CN111814686A - Vision-based power transmission line identification and foreign matter invasion online detection method - Google Patents
Vision-based power transmission line identification and foreign matter invasion online detection method Download PDFInfo
- Publication number
- CN111814686A CN111814686A CN202010657798.2A CN202010657798A CN111814686A CN 111814686 A CN111814686 A CN 111814686A CN 202010657798 A CN202010657798 A CN 202010657798A CN 111814686 A CN111814686 A CN 111814686A
- Authority
- CN
- China
- Prior art keywords
- power transmission
- line
- transmission line
- area
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 164
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 230000009545 invasion Effects 0.000 title description 3
- 230000003287 optical effect Effects 0.000 claims abstract description 46
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000001914 filtration Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000000877 morphologic effect Effects 0.000 claims abstract description 9
- 230000011218 segmentation Effects 0.000 claims abstract description 9
- 230000003595 spectral effect Effects 0.000 claims abstract description 5
- 230000009466 transformation Effects 0.000 claims description 14
- 230000002146 bilateral effect Effects 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 9
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 5
- 150000001875 compounds Chemical class 0.000 claims 1
- 238000007689 inspection Methods 0.000 abstract description 18
- 230000008901 benefit Effects 0.000 abstract description 6
- 238000005286 illumination Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 102000003712 Complement factor B Human genes 0.000 description 1
- 108090000056 Complement factor B Proteins 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000011179 visual inspection Methods 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
- 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
-
- 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
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a vision-based power transmission line identification and foreign object intrusion online detection method, and belongs to the field of image processing and power transmission line detection. The invention preprocesses the image; extracting line segment characteristics by using an EDLines algorithm; classifying the obtained line segment characteristics in an angle interval, extracting parallel line segment groups of the power transmission line, and fusing the parallel line segment groups meeting the preset line segment connection standard; searching and defining a target area of the power transmission line according to the shape and spectral characteristic characteristics of the power transmission line, and marking a suspected area according to the foreign matter characteristics; and calculating the optical flow distribution of the target area of the power transmission line in the adjacent frame images, and performing threshold segmentation, morphological operation and filtering processing on the optical flow images. And comparing the average values of the optical flow velocities of the suspected area and the background area to realize the detection of the foreign matters in the power transmission line. The method has the advantages of no limitation of illumination and complex background, strong robustness, high general use degree, strong real-time performance and suitability for various electric power inspection environments.
Description
Technical Field
The invention relates to a visual-based online detection method for power transmission line identification and foreign object intrusion, in particular to a power transmission line identification and foreign object intrusion online detection method based on an EDLines (edge Drawing lines) and optical flow method combination, and belongs to the field of image processing and power transmission line detection.
Background
In the maintenance work of the power transmission line of the power grid, the analysis of the foreign matter invasion fault of the power transmission line is an important direction. Foreign matters hung on the high-voltage transmission line can cause short-circuit tripping of the transmission line and the transformer substation and the like, great harm is generated, and the safe and stable operation of a power grid is directly influenced.
The traditional manual inspection mode has the problems of high risk, low efficiency and low reliability. Unmanned aerial vehicle patrols and examines relies on its advantage with low costs, efficient, mobility is strong, is applied to the power line field of patrolling and examining gradually.
Unmanned aerial vehicle patrols and examines and gathers power equipment video image through the image equipment that unmanned aerial vehicle carried on, carries out the analysis to video image again, and then monitors the power equipment state so that maintain and repair. However, manually reviewing a large number of images is time and labor consuming. The aerial image is processed by utilizing the computer vision technology, the foreign matter fault of the power transmission line is automatically identified, the inspection efficiency and the inspection intelligent level can be greatly improved, the automatic inspection of the power transmission line fault is effectively promoted, and the method has great significance for maintaining the safe and stable operation of the power transmission line of China.
Foreign object fault detection is often based on the identification of the transmission line. The existing power transmission line identification method based on aerial images mainly considers the influence of factors such as poor image quality and complex shooting background on a linear target, but still has the defects of low accuracy, long time consumption and the like. At present, the line foreign matter detection technology mainly comprises the following three types: one is to segment the image and analyze the edge number, the surrounding abnormal point rate and other line features in the transmission line area identified by Hough transform to judge the existence of foreign matters. The technology has low operation cost, but is easily influenced by factors such as illumination, complex background and the like, so that false detection and missed detection are caused; one is a method combining moving object detection and feature point tracking. However, the problem of large calculation amount exists in the characteristic point tracking, and the real-time requirement of unmanned aerial vehicle routing inspection is difficult to meet; and the other type is to identify the foreign matters by using a deep learning technology, but the detection rate of the foreign matter defects is low because training samples are often insufficient. Therefore, the method for realizing the visual-based power transmission line identification and foreign object intrusion online detection with rapidness, accuracy and strong robustness still has great research value.
Disclosure of Invention
The invention aims to solve the technical problems and provides a visual-based power transmission line identification and foreign object intrusion online detection method. According to the method, after the line segment characteristics of the aerial image are extracted, the power transmission line area is extracted according to the structural characteristics of the power transmission line, the light stream distribution of the power transmission line area in the adjacent frame image is calculated, and the position of the foreign matter is judged according to the pixel light stream information, so that the identification accuracy of the unmanned aerial vehicle on the power transmission line and the foreign matter fault is obviously improved, the possible line safety hidden danger is eliminated, and the method has the advantages of being high in instantaneity, low in cost and high in automation degree.
The purpose of the invention is realized by the following technical scheme.
The invention discloses a vision-based power transmission line identification and foreign matter intrusion online detection method, which comprises the steps of carrying out bilateral filtering and piecewise linear gray level transformation on an aerial image of an unmanned aerial vehicle to obtain a preprocessed image; extracting line segment characteristics by using an EDLines (EdgeDrawing Lines) algorithm; classifying the obtained line segment characteristics in an angle interval, extracting parallel line segment groups of the power transmission line, and fusing the parallel line segment groups meeting the preset line segment connection standard; searching and defining a target area of the power transmission line according to the shape and spectral characteristic characteristics of the power transmission line, and marking a suspected area according to the foreign matter characteristics; and calculating the optical flow distribution of the target area of the power transmission line in the adjacent frame images, and performing threshold segmentation, morphological operation and filtering processing on the optical flow images. And comparing the average value of the optical flow velocities of the suspected area and the background area, if the average value of the optical flow velocities of the suspected area and the background area is greater than a set multiple threshold value of the background area, judging that the foreign matter exists, wherein the suspected area is a foreign matter existing area, and realizing the detection of the foreign matter in the power transmission line.
