CN112395972A - Electric power system insulator string identification method based on unmanned aerial vehicle image processing - Google Patents
Electric power system insulator string identification method based on unmanned aerial vehicle image processing Download PDFInfo
- Publication number
- CN112395972A CN112395972A CN202011276698.1A CN202011276698A CN112395972A CN 112395972 A CN112395972 A CN 112395972A CN 202011276698 A CN202011276698 A CN 202011276698A CN 112395972 A CN112395972 A CN 112395972A
- Authority
- CN
- China
- Prior art keywords
- insulator string
- image
- insulator
- aerial vehicle
- unmanned aerial
- 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
- 239000012212 insulator Substances 0.000 title claims abstract description 160
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012545 processing Methods 0.000 title claims abstract description 21
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 230000011218 segmentation Effects 0.000 claims description 17
- 238000005265 energy consumption Methods 0.000 abstract description 19
- 238000004364 calculation method Methods 0.000 abstract description 9
- 230000008569 process Effects 0.000 abstract description 6
- 238000007689 inspection Methods 0.000 abstract description 4
- 230000005540 biological transmission Effects 0.000 description 12
- 230000006870 function Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 8
- 230000007547 defect Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Water Supply & Treatment (AREA)
- Remote Sensing (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Astronomy & Astrophysics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of industrial Internet of things and edge calculation, in particular to a power system insulator string identification method based on unmanned aerial vehicle image processing. The method comprises the following steps: 1) the unmanned aerial vehicle divides the H channel image by adopting an OTSU algorithm with a threshold value of 3 to obtain a divided result image; 2) processing the result image by the unmanned aerial vehicle to obtain a zero-order moment, a first-order moment and a second-order moment of the result image; acquiring parameter information of an insulator string covering ellipse; 3) estimating a suspected insulator string area by an iterative optimization algorithm according to parameter information of an insulator string covering ellipse, and estimating a possible direction of the insulator string; 4) sending the detected images containing the insulator strings and the estimated directions thereof back to a ground server for processing; 5) and the ground server rotates the image until the insulator string is in a horizontal state, and identifies the insulator string in the image by using a Faster RCNN network. The invention solves the problem of high energy consumption in the inspection process of the unmanned aerial vehicle, and prolongs the flight time of the unmanned aerial vehicle.
Description
Technical Field
The invention relates to the field of industrial Internet of things and edge calculation, in particular to a power system insulator string identification method based on unmanned aerial vehicle image processing.
Background
Along with the development of unmanned aerial vehicle technique, unmanned aerial vehicle receives more and more attention in the application of industry thing networking. The unmanned aerial vehicle can improve the flexibility of target monitoring and assist in completing the communication process, so that the application range of the industrial Internet of things is effectively expanded. The limited energy of the unmanned aerial vehicle brings great challenges to the application of the unmanned aerial vehicle in the industrial internet of things. The flight time of the unmanned aerial vehicle is mainly related to the weight, the flight time, the data transmission energy consumption and the task calculation energy consumption of the unmanned aerial vehicle.
In an electric power system, the safety and reliability of a power transmission line play a crucial role in smooth operation of power transmission. The insulator string is a key component in the power transmission line which plays a role in insulation and support, and once a defect occurs, the whole power transmission line can be in a paralysis state. Insulator string recognition and defect detection are core parts of power transmission line inspection in an electric power system and have very important practical significance, insulator string recognition is the basis of insulator string defect detection, and whether the defect detection is accurate or not is directly determined by the quality of recognition results.
Transmitting the real-time video shot by the drone to a ground server for analysis may consume a large amount of energy. If the unmanned aerial vehicle can execute some data analysis tasks, the transmission of real-time videos is only carried out under necessary conditions, and the energy consumption of data transmission is greatly reduced. Moreover, compared with a general natural object, the insulator string has a very large length-width ratio, and when the target is identified by using the deep neural network, all possible directions of the target need to be detected, which is very large for the energy consumption of calculation. For the target detection of a large aspect ratio based on a deep neural network, an effective solution for reducing the computational energy consumption has not been proposed at present.
Disclosure of Invention
At unmanned aerial vehicle inspection process, no matter be unmanned aerial vehicle and carry out whole image analysis process, still with whole real-time video transmission to ground server, all can consume a large amount of energy to shorten unmanned aerial vehicle's flight time. Aiming at the defects of the prior art, the invention provides a power system insulator string identification method based on unmanned aerial vehicle image processing, which is a method for accurately positioning an insulator string at a server end and solves the problem that the existing deep neural network is large in energy consumption for identifying a target with a large length-width ratio, so that the flight time of an unmanned aerial vehicle is prolonged.
