CN116091494A - Method for measuring distance of hidden danger of external damage of power transmission machinery - Google Patents
Method for measuring distance of hidden danger of external damage of power transmission machinery Download PDFInfo
- Publication number
- CN116091494A CN116091494A CN202310361171.6A CN202310361171A CN116091494A CN 116091494 A CN116091494 A CN 116091494A CN 202310361171 A CN202310361171 A CN 202310361171A CN 116091494 A CN116091494 A CN 116091494A
- Authority
- CN
- China
- Prior art keywords
- hidden danger
- external damage
- mechanical external
- power transmission
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G06T7/85—Stereo camera calibration
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
-
- 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
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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 provides a method for measuring the distance of hidden danger of external damage of a power transmission machine, which belongs to the field of intelligent operation and detection of power transmission lines and comprises the following steps: establishing a model; based on the data such as the visible light pictures of the transmission channel shot by the transmission line visual monitoring equipment and the camera internal reference matrix of the visual monitoring equipment, three-dimensional reconstruction is carried out on the transmission line in the visible light pictures of the transmission channel shot by the transmission line visual monitoring equipment through the line area acquired by the transmission line segmentation model and the depth map acquired by the monocular depth estimation model, the space information of the transmission line mechanical external damage hidden danger target is acquired by using the transmission line mechanical external damage hidden danger target detection model and the depth map in combination with the camera imaging principle, and the minimum distance between the transmission line and the mechanical external damage hidden danger is calculated to realize the distance measurement of the transmission line mechanical external damage hidden danger.
Description
Technical Field
The invention relates to the field of intelligent operation and detection of power transmission lines, in particular to a method for measuring the distance of hidden danger of external damage of power transmission machinery.
Background
With the rapid development of national economy, the stable operation of a power system has been related to daily life of thousands of households. Meanwhile, as the region of China is wide, the power transmission lines are all around, the external environment of the power transmission channel is various, the external damage event of the power transmission lines in the channel is increased year by year, and the mechanical external damage is a serious problem facing the power supply guarantee of power grid enterprises. The problem of power failure caused by improper construction of machinery, particularly ultra-high machinery, not only affects stable transportation of power, but also causes huge economic loss, so that hidden danger of mechanical external damage is already a hidden danger problem to be solved in the power system of China.
With the development of modern science and technology, the detection of the hidden danger of the external damage of the power transmission based on the combination of the remote monitoring equipment and the deep learning algorithm is widely applied to the inspection of the power transmission line, but because the remote monitoring equipment generally shoots a large picture range, the hidden danger identification can not effectively distinguish effective hidden danger from ineffective hidden danger, so that a great number of ineffective early warning is caused, and the detection is a serious problem in the inspection of the power transmission line at present. And the distance between the hidden danger and the wire can be obtained, so that the problem can be effectively solved. Chinese invention patent name: method, equipment and medium for measuring potential hazards of transmission line channel targets, and patent numbers: CN113345019a discloses a method for measuring the potential hazards of transmission line channels, which comprises the following steps: acquiring two-dimensional image data and three-dimensional point cloud data of a transmission line channel, and acquiring two-dimensional position information of a hidden danger target according to a target segmentation network model; if the two-dimensional position information of the hidden danger target is in the preset channel three-dimensional protection area, mapping the two-dimensional position information of the hidden danger target into three-dimensional point cloud data through a preset ground data mapping model to obtain three-dimensional coordinate position information of the hidden danger target; and obtaining the distance information between the hidden danger target and the lead according to the three-dimensional coordinate position information of the hidden danger target and a preset lead data mapping model. The method comprehensively utilizes the dense information advantage of the two-dimensional image data and the precision advantage of the three-dimensional point cloud data, and provides more effective early warning information for the safety of the transmission line channel.
However, the conventional ranging methods are basically based on point clouds and related data, and have the defects of incapability of acquiring data in real time, difficult data acquisition, high processing difficulty, low calculation precision and the like.
Disclosure of Invention
Aiming at the problems in the prior art, the method and the device are based on the data such as the visible light pictures of the transmission channel shot by the transmission line visual monitoring equipment, the camera internal reference matrix of the visual monitoring equipment and the like, and the transmission lines in the visible light pictures of the transmission channel shot by the transmission line visual monitoring equipment are subjected to three-dimensional reconstruction by the depth map of the transmission line visual pictures acquired by the transmission line segmentation model based on deep learning and the transmission line region and the monocular depth estimation model, and the spatial information of the transmission line mechanical external damage hidden danger target is acquired by using the transmission line mechanical external damage hidden danger target detection model and the depth map of the transmission line visual pictures and the camera imaging principle, so that the minimum distance between the transmission lines and the mechanical external damage hidden danger is calculated, and the transmission line mechanical external damage hidden danger is realized.