A visual-based power transmission line identification and foreign object intrusion online detection method comprises the following steps:
the method comprises the following steps: and carrying out bilateral filtering and piecewise linear gray level transformation on the aerial image of the unmanned aerial vehicle to obtain a preprocessed image.
And carrying out bilateral filtering on the aerial image, determining a filter weight coefficient according to the pixel distance and the gray level similarity, removing noise and keeping an edge. And performing piecewise linear transformation based on the gray level characteristics of the power transmission line, calculating the gray level value of the transformed pixel, and expanding the gray level interval of the power transmission line to obtain an enhanced gray level image.
Step two: and extracting line segment characteristics from the image with the enhanced gray level by using an EDLines (edge Drawing lines) algorithm.
In order to avoid interference of non-target objects on the EDLines line segment detection algorithm, the EDLines line detection algorithm is used for extracting line segments to obtain line segment characteristics of the image.
Step three: and classifying the angle intervals of the obtained line segment characteristics, extracting parallel line segment groups of the power transmission line, and fusing the parallel line segment groups meeting the preset line segment connection standard.
In order to extract the power transmission line area, parallel line segment group extraction is carried out by utilizing topological structure characteristics of mutual parallelism among the power transmission lines. Dividing 0-180 degrees into 36 small intervals, and classifying the detected line segments into 36 classes according to angles in every 5 degrees. And counting the sum of the lengths of all the straight line segments of the angle in each interval, wherein the angle interval corresponding to the maximum value is the trend angle range of the power transmission line. And (3) eliminating the residual straight line segments by keeping the straight line segments with the inclination angles within the angle interval, namely obtaining a parallel line group which is approximately consistent with the trend of the power transmission line and comprises n line segments.
The method for realizing the fusing of the discontinuous line segments meeting the line segment fusing standard in the parallel line groups comprises the following steps: sorting according to the coordinates of the end points of the line segments from left to right, judging whether the two line segments are fused or not according to the lengths and the distances of the two line segments adjacent to the sequence numbers and the difference angle between the two line segments, and fusing the two line segments when the distance between the middle points of the two line segments is smaller than a set threshold value and the angle difference value between the two line segments is smaller than a corresponding threshold value, wherein the specific method comprises the following steps:
the distance between the two segment midpoints is represented as:
the coordinate of the end point of the ith line segment is (x)i1,yi1),(xi2,yi2),i∈[0,n-1]The coordinates of the end points of the adjacent jth line segment are (x)j1,yj1),(xj2,yj2),j∈[0,n-1]And j > i;
setting the length threshold value of the distance between the middle points of the two line segments to be lthres＝4max(li,lj). The length of the ith line segment is li(ii) a The length of the jth line segment is lj。
The angle of inclination of the ith line segment is
The inclination angle of the jth line segment is
The threshold value of the inclination angle is thetathres。
When l isi,j≤lthresAnd | θi-θj|＜θthresThen, the two line segments of i and j are fused to obtain an endpoint coordinate (x)i1,yi1),(xj2,yj2),i,j∈[0,n-1]Otherwise, the two line segments are still remained.
And E, processing the n parallel line segments obtained in the step three one by one to obtain a fused straight line segment.
Step four: searching and defining a target area of the power transmission line according to the shape and spectral characteristic characteristics of the power transmission line, and marking a suspected area according to the foreign matter characteristics;
and taking the fused line segment obtained in the third step as a target line segment, filtering based on the shape and the color characteristics of the power transmission line, and removing the line segments which do not meet the characteristic constraint of the power transmission line to obtain the edge straight line of the power transmission line.
The power transmission line characteristic constraint conditions are as follows:
constraint condition one, if the total length of a fused straight line segment is less than the shortest side of the image
The line is considered not to be an edge line of the transmission line.
And secondly, carrying out statistics on pixel color characteristics of the target line segment, calculating the number of pixel points on a straight line, wherein the included angle of a color vector in the RGB space deviating from the gray axis is within a range of 5 degrees, and if the number of the pixels is less than 90% of the total number of the pixels on the straight line, determining that the straight line is not the edge straight line of the power transmission line.
And removing the non-power transmission line target line segment according to the two constraint conditions, and reserving the target line segment meeting the constraint conditions to obtain a power transmission line edge straight line and obtain a power transmission line edge straight line graph.
After the electric transmission line edge line graph is obtained, the inclination correction is carried out on the electric transmission line edge line graph and the unmanned aerial vehicle aerial image obtained in the first step by taking the included angle between the longest edge line and the horizontal direction, namely the inclination angle as the rotating basis;
the method for correcting the inclination comprises the following steps: the coordinate of the end point of the straight line segment with the longest edge is (x)longest1,ylongest1),(xlongest2,ylongest2) The angle of inclination of the longest edge line is
Tilt angle thetalongestAs an input, the target area image is subjected to an inverse inclination correction of the image using an affine transformation matrix so that the power transmission line is in a horizontal direction.
The affine transformation matrix is:
and (3) defining a target area of the power transmission line in the image after the inclination correction, wherein the definition rule of the target area is as follows:
expanding the detected edge straight lines of the two outermost power transmission lines to the vertical direction thereof, namely expanding any one power transmission line to the direction far away from the other power transmission line, wherein the expanded horizontal straight line is a horizontal boundary; the distance between the power transmission line and each horizontal boundary is m; intercepting a maximum inscribed rectangle in a horizontal boundary area; namely, the target area of the transmission line is defined.
And m is 0.8-2 times of the length of the vertical distance between the two outmost power transmission lines.