The technical scheme adopted by the invention for realizing the purpose is as follows: an electric power system insulator string identification method based on unmanned aerial vehicle image processing comprises the following steps:
1) the unmanned aerial vehicle acquires an image of a scene by carrying a camera, converts the image from an RGB space to an HSV space, and divides an H channel image by adopting an OTSU algorithm with a 3 threshold value to obtain a result image divided by the OTSU algorithm;
2) processing the result image by the unmanned aerial vehicle to obtain a zero-order moment, a first-order moment and a second-order moment of the result image; acquiring parameter information of the insulator string covering ellipse according to the zero-order moment, the first-order moment and the second-order moment;
3) estimating a suspected insulator string area according to the parameter information of the ellipse covered by the insulator string and an iterative optimization algorithm, and estimating the possible direction of the insulator string according to the suspected insulator string area;
4) sending the detected images containing the insulator strings and the estimated directions of the insulator string areas back to a ground server for processing;
5) after receiving the insulator string images and the direction information sent by the unmanned aerial vehicle, the ground server firstly rotates the images to the direction in which the insulator string is in a horizontal state, and then uses a Faster RCNN network to identify the insulator string in the images.
The step 2) is specifically as follows:
firstly, using the result image R after the OTSU algorithm segmentation*Is selected as the initial seed region R0;
Resulting image R*Wherein x and y represent pixels in the horizontal direction and the vertical direction in the image, respectively;
from the resulting image R*Obtaining the result image R by the pixel point (x, y) in*The zeroth order moment, the first order moment, and the second order moment of (d) are represented by:
zero order moment M00:
First moment M10And M01:
Second moment M20、M02And M11:
Acquiring parameter information of the insulator string covering ellipse according to the zero order moment, the first order moment and the second order moment, specifically to
Obtaining an initial seed region R according to formula (1) and formula (2)0The center point (X, Y) of the coverage ellipse is:
(X,Y)=(M10/M00,M01/M00)(4)
the length of the long axis of the insulator string covering ellipse obtained according to the formula (3) is as follows:
and (3) obtaining the length of the short axis of the insulator string covering ellipse according to the formula (3):
obtaining the direction of the corresponding insulator string according to the formulas (1), (2) and (3):
wherein, XCIs the abscissa, Y, of the center point of the ellipse in the imageCThe center point of the ellipse is on the ordinate of the image.
The parameter information of the insulator string comprises: initial seed region R0The center point of the cover ellipse, the major axis length L, the minor axis length S and the direction O.
In step 3), the estimating the suspected insulator string region by the iterative optimization algorithm specifically includes:
step (1): by maximizing this algorithm, an iterative objective function T is obtained, namely:
wherein R is*Is the result image after segmentation using the OTSU algorithm, E is the coverage ellipse of the suspected insulator string area, | E | is the area of the ellipse, | R |*N E | is the area of the resulting image covered by the ellipse E;
step (2): estimating a coverage ellipse E of a suspected insulator string region through the parameter information of the insulator string; and obtaining a coverage ellipse E and a result image R of the suspected insulator string region*Is covered by an ellipse EArea of image, i.e. overlap region RF;
And (3): according to the area | E | and the overlapping region R of the ellipseFObtaining a filling rate F;
and (4): according to the filling rate F and the overlapping region RFObtaining a target function T which is continuously iterated;
and (5): starting to reduce according to the value of the target function T which is iterated continuously, judging that the length-width ratio of the current E is larger than a threshold value, and possibly judging that the current E is an insulator string; and on the contrary, if the aspect ratio of the E is smaller than the set value, the E is not an insulator string.
Estimating the possible direction of the insulator string according to the suspected insulator string region in the step 3), specifically:
when only one insulator string exists in the image, directly acquiring the direction of the insulator string in the image;
when a plurality of insulator strings exist in the image, the direction set of the insulator strings in the image is obtained through a direction estimation algorithm.
When a plurality of insulator strings exist in the image, obtaining a direction set of the insulator strings in the image through a direction estimation algorithm, specifically:
A. setting an initial direction candidate set Ori as an empty set, and defining the number of items of the Ori set as | Ori |;
B. converting the shot insulator string image from an RGB space to an HSV space to obtain an H-channel image;
C. obtaining a segmentation threshold T ═ T using a 3-threshold OTSU algorithm1,t2,t3};
D. Using T to segment H channel image, obtaining region set R ═ { R1,r2,r3,r4In which r is1≤t1,t1<r2≤t2,t2<r3≤t3,t3<r4Wherein, t1,t2,t3Three thresholds for OTSU algorithm segmentation respectively;
E. using the get connected Domain function, r is obtained4The area of the noise point is less than 100Removing to obtain a connected domain set C;
F. arranging the items in the C according to the descending order of the area to obtain a region set R ═ { R ═ RFkK, where K is the total number of regions in R, RFkK coincident regions in the region set R.