The technical scheme adopted by the invention is as follows:
the invention provides a method for measuring the distance of hidden danger of external damage of a power transmission machine, which comprises the following steps:
establishing a monocular depth estimation model, a transmission line segmentation model and a transmission line mechanical external damage hidden danger target detection model;
monocular depth estimation of scene scenery is carried out on the visible light picture of the power transmission channel by adopting a monocular depth estimation model, and a depth map of the visible light picture of the power transmission channel is obtained; carrying out image segmentation on the wires in the visible light pictures of the transmission channels by adopting a transmission wire segmentation model, and obtaining wire segmentation pixel positions; the wire segmentation pixel positions are combined with a depth map of a visible light picture of the power transmission channel, and wire depth information is obtained; converting the wire depth information into wire point clouds through a camera internal reference matrix; acquiring a mechanical external broken hidden danger frame in a visible light picture of a power transmission channel by adopting a mechanical external broken hidden danger target detection model of the power transmission line;
the mechanical external damage hidden danger frame is combined with the depth map of the visible light picture of the power transmission channel, and plane position information of a mechanical external damage hidden danger target representative point set is obtained; acquiring the spatial position information of a target representative point set of the hidden danger of mechanical external damage by using the plane position information through a camera imaging principle; and calculating Euclidean distance through the space position information and the wire point cloud, wherein the minimum value in the Euclidean distance is the minimum distance between the hidden danger of mechanical external damage and the transmission wire.
The beneficial effects of the invention are as follows:
according to the invention, the pixel positions of the wires can be well extracted by using the power transmission wire segmentation model based on deep learning, the three-dimensional reconstruction of the wires based on a single picture can be realized by combining the depth map of the visible light picture of the power transmission channel and the camera internal reference matrix, the real-time performance of reconstructing the power transmission wires is effectively improved, and the cost and time of reconstructing the Jing Wudian cloud of the power transmission scene are greatly reduced while the ranging precision is improved;
the invention combines a single picture with the internal parameters of a camera to generate the space coordinates of the hidden danger of the external damage of the power transmission and the wire coordinates with three-dimensional space information by utilizing the monocular depth estimation technology, and measures the distance between the space coordinates and the wire coordinates;
according to the method, the hidden danger in the single picture is selectively acquired, so that the problem of difficulty in ranging of a large amount of point cloud data is avoided, the method has high instantaneity, and the minimum distance between the tree barrier and the wire in the power transmission line can be acquired accurately in real time.
Drawings
Fig. 1 is a flow chart of a method for measuring the distance of hidden danger of external damage of a power transmission machine.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the accompanying drawings: in order to clearly illustrate the technical features of the present solution, the present invention will be described in detail below with reference to the following detailed description and the accompanying drawings. The following disclosure provides many different embodiments, or examples, for implementing different structures of the invention. In order to simplify the present disclosure, components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and processes are omitted so as to not unnecessarily obscure the present invention.
As shown in fig. 1, the invention provides a method for measuring the distance of hidden danger of external damage of a power transmission machine, which comprises the following steps:
establishing a monocular depth estimation model, a transmission line segmentation model and a transmission line mechanical external damage hidden danger target detection model;
monocular depth estimation of scene scenery is carried out on the visible light picture of the power transmission channel by adopting a monocular depth estimation model, and a depth map of the visible light picture of the power transmission channel is obtained; carrying out image segmentation on the wires in the visible light pictures of the transmission channels by adopting a transmission wire segmentation model, and obtaining wire segmentation pixel positions; the wire segmentation pixel positions are combined with a depth map of a visible light picture of the power transmission channel, and wire depth information is obtained; converting the wire depth information into wire point clouds through a camera internal reference matrix; acquiring a mechanical external broken hidden danger frame in a visible light picture of a power transmission channel by adopting a mechanical external broken hidden danger target detection model of the power transmission line; the method for detecting the potential mechanical damage of the power transmission line by using the target detection model of the potential mechanical damage of the power transmission line, such as RCNN, YOLO and other target detection methods, is not limited to the method, and has the capability of identifying and detecting the potential mechanical damage;
the mechanical external damage hidden danger frame is combined with the depth map of the visible light picture of the power transmission channel, and plane position information of a mechanical external damage hidden danger target representative point set is obtained; acquiring the spatial position information of a target representative point set of the hidden danger of mechanical external damage by using the plane position information through a camera imaging principle; and calculating Euclidean distance through the space position information and the wire point cloud, wherein the minimum value in the Euclidean distance is the minimum distance between the hidden danger of mechanical external damage and the transmission wire.
The visible light picture of the power transmission channel refers to an array of W.times.H.times.3, W and H are respectively the sizes of the pictures, the pictures are shot by power transmission line visual monitoring equipment, and the content of the visible light picture comprises a power transmission wire, a pole tower and an off-line area which need to be monitored.