Searching connected domains at the linear intermittent positions of the edges of the power transmission line in a target area of the power transmission line, generating a minimum external rectangle of each connected domain, and marking the minimum external rectangle as a suspected area;
step five: and calculating the optical flow distribution of the target area of the power transmission line in the adjacent frame images, and performing threshold segmentation, morphological operation and filtering processing on the optical flow images. And comparing the average values of the optical flow velocities of the suspected area and the background area, and if the suspected area is larger than a set multiple threshold of the background area, judging that the foreign matter exists, wherein the suspected area is a foreign matter existing area, so that the foreign matter detection of the power transmission line is realized.
And calculating a global optical flow image of the target area of the adjacent frame power transmission line through an optical flow algorithm, wherein the average value of optical flow velocity scales of the area where the foreign matter is located is larger because the moving speed of the foreign matter pixel point relative to the background pixel point is higher. In order to determine the foreign object, the optical flow image needs to be subjected to adaptive threshold segmentation, and the target and the background are separated. Performing morphological closed operation on the divided binary optical flow image, and filling a cavity in the area; and removing background noise by adopting average filtering. And respectively calculating the average values of the optical flow velocity scales of the suspected area and the background area, and determining the existence condition of the foreign matters according to the comparison result.
The method for determining the existence of the foreign matter comprises the following steps: scalar mean v of light flow velocity within the suspected areaobjectV greater than 3 timesbackgroundJudging whether a foreign matter hung on a line exists in the suspected area; when v isobject≤3·vbackgroundIf so, no foreign object is present in the suspect area.
Advantageous effects
1. The invention discloses a vision-based power transmission line identification and foreign matter intrusion online detection method, which is a target detection method based on line segment characteristics and light stream distribution calculation, can avoid the training step of using a complex neural network when carrying out target detection on a power transmission line and a hung foreign matter, has the advantages of no limitation of illumination and a complex background, strong robustness, high universality and strong real-time performance, and can be suitable for various power inspection environments.
2. According to the vision-based power transmission line identification and foreign matter intrusion online detection method disclosed by the invention, the computer vision is utilized to carry out power transmission line foreign matter hanging detection, compared with methods such as manual visual inspection and the like, the subjective misjudgment is reduced, the detection accuracy is obviously improved, and the unmanned aerial vehicle inspection efficiency and the real-time performance are high.
3. According to the vision-based online detection method for power transmission line identification and foreign object intrusion, hardware equipment is few, a server related to deep learning does not need to be purchased, a sensor does not need to be erected on a power transmission line, manpower is saved, and the method has the advantages of low cost and low manual participation degree.
4. The vision-based power transmission line identification and foreign matter intrusion online detection method disclosed by the invention can obviously improve the identification efficiency of an unmanned aerial vehicle on the power transmission line, can judge the foreign matter hanging condition in real time, can eliminate potential safety hazards, and provides support for the maintenance and guarantee of the power transmission line.
Drawings
FIG. 1 is a schematic flow chart of a vision-based method for power transmission line identification and foreign object intrusion online detection according to the present invention;
FIG. 2 is a diagram of image pre-processing effects; fig. 2(a) is a patrol sample image 1, fig. 2(b) is a grayscale image of fig. 2(a), and fig. 2(c) is a piecewise linear graying result of fig. 2 (b);
FIG. 3 is the result of EDLines (edge Drawing lines) extracting line segments, where a red line segment is the line information extracted by EDLines;
fig. 4 is a transmission line edge straight line detection result, in which a red line segment is a transmission line edge straight line;
FIG. 5 is a result of processing an optical flow map of a target area of a power transmission line; FIG. 5(a) shows a target area and a suspected area of the transmission line, wherein red boxes represent the suspected area O1And O2Fig. 5(b) is a light flow diagram of a target area of the transmission line, and fig. 5(c) is a processed binary light flow diagram;
fig. 6 shows the result of detecting a foreign object in the inspection sample image 1: foreign matter is present.
Fig. 7 is a detection result of the patrol inspection sample image 2; fig. 7(a) is a patrol sample image 2, fig. 7(b) is a transmission line target area light flow diagram of fig. 7(a), and fig. 7(c) shows the detection results as: sample image 2 was free of foreign objects.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and examples.
As shown in fig. 1, the method for online detection of power transmission line identification and foreign object intrusion based on vision disclosed in this embodiment includes the following specific steps:
and selecting an actual inspection video of a certain area where the unmanned aerial vehicle is transmitted back to the processing platform, and processing the actual inspection video on the processing platform. The embodiment extracts 2 different images in the patrol video to be respectively used as sample images verified by the method. As shown in fig. 1, the method for online detection of power transmission line identification and foreign object intrusion based on vision disclosed in this embodiment includes the following specific steps:
the method comprises the following steps: carrying out bilateral filtering and piecewise linear gray level transformation on the aerial image of the unmanned aerial vehicle to obtain a preprocessed image, as shown in fig. 2;
the complex weight impact factor B (x, y, p, q) of the bilateral filter is expressed as:
where B (x, y, p, q) represents the integrated weight impact factor for pixel (p, q) at the center pixel (x, y) of the bilateral filter window. SigmadRepresenting the variance, σ, of the spatial distance of the pixels in the neighborhoodgRepresenting the variance of the pixel gray level in the window, calculating the similarity between the gray levels by bilateral filtering according to the brightness information of the pixels in the image window, and combining the similarity with the weight of the distance factor to eliminate the noise of the aerial image and obtain better image edge effect.
Graying the filtered image, and performing weighted average on R, G, B channel components of the RGB color image, wherein the algorithm is as follows:
I(x,y)＝0.2987R(x,y)+0.5870G(x,y)+0.1140B(x,y)
wherein I(x,y)Is a gray value, R, at the gray image (x, y)(x,y)Red channel pixel value at original image (x, y), 0.5870G(x,y)For the green channel pixel value at the original image (x, y), 0.1140B(x,y)Is the original image (x, y) blue channel pixel value.
In order to enhance the contrast of a power transmission line region in an image, the gray value of a pixel point after transformation is calculated by utilizing piecewise linear transformation, and the gray interval of the power transmission line is expanded to obtain the image enhanced based on the characteristics of the power transmission line. Assuming that the gray scale value range of the transmission line is [ a, c ], the following operation is performed for each pixel point (x, y):
where f (x, y) is the pixel gray scale value with coordinates (x, y) after enhancement, and c ', a' represent the gray scale segment points of the transformed image. And obtaining the enhanced image after the gray scale interval of the transmission line is expanded through the calculation of the formula. The preprocessing effect diagram is shown in fig. 2, in which fig. 2(a) is the inspection sample image 1, fig. 2(b) is the grayscale image of fig. 2(a), and fig. 2(c) is the piecewise linear graying result of fig. 2 (b). As can be seen from fig. 2(c), the gray scale range of the transmission line is enhanced, which is beneficial to the extraction of the subsequent edge straight line.
Step two: and extracting line segment characteristics from the image with the enhanced gray level by using an EDLines (edge Drawing lines) algorithm.
The EDLines (edge Drawing lines) line segment extraction method comprises the following steps:
1. smoothing the gray level image by adopting a Gaussian filter with a 5 multiplied by 5 Gaussian kernel;
2. calculating the gradient of each pixel by utilizing gradient operators such as Sobel, Prewitt and the like;
3. generating pixel anchor points with high existence probability in the edge pixels, connecting, and generating a pixel chain by the gradient peak value;
4. fitting and extracting a generated pixel chain line segment through a least square straight line;
5. and eliminating false line segments according to the Helmholtz principle, and acquiring accurate line segments.
And (4) extracting the line segment information in the image after the step one is enhanced by using an EDLines method. The EDLines line segment extraction result is shown in fig. 3, and a red line segment in the drawing is a straight line segment of the inspection sample image 1 detected by the EDLines algorithm.
Step three: and D, carrying out angle interval classification on the line segment characteristics obtained in the step two, extracting parallel line segment groups of the power transmission line, and fusing the parallel line segment groups meeting the preset line segment connection standard.
In order to extract the power transmission line area, parallel line segment group extraction is carried out by utilizing topological structure characteristics of mutual parallelism among the power transmission lines. Dividing 0-180 degrees into 36 small intervals, and classifying the detected line segments into 36 classes according to angles in every 5 degrees. And counting the sum of the lengths of all the straight line segments of the angle in each interval, wherein the angle interval corresponding to the maximum value is the trend angle range of the power transmission line. By keeping the straight line segments with the inclination angles within the angle interval and removing the residual straight line segments, the parallel line group approximately consistent with the trend of the power transmission line can be obtained, and the information of the n line segments is (x)k1,yk1),(xk2,yk2),k∈[0,n-1]。
The method for realizing the fusing of the discontinuous line segments meeting the line segment fusing standard in the parallel line groups comprises the following steps: sorting according to the coordinates of the end points of the line segments from left to right, judging whether the two line segments are fused or not according to the lengths and the distances of the two line segments adjacent to the sequence numbers and the difference angle between the two line segments, and fusing the two line segments when the distance between the middle points of the two line segments is smaller than a set threshold value and the angle difference value between the two line segments is smaller than a corresponding threshold value, wherein the specific method comprises the following steps:
the coordinate of the end point of the ith line segment is (x)i1,yi1),(xi2,yi2),i∈[0,n-1]The coordinates of the end points of the adjacent jth line segment are (x)j1,yj1),(xj2,yj2),j∈[0,n-1]And j > i; the length of the ith line segment is expressed as:
the length of the jth line segment is expressed as:
the distance between the two segment midpoints is represented as:
setting the length threshold value of the distance between the middle points of the two line segments to be lthres＝4max(li,lj)。
The angle of inclination of the ith line segment is
The inclination angle of the jth line segment is
The threshold value of the inclination angle is thetathres。
When l isi,j≤lthresAnd | θi-θj|＜θthresThen, the two line segments of i and j are fused to obtain an endpoint coordinate (x)i1,yi1),(xj2,yj2),i,j∈[0,n-1]Otherwise, the two line segments are still remained.
And E, processing the n parallel line segments obtained in the step three one by one to obtain a fused straight line segment.
Step four: searching and defining a target area of the power transmission line according to the shape and spectral characteristic characteristics of the power transmission line, and marking a suspected area according to the foreign matter characteristics;
and taking the fused line segment obtained in the third step as a target line segment, and performing line segment filtering based on the shape and color characteristic constraint of the power transmission line to obtain a straight line at the edge of the power transmission line.
The power transmission line characteristic constraint conditions are as follows:
constraint condition one, if the total length of a fused straight line segment is less than the shortest side of the image
Then the straight line is considered not to be inputThe edges of the wire are straight.
And secondly, carrying out statistics on pixel color characteristics of the target line segment, calculating the number of pixel points on a straight line, wherein the included angle of a color vector in the RGB space deviating from the gray axis is within a range of 5 degrees, and if the number of the pixels is less than 90% of the total number of the pixels on the straight line, determining that the straight line is not the edge straight line of the power transmission line.
And removing the non-power transmission line target line segment according to the two constraint conditions, reserving the target line segment meeting the constraint conditions to obtain a power transmission line edge straight line, wherein the detection result of the power transmission line edge straight line of the inspection sample image 1 is shown in fig. 4, a red line segment in the graph is the power transmission line edge straight line, and the background interference line segment can be seen from the graph to be removed. Then, the steps of identifying and positioning the power transmission line area are three steps:
1. the included angle between the longest edge straight line and the horizontal direction, namely the inclination angle is used as a rotation basis, and inclination correction is carried out on the power transmission line edge straight line graph and the unmanned aerial vehicle aerial image in the first step;
the method for correcting the inclination comprises the following steps: the coordinate of the end point of the straight line segment with the longest edge is (x)longest1,ylongest1),(xlongest2,ylongest2) The angle of inclination of the longest edge line is
Tilt angle thetalongestAs an input, the target area image is subjected to an inverse inclination correction of the image using an affine transformation matrix so that the power transmission line is in a horizontal direction.
The affine transformation matrix is:
2. and (3) defining a target area of the power transmission line in the image after the inclination correction, wherein the definition rule of the target area is as follows:
expanding the detected edge straight lines of the two outermost power transmission lines to the vertical direction thereof, namely expanding any one power transmission line to the direction far away from the other power transmission line, wherein the expanded horizontal straight line is a horizontal boundary; the distance between the power transmission line and each horizontal boundary is m; intercepting a maximum inscribed rectangle in a horizontal boundary area; namely, the target area of the transmission line is defined. And selecting the length of the vertical distance between the two outmost power transmission lines with m being 1.1 times.
3. And searching connected domains at the linear discontinuous positions of the edges of the power transmission line in the target area of the power transmission line, generating the minimum circumscribed rectangle of each connected domain, and marking the minimum circumscribed rectangle as a suspected area.
Step five: and calculating the optical flow distribution of the target area of the adjacent frame of power transmission line, and performing threshold segmentation, morphological operation and filtering processing on the optical flow image. And comparing the average value of the optical flow velocities of the suspected area and the background area, if the average value of the optical flow velocities of the suspected area and the background area is greater than a set multiple threshold value of the background area, judging that the foreign matter exists, wherein the suspected area is a foreign matter existing area, and realizing the detection of the foreign matter in the power transmission line.
And after the transmission line target area and the suspected foreign matter area are obtained in the last step, calculating an optical flow graph of the adjacent frame transmission line target area through a global optical flow algorithm. The basic constraint equation and the smooth constraint equation of the global optical flow calculation are as follows:
It+Ixu+Iyv＝0
in the formula, u and v are respectively the horizontal component and the vertical component of the luminous flow velocity obtained by the pixel point (x, y) in the optical flow field at the time t, It、IxAnd IyThe method is characterized in that partial derivatives of gray values I at pixel points (x, y) to x, y and t are respectively, E is a defined energy function, and lambda is a coefficient factor of a smooth term, and is a larger value when image noise is larger.
According to the feature that the unmanned aerial vehicle patrols along the horizontal direction, a scalar op of the optical flow velocity needs to be extracted as follows:
and further foreign matter detection is realized according to the following steps:
1. carrying out self-adaptive threshold segmentation on the optical flow image, and separating a target from a background;
the calculation method of the segmentation threshold is as follows:
suppose an image has M rows and N columns, and N, respectively1A target pixel and N2A background pixel. The ratio of the target pixel in the image is
The ratio of background pixels in the image is
Let u1Is the mean gray value of the pixels in the foreground region, mu2If the average gray value of the pixels in the background area is the gray value of the background area, the average gray value of the gray image is
μ＝μ1ω1+μ2ω2
The inter-class variance of the two regions of the target and background can be expressed as
g＝ω1(μ-μ1)2+ω2(μ-μ2)2
Finding the threshold t that maximizes the inter-class variance by traversing the entire imagethMake the gray value greater than the threshold tthIs set to 1, the gray value is less than the threshold value tthThe pixel of (2) is set to be 0, and the foreground and the background of the optical flow image are distinguished to obtain a segmented optical flow image.
2. Performing morphological closed operation on the segmented optical flow image, filling a cavity in the region, and removing background noise by adopting mean filtering;
the closed operation is a composite operation of expansion and corrosion. By sets DbRepresenting objects in images, moving up a given imageAnd (4) moving structural elements. Let b be the structural element, f be the image, (i, j) be the pixel coordinates in structural element b, and (x, y) be the pixel coordinates in the image. Let the expansion operation be
Then there is
When the corrosion operation is expressed as (f Θ b), there is
(fΘb)(x,y)＝max{f(x+i,y+j)-b(i,j)|(i,j)∈Db}
And after closed operation processing, performing mean value filtering processing. And selecting an 8-neighborhood mean filtering template, and calculating the mean value of 8-neighborhood pixels of any point pixel in the image to replace the gray value of the pixel. And finally obtaining the processed binary optical flow image through morphological closed operation and mean filtering.
The processing result of the optical flow diagram of the target area of the power transmission line is shown in fig. 5, wherein fig. 5(a) shows the target area and the suspected area of the power transmission line, and red boxes in the diagram represent the suspected area O1And O2Fig. 5(b) is a light flow diagram of a target area of a power transmission line, in which the relative motion speed of pixels with higher brightness is faster and the light flow scalar value is larger; fig. 5(c) is a processed binary light flow diagram, where white represents the foreground and black represents the background.
3. And judging the distance relation of the target object in the image according to the instantaneous motion speed information of the pixels in the image.
The suspected area is O, the total number of pixels in the area O is NO. The background area of the power transmission line indicates the area except the suspected area in the target area of the power transmission line, and the total number of pixels in the background area B is NB. The mean value of the optical flow velocity scale quantity of the suspected area is vobjectThe average value of the optical flow velocity scale quantity of the background area is vbackgroundThen the expression of the mean of the optical flow scalar is:
according to analysis and statistics of a large number of inspection images, a threshold value is set to be 3 times of the scalar mean value of the light flow speed of the background area of the power transmission line, and foreign matter detection is carried out on the binary light flow diagram shown in the figure 5 (c). Calculated to obtain a suspected area O1Is greater than the threshold, O1The presence of foreign matter; suspected area O2Is below the threshold, O2The suspected area is excluded when no foreign matter exists. The foreign object detection result is shown in fig. 6, a red solid frame in the diagram is a foreign object distribution area, the foreign object is accurately positioned in the complex background, suspected situations caused by the complex background are eliminated, and the occurrence of missing detection situations can be reduced.
Step six: and (5) repeating the processing procedures from the first step to the fifth step on the inspection sample image 2, and detecting the foreign matters. The detection result is shown in fig. 7, where fig. 7(a) is an inspection sample image 2, fig. 7(b) is a transmission line target area light flow graph obtained after the processing of the first step to the fifth step in the diagram (a), and a red frame in the diagram represents a suspected area O generated by background noise1'. Due to the suspected area O1' the average value of the optical flow velocity scale amount of the background area is less than 3 times, and it is determined that no foreign matter exists in the sample image 2, that is, fig. 7(c) shows the detection result: no foreign matter is present in the sample image 2. This method can also eliminate false detection in which a complex background is false detected as a foreign object.
The above detailed description is intended to illustrate the objects, aspects and advantages of the present invention, and it should be understood that the above detailed description is only exemplary of the present invention and is not intended to limit the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (1)
1. A visual-based power transmission line identification and foreign object intrusion online detection method is characterized in that: the method comprises the following steps:
the method comprises the following steps: carrying out bilateral filtering and piecewise linear gray level transformation on the aerial image of the unmanned aerial vehicle to obtain a preprocessed image;
carrying out bilateral filtering on the aerial image, determining a filter weight coefficient according to the pixel distance and the gray level similarity, removing noise and keeping an edge; performing piecewise linear transformation based on the gray characteristic of the power transmission line, calculating the gray value of the transformed pixel, and expanding the gray interval of the power transmission line to obtain an enhanced gray image;
step two: extracting line segment characteristics from the image with enhanced gray level by using an EDLines (edge Drawing lines) algorithm;
in order to avoid interference of non-target objects on the EDLines line segment detection algorithm, the EDLines line detection algorithm is used for extracting line segments to obtain line segment characteristics of the image;
step three: classifying the obtained line segment characteristics in an angle interval, extracting parallel line segment groups of the power transmission line, and fusing the parallel line segment groups meeting the preset line segment connection standard;
in order to extract the power transmission line area, parallel line segment group extraction is carried out by utilizing topological structure characteristics of mutual parallelism among the power transmission lines. Dividing 0-180 degrees into 36 small intervals, and classifying the detected line segments into 36 classes according to angles in every 5 degrees. And counting the sum of the lengths of all the straight line segments of the angle in each interval, wherein the angle interval corresponding to the maximum value is the trend angle range of the power transmission line. And (3) eliminating the residual straight line segments by keeping the straight line segments with the inclination angles within the angle interval, namely obtaining a parallel line group which is approximately consistent with the trend of the power transmission line and comprises n line segments.
The method for realizing the fusing of the discontinuous line segments meeting the line segment fusing standard in the parallel line groups comprises the following steps: sorting according to the coordinates of the end points of the line segments from left to right, judging whether the two line segments are fused or not according to the lengths and the distances of the two line segments adjacent to the sequence numbers and the difference angle between the two line segments, and fusing the two line segments when the distance between the middle points of the two line segments is smaller than a set threshold value and the angle difference value between the two line segments is smaller than a corresponding threshold value, wherein the specific method comprises the following steps:
the distance between the two segment midpoints is represented as:
the coordinate of the end point of the ith line segment is (x)i1,yi1),(xi2,yi2),i∈[0,n-1]The coordinates of the end points of the adjacent jth line segment are (x)j1,yj1),(xj2,yj2),j∈[0,n-1]And j > i; (ii) a
Setting the length threshold value of the distance between the middle points of the two line segments to be lthres＝4max(li,lj) (ii) a The length of the ith line segment is li(ii) a The length of the jth line segment is lj；
The angle of inclination of the ith line segment is
The inclination angle of the jth line segment is
The threshold value of the inclination angle is thetathres；
When l isi,j≤lthresAnd | θi-θj|＜θthresThen, the two line segments of i and j are fused to obtain an endpoint coordinate (x)i1,yi1),(xj2,yj2),i,j∈[0,n-1]Otherwise, the two line segments are still reserved;
processing the n parallel line segments obtained in the step three one by one to obtain fused straight line segments;
step four: searching and defining a target area of the power transmission line according to the shape and spectral characteristic characteristics of the power transmission line, and marking a suspected area according to the foreign matter characteristics;
taking the fused line segment obtained in the third step as a target line segment, filtering based on the shape and color characteristics of the power transmission line, and removing the line segments which do not meet the characteristic constraint of the power transmission line to obtain a straight line at the edge of the power transmission line;
the power transmission line characteristic constraint conditions are as follows:
constraint condition one, if the total length of a fused straight line segment is less than the shortest side of the image
The straight line is not considered as the edge straight line of the transmission line;
the constraint condition II is that pixel color characteristics of the target line segment are counted, the number of pixel points of a straight line, wherein the included angle of a color vector in an RGB space deviates from a gray axis is within a range of 5 degrees, and if the number of the pixels is lower than 90% of the total number of the pixels on the straight line, the straight line is considered not to be the edge straight line of the power transmission line;
removing the non-power transmission line target line segments according to the two constraint conditions, and reserving the target line segments meeting the constraints to obtain a power transmission line edge straight line and obtain a power transmission line edge straight line graph;
after the electric transmission line edge line graph is obtained, the inclination correction is carried out on the electric transmission line edge line graph and the unmanned aerial vehicle aerial image obtained in the first step by taking the included angle between the longest edge line and the horizontal direction, namely the inclination angle as the rotating basis;
the method for correcting the inclination comprises the following steps: the coordinate of the end point of the straight line segment with the longest edge is (x)longest1,ylongest1),(xlongest2,ylongest2) The angle of inclination of the longest edge line is
Tilt angle thetalongestAs input, the target area image carries out reverse inclination correction on the image by using an affine transformation matrix so that the power transmission line is in a horizontal direction;
the affine transformation matrix is:
and (3) defining a target area of the power transmission line in the image after the inclination correction, wherein the definition rule of the target area is as follows:
expanding the detected edge straight lines of the two outermost power transmission lines to the vertical direction thereof, namely expanding any one power transmission line to the direction far away from the other power transmission line, wherein the expanded horizontal straight line is a horizontal boundary; the distance between the power transmission line and each horizontal boundary is m; intercepting a maximum inscribed rectangle in a horizontal boundary area; namely, the target area of the transmission line is defined.
And m is 0.8-2 times of the length of the vertical distance between the two outmost power transmission lines.
Searching connected domains at the linear intermittent positions of the edges of the power transmission line in a target area of the power transmission line, generating a minimum external rectangle of each connected domain, and marking the minimum external rectangle as a suspected area;
step five: calculating the optical flow distribution of the target area of the power transmission line in the adjacent frame images, and performing threshold segmentation, morphological operation and filtering processing on the optical flow images; comparing the average values of the optical flow velocities of the suspected area and the background area, and if the suspected area is larger than a set multiple threshold of the background area, judging that foreign matters exist, wherein the suspected area is a foreign matter existing area, so that the foreign matter detection of the power transmission line is realized;
calculating a global optical flow image of a target area of the adjacent frame power transmission line through an optical flow algorithm, wherein the average value of optical flow velocity scales of an area where the foreign matter is located is larger because the moving speed of the foreign matter pixel point relative to the background pixel point is higher; in order to determine the foreign matter, the optical flow image needs to be subjected to self-adaptive threshold segmentation, and a target and a background are separated; performing morphological closed operation on the divided binary optical flow image, and filling a cavity in the area; removing background noise by mean filtering; respectively calculating the average values of optical flow velocity scales of the suspected area and the background area, and determining the existence condition of foreign matters according to the comparison result;
the method for determining the existence of the foreign matter comprises the following steps: scalar mean v of light flow velocity within the suspected areaobjectV greater than 3 timesbackgroundJudging whether a foreign matter hung on a line exists in the suspected area; when v isobject≤3·vbackgroundThen, there is no radio channel difference in the suspected areaA compound (I) is provided.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010657798.2A CN111814686A (en) | 2020-07-09 | 2020-07-09 | Vision-based power transmission line identification and foreign matter invasion online detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010657798.2A CN111814686A (en) | 2020-07-09 | 2020-07-09 | Vision-based power transmission line identification and foreign matter invasion online detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111814686A true CN111814686A (en) | 2020-10-23 |
Family
ID=72842068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010657798.2A Pending CN111814686A (en) | 2020-07-09 | 2020-07-09 | Vision-based power transmission line identification and foreign matter invasion online detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111814686A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329584A (en) * | 2020-10-29 | 2021-02-05 | 深圳技术大学 | Method, system and equipment for automatically identifying foreign matters in power grid based on machine vision |
| CN112383750A (en) * | 2020-11-04 | 2021-02-19 | 国网山东省电力公司潍坊供电公司 | Power transmission line channel intelligent management and control platform |
| CN113011252A (en) * | 2021-02-04 | 2021-06-22 | 成都希格玛光电科技有限公司 | Track foreign matter intrusion detection system and method |
| CN113160255A (en) * | 2021-02-24 | 2021-07-23 | 国网福建省电力有限公司检修分公司 | Method for monitoring change of environment factor of operating line corridor |
| CN113221685A (en) * | 2021-04-27 | 2021-08-06 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Method and device for identifying tiny foreign matters in power transmission line and computer equipment |
| CN113409362A (en) * | 2021-06-29 | 2021-09-17 | 深圳市商汤科技有限公司 | High altitude parabolic detection method and device, equipment and computer storage medium |
| CN113450372A (en) * | 2021-08-27 | 2021-09-28 | 海门裕隆光电科技有限公司 | Power transmission line image intelligent enhancement method and system based on artificial intelligence |
| CN113673514A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司微山县供电公司 | Method and system for detecting invasion of foreign matters into power transmission line |
| CN114299406A (en) * | 2022-03-07 | 2022-04-08 | 山东鹰联光电科技股份有限公司 | Optical fiber cable line inspection method based on unmanned aerial vehicle aerial photography |
| CN114708439A (en) * | 2022-03-22 | 2022-07-05 | 重庆大学 | Improved EDLines linear extraction method based on PROSAC and screening combination |
| CN115035412A (en) * | 2022-06-23 | 2022-09-09 | 郑州儒慧信息技术有限责任公司 | Method for identifying contact net foreign matter |
| CN115063402A (en) * | 2022-07-26 | 2022-09-16 | 国网山东省电力公司东营供电公司 | Cable foreign matter detection method, system, terminal and medium based on sliding analysis |
| CN115294451A (en) * | 2022-01-14 | 2022-11-04 | 中国铁路兰州局集团有限公司 | Method and device for detecting foreign matters on high-voltage line |
| CN115686073A (en) * | 2023-01-03 | 2023-02-03 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle-based power transmission line inspection control method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778472A (en) * | 2016-11-17 | 2017-05-31 | 成都通甲优博科技有限责任公司 | The common invader object detection and recognition method in transmission of electricity corridor based on deep learning |
| CN106960438A (en) * | 2017-03-25 | 2017-07-18 | 安徽继远软件有限公司 | Method for recognizing impurities to transmission line of electricity is converted based on Hough straight line |
| CN109631848A (en) * | 2018-12-14 | 2019-04-16 | 山东鲁能软件技术有限公司 | Electric line foreign matter intruding detection system and detection method |
| CN110414308A (en) * | 2019-05-16 | 2019-11-05 | 南京理工大学 | A kind of target identification method for dynamic foreign matter on transmission line of electricity |
| US20200104621A1 (en) * | 2017-03-24 | 2020-04-02 | Dalian Czur Tech Co., Ltd. | Marker for occluding foreign matter in acquired image, method for recognizing foreign matter marker in image and book scanning method |
-
2020
- 2020-07-09 CN CN202010657798.2A patent/CN111814686A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106778472A (en) * | 2016-11-17 | 2017-05-31 | 成都通甲优博科技有限责任公司 | The common invader object detection and recognition method in transmission of electricity corridor based on deep learning |
| US20200104621A1 (en) * | 2017-03-24 | 2020-04-02 | Dalian Czur Tech Co., Ltd. | Marker for occluding foreign matter in acquired image, method for recognizing foreign matter marker in image and book scanning method |
| CN106960438A (en) * | 2017-03-25 | 2017-07-18 | 安徽继远软件有限公司 | Method for recognizing impurities to transmission line of electricity is converted based on Hough straight line |
| CN109631848A (en) * | 2018-12-14 | 2019-04-16 | 山东鲁能软件技术有限公司 | Electric line foreign matter intruding detection system and detection method |
| CN110414308A (en) * | 2019-05-16 | 2019-11-05 | 南京理工大学 | A kind of target identification method for dynamic foreign matter on transmission line of electricity |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329584A (en) * | 2020-10-29 | 2021-02-05 | 深圳技术大学 | Method, system and equipment for automatically identifying foreign matters in power grid based on machine vision |
| CN112383750A (en) * | 2020-11-04 | 2021-02-19 | 国网山东省电力公司潍坊供电公司 | Power transmission line channel intelligent management and control platform |
| CN113011252A (en) * | 2021-02-04 | 2021-06-22 | 成都希格玛光电科技有限公司 | Track foreign matter intrusion detection system and method |
| CN113011252B (en) * | 2021-02-04 | 2023-12-05 | 成都希格玛光电科技有限公司 | Rail foreign matter intrusion detection system and method |
| CN113160255A (en) * | 2021-02-24 | 2021-07-23 | 国网福建省电力有限公司检修分公司 | Method for monitoring change of environment factor of operating line corridor |
| CN113221685A (en) * | 2021-04-27 | 2021-08-06 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Method and device for identifying tiny foreign matters in power transmission line and computer equipment |
| CN113409362A (en) * | 2021-06-29 | 2021-09-17 | 深圳市商汤科技有限公司 | High altitude parabolic detection method and device, equipment and computer storage medium |
| CN113409362B (en) * | 2021-06-29 | 2023-02-21 | 深圳市商汤科技有限公司 | High altitude parabolic detection method and device, equipment and computer storage medium |
| CN113673514A (en) * | 2021-08-11 | 2021-11-19 | 国网山东省电力公司微山县供电公司 | Method and system for detecting invasion of foreign matters into power transmission line |
| CN113673514B (en) * | 2021-08-11 | 2024-06-04 | 国网山东省电力公司微山县供电公司 | Foreign matter intrusion detection method and system for power transmission line |
| CN113450372A (en) * | 2021-08-27 | 2021-09-28 | 海门裕隆光电科技有限公司 | Power transmission line image intelligent enhancement method and system based on artificial intelligence |
| CN113450372B (en) * | 2021-08-27 | 2021-11-16 | 海门裕隆光电科技有限公司 | Power transmission line image intelligent enhancement method and system based on artificial intelligence |
| CN115294451B (en) * | 2022-01-14 | 2023-04-28 | 中国铁路兰州局集团有限公司 | Method and device for detecting foreign matters on high-voltage line |
| CN115294451A (en) * | 2022-01-14 | 2022-11-04 | 中国铁路兰州局集团有限公司 | Method and device for detecting foreign matters on high-voltage line |
| CN114299406A (en) * | 2022-03-07 | 2022-04-08 | 山东鹰联光电科技股份有限公司 | Optical fiber cable line inspection method based on unmanned aerial vehicle aerial photography |
| CN114708439A (en) * | 2022-03-22 | 2022-07-05 | 重庆大学 | Improved EDLines linear extraction method based on PROSAC and screening combination |
| CN114708439B (en) * | 2022-03-22 | 2024-05-24 | 重庆大学 | PROSAC and screening combination-based improved EDLines linear extraction method |
| CN115035412B (en) * | 2022-06-23 | 2024-04-12 | 郑州儒慧信息技术有限责任公司 | Method for identifying foreign matters of overhead contact system |
| CN115035412A (en) * | 2022-06-23 | 2022-09-09 | 郑州儒慧信息技术有限责任公司 | Method for identifying contact net foreign matter |
| CN115063402B (en) * | 2022-07-26 | 2022-10-25 | 国网山东省电力公司东营供电公司 | Cable foreign matter detection method, system, terminal and medium based on sliding analysis |
| CN115063402A (en) * | 2022-07-26 | 2022-09-16 | 国网山东省电力公司东营供电公司 | Cable foreign matter detection method, system, terminal and medium based on sliding analysis |
| CN115686073A (en) * | 2023-01-03 | 2023-02-03 | 国网江苏省电力有限公司泰州供电分公司 | Unmanned aerial vehicle-based power transmission line inspection control method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111814686A (en) | Vision-based power transmission line identification and foreign matter invasion online detection method | |
| CN115170576B (en) | Aluminum pipe surface defect detection method based on machine vision | |
| CN111428748B (en) | HOG feature and SVM-based infrared image insulator identification detection method | |
| TWI409718B (en) | Method of locating license plate of moving vehicle | |
| WO2022027931A1 (en) | Video image-based foreground detection method for vehicle in motion | |
| CN113139521B (en) | Pedestrian boundary crossing monitoring method for electric power monitoring | |
| CN109523529B (en) | Power transmission line defect identification method based on SURF algorithm | |
| CN111292321B (en) | Transmission line insulator defect image identification method | |
| Saha et al. | License Plate localization from vehicle images: An edge based multi-stage approach | |
| CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
| CN107179479B (en) | Visible light image-based transmission conductor strand scattering defect detection method | |
| CN105740844A (en) | Insulator cracking fault detection method based on image identification technology | |
| CN110276747B (en) | Insulator fault detection and fault rating method based on image analysis | |
| CN113538503A (en) | Solar panel defect detection method based on infrared image | |
| Luo et al. | Stationary vehicle detection in aerial surveillance with a UAV | |
| CN114298948A (en) | Ball machine monitoring abnormity detection method based on PSPNet-RCNN | |
| CN108596196B (en) | Pollution state evaluation method based on insulator image feature dictionary | |
| CN108961262B (en) | Bar code positioning method in complex scene | |
| CN113435452A (en) | Electrical equipment nameplate text detection method based on improved CTPN algorithm | |
| Su et al. | A new local-main-gradient-orientation HOG and contour differences based algorithm for object classification | |
| CN113313107A (en) | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge | |
| CN116309407A (en) | Method for detecting abnormal state of railway contact net bolt | |
| CN115841633A (en) | Power tower and power line associated correction power tower and power line detection method | |
| CN113947563A (en) | Cable process quality dynamic defect detection method based on deep learning | |
| CN110348307B (en) | Path edge identification method and system for crane metal structure climbing robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220307 Address after: No.844, Longshan Road, Lishi District, Luliang City, Shanxi Province 033000 Applicant after: LVLIANG POWER SUPPLY COMPANY, STATE GRID SHANXI ELECTRIC POWER Co. Address before: No.844, Longshan Road, Lishi District, Luliang City, Shanxi Province 033000 Applicant before: LVLIANG POWER SUPPLY COMPANY, STATE GRID SHANXI ELECTRIC POWER Co. Applicant before: Beijing Institute of Technology |
|
| TA01 | Transfer of patent application right | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20240126 |
|
| AD01 | Patent right deemed abandoned |