Analyzing the insulator string in the step 5) according to the parameter information of the insulator string and the possible direction of the estimated insulator string to determine the direction of the insulator string, which specifically comprises the following steps:
when K ≦ K and the number of terms of | Ori | is < 6, the following loop is performed:
obtaining each R in the region set R through the target function TFkThe filling factor F, the length L of the long axis, the length S of the short axis, the direction O and the center points (X, Y) of the insulator strings, and determining the direction set of the insulator strings.
The determining of the direction set of the insulator string specifically includes:
when L/S > 4 and M00> 100 time
Judging the current RFkDirection O ofkIf the difference from all the terms in Ori is greater than 15 °, the direction of the current insulator string is added to the set of candidate directions Ori, where Ori ═ { O }k}; on the contrary, if the difference between the direction of the insulator string and the terms in Ori is less than or equal to 15 °, the terms in the original candidate direction set are still kept unchanged.
The invention has the following beneficial effects and advantages:
the invention provides a method for identifying an insulator string in an electric power system based on cloud edge fusion.
Drawings
FIG. 1 is a block diagram of the overall architecture of the present invention;
FIG. 2 is a sample image of 3 different channels of the present invention;
fig. 3 shows a rough insulator string recognition example according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as modified in the spirit and scope of the present invention as set forth in the appended claims.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
In the invention, the navigation and obstacle avoidance of the unmanned aerial vehicle depend on GPS signals and visual analysis. And only when a suspected target is detected, the real-time video is transmitted back to the cloud server side for further analysis. On the basis, a novel insulator string identification framework is provided.
As shown in fig. 1, the proposed framework comprises two steps. The first step is to roughly detect the insulator string and estimate the possible direction of the insulator string; and if one or more insulator strings are detected in the image, sending the image and the estimated direction back to the ground server, and then adopting a fast RCNN algorithm to accurately detect the insulator strings.
For the same insulator string, the colors of all insulators are the same, and the insulator string can be segmented according to the color information. As shown in fig. 2, different channel images of one image are given, which are RGB images, R-channel, G-channel, B-channel, H-channel, S-channel, and V-channel images, respectively, from left to right. The insulator strings in the three figures are different in color. As can be seen from fig. 2, the H-channel image is more suitable for segmentation using a threshold-based segmentation algorithm. In the invention, firstly, the OTSU algorithm is adopted to calculate the threshold, and then the OTSU algorithm with 3 thresholds is adopted to segment the insulator string in the image.
Although the insulator string region can be obtained by the method, the threshold-based segmentation algorithm is sensitive to background noise under the condition of complex background. The present invention thus proposes a new regional filtering method. The core idea of the method is that compared with background noise, the pixel distribution of the insulator string region is denser, and the overall shape is close to a rectangle. Also, for a rectangle, when the length and width are large, it can be approximated to an ellipse. Based on the above analysis, the present invention estimates the suspected insulator string area by maximizing the following equation.
Wherein R is*Is the result of segmentation using the OTSU algorithm, E is the coverage ellipse of the suspected insulator string region, | E | is the area of the ellipse, | R |*And # E | is the area of the segmentation result covered by the ellipse E.
The objective function T may be continuously iteratively optimized using the following algorithm:
first, the segmentation result R*Is selected as the initial seed region R0;
Second, R0The center point (X, Y), the major axis length L, the minor axis length S, and the direction O of (a) can be calculated from the first moment, the second moment, and the zero moment of the image, and the corresponding calculation formula is as follows:
zero order moment M00:
First moment M10And M01:
Second moment M20、M02And M11:
Acquiring parameter information of the insulator string covering ellipse according to the zero order moment, the first order moment and the second order moment, specifically to
Obtaining an initial seed region R according to formula (1) and formula (2)0The center point (X, Y) of the coverage ellipse is:
(X,Y)=(M10/M00,M01/M00) (4)
the length of the long axis of the insulator string covering ellipse obtained according to the formula (3) is as follows:
and (3) obtaining the length of the short axis of the insulator string covering ellipse according to the formula (3):
obtaining the direction of the corresponding insulator string according to the formulas (1), (2) and (3):
wherein, XCIs the abscissa, Y, of the center point of the ellipse in the imageCThe center point of the ellipse is on the ordinate of the image.