The monocular depth estimation model refers to a depth neural network model, has the capability of predicting the depth of a scene pixel in a visible light picture of a power transmission channel, and is output as a depth map of the visible light picture of the power transmission channel.
The depth map of the transmission channel visible light picture refers to a 2-dimensional array formed by horizontal distances from scenes in a three-dimensional space displayed by each pixel in the transmission channel visible light picture to an image plane of a camera for shooting the transmission channel visible light picture, and the size of the array is the same as the length and the width of the transmission channel visible light picture.
The deep learning-based transmission line segmentation model refers to a neural network model capable of segmenting transmission line capacity of equipment orientation in a tower of a line where transmission line visual monitoring equipment is located in a transmission channel visible light picture, and output of the neural network model is pixel positions of the transmission lines in the transmission channel visible light picture.
The position of the lead split pixel is a two-dimensional array, the first dimension is the transverse position of the lead pixel in the visible light picture of the power transmission channel, the second dimension represents the longitudinal position of the lead pixel in the visible light picture of the power transmission channel, and the two dimensions can describe the position information of the lead in the visible light picture of the power transmission channel.
The camera internal reference matrix is a 3*3 matrix, which represents the imaging process of the visible light picture of the power transmission channel, and through the matrix, the interconversion of the point cloud and the depth map (the depth map has no directivity meaning and is a proper noun) can be realized. The specific form is as follows, which can be calculated by camera parameter calibration (for example Zhang Dingyou calibration method) or by visible light picture size, camera focal length f and camera CCD or COMS sensor size:
wherein M is a camera reference matrix, fx is a camera horizontal focal length, fy is a camera vertical focal length, cx is a center point in the horizontal direction of the image coordinate system, and cy is a center point in the vertical direction of the image coordinate system.
The wire point cloud consists of a two-dimensional array of 3*N, wherein the first column describes the position of the point cloud on the x axis in a Cartesian space coordinate system, the second column describes the position of the point cloud on the y axis in the Cartesian space coordinate system, and the third column describes the position of the point cloud on the z axis in the Cartesian space coordinate system. Each 3*1 row vector thus describes the spatial location of one of the wire point clouds. The acquisition process is as follows:
wherein, PC L Is a two-dimensional matrix with 3 rows, M is a camera internal reference matrix, D L Is wire depth information.
The wire depth information is an array with 3 rows, the first row represents the transverse position of each pixel of the wire in the visible light picture, the second row represents the longitudinal position of each pixel of the wire in the visible light picture, and the third row represents the pixel depth information on the depth map position of the visible light picture of the power transmission channel, which is determined by the previous two-dimensional data and is obtained from the depth map of the visible light picture of the power transmission channel.
The power transmission line mechanical external damage hidden danger target detection model is capable of identifying mechanical external damage hidden dangers (various engineering vehicles) in a power transmission line visible light picture.
The mechanical external broken hidden danger frame is a rectangular frame and consists of a one-dimensional array of 1*4. The position of the mechanical external break in the visible light picture of the power transmission channel can be represented. The first number of the array is the abscissa of the upper left corner of the hidden danger frame, the second number is the ordinate of the upper left corner of the hidden danger frame, the third number is the width of the hidden danger frame, namely the number of pixels occupied by the rectangular frame transversely, and the fourth number is the height of the hidden danger frame, namely the number of pixels occupied by the rectangular frame longitudinally.
The mechanical external damage hidden danger target representative point set is composed of pixel points which are uniformly distributed or distributed in some other way in a mechanical external damage hidden danger frame, the position of the mechanical external damage hidden danger in the visible light picture of the power transmission channel can be expressed briefly to a certain extent, the mechanical external damage hidden danger target representative point set is a two-dimensional array with the number of lines of 2, the first line represents the transverse position of the mechanical external damage hidden danger target representative point in the visible light picture of the power transmission channel, and the second line represents the longitudinal position of the mechanical external damage hidden danger target representative point in the visible light picture of the power transmission channel. In the invention, the method for acquiring the target representative point of the hidden danger of mechanical external damage is as follows:
wherein Re is （i）（j） For a mechanical external damage hidden danger target representative point of a certain mechanical external damage hidden danger frame of a visible light picture of a power transmission channel, a mechanical external damage hidden danger target representative point set R is a mechanical external damage hidden danger target representative point Re （i）（j） A two-dimensional array with 2 rows and xl isThe left upper-corner abscissa of the mechanical external damage hidden danger frame, yu is the left upper-corner ordinate of the mechanical external damage hidden danger frame, xr is the right lower-corner abscissa of the mechanical external damage hidden danger frame, yd is the right lower-corner ordinate of the mechanical external damage hidden danger frame, w is the width of the mechanical external damage hidden danger frame, h is the height of the mechanical external damage hidden danger frame, i is the whole number of 0 to w/5 plus one, j is the whole number of 0 to h/5 plus one, [ xl, yu, w, h ]]And forming a certain mechanical external damage hidden danger frame in the visible light picture of the power transmission channel.
