the like, and the MAX-MIN filter reassigns the pixels in the grid by using the difference value between the maximum value and the minimum value of the pixels in the grid.
Further, the step of detecting the electronic fence line specifically includes:
s11, controlling the unmanned aerial vehicle to fly to a first threshold value from the electronic fence;
s12, acquiring a fence image shot by the unmanned aerial vehicle, and marking all electronic fence lines in the fence image;
s13, adjusting the vertical height of the unmanned aerial vehicle so that the flexible metal wire intersects one of the marked electronic fence lines in the fence image in the step S12;
s14, controlling the unmanned aerial vehicle to fly parallel to the electronic fence lines marked in the step S12, continuously shooting electronic fence line images to form an electronic fence line image set until the electronic fence is completely shot, wherein the height of the unmanned aerial vehicle is kept stable, and the distance between the unmanned aerial vehicle and the electronic fence lines is kept stable;
S15, analyzing an included angle between the electronic fence line and the flexible metal wire for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard;
s16, adjusting the vertical height of the unmanned aerial vehicle to enable the flexible metal wire to be intersected with the marked other electronic fence line, and executing the steps S14 and S15 to judge whether the levelness of the current electronic fence line meets the standard; and repeatedly executing the step until the levelness of each electronic fence line is detected.
The first threshold is 20 cm-100 cm. The levelness standard of the electronic fence line is as follows: keep level and steady, do not rock.
Further, the step of detecting the electronic fence line may further include:
s21, controlling the distance from the unmanned aerial vehicle to the electronic fence to be a second threshold value;
s22, acquiring a fence image shot by the unmanned aerial vehicle, and marking all electronic fence lines in the fence image;
s23, adjusting the vertical height of the unmanned aerial vehicle so that the flexible metal wires intersect all marked electronic fence lines in the fence image in the step S22;
s24, controlling the unmanned aerial vehicle to fly parallel to the electronic fence lines marked in the step S22, and continuously shooting electronic fence line images to form an electronic fence line image set until all electronic fences are shot, wherein the height of the unmanned aerial vehicle is kept stable, and the distance between the unmanned aerial vehicle and the electronic fence lines is kept stable;
S25, analyzing the included angle between each electronic fence line and the flexible metal wire for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard.
The second threshold is 200 cm-500 cm.
Further, the step of maintaining the high stability of the unmanned aerial vehicle specifically includes:
detecting position information of an object behind an electronic fence line in a picture shot by a camera by using a target detection algorithm, and marking a boundary frame of the object in a video frame;
extracting the outline of the object from each video frame by using an edge detection algorithm through the boundary frame of the object;
calculating the circle center of the minimum circumscribing circle of the outline, and marking the circle center as a first center coordinate;
step four, repeating the steps in the next video frame to obtain a second center coordinate;
and fifthly, calculating the distance between the first center coordinate and the second center coordinate, and if the distance is smaller than 5mm, considering the unmanned aerial vehicle to be highly stable.
The edge detection algorithm can adopt Canny edge detection and can also adopt segmentation algorithms such as GrabCut, mask R-CNN and the like.
Further, the step of keeping the distance between the unmanned aerial vehicle and the electronic fence line stable specifically includes:
Firstly, acquiring a continuous video frame sequence of an object after an electronic fence line by using a camera;
the second step, carrying out feature extraction on each video frame, including: size, shape, corner points, edges, texture, and sharpness of the object;
thirdly, matching the characteristics of the current video frame with the characteristics of the previous frame, and using a characteristic descriptor and a matching algorithm;
calculating an estimated value of the object distance of the object by using camera calibration parameters and a triangulation principle based on a result of feature matching;
and fifthly, judging whether the distance between the object and the electronic fence line changes or not according to the estimated distance, and if the estimated distance does not exceed 5mm, stabilizing the distance between the unmanned aerial vehicle and the electronic fence line.
Common matching algorithms include: SIFT (scale invariant feature transform), SURF (accelerated robust feature), ORB, etc.
Further, the step of analyzing the included angle between the electronic fence line and the flexible metal wire for each electronic fence line image in the electronic fence line image set specifically includes:
acquiring continuous video frames for a period of time, and calculating the mean value and variance of each pixel point to obtain a background model;
comparing each video frame with a background model, calculating the difference between the pixels of the current video frame and the background model, and for each pixel point, if the difference between the pixels of the current video frame and the background model exceeds a third threshold value obtained by an adaptive threshold value processing method, dynamically adjusting the threshold value according to the local characteristics of the image, marking the pixels of the marked foreground as a foreground, and forming a foreground image by the marked foreground pixels;
Updating the background model by adopting a rolling average method, and enabling pixel values of the video frame to participate in the updating of the background model;
filtering the background model by using a median filtering method;
detecting position information of the electronic fence line and the flexible metal wire by adopting a target detection algorithm according to the foreground image, and tracking the change of the electronic fence line and the flexible metal wire in time by using a tracking algorithm;
identifying shape boundaries of the electronic fence lines and the flexible metal wires by adopting a Canny edge detection algorithm;
based on the outlines of the electronic fence line and the flexible metal wire, two position vectors of the electronic fence line and the flexible metal wire are obtained according to the coordinates of the central points of the electronic fence line and the flexible metal wire, and an included angle of the two vectors is calculated.