Then, a coverage ellipse E is estimated using the center points (X, Y), L, S and O;
after that, according to R0=R*Andu E to obtain R*Overlap region R with E0;
Finally, according to F ═ R0Calculating filling rate F by I/E;
at this time, the objective function T can be based on F | R0And | calculating to obtain. This process is iterated until T starts to decrease. The region E with larger area and higher filling rate can be obtained simultaneously by using the algorithm.
If the aspect ratio of E is large, it may be an insulator string. Conversely, if the aspect ratio of E is small, even close to 1, it should not be an insulator string. The result shows that compared with a deep learning method with higher calculation complexity, the algorithm can be used for screening out images possibly containing insulator strings at lower calculation cost.
Considering that there may be multiple insulator strings in the image, the following method for obtaining the possible directions of the multiple insulator strings through a direction estimation algorithm is proposed. Since the number of insulator strings is relatively small in an actual scene and the directions of some insulator strings are the same, 6 main directions are sufficient for most scenes. Thus, the present invention results in up to 6 possible directions.
In this algorithm, L/S > 4 is because the length of the insulator string is typically much greater than the width of the insulator sheet, so if the L/S of the region obtained using the above-described optimization algorithm is very small, e.g., close to 1, then this region should not be the insulator string region. At the same time, M00The meaning of > 100 is when M00When the noise point is less than 100 ℃, the small noise point can be removed by using a method for removing the small connected domain.
Inputting: RGB image Img.
A. Setting an initial direction candidate set Ori as an empty set, and defining the number of items of the Ori set as | Ori |;
B. converting the shot insulator string image from an RGB space to an HSV space to obtain an H-channel image;
C. obtaining a segmentation threshold T ═ T using a 3-threshold OTSU algorithm1,t2,t3};
D. Using T to segment H channel image, obtaining region set R ═ { R1,r2,r3,r4In which r is1≤t1,t1<r2≤t2,t2<r3≤t3,t3<r4Wherein, t1,t2,t3Three thresholds for OTSU algorithm segmentation respectively;
E. using the get connected Domain function, r is obtained4Removing the area of the noise point smaller than 100 to obtain a connected domain set C;
F. arranging the items in the C according to the descending order of the area to obtain a region set R ═ { R ═ RFkK, where K is the total number of regions in R, RFkK coincident regions in the region set R.
Analyzing the insulator string in the step 5) according to the parameter information of the insulator string and the possible direction of the estimated insulator string to determine the direction of the insulator string, which specifically comprises the following steps:
when K ≦ K and the number of terms of | Ori | is < 6, the following loop is performed:
obtaining each R in the region set R through the target function TFkThe filling factor F, the length L of the long axis, the length S of the short axis, the direction O and the center points (X, Y) of the insulator strings, and determining the direction set of the insulator strings.
The determining of the direction set of the insulator string specifically includes:
when L/S > 4 and M00> 100 time
Judging the current RFkDirection O ofkIf the difference from all the terms in Ori is greater than 15 °, the direction of the current insulator string is added to the set of candidate directions Ori, where Ori ═ { O }k}; on the contrary, if the difference between the direction of the insulator string and the terms in Ori is less than or equal to 15 °, the terms in the original candidate direction set are still kept unchanged.
And (3) outputting: the directional candidate set Ori.
Finally, the image is rotated to the direction that the insulator string is in a horizontal state through the direction candidate set Ori, and then the insulator string in the image is identified by using a Faster RCNN network.
An example of coarse identification of an insulator string is given in fig. 3, where fig. 3(d) is the result of algorithm 2, with different insulator string regions highlighted in different colors. As can be seen from fig. 3(d), the direction estimation algorithm proposed in the present invention can successfully identify the insulator string in the image, so as to estimate the direction of the insulator string.
When the drone identifies the presence of one or more insulator strings in the image, the corresponding image and the estimated direction will be transmitted back to the cloud server for further analysis. On the server side, the insulator string images are firstly rotated to the horizontal direction according to the estimated direction, then the insulator strings are identified by using an improved Faster RCNN deep learning framework, the length-width ratio of the insulator strings is large, the proportion of candidate area networks in the Faster RCNN is changed to 1:8,1:4,1:2,1:1,2:1,4:1,8:1, the scale is changed to 64,128,256 and 512, the experiment is carried out on ubuntu18.04, python3.6, rtx2080ti, the deep learning framework uses cafe, in the training stage, the initial learning rate is set to 0.001, the weight attenuation coefficient is set to 0.0005, and the momentum value is set to 0.9.