The plane position information of the mechanical external damage hidden danger target representative point set refers to the position information of the mechanical external damage hidden danger target representative point set, wherein the position information of the spatial position information of the mechanical external damage hidden danger target representative point set is obtained by removing the height information, and the method comprises the following steps:
for a certain mechanical external damage hidden danger target representative point set R, taking out one mechanical external damage hidden danger target representative point Re （i）（j） Selecting a target representative point Re of hidden danger of mechanical external damage from a depth map of a visible light picture of a power transmission channel （i）（j） The depth of the pixels at the adjacent positions serving as the centers and the pixel positions form a mechanical external damage hidden danger target representative point Re （i）（j） Region depth information Dr of (2) （i）（j） ，
Wherein Lo （i）（j） Is the target representative point Re of hidden danger of mechanical external damage （i）（j） Plane position information of Z （i）（0） Is the height value PCr of the bottom of the mechanical external broken hidden danger frame （i）（j） Is the target representative point Re of hidden danger of mechanical external damage （i）（j） M is a camera internal reference matrix of a camera for shooting a picture of the hidden danger of mechanical external damage, mean is average operation, w is the width of a hidden danger frame of mechanical external damage, h is the height of the hidden danger frame of mechanical external damage, the value range of i is 0 to w/5, the whole value of i is added by one, and the value of j is 0 to h/5, the whole value of j is added by one;
the plane position information RLo of the target representative point set of the hidden danger of mechanical external damage is expressed as follows:
and RLo is plane position information of the target representative point set of the hidden danger of mechanical external damage.
Region depth information Dr （i）（j） The first row represents the target representing point Re of hidden danger of mechanical external damage （i）（j） The second row represents the target representing point Re of hidden danger of mechanical external damage （i）（j） The third row represents the mechanical external damage hidden danger target representing point Re for the longitudinal pixel position of the pixels at the adjacent positions of the center （i）（j） Is the depth value of the center adjacent pixel.
The spatial position information of the mechanical external damage hidden danger target representative point set refers to the spatial position of the mechanical external damage hidden danger target representative point under a three-dimensional spatial coordinate system, and is obtained by adding the planar position information of the corresponding mechanical external damage hidden danger target representative point and the corresponding pixel distance between the mechanical external damage hidden danger target representative point and the bottom frame of the mechanical external damage hidden danger frame through a camera imaging principle, wherein the expression form of the spatial position information is a two-dimensional array with the number of lines of 3. The specific acquisition mode is as follows:
first, the hidden danger target representative point Re is broken from the outside of the machine （i）（j） Obtain a Lo from the plane position information of (2) （i）（j） Taking out one of the target representative points Re from the target representative point set R （i）（j） Re and Re （i）（j） The corresponding mechanical external breaking hidden danger target representative point Re at the lower edge of the mechanical external breaking hidden danger target frame （i）（0） Which corresponds to the plane position information Lo （i）（0） Re and corresponding space （i）（0） ，Lo （i）（0） Refers to a value taken when j=0: lo (Lo) （i）（0） Refers to a mechanical external damage hidden danger target representative point Re at the lower edge of the mechanical external damage hidden danger target frame （i）（0） Is a plane position information of:
therein, PCz （i）（j） The value of the space position information Z direction of the representing point of the hidden danger target of the mechanical external damage is the height information of the representing point of the hidden danger target of the mechanical external damage, namely Imgh （i）（j） For the pixel distance between the representative point of the mechanical external damage hidden danger target and the representative point of the mechanical external damage hidden danger target at the lower edge of the corresponding mechanical external damage hidden danger target frame, dist （i）（j） The Euclidean distance between the representative point of the mechanical external breaking hidden danger target and the origin of the space coordinate system is M [1,1]For the values of the first row and first column of the camera internal parameter matrix, PCda （i）（j） For the spatial position information of the target representative point of the hidden danger of the mechanical external damage, w is the width of the hidden danger frame of the mechanical external damage, h is the height of the hidden danger frame of the mechanical external damage, the value range of i is 0 to w/5, the value of j is 0 to h/5, the value of i is one;
and then, the spatial position information of all the mechanical external damage hidden danger target representative points of each hidden danger frame jointly form the spatial position information of the mechanical external damage hidden danger target representative point set:
RPCda is the spatial position information of the target representative point set of the hidden danger of mechanical external damage.