The unmanned aerial vehicle is further provided with a colored light source emitter, and the colored light source emitter is used for irradiating the electronic fence line.
Through the arrangement of the colored light source, rays emitted by the colored light source can be captured by the camera, so that the electronic fence line can be accurately positioned.
Wherein, still include: the handheld voltage tester is used for receiving images transmitted by the unmanned aerial vehicle and prompting whether the electronic fence line is inclined or not.
Compared with the prior art, the levelness detection device for the transformer substation electronic fence line has the beneficial effects that: including unmanned aerial vehicle, cloud platform and camera, the cloud platform sets up in unmanned aerial vehicle middle part below and with unmanned aerial vehicle fixed connection, the camera is installed on the cloud platform, the lens-barrel top of camera hangs through the weight and has the flexible wire that appears on the camera, be provided with memory, controller, radio communication ware in the unmanned aerial vehicle, memory, controller and radio communication ware with unmanned aerial vehicle's power electricity is connected, the controller with memory radio communication ware and cloud platform with unmanned aerial vehicle's control system electricity is connected, the memory stores the program command, when the controller is carried out the program command, be used for realizing the following step: the electronic fence line identification step is used for identifying the electronic fence line and the image of the flexible metal wire used for assisting calculation from the shooting image acquired by the camera; the electronic fence line detection step is used for judging whether the levelness of the fence line of the electronic fence meets the standard or not according to the included angle between the flexible metal wire and the electronic fence line in the image based on the shooting image acquired by the camera, and whether the distance is too close or the sagging is too low, so that the hidden danger of other operations is formed; and the step of data communication is used for communicating the detection result with the outside and controlling the flight command of the unmanned aerial vehicle. Through the arrangement of the unmanned aerial vehicle, the cradle head and the camera are suspended in the air and are stabilized at fixed positions; capturing the position of the electronic fence line and the image of the flexible metal wire through the camera; a vertical line appears in a picture shot by the camera through the arrangement of the flexible metal wire, and the electronic fence line is judged to be in accordance with the levelness standard according to the included angle at the intersection of the flexible metal wire and the electronic fence line; the weight of the flexible metal wire can be given by arranging the heavy hammer so as to ensure that the flexible metal wire is vertical to the ground. The device can be used for rapidly and accurately detecting whether the electronic fence line is straight or damaged.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a transformer substation electronic fence line levelness detection device;
FIG. 2 is a flowchart of a first embodiment of a method of detecting an electronic fence line according to the present invention;
FIG. 3 is a flowchart of a second embodiment of a method of detecting an electronic fence line according to the present invention;
FIG. 4 is a schematic view of an image of a flexible wire intersecting an electronic fence line in an image captured by a camera according to a first embodiment of the present invention;
FIG. 5 is a schematic view of an image of a flexible wire intersecting an electronic fence line in an image captured by a camera according to a second embodiment of the present invention;
in the drawings, the list of components represented by the various numbers is as follows:
10. unmanned plane; 20. a cradle head; 30. a camera; 31. a lens barrel; 32. a flexible wire; 33. and a heavy hammer.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Embodiment one:
as shown in fig. 1 and 2, in the first embodiment of the transformer substation electronic fence line levelness detection device provided by the present invention, the device includes an unmanned aerial vehicle 10, a cradle head 20 and a camera 30, the cradle head 20 is disposed below the middle portion of the unmanned aerial vehicle 10 and is fixedly connected with the unmanned aerial vehicle 10, the camera 30 is mounted on the cradle head 20, a flexible wire 32 that images on the camera 30 is suspended on the top of a lens barrel 31 of the camera 30 through a weight 33, a memory, a controller, a wireless communicator are disposed in the unmanned aerial vehicle 10, the memory, the controller and the wireless communicator are electrically connected with a power supply of the unmanned aerial vehicle 10, the controller is electrically connected with the memory, the wireless communicator and a control system of the cradle head 20 and the unmanned aerial vehicle 10, program instructions are stored in the memory, and when the controller executes the program instructions, the controller is used for implementing the following steps: a step of recognizing an electronic fence line, for recognizing the electronic fence line and an image of the flexible wire 32 for assisting calculation from the photographed image acquired by the camera 30; a step of detecting the electronic fence line, which is used for judging whether the levelness of the fence line of the electronic fence meets the standard according to the included angle between the flexible metal wire 32 and the electronic fence line in the image based on the shooting image acquired by the camera 30; and a step of data communication, which is used for communicating the detection result and the flight control instruction of the unmanned aerial vehicle 10 with the outside.