Modifying the RPN part in a target detection algorithm Faster RCNN according to the characteristic that the length-width ratio of the insulator string is large:
the proportion of the RPN network is changed from the original 1:1,1:2,2:1 to: 1:8,1:4,1:2,1:1,2:1,4:1,8:1, the scale was changed from original 128,256,512 to 64,128,256 and 512.
Example 1:
by the identification method, the power consumption of the unmanned aerial vehicle is evaluated, and the method specifically comprises the following steps:
at present, in the electric power inspection process, the unmanned aerial vehicle needs to transmit the shot video and the real-time position data of the unmanned aerial vehicle back to the cloud server. By adopting the method provided by the invention, when no target or obstacle appears, the unmanned aerial vehicle only needs to send the position data back to the operator, so that the operator can control the flight path by utilizing the electronic map. When a suspected target is found using the coarse recognition algorithm, no one will transmit the video back to the operator. Then, the operator can control the unmanned aerial vehicle to take detailed videos or photos of the suspected target, and further analysis is carried out by using a fine recognition algorithm at the cloud server end. Therefore, if the method provided by the present invention is adopted, the energy consumption Ecol of the drone will be modeled as:
Ecol=(F×W+K+I1+mean(CB))×T+mean(CI)×t
f is the energy consumption per unit weight per unit time in the flight process of the unmanned aerial vehicle, W is the weight of the unmanned aerial vehicle, K is the energy consumption of common calculation tasks per unit time, I1Energy consumption for rough image recognition, mean (C)B) Is the average energy consumption (e.g., location data transmission), mean (C) of the underlying communicationI) The average energy consumption for transmitting the image in unit time, T is total flight time, and T is the image transmission time after the suspected target is found.
For comparison, the method of transmitting all videos and images to the operator in real time for intelligent analysis is called a centralized method, and the method of fully performing intelligent analysis by the unmanned aerial vehicle is called a distributed method. For a centralized approach, the drone energy consumption E may be reducedcenModeling is as follows:
Ecen=(F×W+K+mean(CB)+mean(CI))×T
in addition, for a distributed method, the energy consumption E of the unmanned aerial vehicle can be reduceddecModeling is as follows:
Edec=(F×(W+S)+K+Iall+mean(CB))×T
where S is the quality of the extra special hardware supporting deep learning computation, IallIs the energy consumption per unit time to run the entire target recognition algorithm.
In order to compare the method of the present invention with the centralized method and the decentralized method, respectively, the difference between the method of the present invention and the other two methods is calculated:
Ecol-Ecen=I1×T+mean(CI)×(t-T)
Ecol-Edec=-(I2+F×S)×T+mean(CI)×t
wherein I2Representing the energy consumption at which the target is accurately identified.
Typically T < T, so:
Ecol-Ecen≈(I1-mean(CI))×T
Ecol-Edec=-(I2+F×S)×T
the method provided by the invention only carries out simple linear conversion, threshold-based segmentation and pixel statistical operation, so that the calculation amount of rough identification is small, and I is1Is very small. Furthermore, the typical video data transmission energy consumption of the drone is about 1W, so that E in the above equationcol-Ecen<0。
From Ecol-EdecIt can also be seen from this formula that the method proposed by the present invention is superior to the discretization method, and is particularly represented by- (I)2+I3The + FxS) xT is less than 0, so that the electric power system insulator string identification method based on unmanned aerial vehicle image processing has the effect of reducing energy consumption.
The algorithm provided by the invention can well identify the insulator strings in the image and has great advantages in the aspects of accuracy and recall rate.
The above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (8)
1. An electric power system insulator string identification method based on unmanned aerial vehicle image processing is characterized by comprising the following steps:
1) the unmanned aerial vehicle acquires an image of a scene by carrying a camera, converts the image from an RGB space to an HSV space, and divides an H channel image by adopting an OTSU algorithm with a 3 threshold value to obtain a result image divided by the OTSU algorithm;
2) processing the result image by the unmanned aerial vehicle to obtain a zero-order moment, a first-order moment and a second-order moment of the result image; acquiring parameter information of the insulator string covering ellipse according to the zero-order moment, the first-order moment and the second-order moment;
3) estimating a suspected insulator string area according to the parameter information of the ellipse covered by the insulator string and an iterative optimization algorithm, and estimating the possible direction of the insulator string according to the suspected insulator string area;
4) sending the detected images containing the insulator strings and the estimated directions of the insulator string areas back to a ground server for processing;
5) after receiving the insulator string images and the direction information sent by the unmanned aerial vehicle, the ground server firstly rotates the images to the direction in which the insulator string is in a horizontal state, and then uses a Faster RCNN network to identify the insulator string in the images.