The minimum distance between the mechanical external damage hidden danger and the power transmission wire is the minimum Euclidean distance between the mechanical external damage hidden danger and the power transmission wire in a three-dimensional space coordinate system. The specific calculation mode is as follows:
firstly, acquiring the number lm of wire point clouds and respectively obtaining the spatial position information PCda of the mechanical external damage hidden danger target representative points from the spatial position information PRCda of the mechanical external damage hidden danger target representative point set （i）（j） 。
Therein, PCx （i）（j）（k） 、PCy （i）（j）（k） 、PCz （i）（j）（k） Spatial position information PCda of kth wire point cloud and mechanical external damage hidden danger target representative point respectively （i）（j） Square of distance in three coordinate axis directions of three-dimensional space coordinate system, PCdist （i）（j）（k） Spatial position information PCda for kth wire point cloud and mechanical external damage hidden danger target representative point （i）（j） The Euclidean distance of the three-dimensional space coordinate system is as follows: 0 to 0
，LDdist （i）（j） Spatial position information PCda for representing point of mechanical external damage hidden danger target （i）（j） Euclidean distance between the LDdist and wire point cloud min Spatial position information PCda for representing point of mechanical external damage hidden danger target （i）（j） The minimum Euclidean distance between the potential broken frame and the wire point cloud is that w is the width of the potential broken frame, h is the height of the potential broken frame, i is the value range of 0 to w/5, and is the value of 0 to h/5.
The invention is further illustrated by the following examples:
example 1
The embodiment uses a picture shot by using a transmission line monitoring device of a certain transmission line maintenance company, the picture size is 1408 x 3, and a camera internal reference matrix corresponding to the transmission line monitoring device is:
a. monocular depth estimation of scene scenery is carried out on the visible light pictures of the transmission channel shot by the transmission line visual monitoring equipment by using the monocular depth estimation model, and a depth map of the visible light pictures of the transmission channel is obtained, wherein the depth map of the visible light pictures of the transmission channel is as follows:
b. conducting wire segmentation in a visible light picture of a transmission channel shot by visual monitoring equipment of a transmission line is subjected to image segmentation by using a deep learning-based transmission wire segmentation model to obtain a wire segmentation pixel position D L The following is shown:
the position of the split pixels of the lead is combined with the depth map of the visible light picture of the transmission channel, and then the lead is reconstructed by combining with the camera internal reference matrix to obtain a lead point cloud PC L The following is shown:
c. the method comprises the steps of obtaining a mechanical external broken hidden danger frame in a visible light picture of a power transmission channel shot by visual monitoring equipment of a power transmission line by using a mechanical external broken hidden danger target detection model of the power transmission line:
and obtaining the plane position of the target representative point set of the hidden danger of mechanical external damage through combining with the depth map of the visible light picture of the power transmission channel. Taking a hidden danger frame Y1 as an example, calculating the plane position of the obtained mechanical external damage hidden danger target representative point set;
d. acquiring spatial position information of a target representative point set of the hidden danger of mechanical external damage by combining with a camera imaging principle;
e. finally, calculating the minimum Euclidean distance by using the space position information of the mechanical external damage hidden danger target representative point set and the wire point cloud as the minimum distance between the mechanical external damage hidden danger in the hidden danger frame and the transmission wire
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (10)
1. The method for measuring the distance of the hidden danger of the external damage of the power transmission machinery is characterized by comprising the following steps:
establishing a monocular depth estimation model, a transmission line segmentation model and a transmission line mechanical external damage hidden danger target detection model;
monocular depth estimation of scene scenery is carried out on the visible light picture of the power transmission channel by adopting a monocular depth estimation model, and a depth map of the visible light picture of the power transmission channel is obtained; carrying out image segmentation on the wires in the visible light pictures of the transmission channels by adopting a transmission wire segmentation model, and obtaining wire segmentation pixel positions; the wire segmentation pixel positions are combined with a depth map of a visible light picture of the power transmission channel, and wire depth information is obtained; converting the wire depth information into wire point clouds through a camera internal reference matrix; acquiring a mechanical external broken hidden danger frame in a visible light picture of a power transmission channel by adopting a mechanical external broken hidden danger target detection model of the power transmission line;
combining the mechanical external damage hidden danger frame with the depth map to acquire plane position information of a mechanical external damage hidden danger target representative point set; acquiring the spatial position information of a target representative point set of the hidden danger of mechanical external damage by using the plane position information through a camera imaging principle; and calculating Euclidean distance through the space position information and the wire point cloud, wherein the minimum value in the Euclidean distance is the minimum distance between the hidden danger of mechanical external damage and the transmission wire.
2. The method for measuring the distance of the hidden danger of the external damage of the power transmission machine according to claim 1, wherein,
the monocular depth estimation model is a depth neural network model and is used for predicting the depth of a scene pixel in a visible light picture of the power transmission channel and outputting a depth map of the visible light picture of the power transmission channel;
the transmission line segmentation model is used for segmenting a transmission line oriented by the visual monitoring equipment in the transmission channel visible light picture, and outputting the pixel position of the transmission line in the transmission channel visible light picture;
the power transmission line mechanical external damage hidden danger target detection model is used for identifying mechanical external damage hidden danger in a power transmission line visible light picture.