Preferably, the weight 33 and the wire 32 may be fixed on the surface of the lens barrel 31, and the end of the lens barrel 31 may be made into a rotatable ring shape, so as to avoid the excessive swing of the weight 33 during the flight of the unmanned aerial vehicle. The contact between the annular component and the lens barrel can adopt a contact surface with high smoothness and be matched with the ball so as to ensure the flying, keep vertical and stable, and can quickly swing the heavy hammer 33 to restore to the vertical state after the aircraft swings.
In the above technical solution, the step of electronic fence line identification specifically includes performing electronic fence line identification on a photographed image by using a pre-trained fence line identification model, and marking the identified electronic fence line in the photographed image.
Further, in the above technical solution, the steps of establishing and training the fence line identification model specifically include:
step 1, acquiring a history fence image, preprocessing the history fence image, denoising the history fence image by adopting a median filtering method to obtain a first denoising image, and performing gray processing by adopting an average value method, wherein the calculation formula is as follows:
where Gray represents a Gray value, R represents a red value in an RGB color mode, G represents a green value in the RGB color mode, and B represents a blue value in the RGB color mode.
The first denoised image after gray processing is enhanced by an edge enhancement filter, which is commonly used as a MAX-MIN filter, a differential filter, etc., to improve the quality, enhancement characteristics or sharpness of the image. The present embodiment employs a MAX-MIN filter to reassign pixels in the grid using the difference between the maximum and minimum values of the pixels in the grid. The details of the image belong to low-frequency information, the edges of the image belong to high-frequency information, a Max-Min filter is applied to the image, and when the filter is applied to the details of the image, the output result tends to be 0 and is black; whereas the Max-Min output tends to be 255 when the filter acts on the image edges, which is white.
Step 2, taking the preprocessed history fence image as a training sample, and dividing the training sample into a training set, a verification set and a test set, wherein the ratio of the training set to the verification set to the test set is 7:2:1;
step 3, selecting a feedforward neural network model as a basic model, selecting an activation function, defining a loss function and an optimization algorithm, wherein the activation function selects a ReLU activation function, the loss function selects a cross entropy loss function, and the optimization algorithm selects a random gradient descent method;
the feedforward neural network model further includes: an input layer, three convolution layers, an activation function layer, a pooling layer, a full connection layer, and an output layer.
Step 4, selecting an Xavier initialization mode to initialize weights and bias parameters;
and step 5, training the defined feedforward neural network model by using a training sample, performing forward propagation by inputting the training sample into the defined feedforward neural network model, calculating a gradient by using backward propagation, and updating network parameters by using an optimization algorithm to obtain a trained fence line identification model.
Further, in the above technical solution, the step of detecting the electronic fence line specifically includes:
s11, controlling the unmanned aerial vehicle 10 to fly to a distance of 20 cm-100 cm from the electronic fence;
s12, acquiring a fence image shot by the unmanned aerial vehicle 10, and marking all electronic fence lines in the fence image;
s13, adjusting the vertical height of the unmanned aerial vehicle 10 so that the flexible metal wire 32 intersects one of the marked electronic fence lines in the fence image in the step S12;
s14, controlling the unmanned aerial vehicle 10 to fly parallel to the electronic fence lines marked in the step S12, continuously shooting the electronic fence line images to form an electronic fence line image set until the electronic fences are completely shot, wherein the height of the unmanned aerial vehicle 10 is kept stable, and the distance between the unmanned aerial vehicle 10 and the electronic fence lines is kept stable;
S15, analyzing the included angle between the electronic fence line and the flexible metal wire 32 for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard;
for example, an image of an electronic fence line is shown in fig. 4, wherein a vertical straight line is a flexible metal wire 32, a horizontal straight line is an electronic fence line, and an included angle between the flexible metal wire 32 and the electronic fence line is 89 ° on one side and 91 ° on the other side, so that the electronic fence line meets the levelness standard.
S16, adjusting the vertical height of the unmanned aerial vehicle 10 to enable the flexible metal wire 32 to intersect with the marked other electronic fence line, executing steps S14 and S15, and judging whether the levelness of the current electronic fence line meets the standard; and repeatedly executing the step until the levelness of each electronic fence line is detected.
Further, in the above technical solution, the step of maintaining the high stability of the unmanned aerial vehicle 10 specifically includes:
step one, detecting the position information of an object behind an electronic fence line in a picture shot by a camera 30 by using a target detection algorithm, and marking a boundary frame of the object in a video frame;
and secondly, calculating the gradient of the image by using a Canny edge detection algorithm, and screening non-maximum value inhibition, a high threshold value and a low threshold value. The high threshold is 2-3 times of the low threshold, the high threshold is selected to be the pixel with gray change larger than the high threshold and is set as the strong edge pixel, and the pixel is rejected lower than the low threshold. The setting between the high and low thresholds is a weak edge. Further judging that if the field has strong edge pixels, reserving, and if not, rejecting. The outline of the object is extracted in each video frame by the bounding box of the object.
The edge detection algorithm may also employ segmentation algorithms such as GrabCot, mask R-CNN, and the like.
Calculating the circle center of the minimum circumscribing circle of the outline, and marking the circle center as a first center coordinate;
step four, repeating the steps in the next video frame to obtain a second center coordinate;
and fifthly, calculating the distance between the first center coordinate and the second center coordinate, and if the distance is smaller than 5mm, considering the height of the unmanned aerial vehicle 10 to be stable.
Further, in the above technical solution, the step of keeping the distance between the unmanned aerial vehicle 10 and the electronic fence line stable specifically includes:
a first step of acquiring a continuous video frame sequence of the object after the electronic fence line by using the camera 30;
the second step, carrying out feature extraction on each video frame, including: size, shape, corner points, edges, texture, and sharpness of the object;
thirdly, matching the characteristics of the current video frame with the characteristics of the previous frame, and using a characteristic descriptor and a matching algorithm;
calculating an estimated value of the object distance by using camera calibration parameters and a triangulation principle based on a result of feature matching;
and fifthly, judging whether the distance between the object and the object changes according to the estimated distance, and if the estimated distance does not exceed 5mm, stabilizing the distance between the unmanned aerial vehicle 10 and the electronic fence line.
Further, in the above technical solution, the step of analyzing the included angle between the electronic fence line and the flexible wire 32 for each electronic fence line image in the electronic fence line image set specifically includes:
acquiring continuous video frames for a period of time, and calculating the mean value and variance of each pixel point to obtain a background model;
comparing each video frame with a background model, calculating the difference between the pixels of the current video frame and the background model, and for each pixel point, if the difference between the pixels of the current video frame and the background model exceeds a third threshold value obtained by an adaptive threshold value processing method, dynamically adjusting the threshold value according to the local characteristics of the image, marking the pixels of the marked foreground as a foreground, and forming a foreground image by the marked foreground pixels;
updating the background model by adopting a rolling average method, and enabling pixel values of the video frame to participate in the updating of the background model;
filtering the background model by using a median filtering method;
detecting position information of the electronic fence line and the flexible metal wire 32 by adopting a target detection algorithm according to the foreground image, and tracking the change of the electronic fence line and the flexible metal wire 32 in time by using a tracking algorithm;
identifying shape boundaries for the electronic fence lines and the flexible wires 32 by using a Canny edge detection algorithm;
Based on the outline of the electronic fence line and the flexible metal wire 32, two position vectors of the electronic fence line and the flexible metal wire 32 are obtained according to the coordinates of the central points of the electronic fence line and the flexible metal wire 32, and the included angle of the two vectors is calculated.
In the above technical scheme, the unmanned aerial vehicle 10 is further provided with a colored light source emitter, and the colored light source emitter is used for irradiating the electronic fence line and assisting in marking the electronic fence line.
Wherein, in above-mentioned technical scheme, still include: the handheld voltage tester is used for receiving the image transmitted by the unmanned aerial vehicle 10 and prompting whether the electronic fence line is inclined.
For example, the shot electronic fence line image and/or the analysis result are/is horizontal, and are transmitted to the handheld voltage tester in a wireless communication mode, such as Bluetooth, and the handheld voltage tester displays the test result to a user, such as a text form of a liquid crystal display. When the handheld voltage tester receives the image shot by the camera 30 and the calculated levelness detection result, if the electronic fence line does not meet the levelness standard, the operator is reminded by means of a liquid crystal screen or an indicator lamp.
Hand-held voltage testers are common test components with handles and display screens that detect the live condition of an electronic fence to verify whether it is in operation.
Embodiment two:
as shown in fig. 1 and 3, in a second embodiment of a transformer substation electronic fence line levelness detection device provided by the present invention, the device includes an unmanned aerial vehicle 10, a cradle head 20 and a camera 30, wherein the cradle head 20 is disposed below the middle portion of the unmanned aerial vehicle 10 and is fixedly connected with the unmanned aerial vehicle 10, the camera 30 is mounted on the cradle head 20, a flexible wire 32 that images on the camera 30 is suspended on the top of a lens barrel 31 of the camera 30 through a weight 33, a memory, a controller, a wireless communicator are disposed in the unmanned aerial vehicle 10, the memory, the controller and the wireless communicator are electrically connected with a power supply of the unmanned aerial vehicle 10, the controller is electrically connected with the memory, the wireless communicator and a control system of the cradle head 20 and the unmanned aerial vehicle 10, program instructions are stored in the memory, and when the controller executes the program instructions, the controller is used for implementing the following steps: a step of recognizing an electronic fence line, for recognizing the electronic fence line and an image of the flexible wire 32 for assisting calculation from the photographed image acquired by the camera 30; a step of detecting an electronic fence line, for detecting whether the levelness of the fence line of the electronic fence meets a standard based on the photographed image acquired by the camera 30; and a step of data communication, which is used for communicating the detection result and the flight control instruction of the unmanned aerial vehicle 10 with the outside.
In the above technical solution, the step of electronic fence line identification specifically includes performing electronic fence line identification on a photographed image by using a pre-trained fence line identification model, and marking the identified electronic fence line in the photographed image.
Further, in the above technical solution, the steps of establishing and training the fence line identification model specifically include:
step 1, acquiring a history fence image, preprocessing the history fence image, denoising the history fence image by adopting a median filtering method to obtain a first denoising image, and performing gray processing by adopting an average value method, wherein the calculation formula is as follows:
where Gray represents a Gray value, R represents a red value in an RGB color mode, G represents a green value in the RGB color mode, and B represents a blue value in the RGB color mode.
The gray-scale processed image is enhanced by an edge enhancement filter to improve the quality, enhancement characteristics or sharpness of the image. And (3) adopting a MAX-MIN filter, and reassigning the pixels in the grid by using the difference value between the maximum value and the minimum value of the pixels in the grid.
Step 2, taking the preprocessed history fence image as a training sample, and dividing the training sample into a training set, a verification set and a test set, wherein the ratio of the training set to the verification set to the test set is 7:2:1;
Step 3, selecting a feedforward neural network model as a basic model, selecting a ReLU activation function, defining a loss function and an optimization algorithm, selecting a cross entropy loss function, and selecting a random gradient descent method by the optimization algorithm;
step 4, selecting an Xavier initialization mode to initialize weights and bias parameters;
and step 5, training the defined feedforward neural network model by using a training sample, performing forward propagation by inputting the training sample into the defined feedforward neural network model, calculating a gradient by using backward propagation, and updating network parameters by using an optimization algorithm to obtain a trained fence line identification model.
In the above technical solution, the step of electronic fence line identification specifically includes performing electronic fence line identification on a photographed image by using a pre-trained fence line identification model, and marking the identified electronic fence line in the photographed image.
Further, in the above technical solution, the step of detecting the electronic fence line may further adopt a second method, which specifically includes:
s21, controlling the distance from the flying of the unmanned aerial vehicle 10 to the electronic fence to be 200 cm-500 cm;
s22, acquiring a fence image shot by the unmanned aerial vehicle 10, and marking all electronic fence lines in the fence image;
S23, adjusting the vertical height of the unmanned aerial vehicle 10 so that the flexible metal wires 32 intersect all marked electronic fence lines in the fence image in the step S22;
s24, controlling the unmanned aerial vehicle 10 to fly parallel to the electronic fence line marked in the step S22, and continuously shooting electronic fence line images to form an electronic fence line image set until the height of the unmanned aerial vehicle 10 is kept stable and the distance between the unmanned aerial vehicle 10 and the electronic fence line is kept stable;
s25, analyzing the included angle between each electronic fence line and the flexible metal wire 32 for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard. For example, an image of one electronic fence line is shown in fig. 5, in which a vertical straight line is a flexible wire 32, and a horizontal straight line is a plurality of electronic fence lines, and each included angle is calculated.
Further, in the above technical solution, the step of analyzing the included angle between the electronic fence line and the flexible wire 32 for each electronic fence line image in the electronic fence line image set specifically includes:
acquiring continuous video frames for a period of time, and calculating the mean value and variance of each pixel point to obtain a background model;
Comparing each video frame with a background model, calculating the difference between the pixels of the current video frame and the background model, and for each pixel point, if the difference between the pixels of the current video frame and the background model exceeds a third threshold value obtained by an adaptive threshold value processing method, dynamically adjusting the threshold value according to the local characteristics of the image, marking the pixels of the marked foreground as a foreground, and forming a foreground image by the marked foreground pixels;
updating the background model by adopting a rolling average method, and enabling pixel values of the video frame to participate in the updating of the background model;
filtering the background model by using a median filtering method;
detecting position information of the electronic fence line and the flexible metal wire 32 by adopting a target detection algorithm according to the foreground image, and tracking the change of the electronic fence line and the flexible metal wire 32 in time by using a tracking algorithm;
identifying shape boundaries for the electronic fence lines and the flexible wires 32 by using a Canny edge detection algorithm;
based on the outline of the electronic fence line and the flexible metal wire 32, two position vectors of the electronic fence line and the flexible metal wire 32 are obtained according to the coordinates of the central points of the electronic fence line and the flexible metal wire 32, and the included angle of the two vectors is calculated.
As shown in fig. 5, all three electronic fence lines are calculated to meet the levelness standard.
In the above technical scheme, the unmanned aerial vehicle 10 is further provided with a colored light source emitter, and the colored light source emitter is used for irradiating the electronic fence line and assisting in marking the electronic fence line.
For example, the shot electronic fence line image and/or the analysis result are/is horizontal, and are transmitted to the handheld voltage tester in a wireless communication mode, such as Bluetooth, and the handheld voltage tester displays the test result to a user, such as a text form of a liquid crystal display. When the handheld voltage tester receives the image shot by the camera 30 and the calculated levelness detection result, if the electronic fence line does not meet the levelness standard, the operator is reminded by means of a liquid crystal screen or an indicator lamp.
Hand-held voltage testers are common test components with handles and display screens that detect the live condition of an electronic fence to verify whether it is in operation.
Preferably, in the above embodiment, besides being used for electronic fence, the method flow of the above embodiment is also applicable to various drainage lines vertically arranged in multiple rows/single row. Because of the drainage wire or the electronic fence, discharge interference is caused when the distance is relatively short, so that safety accidents are caused, and the detection method is also applicable.
The method is also suitable for detecting the sagging of the drainage line, and can judge whether the drainage line meets the preset design requirement by scanning the included angle between the drainage line and the auxiliary reference line in a section with a certain length in the image. If the angle of the drainage wire in the converter station is smaller, if the angle is smaller than 80 degrees, the sagging is considered to be too low, and false touch in the operation process in the converter station is easy to occur.
Specifically, the principle of the invention is as follows: including unmanned aerial vehicle 10, cloud platform 20 and camera 30, cloud platform 20 sets up in unmanned aerial vehicle 10 middle part below and with unmanned aerial vehicle 10 fixed connection, camera 30 installs on cloud platform 20, the lens cone 31 top of camera 30 hangs through weight 33 has the flexible wire 32 that appears on camera (30), be provided with memory, controller, wireless communicator in unmanned aerial vehicle 10, memory, controller and wireless communicator with unmanned aerial vehicle 10's power electricity is connected, the controller with memory wireless communicator and cloud platform 20 the control system electricity of unmanned aerial vehicle 10 is connected, the memory stores has the program command, when the controller carries out the program command, be used for realizing the following steps: a step of electronic fence line recognition for recognizing an electronic fence line and an image of the flexible wire 32 for assisting calculation from the photographed image acquired by the camera 30; a step of detecting an electronic fence line, which is used for detecting a shooting image acquired based on the camera 30, and judging whether the levelness of the fence line of the electronic fence meets the standard according to the included angle between the flexible metal wire 32 and the electronic fence line in the image; and a step of data communication, which is used for communicating the detection result and the flight control instruction of the unmanned aerial vehicle 10 with the outside. Through the arrangement of the unmanned aerial vehicle 10, the cradle head 20 and the camera 30 are suspended in the air and are stabilized at fixed positions; capturing an image of the position of the electronic fence line and the flexible wire 31 by arranging the camera 30; a vertical line appears in a picture shot by the camera 30 through the arrangement of the flexible metal wire 31, and the electronic fence line is judged to be in accordance with the levelness standard according to the included angle at the intersection of the flexible metal wire 31 and the electronic fence line; the weight of the flexible wire 31 can be given by providing the weight 32 to ensure that the flexible wire 31 is perpendicular to the ground. The device can be used for rapidly and accurately detecting whether the electronic fence line is straight or not and whether the falling danger occurs or not.
Claims (10)
1. The utility model provides a transformer substation's electronic fence line levelness detection device, its characterized in that includes unmanned aerial vehicle (10), cloud platform (20) and camera (30), cloud platform (20) set up in unmanned aerial vehicle (10) middle part below and with unmanned aerial vehicle (10) fixed connection, camera (30) are installed on cloud platform (20), lens cone (31) top of camera (30) hangs flexible wire (32) that appear on camera (30) through weight (33), be provided with memory, controller, radio communication ware in unmanned aerial vehicle (10), memory, controller and radio communication ware with the power electricity of unmanned aerial vehicle (10) is connected, the controller with memory radio communication ware and cloud platform (20) the control system electricity of unmanned aerial vehicle (10) is connected, the memory stores the program command, when the controller is carried out the program command is used for realizing following steps:
a step of electronic fence line identification, which is used for identifying the electronic fence line and an image of the flexible metal wire (32) for assisting calculation from the shooting image acquired by the camera (30);
the electronic fence line detection step is used for judging whether the levelness of the fence line of the electronic fence meets the standard or not according to the included angle between the flexible metal wire (32) and the electronic fence line in the image based on the shooting image acquired by the camera (30);
And the step of data communication is used for communicating the detection result with the outside and controlling the flight command of the unmanned aerial vehicle (10).
2. The substation electronic fence line levelness detection device according to claim 1, wherein the electronic fence line identification step specifically includes performing electronic fence line identification on the photographed image by using a pre-trained fence line identification model, and marking the identified electronic fence line in the photographed image.
3. The substation electronic fence line levelness detection device according to claim 2, wherein the steps of building and training the fence line identification model specifically comprise:
step 1, acquiring a history fence image and preprocessing the history fence image, wherein the preprocessing comprises the following steps: denoising, gray level processing and enhancement;
step 2, taking the preprocessed historical fence image as a training sample, and dividing the training sample into a training set, a verification set and a test set, wherein the ratio of the training sample to the verification set to the test set is 7:2:1;
step 3, selecting a feedforward neural network model as a basic model, selecting an activation function, and defining a loss function and an optimization algorithm, wherein the activation function selects a ReLU activation function, the loss function selects a cross entropy loss function, and the optimization algorithm selects a random gradient descent method;
Step 4, selecting an Xavier initialization mode to initialize weights and bias parameters;
and 5, training the defined feedforward neural network model by using a training sample, performing forward propagation by inputting the training sample into the defined feedforward neural network model, calculating a gradient by using backward propagation, and updating network parameters by using an optimization algorithm to obtain a trained fence line identification model.
4. The substation electronic fence line levelness detection device according to claim 2, wherein the electronic fence line detection step specifically includes:
s11, controlling the unmanned aerial vehicle (10) to fly to a first threshold value of the distance from the electronic fence;
s12, acquiring a fence image shot by the unmanned aerial vehicle (10), and marking all electronic fence lines in the fence image;
s13, adjusting the vertical height of the unmanned aerial vehicle (10) so that the flexible metal wire (32) is intersected with one of the marked electronic fence lines in the fence image in the step S12;
s14, controlling the unmanned aerial vehicle (10) to fly parallel to the electronic fence lines marked in the step S12 and continuously shooting the electronic fence line images to form an electronic fence line image set until the electronic fences are completely shot, wherein the height of the unmanned aerial vehicle (10) is kept stable, and the distance between the unmanned aerial vehicle (10) and the electronic fence lines is kept stable;
S15, analyzing an included angle between the electronic fence line and the flexible metal wire (32) for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard;
s16, adjusting the vertical height of the unmanned aerial vehicle (10) to enable the flexible metal wire (32) to be intersected with the marked other electronic fence line, and executing steps S14 and S15 to judge whether the levelness of the current electronic fence line meets the standard; and repeatedly executing the step until the levelness of each electronic fence line is detected.
5. The substation electronic fence line levelness detection device according to claim 2, wherein the electronic fence line detection step specifically includes:
s21, controlling the distance from the unmanned aerial vehicle (10) to the electronic fence to be a second threshold value;
s22, acquiring a fence image shot by the unmanned aerial vehicle (10), and marking all electronic fence lines in the fence image;
s23, adjusting the vertical height of the unmanned aerial vehicle (10) so that the flexible metal wires (32) intersect all marked electronic fence lines in the fence image in the step S22;
s24, controlling the unmanned aerial vehicle (10) to fly parallel to the electronic fence line marked in the step S22, and continuously shooting electronic fence line images to form an electronic fence line image set until all electronic fences are shot, wherein the height of the unmanned aerial vehicle (10) is kept stable, and the distance between the unmanned aerial vehicle (10) and the electronic fence line is kept stable;
S25, analyzing the included angle between each electronic fence line and the flexible metal wire (32) for each electronic fence line image in the electronic fence line image set, and if the included angle is 89-91 degrees, considering that the levelness of the electronic fence line meets the standard.
6. A substation electronic fence line levelness detection arrangement according to claim 4 or 5, characterized in that the step of keeping the height of the drone (10) stable comprises in particular:
detecting position information of an object behind an electronic fence line in a picture shot by a camera (30) by using a target detection algorithm, and marking a boundary frame of the object in a video frame;
extracting the outline of the object from each video frame by using an edge detection algorithm through the boundary frame of the object;
calculating the circle center of the minimum circumscribing circle of the outline, and marking the circle center as a first center coordinate;
step four, repeating the steps in the next video frame to obtain a second center coordinate;
and fifthly, calculating the distance between the first center coordinate and the second center coordinate, and if the distance is smaller than 5mm, considering the height of the unmanned aerial vehicle (10) to be stable.
7. The substation electronic fence line levelness detection device according to claim 4 or 5, characterized in that the step of keeping the distance between the unmanned aerial vehicle (10) and the electronic fence line stable comprises:
Firstly, acquiring a continuous video frame sequence of an object behind an electronic fence line by using a camera (30);
the second step, carrying out feature extraction on each video frame, including: size, shape, corner points, edges, texture, and sharpness of the object;
thirdly, matching the characteristics of the current video frame with the characteristics of the previous frame, and using a characteristic descriptor and a matching algorithm;
calculating an estimated value of the object distance of the object by using camera calibration parameters and a triangulation principle based on a result of feature matching;
and fifthly, judging whether the distance between the object and the object is changed according to the estimated distance, and if the estimated distance is not more than 5mm, stabilizing the distance between the unmanned aerial vehicle (10) and the electronic fence line.
8. The transformer substation electronic fence line levelness detection device according to claim 4 or 5, wherein the step of analyzing an angle between the electronic fence line and the flexible wire (32) for each electronic fence line image in the electronic fence line image set specifically comprises:
acquiring continuous video frames for a period of time, and calculating the mean value and variance of each pixel point to obtain a background model;
for each video frame, comparing the video frame with a background model, calculating the difference between the pixels of the current video frame and the background model, and for each pixel point, if the difference between the pixel point and the background model exceeds a third threshold value obtained by an adaptive threshold value processing method, dynamically adjusting the threshold value according to the local characteristics of the image, marking the pixel point as a foreground, and forming the marked foreground pixels into a foreground image;
Updating the background model by adopting a rolling average method, and enabling pixel values of the video frame to participate in the updating of the background model;
filtering the background model by using a median filtering method;
detecting position information of the electronic fence line and the flexible metal wire (32) by adopting a target detection algorithm according to the foreground image, and tracking the time change of the electronic fence line and the flexible metal wire (32) by using a tracking algorithm;
identifying shape boundaries of the electronic fence lines and the flexible metal wires (32) by adopting a Canny edge detection algorithm;
based on the outlines of the electronic fence line and the flexible metal wire (32), two position vectors of the electronic fence line and the flexible metal wire (32) are obtained according to the coordinates of the central points of the electronic fence line and the flexible metal wire (32), and the included angle of the two vectors is calculated.
9. The substation electronic fence line levelness detection device according to claim 1, wherein a colored light source emitter is further arranged on the unmanned aerial vehicle (10), and the colored light source emitter is used for irradiating the electronic fence line.
10. The substation electronic fence line levelness detection device according to claim 1, further comprising: the handheld voltage tester is used for receiving images transmitted by the unmanned aerial vehicle (10) and prompting whether the electronic fence line is inclined or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311140283.5A CN117168411A (en) | 2023-09-05 | 2023-09-05 | Electronic fence line levelness detection device of transformer substation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311140283.5A CN117168411A (en) | 2023-09-05 | 2023-09-05 | Electronic fence line levelness detection device of transformer substation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117168411A true CN117168411A (en) | 2023-12-05 |
Family
ID=88935045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311140283.5A Pending CN117168411A (en) | 2023-09-05 | 2023-09-05 | Electronic fence line levelness detection device of transformer substation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117168411A (en) |
-
2023
- 2023-09-05 CN CN202311140283.5A patent/CN117168411A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10269138B2 (en) | UAV inspection method for power line based on human visual system | |
| CN107782733B (en) | Image recognition nondestructive detection device and method for metal surface defects | |
| CN103745226B (en) | Dressing safety detection method for worker on working site of electric power facility | |
| CN109635758B (en) | Intelligent building site video-based safety belt wearing detection method for aerial work personnel | |
| CN108846418A (en) | A kind of positioning of cable machinery temperature anomaly and recognition methods | |
| US9934578B2 (en) | Method for edge detection | |
| CN106162124B (en) | Projector exports the calibration method of image | |
| CN108307146A (en) | A kind of ultra-high-tension power transmission line Security Vulnerability Detecting System and method | |
| CN110837822B (en) | Fire-fighting robot injection curve adjusting method and device based on multi-view vision | |
| CN107085714A (en) | A kind of forest fire detection method based on video | |
| CN103442209A (en) | Video monitoring method of electric transmission line | |
| CN112149543B (en) | Building dust recognition system and method based on computer vision | |
| CN104198752B (en) | Many rate detection methods of high temperature billet steel motion state based on machine vision | |
| CN107977639A (en) | A kind of face definition judgment method | |
| CN109472261A (en) | A kind of quantity of stored grains in granary variation automatic monitoring method based on computer vision | |
| CN109413411A (en) | A kind of blank screen recognition methods, device and the server of monitoring circuit | |
| CN107316024B (en) | Perimeter alarm algorithm based on deep learning | |
| CN113706566B (en) | Edge detection-based perfuming and spraying performance detection method | |
| CN116778348B (en) | Machine vision-based method, device, equipment and medium for monitoring falling off of outer wall | |
| CN107688793A (en) | A kind of outside transformer substation fire automatic monitoring method for early warning | |
| CN104966302B (en) | A kind of detection localization method of any angle laser cross | |
| CN112101260A (en) | Method, device, equipment and storage medium for identifying safety belt of operator | |
| CN115841633A (en) | Power tower and power line associated correction power tower and power line detection method | |
| CN117474321B (en) | BIM model-based construction site risk intelligent identification method and system | |
| CN110276787B (en) | Conductor galloping monitoring method based on marker image detection |
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 |