2. The method for identifying the insulator string of the power system based on the unmanned aerial vehicle image processing as claimed in claim 1, wherein the step 2) is specifically as follows:
firstly, using the result image R after the OTSU algorithm segmentation*Is selected as the initial seed region R0;
Resulting image R*Wherein x and y represent pixels in the horizontal direction and the vertical direction in the image, respectively;
from the resulting image R*Obtaining the result image R by the pixel point (x, y) in*The zeroth order moment, the first order moment, and the second order moment of (d) are represented by:
zero order moment M00:
First moment M10And M01:
Second moment M20、M02And M11:
Acquiring parameter information of the insulator string covering ellipse according to the zero order moment, the first order moment and the second order moment, specifically to
Obtaining an initial seed region R according to formula (1) and formula (2)0The center point (X, Y) of the coverage ellipse is:
(X,Y)=(M10/M00,M01/M00)(4)
the length of the long axis of the insulator string covering ellipse obtained according to the formula (3) is as follows:
and (3) obtaining the length of the short axis of the insulator string covering ellipse according to the formula (3):
obtaining the direction of the corresponding insulator string according to the formulas (1), (2) and (3):
wherein, XCIs the abscissa, Y, of the center point of the ellipse in the imageCThe center point of the ellipse is on the ordinate of the image.
3. The unmanned aerial vehicle image processing-based power system insulator string recognition method as claimed in claim 1, wherein the parameter information of the insulator string comprises: initial seed region R0The center point of the cover ellipse, the major axis length L, the minor axis length S and the direction O.
4. The method for identifying insulator strings of the power system based on unmanned aerial vehicle image processing as claimed in claim 1, wherein in step 3), the suspected insulator string region is estimated through an iterative optimization algorithm, specifically:
step (1): by maximizing this algorithm, an iterative objective function T is obtained, namely:
wherein R is*Is the result image after segmentation using the OTSU algorithm, E is the coverage ellipse of the suspected insulator string area, | E | is the area of the ellipse, | R |*N E | is the area of the resulting image covered by the ellipse E;
step (2): estimating a coverage ellipse E of a suspected insulator string region through the parameter information of the insulator string; and obtaining a coverage ellipse E and a result image R of the suspected insulator string region*The area of the resulting image covered by the ellipse E, i.e. the overlap region RF;
And (3): according to the area | E | and the overlapping region R of the ellipseFObtaining a filling rate F;
and (4): according to the filling rate F and the overlapping region RFObtaining a target function T which is continuously iterated;
and (5): starting to reduce according to the value of the target function T which is iterated continuously, judging that the length-width ratio of the current E is larger than a threshold value, and possibly judging that the current E is an insulator string; and on the contrary, if the aspect ratio of the E is smaller than the set value, the E is not an insulator string.
5. The method for identifying an insulator string of an electric power system based on unmanned aerial vehicle image processing according to claim 1, wherein the estimating of the possible direction of the insulator string according to the suspected insulator string region in step 3) specifically comprises:
when only one insulator string exists in the image, directly acquiring the direction of the insulator string in the image;
when a plurality of insulator strings exist in the image, the direction set of the insulator strings in the image is obtained through a direction estimation algorithm.
6. The electric power system insulator string recognition method based on unmanned aerial vehicle image processing as claimed in claim 5, wherein when there are a plurality of insulator strings in the image, the direction set of the insulator strings in the image is obtained by a direction estimation algorithm, specifically:
A. setting an initial direction candidate set Ori as an empty set, and defining the number of items of the Ori set as | Ori |;
B. converting the shot insulator string image from an RGB space to an HSV space to obtain an H-channel image;
C. obtaining a segmentation threshold T ═ T using a 3-threshold OTSU algorithm1,t2,t3};
D. Using T to segment H channel image, obtaining region set R ═ { R1,r2,r3,r4In which r is1≤t1,t1<r2≤t2,t2<r3≤t3,t3<r4Wherein, t1,t2,t3Three thresholds for OTSU algorithm segmentation respectively;
E. using the get connected Domain function, r is obtained4Removing the area of the noise point smaller than 100 to obtain a connected domain set C;
F. arranging the items in the C according to the descending order of the area to obtain a region set R ═ { R ═ RFkK, where K is the total number of regions in R, RFkK coincident regions in the region set R.