3. The method for measuring the distance of potential external damage hazards of power transmission machinery according to claim 1, wherein the position of the divided pixels of the wire is a two-dimensional array, the first-dimensional data is the transverse position of the pixels of the wire in the visible light picture of the power transmission channel, the second-dimensional data is the longitudinal position of the pixels of the wire in the visible light picture of the power transmission channel, and the transverse position and the longitudinal position are used for describing the position information of the wire in the visible light picture of the power transmission channel.
4. The method for measuring the distance of hidden danger of power transmission machinery external damage according to claim 1, wherein the camera internal reference matrix is a 3*3 matrix, which is used for representing the imaging process of the visible light picture of the power transmission channel and realizing the interconversion of the point cloud and the depth map, and the specific calculation method is as follows:
wherein M is a camera reference matrix, fx is a camera horizontal focal length, fy is a camera vertical focal length, cx is a center point in the horizontal direction of the image coordinate system, and cy is a center point in the vertical direction of the image coordinate system.
5. The method for ranging potential external damage hazards of power transmission machinery according to claim 1, wherein the wire point cloud is composed of a two-dimensional array of 3*N, the first column is used for describing the position of the wire point cloud in an x axis in a cartesian coordinate system, the second column is used for describing the position of the wire point cloud in a y axis in the cartesian coordinate system, and the third column is used for describing the position of the wire point cloud in a z axis in the cartesian coordinate system, and the specific calculation method is as follows:
wherein, PC L Is a two-dimensional matrix with 3 rows, M is a camera internal reference matrix, D L Is wire depth information.
6. The method for measuring the distance of the potential external damage hazard of the power transmission machine according to claim 1, wherein the set of the potential external damage hazard target points consists of pixel point positions in a frame of the potential external damage hazard of the power transmission machine, and is a two-dimensional array with the number of lines being 2, the first line of data represents the transverse position of the potential external damage hazard target points of the power transmission machine in the visible light picture of the power transmission channel, and the second line represents the longitudinal position of the potential external damage hazard target points of the power transmission machine in the visible light picture of the power transmission channel, and the specific calculation method is as follows:
wherein Re is （i）（j） The mechanical external damage hidden danger target representative point is a mechanical external damage hidden danger target representative point, and the mechanical external damage hidden danger target representative point set R is a mechanical external damage hidden danger target representative point Re （i）（j） The two-dimensional array with the number of 2 is formed, xl is the abscissa of the upper left corner of the mechanical external broken hidden danger frame, yu is the ordinate of the upper left corner of the mechanical external broken hidden danger frame, xr is the abscissa of the lower right corner of the mechanical external broken hidden danger frame, yd is the ordinate of the lower right corner of the mechanical external broken hidden danger frame, w is the width of the mechanical external broken hidden danger frame, and h is the mechanical external broken hidden danger frameThe height of the external broken hidden danger frame is that the value range of i is 0 to w/5, the whole value is added by one, the value of j is 0 to h/5, the whole value is added by one, and xl, yu, w, h forms the mechanical external broken hidden danger frame of the visible light picture of the power transmission channel.
7. The power transmission machinery external damage hidden danger ranging method of claim 1, wherein the plane position information of the machinery external damage hidden danger target representative point set represents: the position information of the height information is removed from the spatial position information of the mechanical external damage hidden danger target point set, and the calculation mode is as follows:
selecting a target representative point Re of hidden danger of mechanical external damage from a depth map of a visible light picture of a power transmission channel （i）（j） The depth of the pixels at the adjacent positions serving as the centers and the pixel positions form a mechanical external damage hidden danger target representative point Re （i）（j） Region depth information Dr of (2) （i）（j） ，
Wherein Lo （i）（j） Is the target representative point Re of hidden danger of mechanical external damage （i）（j） Plane position information of Z （i）（0） Is the height value PCr of the bottom of the mechanical external broken hidden danger frame （i）（j） Is the target representative point Re of hidden danger of mechanical external damage （i）（j） Is the initial spatial position of PCr （i）（0） Refers to when PCr （i）（j） Value of j=0, PCr （i）（0） Refers to a mechanical external damage hidden danger target representative point Re at the lower edge of the mechanical external damage hidden danger target frame （i）（0） M is an internal reference matrix of the camera, mean is an average value operation, w is the width of the mechanical external damage hidden danger frame, h is the height of the mechanical external damage hidden danger frame, the value range of i is 0 to w/5, the whole value is added by one, and the value of j is 0 to h/5, the whole value is added by one;
the specific calculation mode of the plane position information of the mechanical external damage hidden danger target representative point set is as follows:
and RLo is plane position information of the target representative point set of the hidden danger of mechanical external damage.
8. The method for measuring distance of potential external damage of power transmission machine as recited in claim 7, wherein the regional depth information Dr （i）（j） The first line data represents the target representing point Re of hidden danger of mechanical external damage for a two-dimensional array with 3 lines （i）（j） The second row of data represents the target representing point Re of hidden danger of mechanical external damage for the horizontal pixel position of the pixels at the adjacent positions of the center （i）（j） The longitudinal pixel position of the pixel at the adjacent position of the center is the longitudinal pixel position of the pixel at the adjacent position, and the third data represents the mechanical external damage hidden danger target representative point Re （i）（j） Is the depth value of the center adjacent pixel.
9. The method for measuring the distance of the potential external damage hazards of the power transmission machinery according to claim 7, wherein the spatial position information of the potential external damage hazard target representative point set of the power transmission machinery represents the spatial position of the potential external damage hazard target representative point of the power transmission machinery in a three-dimensional spatial coordinate system, and the specific calculation method is as follows:
therein, PCz （i）（j） The value of the space position information Z direction of the representing point of the hidden danger target of the mechanical external damage is the height information of the representing point of the hidden danger target of the mechanical external damage, namely Imgh （i）（j） For the pixel distance between the representative point of the mechanical external damage hidden danger target and the representative point of the mechanical external damage hidden danger target at the lower edge of the corresponding mechanical external damage hidden danger target frame, dist （i）（j） The Euclidean distance between the representative point of the mechanical external breaking hidden danger target and the origin of the space coordinate system is M [1,1]Is an internal reference of the cameraNumerical value of first column of first row of matrix, pda （i）（j） For the spatial position information of the target representative point of the hidden danger of the mechanical external damage, w is the width of the hidden danger frame of the mechanical external damage, h is the height of the hidden danger frame of the mechanical external damage, the value range of i is 0 to w/5, the value of j is 0 to h/5, the value of i is one;
the space position information of the representative click of the mechanical external damage hidden danger target consists of the space position information of the representative point of the mechanical external damage hidden danger target, and specifically comprises the following steps:
RPCda is the spatial position information of the target representative point set of the hidden danger of mechanical external damage.
10. The method for measuring the distance between the hidden danger of the external damage of the power transmission machine according to claim 1, wherein the specific calculation method for the minimum distance between the hidden danger of the external damage of the power transmission machine and the power transmission wire is as follows:
therein, PCx （i）（j）（k） 、PCy （i）（j）（k） 、PCz （i）（j）（k） Spatial position information PCda of kth wire point cloud and mechanical external damage hidden danger target representative point respectively （i）（j） Square of distance in three coordinate axis directions of three-dimensional space coordinate system, PCdist （i）（j）（k） Spatial position information PCda for kth wire point cloud and mechanical external damage hidden danger target representative point （i）（j） The Euclidean distance of the three-dimensional space coordinate system is as follows: 0 to 0
Lm is the number of wire point clouds, LDdist （i）（j） Spatial position information PCda for representing point of mechanical external damage hidden danger target （i）（j） Euclidean distance between the LDdist and wire point cloud min Spatial position information PCda for representing point of mechanical external damage hidden danger target （i）（j） The minimum Euclidean distance between the potential broken frame and the wire point cloud is that w is the width of the potential broken frame, h is the height of the potential broken frame, i is the value range of 0 to w/5, and is the value of 0 to h/5. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310361171.6A CN116091494B (en) | 2023-04-07 | 2023-04-07 | Method for measuring distance of hidden danger of external damage of power transmission machinery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310361171.6A CN116091494B (en) | 2023-04-07 | 2023-04-07 | Method for measuring distance of hidden danger of external damage of power transmission machinery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116091494A true CN116091494A (en) | 2023-05-09 |
| CN116091494B CN116091494B (en) | 2023-06-27 |
Family
ID=86202926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310361171.6A Active CN116091494B (en) | 2023-04-07 | 2023-04-07 | Method for measuring distance of hidden danger of external damage of power transmission machinery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116091494B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245932A (en) * | 2023-05-10 | 2023-06-09 | 智洋创新科技股份有限公司 | Hidden danger ranging method for power transmission line based on grid idea |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018029270A (en) * | 2016-08-18 | 2018-02-22 | キヤノン株式会社 | Image processing apparatus, control method thereof, imaging apparatus, and program |
