CN114663672B - Method and system for detecting corrosion of steel member of power transmission line tower - Google Patents
Method and system for detecting corrosion of steel member of power transmission line tower Download PDFInfo
- Publication number
- CN114663672B CN114663672B CN202210243912.6A CN202210243912A CN114663672B CN 114663672 B CN114663672 B CN 114663672B CN 202210243912 A CN202210243912 A CN 202210243912A CN 114663672 B CN114663672 B CN 114663672B
- Authority
- CN
- China
- Prior art keywords
- image
- corrosion
- steel member
- transmission line
- power transmission
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- 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
Abstract
The invention discloses a method and a system for detecting corrosion of a steel member of a power transmission line tower, which comprises the following steps of firstly, obtaining an image of the steel member of the power transmission line tower; then, acquiring a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the two images; then, accurate mapping between the corrosion area and the corresponding linear angle steel is realized; then, acquiring the corrosion degree of the image of the steel member of the power transmission line tower; and finally, analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image, and giving an evaluation state, a corrosion grade and a treatment measure. The method can effectively improve the efficiency and accuracy of corrosion detection on the steel members of the power transmission line tower, thereby reducing safety accidents and economic losses caused by corrosion and having important significance on safe operation of a power grid.
Description
Technical Field
The invention belongs to the technical field of defect diagnosis of electric power systems, relates to defect diagnosis of a power transmission line tower steel member, and particularly relates to a corrosion detection technology of the power transmission line tower steel member based on machine vision.
Background
As cities develop and electrical loads grow, more and more transmission towers are out of the ground and are increasingly used for steel members.
The detection, repair and safety problems of electric tower steel components brought by the large-scale construction and use of the power transmission tower cannot be ignored. Most of power transmission towers are in severe working environments, for example, power transmission towers located in the field are greatly influenced by environmental changes, power transmission towers in industrial areas are easily corroded by emissions, and the like, and the power transmission towers must be timely and effectively maintained and protected, otherwise, sudden structural damage may occur due to corrosion, for example, steel members fail to break due to corrosion penetration, bolts fail to corrode, tower foot corrosion breaks due to corrosion, and the like, so that the stability of a power transmission line is seriously influenced, and even collapse, electric wire breakage and other major safety accidents threatening the safety and ecological environment of surrounding residents occur. Therefore, an electric tower anti-corrosion maintenance plan is formulated, all-dimensional detection of corrosion failure conditions of electric tower steel members of the power transmission line is regularly performed, effective protective measures are taken according to detection results, and the method has great significance for guaranteeing safe work of the power transmission line and stable operation of a power grid.
The traditional methods for detecting corrosion of steel members comprise an alternating current impedance test method, a magnetic flux leakage detection method, an eddy current detection method, a ray detection method, an ultrasonic detection method and the like, and detection is mainly realized based on the emission and the reception of signals such as electromagnetic waves, sound waves and the like. The technology can carry out preliminary analysis on the corrosion condition, but cannot acquire an image which can be directly observed, and also cannot judge and analyze the corrosion condition of the steel member and determine the corrosion grade through an actual image. This disadvantage directly leads to the cleavage of the detection theory and the practical operation.
The computer vision and digital image processing technology with higher intuition and accuracy is applied to actual measurement, and the developed intelligent steel member corrosion identification technology based on the computer vision is beneficial to solving the technical pain points. In the actual application of technique, the collection of steel member corruption image generally relies on professional unmanned aerial vehicle pilot to follow certain requirement of patrolling the line, uses unmanned aerial vehicle to carry on the camera and shoots the transmission line and obtain, and a large amount of images and the video of gathering need be by professional engineering technical personnel analysis and evaluation. The difficulties in this process are: the operation that unmanned aerial vehicle gathered the image near the electricity tower is extremely dangerous, has very high requirement to pilot's operating technique, and simultaneously, too huge image data volume has increaseed the work load of manual analysis review, and extension detection cycle probably leads to can't be to the emergence of timely effectual emergent anticorrosive treatment of carrying on of transmission line shaft tower, lead to the incident. In addition, the traditional image detection adopted by the existing corrosion identification detection technology is based on the extraction of edge and contour characteristics of a target shape, and sliding detection is performed by using sliding windows with different sizes, but the method is not applicable to the analysis of corrosion images containing a plurality of steel members in a power transmission line tower.
In conclusion, in the process of urban high-speed and modern construction, the erection scale of the power transmission line is enlarged, the working environment is more complex, serious technical pain points exist in the existing traditional steel component corrosion detection technology and scheme in the market, the identification efficiency and the detection precision cannot meet the increasing market demand, an intelligent corrosion detection method suitable for power transmission line tower surface corrosion detection is urgently needed to be developed, the efficiency and the precision of corrosion detection are effectively improved based on the application of deep learning image processing technology and intelligent visual equipment, the comprehensive detection is regularly carried out on the corrosion condition of the power tower by combining with an anticorrosion maintenance plan, treatment measure suggestions are given, the service life of the power transmission line tower is prolonged, and safety accidents and economic losses caused by corrosion are reduced.
Disclosure of Invention
The invention aims to provide a method for detecting corrosion of a steel member of a power transmission line tower, aiming at the problems of low efficiency, non-visual display effect, high complexity and the like of the detection of the corrosion of the steel member on the power transmission line tower, which can realize the high-efficiency and high-precision detection of the corrosion degree of the steel member of the power transmission line tower and can further provide a processing suggestion.
The invention also aims to provide a corrosion detection system for the steel member of the power transmission line tower.