7. The method for identifying insulator strings of the power system based on unmanned aerial vehicle image processing according to claim 5, wherein in the step 5), the insulator strings are analyzed according to parameter information of the insulator strings and possible directions of the estimated insulator strings, and the directions of the insulator strings are determined, specifically:
when K ≦ K and the number of terms of | Ori | is < 6, the following loop is performed:
passing through the targetThe function T obtains each R in the region set RFkThe filling factor F, the length L of the long axis, the length S of the short axis, the direction O and the center points (X, Y) of the insulator strings, and determining the direction set of the insulator strings.
8. The method for identifying insulator strings of the power system based on unmanned aerial vehicle image processing according to claim 7, wherein the determining of the direction set of the insulator strings specifically comprises:
when L/S > 4 and M00> 100 time
Judging the current RFkDirection O ofkIf the difference from all the terms in Ori is greater than 15 °, the direction of the current insulator string is added to the set of candidate directions Ori, where Ori ═ { O }k}; on the contrary, if the difference between the direction of the insulator string and the terms in Ori is less than or equal to 15 °, the terms in the original candidate direction set are still kept unchanged.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011276698.1A CN112395972B (en) | 2020-11-16 | 2020-11-16 | Unmanned aerial vehicle image processing-based insulator string identification method for power system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011276698.1A CN112395972B (en) | 2020-11-16 | 2020-11-16 | Unmanned aerial vehicle image processing-based insulator string identification method for power system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112395972A true CN112395972A (en) | 2021-02-23 |
CN112395972B CN112395972B (en) | 2023-07-14 |
Family
ID=74599516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011276698.1A Active CN112395972B (en) | 2020-11-16 | 2020-11-16 | Unmanned aerial vehicle image processing-based insulator string identification method for power system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112395972B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113255434A (en) * | 2021-04-08 | 2021-08-13 | 淮阴工学院 | Apple identification method fusing fruit features and deep convolutional neural network |
CN114140679A (en) * | 2021-10-26 | 2022-03-04 | 中科慧远视觉技术(北京)有限公司 | Defect fusion method, device, recognition system and storage medium |
CN114460086A (en) * | 2021-12-15 | 2022-05-10 | 国网河北省电力有限公司电力科学研究院 | Detection method for inclined strain insulator string |
CN116846789A (en) * | 2023-09-01 | 2023-10-03 | 国网四川省电力公司信息通信公司 | Operation and maintenance management system for communication link |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104504696A (en) * | 2014-12-16 | 2015-04-08 | 江南大学 | Embedded parallel optimization method for image salient region detection |
CN105389555A (en) * | 2015-11-06 | 2016-03-09 | 天津航天中为数据***科技有限公司 | Aerial-image-based hidden-danger target analysis method and system |
CN105549603A (en) * | 2015-12-07 | 2016-05-04 | 北京航空航天大学 | Intelligent road tour inspection control method for multi-rotor-wing unmanned aerial vehicle |
CN105740843A (en) * | 2016-03-02 | 2016-07-06 | 成都翼比特自动化设备有限公司 | Insulator complete segmentation algorithm based on image recognition technology |
CN107507172A (en) * | 2017-08-08 | 2017-12-22 | 国网上海市电力公司 | Merge the extra high voltage line insulator chain deep learning recognition methods of infrared visible ray |
CN107610128A (en) * | 2017-09-26 | 2018-01-19 | 山东鲁能智能技术有限公司 | The method for inspecting and device of a kind of oil level indicator |
CN110222683A (en) * | 2019-06-11 | 2019-09-10 | 云南电网有限责任公司曲靖供电局 | A kind of quick-fried defect recognition localization method of electric transmission line isolator component based on depth convolutional neural networks |
US20190362185A1 (en) * | 2018-05-09 | 2019-11-28 | Figure Eight Technologies, Inc. | Aggregated image annotation |
-
2020
- 2020-11-16 CN CN202011276698.1A patent/CN112395972B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104504696A (en) * | 2014-12-16 | 2015-04-08 | 江南大学 | Embedded parallel optimization method for image salient region detection |
CN105389555A (en) * | 2015-11-06 | 2016-03-09 | 天津航天中为数据***科技有限公司 | Aerial-image-based hidden-danger target analysis method and system |
CN105549603A (en) * | 2015-12-07 | 2016-05-04 | 北京航空航天大学 | Intelligent road tour inspection control method for multi-rotor-wing unmanned aerial vehicle |
CN105740843A (en) * | 2016-03-02 | 2016-07-06 | 成都翼比特自动化设备有限公司 | Insulator complete segmentation algorithm based on image recognition technology |