| CN113345019A (en) * | 2021-06-09 | 2021-09-03 | 山东信通电子股份有限公司 | Power transmission line channel hidden danger target ranging method, equipment and medium |
| CN114240911A (en) * | 2021-12-21 | 2022-03-25 | 广东电网有限责任公司 | Method and system for measuring and calculating distance of broken object outside power transmission line |
| CN115311560A (en) * | 2022-07-29 | 2022-11-08 | 国网江苏省电力有限公司电力科学研究院 | Method and device for identifying hidden danger of external damage of power transmission channel |
| CN115902870A (en) * | 2022-11-09 | 2023-04-04 | 广东电网有限责任公司广州供电局 | Visual monitoring method and system for distance of hidden danger of external damage of power transmission line |
| CN115909092A (en) * | 2022-09-29 | 2023-04-04 | 广西大学 | Light-weight power transmission channel hidden danger distance measuring method and hidden danger early warning device |
-
2023
- 2023-04-07 CN CN202310361171.6A patent/CN116091494B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018029270A (en) * | 2016-08-18 | 2018-02-22 | キヤノン株式会社 | Image processing apparatus, control method thereof, imaging apparatus, and program |
| CN113345019A (en) * | 2021-06-09 | 2021-09-03 | 山东信通电子股份有限公司 | Power transmission line channel hidden danger target ranging method, equipment and medium |
| CN114240911A (en) * | 2021-12-21 | 2022-03-25 | 广东电网有限责任公司 | Method and system for measuring and calculating distance of broken object outside power transmission line |
| CN115311560A (en) * | 2022-07-29 | 2022-11-08 | 国网江苏省电力有限公司电力科学研究院 | Method and device for identifying hidden danger of external damage of power transmission channel |
| CN115909092A (en) * | 2022-09-29 | 2023-04-04 | 广西大学 | Light-weight power transmission channel hidden danger distance measuring method and hidden danger early warning device |
| CN115902870A (en) * | 2022-11-09 | 2023-04-04 | 广东电网有限责任公司广州供电局 | Visual monitoring method and system for distance of hidden danger of external damage of power transmission line |
Non-Patent Citations (2)
| Title |
|---|
| LI CHENG ET.AL: "Obstacles detection and depth estimation from monocular vision for inspection robot of high voltage transmission line", 《CLUSTER COMPUTING》, pages 2611 * |
| 张骥;余娟;汪金礼;谭守标;: "基于深度学习的输电线路外破图像识别技术", 计算机系统应用, no. 08, pages 180 - 183 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245932A (en) * | 2023-05-10 | 2023-06-09 | 智洋创新科技股份有限公司 | Hidden danger ranging method for power transmission line based on grid idea |
| CN116245932B (en) * | 2023-05-10 | 2023-08-22 | 智洋创新科技股份有限公司 | Hidden danger ranging method for power transmission line based on grid idea |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116091494B (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110264416B (en) | Sparse point cloud segmentation method and device | |
| CN106356757B (en) | A kind of power circuit unmanned plane method for inspecting based on human-eye visual characteristic | |
| CN109544679B (en) | Three-dimensional reconstruction method for inner wall of pipeline | |
| CN110288586A (en) | A kind of multiple dimensioned transmission line of electricity defect inspection method based on visible images data | |
| CN113345019B (en) | Method, equipment and medium for measuring potential hazards of transmission line channel target | |
| CN107560592B (en) | Precise distance measurement method for photoelectric tracker linkage target | |
| CN116091494B (en) | Method for measuring distance of hidden danger of external damage of power transmission machinery | |
| CN110059602B (en) | Forward projection feature transformation-based overlook human face correction method | |
| CN113192646B (en) | Target detection model construction method and device for monitoring distance between different targets | |
| TWI745204B (en) | High-efficiency LiDAR object detection method based on deep learning | |
| CN111046877A (en) | Millimeter wave image suspicious article detection method and system | |
| Xu et al. | Dynamic obstacle detection based on panoramic vision in the moving state of agricultural machineries | |
| CN115641553A (en) | Online detection device and method for invaders in heading machine working environment | |
| CN114396875B (en) | Rectangular package volume measurement method based on vertical shooting of depth camera | |
| CN106709432B (en) | Human head detection counting method based on binocular stereo vision | |
| CN113970734B (en) | Method, device and equipment for removing snowfall noise points of road side multi-line laser radar | |
| CN116091493B (en) | Distance measurement method for hidden danger of tree obstacle of power transmission line | |
| CN114066999A (en) | Target positioning system and method based on three-dimensional modeling | |
| CN108805218B (en) | Optical target association method based on deviation mapping clustering | |
| CN116883483A (en) | Fish body measuring method based on laser camera system | |
| CN116895059A (en) | BEV space target detection method and device for multi-view perspective image | |
| CN109360269B (en) | Ground three-dimensional plane reconstruction method based on computer vision | |
| CN113963107B (en) | Binocular vision-based large-scale target three-dimensional reconstruction method and system | |
| CN109493354B (en) | Target two-dimensional geometric shape reconstruction method based on multi-view images | |
| CN114494342A (en) | Method for detecting and tracking marine target of visible light sequence image of synchronous orbit satellite |
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 |