In order to achieve the purpose, the invention provides a method for detecting corrosion of a steel member of a power transmission line tower, which comprises the following steps:
s1, acquiring an image of a steel member of a power transmission line tower;
s2, processing the acquired electric transmission line tower steel member image to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the steel member mask image and the corrosion area mask image; the method comprises the following steps:
s21, processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the steel member without a background at the position of the steel member and an image of a corrosion area without a background at the corrosion position of the steel member;
s22, thresholding is respectively carried out on the image without the background steel member and the image without the background corrosion area, the gradient amplitude and the gradient direction of the image are calculated after filtering, and non-edge pixels are eliminated to obtain edge pixels of the image without the background steel member and the image without the background corrosion area;
s23, obtaining steel member edge lines by using edge pixels of the background-free steel member image, clustering the steel member edge lines according to the slope of the steel member edge lines to obtain two parallel lines of each linear angle steel forming the steel member, and forming effective parallel line pairs;
s24, constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image without the background corrosion area; matching and fusing the steel member mask image and the corrosion area mask image;
s3, traversing the effective parallel line pairs, selecting a linear angle steel mask image and a corresponding corrosion area mask image, wherein the number of corrosion pixels is greater than a threshold value of the number of corrosion pixels, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto an electric transmission line tower steel member image to realize accurate mapping of the corrosion area and the corresponding linear angle steel;
s4, obtaining the corrosion degree of each straight angle steel, and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower;
and S5, analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
In the step S1, acquiring images of the steel members of the power transmission line tower from top to bottom or from bottom to top along the power transmission line tower by using a 360-degree panoramic camera; and further carrying out preliminary analysis on the collected images, carrying out close-range or even local image collection on the steel member position with corrosion phenomenon, and obtaining the power transmission line tower steel member image for carrying out corrosion analysis on the power transmission line tower steel member. This with can effectively reduce the process work load of shooting process in the tower pole detects, improve detection efficiency.
In the step S2, the steel member image and the corrosion area image are analyzed respectively, respective mask images are constructed, and then the steel member image and the corrosion area image are matched and fused, so that the accuracy of boundary segmentation is improved.
In the step S21, the image is divided by using a conventional image segmentation method, and the position of the steel member and the position of the corrosion region on the steel member are represented by forming a pixel geometry, so as to obtain a background-free steel member image and a background-free corrosion region image of the corrosion position on the steel member. Furthermore, the method comprises the steps of firstly, intelligently dividing the steel component of the power transmission line tower by using a conventional image segmentation method (such as a grid method YOLO-V4 or DeeplabV3+ and the like) which is widely used in the field, deducting a background part, only reserving the structural part of the steel component to obtain a background-free steel component image, and then, intelligently dividing the steel component by using a conventional image segmentation method (such as a grid method YOLO-V4 or DeeplabV3+ and the like) to divide a corrosion region part in the steel component of the iron tower to obtain a background-free corrosion region image.
In the step S22, a self-adaptive threshold algorithm is adopted to perform thresholding on the image without the background steel member and the image without the background corrosion region, and then a normalized mean filter is used to process the image after thresholding so as to reduce the noise of the image; calculating gradient amplitude and direction of the denoised image, eliminating non-edge pixels in the image based on the calculation result, and reserving a plurality of thin lines as candidate edges; and finally, reserving the pixels with the gradient amplitudes exceeding the set amplitude threshold value as edge pixels.
In the step S23, a majority voting algorithm is used to detect edge pixels to obtain edge lines of the steel member, the edge lines of the steel member are screened out in the form of angles in a polar coordinate system to obtain lines with a slope of 0 (i.e., a vertical line) and a slope of k (i.e., a slope deviated from the vertical direction), and then a mean shift algorithm is used to group two parallel lines with the same inclination angle into one class, so as to obtain two parallel lines of a straight angle iron forming the steel member and form a parallel line pair; by analogy, the parallel line pairs of each straight angle steel can be obtained. And further comparing the parallel line pair of each linear angle steel with the edge of the corrosion image obtained by the edge pixel of the image without the background corrosion area, and finding out an effective parallel line pair, namely obtaining the linear angle steel related to the corrosion area.
In the step S24, a linear angle steel mask threshold and a corrosion area mask threshold are set, and a steel member mask image and a corrosion area mask image including linear angle steel are respectively drawn according to the found effective parallel line pairs and edge pixel points of the internal corresponding corrosion area image. And drawing the mask image of the corrosion area on the mask image of the steel member to realize matching fusion of the mask image of the corrosion area and the mask image of the steel member.
In the step S3, the effective parallel line pairs are traversed, only the linear angle steel (including the fused corrosion area mask image) with the corrosion pixel number larger than the corrosion pixel number threshold is selected on the steel member mask image, the linear angle steel is cut, the cutting result of each single linear angle steel is drawn on the electric transmission line tower steel member image as the linear angle steel mask matching result, the accurate mapping of the corrosion part and the steel member is realized, and the linear angle steel mask image (including the fused corrosion area mask image) of the single structure is used as an analysis object.
In the step S4, the number of pixels in the corrosion area (i.e., the corrosion area) in the linear angle steel mask image of the single structure is divided by the number of pixels in the linear angle steel mask image, and the number is used as the corrosion degree of each linear angle steel, and the highest corrosion degree of each linear angle steel is selected as the corrosion degree of the power transmission line tower steel member image.
In the step S5, the corrosion degree grade standard is constructed based on the power transmission line tower trunk, the corrosion degree of the power transmission line tower steel member image obtained in the step S4 is combined, the corrosion grade of the current power transmission line tower steel member is determined, and the current power transmission line tower steel member image is processed according to the given corresponding measures.
The grade standard of the corrosion degree of the power transmission line tower trunk construction can be set according to past experience, and relevant standards include DL/T2055-2019 guide rules for evaluating corrosion safety of power transmission line steel structures, DL/T1424-2015 Power grid Metal technology supervision regulations, DL/T1424-2015 anticorrosion protection coating of power transmission line iron towers and the like.
Based on the set grade standard of the corrosion degree of the power transmission line tower dry construction, the corrosion grade degree of the power transmission line tower steel member is divided into 6 grades, including A-grade micro corrosion 0, B-grade weak corrosion ((0, 0.001%), C-grade light corrosion ((0.001%, 3%), D-grade medium corrosion ((3%, 10%), E-grade heavy corrosion ((10%, 33%), F-grade extremely heavy corrosion ((33, 100%)), the corresponding treatment measures of each grade are as follows:
the evaluation state of the A level is normal, and the processing measure is 'no processing is needed and can be continuously used';
the B-level evaluation state is normal, and the given treatment measures are 'treatment is not needed temporarily, and attention can be paid during patrol';
c-level evaluation state 'attention', giving processing measures of 'monitoring use is needed', and attention is needed during patrol; the maintenance plan for performing anticorrosive coating construction within 3 years is preferably arranged, and the corrosion prevention can be conditionally developed in advance; the corrosion evaluation period is shortened to 1 year';
the grade D evaluation state is abnormal, and the processing measure is given as that the important attention is paid when the patrol is performed; the maintenance plan of anticorrosive coating construction within 2 years is preferably arranged, and the corrosion prevention can be carried out conditionally as soon as possible. The corrosion evaluation period is shortened to 1 year';
the grade E evaluation state is abnormal, the processing measure is given as 'further corrosion measurement should be carried out by tower climbing inspection', and a sampling mechanical test is carried out by combining power failure maintenance; the replacement condition is achieved as soon as possible, and the maintenance plan of the anticorrosion coating construction within 1 year is arranged when the replacement condition is not achieved; the corrosion evaluation period is shortened to 1 year';
the F-level evaluation state is 'severe', and the treatment measures are 'further corrosion measurement should be carried out immediately', and whether replacement is needed or not is evaluated; the replacement condition is achieved as soon as possible, and an anti-corrosion maintenance plan is arranged immediately when the replacement condition is not achieved; before the updating and the reconstruction are completed, the corrosion evaluation period is shortened to be less than half a year.
The invention further provides a corrosion detection system for the steel member of the power transmission line tower, which comprises an image acquisition unit and a main controller; the main controller comprises an image processing unit, an image mapping unit, a calculating unit and a data analyzing unit;
the image acquisition unit is used for acquiring an image of the power transmission line tower steel member;
the image processing unit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the two images; the unit comprises the following subunits:
the image segmentation subunit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the background-free steel member at the position of the steel member and an image of a background-free corrosion area at the corrosion position on the steel member;
the first image analysis subunit is used for thresholding the image without the background steel member and the image without the background corrosion area respectively, calculating the gradient amplitude and direction of the image after filtering, and eliminating non-edge pixels to obtain edge pixels of the image without the background steel member and the image without the background corrosion area;
the second image analysis subunit is used for obtaining edge lines of the steel member by using edge pixels of the background-free steel member image, classifying the edge lines of the steel member according to the slope of the edge lines of the steel member to obtain two parallel lines of each straight angle steel forming the steel member and form effective parallel line pairs;
the image matching subunit is used for constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image without the background corrosion area; matching and fusing the steel member mask image and the corrosion area mask image;
the image mapping unit is used for traversing the effective parallel line pairs, selecting the linear angle steel mask image with the number of corrosion pixels larger than the threshold value of the number of corrosion pixels and the corresponding corrosion area mask image, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto the power transmission line tower steel member image to realize accurate mapping of the corrosion area and the corresponding linear angle steel;
the calculation unit is used for acquiring the corrosion degree of each straight angle steel and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower;
and the data analysis unit is used for analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
The image acquisition unit can be set as a shooting cloud platform (for example, a 360-degree panoramic camera), the shooting cloud platform is installed on a transmission mechanism arranged between the top of a tower and the top of the tower, and the transmission mechanism drives the shooting cloud platform to collect images of the steel member of the power transmission line tower from top to bottom or from bottom to top along the power transmission line tower; and further carrying out preliminary analysis on the collected images, carrying out close-range or even local image collection on the steel member position with corrosion phenomenon, and obtaining the power transmission line tower steel member image for carrying out corrosion analysis on the power transmission line tower steel member. The image acquisition unit acquires the tower number, the line to which the tower belongs and the tower name corresponding to the image of the power transmission line tower steel member, in addition to the acquired image of the power transmission line tower steel member. The image of the power transmission line tower steel member is acquired through the image acquisition unit.
The application provides a transmission line tower steel member corrosion detection system still includes database unit, data display element and backend server. The database unit is used for storing the grade standard of the corrosion degree of the steel member of the power transmission line tower and corresponding treatment measures. And the data display unit is used for displaying the processing result of the main controller. The background server is connected with the user terminal and used for sending the operation instruction from the user terminal to the image acquisition unit and the main controller, and the background server is used for storing the image of the power transmission line tower steel member acquired by the image acquisition unit.
In the invention, the master controller, the database unit and the background server can be integrated on the computer, wherein the database unit can also be set as a cloud storage. The master controller and the background server can be integrated on one computer or different computers. The data display unit may employ a display connected to a computer.
The method and the system for detecting the corrosion of the steel member of the power transmission line tower have the following beneficial effects that:
1. the method comprises the steps of firstly segmenting a steel member and an image of a corrosion area, respectively carrying out edge detection and positioning, then carrying out matching and mapping, further fusing bottom layer characteristics (such as the outline of the corrosion area) and high-layer characteristics (such as a steel member of a power transmission line tower), analyzing the corrosion degree based on the obtained mapping image, and giving out evaluation states, corrosion grades and treatment measures; the invention improves the accuracy of the segmentation boundary, thereby effectively improving the efficiency and the accuracy of the corrosion detection of the steel members of the power transmission line tower, reducing safety accidents and economic losses caused by corrosion and having important significance on the safe operation of a power grid.
2. The method is based on a self-adaptive threshold method and a contour extraction algorithm, and has completed the identification modeling of the corrosion degree and the identification of the corrosion category based on the standard image of the corrosion degree adopted in each large standard, and the identification accuracy is obviously higher than that of the traditional corrosion identification method.
3. The method has wider application scene and can be applied to almost all iron towers formed by steel components.
Drawings
FIG. 1 is a schematic flow chart of a method for detecting corrosion of a steel member of a power transmission line tower provided by the invention.
Fig. 2 is a schematic flow chart of step S2.
Fig. 3 is an original image of a steel member of a power transmission line tower photographed in an embodiment of the present invention.
Fig. 4 is an image of a non-background steel member divided in the embodiment of the present invention.
FIG. 5 is an image of a steel member mask after treatment in an embodiment of the present invention.
FIG. 6 is an image of a region without background erosion according to an embodiment of the present invention.
FIG. 7 is a mask image of an etched region after treatment in an embodiment of the invention.
FIG. 8 shows the corrosion results of a steel member in an example of the invention.
FIG. 9 is a schematic diagram of a corrosion detection system framework for a steel member of a power transmission line tower provided by the invention.
FIG. 10 is a schematic diagram of the master controller.
Detailed Description
The following provides an embodiment of the present invention with reference to the accompanying drawings, and further describes and explains the technical solution of the present invention in detail through the embodiment. The embodiments described below are only a part of the embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the disclosure of the invention without any inventive step, are within the scope of the invention.
Example 1
The method for detecting corrosion of a steel member of a power transmission line tower provided by the embodiment is shown in fig. 1 and 2, and comprises the following steps:
s1, acquiring an image of the steel member of the power transmission line tower.
The embodiment takes a typical power transmission line tower consisting of steel structure rods as a target.
In the embodiment, a360 ° panoramic camera (for example, an instata 360 camera or a google gopro series camera) is used as a shooting cloud platform to perform image acquisition on the transmission line tower. The 360-degree panoramic camera is installed on a transmission mechanism arranged between the top of the tower and the top of the tower, and the transmission mechanism drives the 360-degree panoramic camera to collect images of the steel members of the power transmission line tower from top to bottom or from bottom to top along the power transmission line tower.
Due to the fact that the size of the power transmission line tower is huge, the corrosion condition and the clear image of the whole iron tower are difficult to obtain through single shooting. The shooting process adopted in this embodiment is a progressive mode of "long shot view-short shot view-local view", so as to obtain the required images with different viewing angles and distances.
In order to reduce the workload of the shooting process, the detection efficiency is improved. An operator can firstly shoot a long-range image of the power transmission line tower from bottom to top by using the 360-degree panoramic camera, corrode and preliminarily position the steel member according to the long-range image, move the 360-degree panoramic camera to the height of the power transmission line tower where the corrosion position is located, adjust the amplification factor of the 360-degree panoramic camera, collect a close-range image or/and a locally amplified image of the corresponding position, and use the two images as images of the power transmission line tower steel member for analyzing the specific corrosion condition.
The image of the power transmission line tower steel member obtained by the step is shown in fig. 3.
S2, processing the acquired electric transmission line tower steel member image to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the steel member mask image and the corrosion area mask image; the method comprises the following steps:
s21, processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the non-background steel member of the steel member position and an image of a non-background corrosion area of the corrosion position on the steel member.
In the step, firstly, a conventional image segmentation method DeeplabV3+ algorithm widely used in the field is used for intelligently segmenting the steel members of the power transmission line tower, the background part is deducted, only the structural part of the steel member is reserved to obtain a background-free steel member image, then the DeeplabV3+ algorithm is further used for intelligently segmenting the steel members, and the corrosion region part in the steel structure of the iron tower is segmented to obtain a background-free corrosion region image.
The model constructed by the Deeplab V3+ algorithm is obtained by training a corresponding electric transmission line tower steel member image training set or a steel member image training set containing a corrosion area.
The background-free steel member image and the background corrosion-free area image obtained by processing the power transmission line tower steel member image given in fig. 1 through the present step are shown in fig. 4 and 6.
S22, thresholding is carried out on the image without the background steel member and the image without the background corrosion area respectively, the gradient amplitude and the gradient direction of the image are calculated after filtering, and non-edge pixels are eliminated to obtain the edge pixels of the image without the background steel member and the image without the background corrosion area.
In the step, a self-adaptive threshold algorithm is adopted to perform thresholding treatment on the image without the background steel member and the image without the background corrosion area respectively, and then a normalized mean filter is used to treat the image after thresholding treatment, so that the image noise is reduced; calculating gradient amplitude and direction of the denoised image by using a canny edge detection method, removing non-edge pixels in the image based on a calculation result, and reserving a plurality of thin lines as candidate edges; and finally, reserving the pixels with the gradient amplitude values exceeding the set amplitude threshold value as edge pixels.
S23, obtaining edge lines of the steel member by using edge pixels of the background-free steel member image, classifying the edge lines of the steel member according to the slope of the edge lines of the steel member, obtaining two parallel lines of each straight angle steel forming the steel member, and forming effective parallel line pairs.
In the step, a conventional majority voting algorithm is adopted to detect edge pixels to obtain edge lines of the steel member, the edge lines of the steel member are screened out to be straight lines with a slope of 0 (namely a vertical straight line) and a slope of k (namely a slope deviated from the vertical direction) in an angle mode in a polar coordinate system, then a mean shift algorithm is adopted to gather two parallel lines with the same slope angle into one class, and two parallel lines of a straight line angle steel forming the steel member are obtained and form a parallel line pair (refer to Jinyanwei, and the research of an object-oriented graph theory image segmentation algorithm based on the mean shift, resource information and engineering are carried out in volume 3 of 2019, volume 34); by analogy, the parallel line pairs of each straight angle steel can be obtained. And further comparing the parallel line pair of each linear angle steel with the edge of the corrosion image obtained by the edge pixel of the image without the background corrosion area, and finding out an effective parallel line pair, namely obtaining the linear angle steel related to the corrosion area.
S24, constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image of the background-free corrosion area; and matching and fusing the steel member mask image and the corrosion area mask image.
In the step, a linear angle steel mask threshold value and a corrosion area mask threshold value are set, and a steel member mask image and a corrosion area mask image containing linear angle steel are respectively drawn according to the found effective parallel line pairs and edge pixel points of the internal corresponding corrosion area image. And drawing the mask image of the corrosion area on the mask image of the steel member to realize matching fusion of the mask image of the corrosion area and the mask image of the steel member.
Fig. 5 and 7 show a steel member mask image and an erosion area mask image including straight angle steel obtained by processing the background-free steel member image and the gum-free erosion area image in step S21 in steps S22 to S24.
And S3, traversing the effective parallel line pairs, selecting the linear angle steel mask image with the number of corrosion pixels larger than the threshold value of the number of corrosion pixels and the corresponding corrosion area mask image, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto the power transmission line tower steel member image to realize accurate mapping of the corrosion area and the corresponding linear angle steel.
In the step, effective parallel line pairs are traversed, only linear angle steel (containing fused corrosion area mask images) with the number of corrosion pixels larger than a corrosion pixel number threshold value is selected on the steel member mask image, the linear angle steel is cut, the cutting result of each single linear angle steel is drawn on the power transmission line tower steel member image as a linear angle steel mask matching result, accurate mapping of corrosion parts and the steel members is achieved, and the linear angle steel mask image (containing the fused corrosion area mask images) of the single structure is used as an analysis object.
The result of drawing the linear angle steel mask matching result on the electric transmission line tower steel member image through the processing of the step S3 is shown in fig. 8. The corrosion area and the corrosion condition of the steel member of the tower pole of the power transmission line can be visually seen from fig. 8.
And S4, acquiring the corrosion degree of each linear angle steel, and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower.
In the step, the number of pixels in a corrosion area on the linear angle steel mask image with a single structure is divided from the number of pixels on the linear angle steel mask image, the corrosion degree of each linear angle steel is used, and the highest corrosion degree of a single linear angle steel is selected as the corrosion degree of the image of the steel member of the power transmission line tower.
According to the step, the corrosion degree of the single straight angle steel given in fig. 8 is calculated to obtain the corrosion degree of 20.91%. And the corrosion degree of the steel member image of the power transmission line tower given in figure 3 is taken as the corrosion degree.
And S5, analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
The grade standard of the corrosion degree of the power transmission line tower dry construction is given, namely, the corrosion grade degree of the power transmission line tower steel member is divided into 6 grades, including grade A micro corrosion 0, grade B weak corrosion ((0, 0.001%), grade C light corrosion ((0.001%, 3%), grade D medium corrosion ((3%, 10%), grade E heavy corrosion ((10%, 33%), and grade F extremely heavy corrosion ((33, 100%).
And (4) based on the power transmission line tower trunk construction corrosion degree grade standard, determining the corrosion grade of the current power transmission line tower steel member by combining the corrosion degree of the power transmission line tower steel member image obtained in the step (S4), and processing according to the given corresponding measures.
In this embodiment, based on the corrosion degree of the image of the power transmission line tower steel member calculated in step S4, it may be determined that the power transmission line tower steel member belongs to the E-level heavy corrosion, and the processing is further performed according to the processing measure corresponding to the level of corrosion.
Example 2
The system for detecting corrosion of a steel member of a power transmission line tower provided by the embodiment mainly comprises an image acquisition unit and a main controller in communication connection with the image acquisition unit as shown in fig. 9. The main controller is mainly an analysis device for performing intelligent processing and analysis according to the method given in embodiment 1 based on the image data acquired by the image acquisition unit.
Image acquisition unit
The image acquisition unit is used for acquiring an image of the steel member of the power transmission line tower. The image acquisition unit may be provided as a360 ° panoramic camera (e.g., an insta360 camera or google gopro series camera).
The 360-degree panoramic camera can be arranged on a monitor body through the clamping device, so that the iron tower can be flexibly shot in a multi-angle clear image flexibly.
The 360-degree panoramic camera can also be arranged on a transmission mechanism arranged between the top of the tower and the top of the tower, and the transmission mechanism drives the 360-degree panoramic camera to collect images of the steel members of the power transmission line tower from top to bottom or from bottom to top along the power transmission line tower.
(II) master controller
The master controller may be integrated in a computer having certain image processing capabilities. The main controller is mainly a software program aiming at the corrosion detection of the iron tower.
The quantitative analysis of the corrosion image is a problem to be mainly solved by the invention, and the main controller mainly realizes the analysis process in the embodiment, which comprises the following steps: 1) Acquiring a straight angle iron edge parallel line pair; 2) Matching the image of the corrosion area with the image of the straight angle steel; 3) Calculating the corrosion degree; 4) And matching corrosion grades, grading evaluation and giving preliminary treatment measures.
Based on the above analysis, as shown in fig. 10, the master includes an image processing unit, an image mapping unit, a calculation unit, and a data analysis unit.
The image processing unit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the two images; the unit comprises the following subunits:
the image segmentation subunit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the background-free steel member at the position of the steel member and an image of a background-free corrosion area at the corrosion position on the steel member;
the first image analysis subunit is used for thresholding the image without the background steel member and the image without the background corrosion area respectively, calculating the gradient amplitude and direction of the image after filtering, and eliminating non-edge pixels to obtain edge pixels of the image without the background steel member and the image without the background corrosion area;
the second image analysis subunit is used for obtaining edge lines of the steel member by using edge pixels of the background-free steel member image, classifying the edge lines of the steel member according to the slope of the edge lines of the steel member to obtain two parallel lines of each straight angle steel forming the steel member and form effective parallel line pairs;
the image matching subunit is used for constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image without the background corrosion area; matching and fusing the steel member mask image and the corrosion area mask image;
the image mapping unit is used for traversing the effective parallel line pairs, selecting the linear angle steel mask image with the number of corrosion pixels larger than the threshold value of the number of corrosion pixels and the corresponding corrosion area mask image, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto the power transmission line tower steel member image to realize accurate mapping of the corrosion area and the corresponding linear angle steel;
the calculation unit is used for acquiring the corrosion degree of each straight angle steel and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower;
and the data analysis unit is used for analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
The corrosion detection system for the steel member of the power transmission line tower provided by the embodiment further comprises a database unit, a data display unit and a background server, as shown in fig. 9. The database unit is used for storing the corrosion degree grade standard of the power transmission line tower construction and corresponding processing measures. The data display unit is used for displaying the processing result of the main controller. The background server is connected with the user terminal and used for sending the operation instruction from the user terminal to the image acquisition unit and the main controller, and the background server is used for storing the image of the power transmission line tower steel member acquired by the image acquisition unit.
In a specific use, the master controller, the database unit and the background server can be integrated on a computer, wherein the database unit can also be set as a cloud storage. The master controller and the background server can be integrated on one computer or different computers. The data display unit may employ a display connected to a computer.
The system for detecting corrosion of a power transmission line tower steel member provided by the embodiment is mainly used for detecting the power transmission line tower steel according to the method for detecting corrosion of the power transmission line tower steel member provided by the embodiment 1. Attention needs to be paid to the specific detection process:
(1) Image acquisition: because the power transmission and transformation steel structural member is huge in size, the corrosion condition and clear images of the whole iron tower are difficult to obtain through single shooting, and certain standards and sequences are needed. The shooting process adopted in this embodiment is a progressive mode of "long shot view-short shot view-local view", so as to obtain the required images with different viewing angles and distances.
(2) Coarse positioning: and carrying out segmentation processing on the shot image, and carrying out corrosion positioning identification. When the distance between a shooting person and the tower is far, the erosion area is roughly positioned by using the long-range image.
(3) Quantitatively identifying the corrosion of the close-range image: after the rough positioning result of the corroded part is obtained, the corresponding close-range image or local image is further input into the main controller of the embodiment to carry out quantitative corrosion identification on the close-range image or local image.
(4) Matching steel members: the quantitative analysis object of the corrosion degree is a single straight angle steel, so that a certain image matching method is needed to realize accurate mapping of a corrosion part and a steel member of the corrosion part according to a segmentation result of the iron tower.
(5) Grading evaluation: according to the obtained segmentation results of the single straight angle steel image and the corroded area image of the corroded part, the corrosion degree of the steel member can be calculated according to a pixel counting method, and a grading evaluation result of the steel member is given according to the grade standard of the related corrosion degree.
The PC end detection software page designed based on the power transmission line tower steel member corrosion detection system provided by the invention basically comprises the following components:
(1) Image display area of the power transmission line tower: namely, images which need to be displayed in the detection process, such as an original image, a detection process image, a corrosion result image and the like;
(2) Basic information display area of the tower: the region can provide the filling of the tower geographic position information for operators, comprises an administrative region and a tower line, and is convenient for image storage;
(3) Corrosion area calculation setup area: the area completes the calculation and setting of the corrosion area of the standard corrosion area map and basic control buttons (operation, reset and storage);
(4) Identifying a calculation result display area: the main function of the area is to display a detection result, and establish a corresponding corrosion degree standard and a corresponding implementation suggestion according to DLT 2055-2019 transmission line steel structure corrosion safety assessment guide rule.
Images of various visual angles and distances shot by a shooting process can be selected on a User Interface (UI) of software, corrosion area calculation setting is carried out, then a main controller processes image data, the corrosion degree of a corrosion area in a steel member image is calculated, and a certain treatment measure suggestion is given.
In conclusion, the method and the system for detecting the corrosion of the steel member of the power transmission line tower provided by the invention have the advantages that the used structure and the network algorithm are independently researched and developed, the scheme is novel, and the method for intelligently detecting the corrosion of the iron tower by using the image is provided to a certain extent. The method can replace manual work to carry out efficient, intelligent and accurate detection, and can avoid various safety accidents caused by corrosion to a certain extent. Meanwhile, compared with the traditional ultrasonic detection method, the eddy current detection method, the manual detection method and the like, the image analysis method provided by the invention has the advantages of higher accuracy, higher safety, higher efficiency and the like.
The above-mentioned case is only an illustration of the present invention, and does not constitute a limitation to the scope of the present invention, and all designs identical or similar to the present invention are within the scope of the present invention.
Claims (9)
1. The method for detecting corrosion of the steel member of the power transmission line tower is characterized by comprising the following steps of:
s1, acquiring an image of a steel member of a power transmission line tower;
s2, processing the acquired electric transmission line tower steel member image to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the steel member mask image and the corrosion area mask image; the method comprises the following steps:
s21, processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the steel member without a background at the position of the steel member and an image of a corrosion area without a background at the corrosion position of the steel member;
s22, thresholding is respectively carried out on the image without the background steel member and the image without the background corrosion area, the gradient amplitude and the gradient direction of the image are calculated after filtering, and non-edge pixels are eliminated to obtain edge pixels of the image without the background steel member and the image without the background corrosion area;
s23, obtaining steel member edge lines by using edge pixels of the background-free steel member image, clustering the steel member edge lines according to the slope of the steel member edge lines to obtain two parallel lines of each linear angle steel forming the steel member, and forming effective parallel line pairs;
s24, constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image without the background corrosion area; matching and fusing the steel member mask image and the corrosion area mask image;
s3, traversing the effective parallel line pairs, selecting a linear angle steel mask image and a corresponding corrosion area mask image, wherein the number of corrosion pixels is greater than a corrosion pixel number threshold value, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto a power transmission line tower steel member image to realize accurate mapping of a corrosion area and the corresponding linear angle steel;
s4, obtaining the corrosion degree of each straight angle steel, and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower;
and S5, analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
2. The method for detecting corrosion of a steel member of a power transmission line tower according to claim 1, wherein in step S21, the steel member of the power transmission line tower is divided by using a grid method YOLO-V4 or Deeplab V3+ algorithm to obtain an image of the steel member without a background, and then the steel member is further divided by using the grid method YOLO-V4 or Deeplab V3+ algorithm to obtain an image of a region without the background corrosion.
3. The method for detecting corrosion of a steel member of an electric transmission line tower according to claim 1 or 2, characterized in that in step S22, an adaptive threshold algorithm is adopted to perform thresholding on an image of a background-free steel member and an image of a background-free corrosion area respectively, and then a normalized mean filter is used to process the thresholded image so as to reduce image noise; then, gradient amplitude and direction are calculated for the image after noise reduction, and non-edge pixels in the image are eliminated based on the calculation result; and finally, reserving the pixels with the gradient amplitude values exceeding the set amplitude threshold value as edge pixels.
4. The method for detecting corrosion of a steel member of a power transmission line tower according to claim 3, wherein in the step S23, a majority voting algorithm is adopted to detect edge pixels to obtain an edge line of the steel member, the edge line of the steel member is screened out to be a straight line with a slope of 0 and a slope of k in a polar coordinate system in an angle form, and then a mean shift algorithm is adopted to gather two parallel lines with the same inclination angle into one class, so that two parallel lines of a straight angle steel forming the steel member are obtained and form a parallel line pair; by analogy, the parallel line pairs of each straight angle steel can be obtained; and comparing the parallel line pair of each linear angle steel with the edge of the corrosion image obtained by the edge pixel of the image without the background corrosion area, and finding out an effective parallel line pair, namely obtaining the linear angle steel related to the corrosion area.
5. The method for detecting corrosion of a steel member of a power transmission line tower according to claim 4, wherein in step S24, a linear angle steel mask threshold value and a corrosion region mask threshold value are set, and a steel member mask image and a corrosion region mask image containing linear angle steel are respectively drawn according to the found effective parallel line pairs and pixels of the internal corresponding corrosion region image; and drawing the mask image of the corrosion area on the mask image of the steel member to realize matching fusion of the mask image of the corrosion area and the mask image of the steel member.
6. The method for detecting corrosion of a steel member of an electric transmission line tower according to claim 1, wherein in step S4, the number of pixels in a corrosion area on a mask image of a single-structure straight angle steel is divided by the number of pixels on the mask image of the single-structure straight angle steel, and the corrosion degree of each straight angle steel is determined, and the highest corrosion degree of the single straight angle steel is selected as the corrosion degree of the steel member image of the electric transmission line tower.
7. The method for detecting corrosion of a power transmission line tower steel member according to claim 1, wherein in the step S5, according to a set power transmission line tower structure corrosion degree grade standard, the corrosion degree of the power transmission line tower steel member image obtained in the step S4 is combined to determine the corrosion degree of the current power transmission line tower steel member, and the current corrosion degree is processed according to corresponding measures.
8. A corrosion detection system for a steel member of a power transmission line tower is characterized by comprising an image acquisition unit and a main controller; the main controller image processing unit, the image mapping unit, the calculating unit and the data analyzing unit;
the image acquisition unit is used for acquiring an image of the steel member of the power transmission line tower;
the image processing unit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain a steel member mask image and a corrosion area mask image corresponding to the position of the steel member, and matching and fusing the two images; the unit comprises the following subunits:
the image segmentation subunit is used for processing the acquired image of the steel member of the power transmission line tower to respectively obtain an image of the background-free steel member at the position of the steel member and an image of a background-free corrosion area at the corrosion position on the steel member;
the first image analysis subunit is used for thresholding the image without the background steel member and the image without the background corrosion area respectively, calculating the gradient amplitude and direction of the image after filtering, and eliminating non-edge pixels to obtain edge pixels of the image without the background steel member and the image without the background corrosion area;
the second image analysis subunit is used for obtaining edge lines of the steel members by using edge pixels of the background-free steel member image, clustering the edge lines of the steel members according to the slope of the edge lines of the steel members to obtain two parallel lines of each straight angle steel forming the steel members and form effective parallel line pairs;
the image matching subunit is used for constructing a steel member mask image and a corrosion area mask image according to the effective parallel line pairs and the edge pixels of the image without the background corrosion area; matching and fusing the steel member mask image and the corrosion area mask image;
the image mapping unit is used for traversing the effective parallel line pairs, selecting the linear angle steel mask image with the number of corrosion pixels larger than the threshold value of the number of corrosion pixels and the corresponding corrosion area mask image, and drawing the linear angle steel mask image and the corresponding corrosion area mask image onto the power transmission line tower steel member image to realize accurate mapping of the corrosion area and the corresponding linear angle steel;
the calculation unit is used for acquiring the corrosion degree of each straight angle steel and taking the highest corrosion degree as the corrosion degree of the image of the steel member of the power transmission line tower;
and the data analysis unit is used for analyzing the corrosion condition of the power transmission line tower steel member according to the corrosion degree of the power transmission line tower steel member image.
9. The system for detecting corrosion of a steel member of an electric transmission line tower according to claim 8, further comprising a database unit, a data display unit and a background server; the database unit is used for storing the grade standard of the corrosion degree of the steel member of the power transmission line tower and corresponding treatment measures; the data display unit is used for displaying the processing result of the main controller; the background server is connected with the user terminal and used for sending the operation instruction from the user terminal to the image acquisition unit and the main controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210243912.6A CN114663672B (en) | 2022-03-11 | 2022-03-11 | Method and system for detecting corrosion of steel member of power transmission line tower |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210243912.6A CN114663672B (en) | 2022-03-11 | 2022-03-11 | Method and system for detecting corrosion of steel member of power transmission line tower |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114663672A CN114663672A (en) | 2022-06-24 |
| CN114663672B true CN114663672B (en) | 2023-04-07 |
Family
ID=82029702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210243912.6A Active CN114663672B (en) | 2022-03-11 | 2022-03-11 | Method and system for detecting corrosion of steel member of power transmission line tower |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114663672B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114486925B (en) * | 2021-12-25 | 2023-10-31 | 国网四川省电力公司电力科学研究院 | Intelligent detection method for corrosion on inner side of power transmission line pole tower based on panoramic information |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104362545A (en) * | 2014-11-20 | 2015-02-18 | 河南送变电工程公司 | Routing inspection air vehicle with multiple rotor wings and electric transmission line hanging method thereof |
| CN105572541A (en) * | 2015-12-07 | 2016-05-11 | 浙江大学 | High-voltage line patrol fault detection method and system based on visual attention mechanism |
| CN109520913A (en) * | 2018-11-22 | 2019-03-26 | 广西电网有限责任公司电力科学研究院 | A kind of appraisal procedure of in-service electric power line pole tower and metal frame etch state |
| CN109544497A (en) * | 2018-11-21 | 2019-03-29 | 国网江苏省电力有限公司扬州供电分公司 | Image interfusion method and electronic equipment for transmission line faultlocating |
| WO2021119739A1 (en) * | 2019-12-17 | 2021-06-24 | Abyss Solutions Pty Ltd | Method and system for detecting physical features of objects |
| CN113409314A (en) * | 2021-08-18 | 2021-09-17 | 南京市特种设备安全监督检验研究院 | Unmanned aerial vehicle visual detection and evaluation method and system for corrosion of high-altitude steel structure |
| WO2021243743A1 (en) * | 2020-06-04 | 2021-12-09 | 青岛理工大学 | Deep convolutional neural network-based submerged oil sonar detection image recognition method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7783433B2 (en) * | 2007-05-22 | 2010-08-24 | Honeywell International Inc. | Automated defect detection of corrosion or cracks using SAFT processed Lamb wave images |
-
2022
- 2022-03-11 CN CN202210243912.6A patent/CN114663672B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104362545A (en) * | 2014-11-20 | 2015-02-18 | 河南送变电工程公司 | Routing inspection air vehicle with multiple rotor wings and electric transmission line hanging method thereof |
| CN105572541A (en) * | 2015-12-07 | 2016-05-11 | 浙江大学 | High-voltage line patrol fault detection method and system based on visual attention mechanism |
| CN109544497A (en) * | 2018-11-21 | 2019-03-29 | 国网江苏省电力有限公司扬州供电分公司 | Image interfusion method and electronic equipment for transmission line faultlocating |
| CN109520913A (en) * | 2018-11-22 | 2019-03-26 | 广西电网有限责任公司电力科学研究院 | A kind of appraisal procedure of in-service electric power line pole tower and metal frame etch state |
| WO2021119739A1 (en) * | 2019-12-17 | 2021-06-24 | Abyss Solutions Pty Ltd | Method and system for detecting physical features of objects |
| WO2021243743A1 (en) * | 2020-06-04 | 2021-12-09 | 青岛理工大学 | Deep convolutional neural network-based submerged oil sonar detection image recognition method |
| CN113409314A (en) * | 2021-08-18 | 2021-09-17 | 南京市特种设备安全监督检验研究院 | Unmanned aerial vehicle visual detection and evaluation method and system for corrosion of high-altitude steel structure |
Non-Patent Citations (4)
| Title |
|---|
| a cnn-based defect inspection method for catenary split pins in high speed railway;junping zhong等;《IEEE transactions on instrumentation and measurement》;第68卷(第8期);2849-2860 * |
| 基于图像处理技术的配电线路导线断股散股缺陷检测;王照等;《电视技术》(第06期);79-83+88 * |
| 基于视觉的输电线路自动监测技术综述;李德海等;《黑龙江电力》(第06期);559-564 * |
| 高压输电线路航拍图像目标边缘检测;赵江曼等;《中国电力》(第02期);27-32 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114663672A (en) | 2022-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108037133B (en) | Intelligent electric power equipment defect identification method and system based on unmanned aerial vehicle inspection image | |
| KR102008973B1 (en) | Apparatus and Method for Detection defect of sewer pipe based on Deep Learning | |
| CN108230344B (en) | Automatic identification method for tunnel water leakage diseases | |
| Schmugge et al. | Crack segmentation by leveraging multiple frames of varying illumination | |
| CN110874596B (en) | Intelligent recognition early warning method and system for instrument | |
| CN111768417B (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
| CN112698303A (en) | Method and system for measuring point cloud tower inclination parameters based on unmanned aerial vehicle laser radar | |
| CN114663672B (en) | Method and system for detecting corrosion of steel member of power transmission line tower | |
| CN113379712A (en) | Steel bridge bolt disease detection method and system based on computer vision | |
| CN108846374B (en) | Power equipment inspection method and device | |
| CN115993096A (en) | High-rise building deformation measuring method | |
| CN113361420A (en) | Mine fire monitoring method, device and equipment based on robot and storage medium | |
| CN116486289A (en) | Gas pipeline high-consequence area identification method driven by multi-source data and knowledge | |
| CN115825984A (en) | Power transmission line inspection system and inspection method based on laser point cloud | |
| CN112455676A (en) | Intelligent monitoring and analyzing system and method for health state of photovoltaic panel | |
| CN116912805B (en) | Well lid abnormity intelligent detection and identification method and system based on unmanned sweeping vehicle | |
| CN109948629A (en) | A kind of GIS equipment X-ray image failure detection method based on SIFT feature | |
| CN114330477B (en) | Power equipment defect detection system and method based on mixed reality equipment | |
| CN116231504A (en) | Remote intelligent inspection method, device and system for booster station | |
| CN114764795A (en) | Crane defect detection method based on unmanned aerial vehicle vision | |
| CN110414606B (en) | Method for determining division number of visual alarm area of power transmission line channel | |
| CN114324401A (en) | Full-coverage type pipeline detection system based on annular multi-beam sonar | |
| CN112198170A (en) | Detection method for identifying water drops in three-dimensional detection of outer surface of seamless steel pipe | |
| CN116109259B (en) | Intelligent auditing method and device for spatial data | |
| CN110688919A (en) | Intelligent line patrol fault identification method |
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 |