CN107507172A (en) * | 2017-08-08 | 2017-12-22 | 国网上海市电力公司 | Merge the extra high voltage line insulator chain deep learning recognition methods of infrared visible ray |
CN107610128A (en) * | 2017-09-26 | 2018-01-19 | 山东鲁能智能技术有限公司 | The method for inspecting and device of a kind of oil level indicator |
US20190362185A1 (en) * | 2018-05-09 | 2019-11-28 | Figure Eight Technologies, Inc. | Aggregated image annotation |
CN110222683A (en) * | 2019-06-11 | 2019-09-10 | 云南电网有限责任公司曲靖供电局 | A kind of quick-fried defect recognition localization method of electric transmission line isolator component based on depth convolutional neural networks |
Non-Patent Citations (2)
Title |
---|
BABAK PORKAR 等: "New mathematical formulations for calculating residual resistance in a static arc model of ice-covered insulators", 《COLD REGIONS SCIENCE AND TECHNOLOGY》, pages 34 - 42 * |
赵振兵 等: "输电线路绝缘子、金具和导地线可靠度校准", 《电力建设》, pages 205 - 211 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113255434A (en) * | 2021-04-08 | 2021-08-13 | 淮阴工学院 | Apple identification method fusing fruit features and deep convolutional neural network |
CN113255434B (en) * | 2021-04-08 | 2023-12-19 | 淮阴工学院 | Apple identification method integrating fruit characteristics and deep convolutional neural network |
CN114140679A (en) * | 2021-10-26 | 2022-03-04 | 中科慧远视觉技术(北京)有限公司 | Defect fusion method, device, recognition system and storage medium |
CN114140679B (en) * | 2021-10-26 | 2022-07-01 | 中科慧远视觉技术(北京)有限公司 | Defect fusion method, device, recognition system and storage medium |
CN114460086A (en) * | 2021-12-15 | 2022-05-10 | 国网河北省电力有限公司电力科学研究院 | Detection method for inclined strain insulator string |
CN114460086B (en) * | 2021-12-15 | 2023-08-15 | 国网河北省电力有限公司电力科学研究院 | Detection method of inclined tension insulator string |
CN116846789A (en) * | 2023-09-01 | 2023-10-03 | 国网四川省电力公司信息通信公司 | Operation and maintenance management system for communication link |
CN116846789B (en) * | 2023-09-01 | 2023-11-14 | 国网四川省电力公司信息通信公司 | Operation and maintenance management system for communication link |
Also Published As
Publication number | Publication date |
---|---|
CN112395972B (en) | 2023-07-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112395972B (en) | Unmanned aerial vehicle image processing-based insulator string identification method for power system | |
Patil et al. | MSFgNet: A novel compact end-to-end deep network for moving object detection | |
CN109635685B (en) | Target object 3D detection method, device, medium and equipment | |
CN108665487B (en) | Transformer substation operation object and target positioning method based on infrared and visible light fusion | |
CN113205176B (en) | Method, device and equipment for training defect classification detection model and storage medium | |
Varadarajan et al. | Spatial mixture of Gaussians for dynamic background modelling | |
Sohn et al. | Automatic powerline scene classification and reconstruction using airborne lidar data | |
CN112183788A (en) | Domain adaptive equipment operation detection system and method | |
WO2004001672A1 (en) | Digital image edge detection and road network tracking method and system | |
CN111259706B (en) | Lane line pressing judgment method and system for vehicle | |
CN112488071B (en) | Method, device, electronic equipment and storage medium for extracting pedestrian features | |
CN111259704B (en) | Training method of dotted lane line endpoint detection model | |
KR102270834B1 (en) | Method and system for recognizing marine object using hyperspectral data | |
CN113037783B (en) | Abnormal behavior detection method and system | |
CN111353496B (en) | Real-time detection method for infrared dim targets | |
Eriksson et al. | Prediction-based load control and balancing for feature extraction in visual sensor networks | |
Funde et al. | Object detection and tracking approaches for video surveillance over camera network | |
CN113284144A (en) | Tunnel detection method and device based on unmanned aerial vehicle | |
CN112164093A (en) | Automatic person tracking method based on edge features and related filtering | |
CN114048546A (en) | Graph convolution network and unsupervised domain self-adaptive prediction method for residual service life of aircraft engine | |
Jaw et al. | Wind adaptive modeling of transmission lines using minimum description length | |
CN115761647A (en) | Computer vision-based method and system for detecting motion trail of object in transformer substation | |
CN112069997B (en) | Unmanned aerial vehicle autonomous landing target extraction method and device based on DenseHR-Net | |
CN113920254A (en) | Monocular RGB (Red Green blue) -based indoor three-dimensional reconstruction method and system thereof | |
JP2001307104A (en) | Object extraction device for moving image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |