CN117268344A

CN117268344A - Method, device, equipment and medium for predicting high-risk source of electric tower line tree

Info

Publication number: CN117268344A
Application number: CN202311532812.6A
Authority: CN
Inventors: 黄全进; 王宇翔; 张攀; 沈均平
Original assignee: Aerospace Hongtu Information Technology Co Ltd
Current assignee: Aerospace Hongtu Information Technology Co Ltd
Priority date: 2023-11-17
Filing date: 2023-11-17
Publication date: 2023-12-22
Anticipated expiration: 2043-11-17
Also published as: CN117268344B

Abstract

The invention provides a method, a device, equipment and a medium for predicting high risk sources of a tower line tree, which comprise the following steps: acquiring remote sensing image data and tree height image data of a region to be detected; identifying a target tower detection frame contained in the region to be detected based on the remote sensing image data, and identifying tree height distribution data corresponding to the region to be detected based on the tree height image data; reasoning the electric tower lines in the area to be detected according to the target electric tower detection frame so as to determine an electric tower pair list corresponding to the area to be detected; and predicting whether a tree height dangerous source exists on a tower line in the region to be detected by using elevation information, tree height distribution data and a tower pair list corresponding to the region to be detected, so as to obtain a dangerous tree height position corresponding to the region to be detected. The invention can obviously improve the automation degree of predicting the tree high dangerous sources existing in the electric tower line, thereby achieving the purposes of reducing the inspection workload, the inspection cost, the inspection difficulty, improving the inspection efficiency and the like.

Description

Method, device, equipment and medium for predicting high-risk source of electric tower line tree

Technical Field

The invention relates to the technical field of dangerous source prediction, in particular to a method, a device, equipment and a medium for predicting a high dangerous source of a tower line tree.

Background

As the scale of the power grid is continuously increased, the complexity of the line is rapidly increased, and a great challenge is brought to the reliability maintenance of the power grid. In order to ensure safe and stable supply of electric power, regular inspection of the transmission line is required. At present, the high-risk source inspection of the electric tower line tree still adopts a manual inspection mode, and the problems of large inspection workload, higher inspection cost, large inspection difficulty, low inspection efficiency and the like exist.

Disclosure of Invention

In view of the above, the present invention aims to provide a method, a device and a medium for predicting a tree high risk source of a tower line, which can significantly improve the automation degree of predicting the tree high risk source existing in the tower line, thereby achieving the purposes of reducing the inspection workload, the inspection cost, the inspection difficulty, improving the inspection efficiency, etc.

In a first aspect, an embodiment of the present invention provides a method for predicting a high risk source of a tower line tree, including:

acquiring remote sensing image data and tree height image data of a region to be detected;

identifying a target tower detection frame contained in the to-be-detected area based on the remote sensing image data, and identifying tree height distribution data corresponding to the to-be-detected area based on the tree height image data;

Reasoning the electric tower lines in the to-be-detected area according to the target electric tower detection frame so as to determine an electric tower pair list corresponding to the to-be-detected area; the electric tower pair list is used for recording coordinate information of a plurality of electric tower pairs, and the electric tower circuit is arranged between two electric towers contained in the electric tower pairs;

and predicting whether a tree height dangerous source exists on the electric tower line in the to-be-detected area by utilizing the elevation information, the tree height distribution data and the electric tower pair list corresponding to the to-be-detected area, so as to obtain a dangerous tree height position corresponding to the to-be-detected area.

In one embodiment, the step of identifying the target tower detection frame included in the to-be-detected area based on the remote sensing image data includes:

intercepting a plurality of images to be detected from the remote sensing image data according to a preset overlapping rate and a preset image size;

performing electric tower detection on the image to be detected through an electric tower detection model obtained through pre-training to obtain an initial electric tower detection frame contained in the image to be detected and a confidence corresponding to the initial electric tower detection frame;

screening the initial electric tower detection frame based on the confidence coefficient to obtain a target electric tower detection frame contained in the region to be detected;

The step of identifying the tree height distribution data corresponding to the region to be detected based on the tree height image data comprises the following steps:

and predicting the tree height of the region to be detected based on the tree height image data through a pre-trained tree height prediction model so as to obtain tree height distribution data corresponding to the region to be detected.

In one embodiment, the step of screening the initial tower detection frame based on the confidence level to obtain a target tower detection frame contained in the to-be-detected area includes:

if the confidence coefficient corresponding to the initial electric tower detection frame is larger than a preset confidence coefficient threshold value, determining the initial electric tower detection frame as a target electric tower detection frame;

and/or, for any two of the initial tower detection frames, detecting an intersection region between the two initial tower detection frames to determine the detection frame overlap rate or detection frame similarity between the two initial tower detection frames based on the intersection region;

and if the overlapping rate of the detection frames is larger than a preset overlapping rate threshold value or if the similarity of the detection frames is larger than a preset similarity threshold value, determining the initial electric tower detection frame corresponding to the maximum confidence as a target electric tower detection frame.

In one embodiment, according to the target tower detection frame, reasoning is performed on the tower lines in the to-be-detected area to determine a tower pair list corresponding to the to-be-detected area, including:

determining a designated point of the electric tower detection frame;

creating a first set and a second set; wherein the first set is used for storing the specified points of the determined shortest paths, and the second set is used for storing the specified points of the undetermined shortest paths;

selecting a current target specified point closest to a starting point from the second set, and storing the current target specified point into the first set;

determining a current adjacent designated point corresponding to the current target designated point from the second set, and updating a first distance between the current target designated point and the current adjacent designated point when the first distance is smaller than a second distance between the starting point and the current adjacent designated point;

determining a next target designated point from the current adjacent designated points based on the updated second distance, storing the next target designated point into the first set, and continuing to determine the next adjacent designated point corresponding to the next target designated point from the second set until the second set is empty, so as to obtain electric tower connection data in the region to be detected; wherein the tower wiring data is used for characterizing the tower wiring;

And storing the coordinate information of the two towers corresponding to the tower connection data as a tower pair list.

In one embodiment, the step of predicting whether a tree height dangerous source exists on the tower line in the to-be-detected area by using elevation information corresponding to the to-be-detected area, the tree height distribution data and the tower pair list to obtain a dangerous tree height position corresponding to the to-be-detected area includes:

based on the tree height distribution data, the electric tower pair list is adjusted to obtain a target electric tower pair list;

for each electric tower pair in the target electric tower pair list, determining the height of an actual point corresponding to any point on a line segment between a first electric tower and a second electric tower in the electric tower pair according to the elevation information corresponding to the area to be detected;

extracting a tree height value corresponding to the point from the tree height distribution data, and taking the sum value of the elevation information corresponding to the point and the tree height value as an actual tree height value corresponding to the point;

and if the difference value between the actual point height corresponding to the point and the actual tree height value is smaller than a preset difference value threshold, determining that a tree height dangerous source exists at the position of the point, and determining the position of the point as a dangerous tree height position.

In one embodiment, the step of adjusting the tower pair list based on the tree height distribution data to obtain a target tower pair list includes:

for each electric tower pair in the electric tower pair list, constructing a linear equation of the electric tower pair, and selecting a plurality of reference points from the linear equation;

extracting a tree height value corresponding to each reference point from the tree height distribution data;

if the tree height value corresponding to any reference point is greater than 0, reserving the electric tower pair; and if the tree height value corresponding to each reference point is equal to 0, eliminating the tower pairs to obtain a target tower pair list.

In one embodiment, after the step of predicting whether a tree height dangerous source exists on the tower line in the to-be-detected area by using the elevation information corresponding to the to-be-detected area, the tree height distribution data and the tower pair list to obtain a dangerous tree height position corresponding to the to-be-detected area, the method further includes:

determining a dangerous level corresponding to the dangerous tree high position according to a threshold interval where a difference value between the actual point height corresponding to the dangerous tree high position and the actual tree high value is located;

Wherein, there is a mapping relationship between the threshold interval and the risk level.

In a second aspect, an embodiment of the present invention further provides a device for predicting a high risk source of a tower line tree, including:

the data acquisition module is used for acquiring remote sensing image data and tree height image data of the area to be detected;

the electric tower and tree height identification module is used for identifying a target electric tower detection frame contained in the to-be-detected area based on the remote sensing image data and identifying tree height distribution data corresponding to the to-be-detected area based on the tree height image data;

the electric tower pair determining module is used for reasoning electric tower lines in the to-be-detected area according to the target electric tower detection frame so as to determine an electric tower pair list corresponding to the to-be-detected area; the electric tower pair list is used for recording coordinate information of a plurality of electric tower pairs, and the electric tower circuit is arranged between two electric towers contained in the electric tower pairs;

and the dangerous source prediction module is used for predicting whether a tree-height dangerous source exists on the electric tower line in the to-be-detected area by utilizing the elevation information, the tree-height distribution data and the electric tower pair list corresponding to the to-be-detected area, so as to obtain a dangerous tree-height position corresponding to the to-be-detected area.

In a third aspect, an embodiment of the present invention further provides an electronic device comprising a processor and a memory storing computer-executable instructions executable by the processor to implement the method of any one of the first aspects.

In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions which, when invoked and executed by a processor, cause the processor to implement the method of any one of the first aspects.

According to the method, the device, the equipment and the medium for predicting the tree height dangerous source of the electric tower line, remote sensing image data and tree height image data of an area to be detected are firstly obtained, a target electric tower detection frame contained in the area to be detected is identified based on the remote sensing image data, and tree height distribution data corresponding to the area to be detected is identified based on the tree height image data; then, according to a target electric tower detection frame, reasoning electric tower lines in the to-be-detected area to determine an electric tower pair list corresponding to the to-be-detected area, wherein the electric tower pair list is used for recording coordinate information of a plurality of electric tower pairs, and electric tower lines are arranged between two electric towers contained in the electric tower pairs; and finally, predicting whether a tree height dangerous source exists on a tower line in the region to be detected by using elevation information, tree height distribution data and a tower pair list corresponding to the region to be detected, so as to obtain a dangerous tree height position corresponding to the region to be detected. According to the method, the target tower detection frame and the tree height distribution data are respectively identified by utilizing the remote sensing image data and the tree height image data, then the tower line is inferred on the basis of the target tower detection frame to obtain the corresponding tower pair list, so that the tree height dangerous sources on the tower line are predicted by combining the tree height distribution data and the height data.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for predicting a high risk source of a tower line tree according to an embodiment of the present invention;

FIG. 2 is a diagram of the overall result of the detection of the electric tower according to the embodiment of the present invention;

FIG. 3 is a diagram of a partial result of tower detection according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of tree height image data according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of tree height distribution data according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a connection line for a center point of a tower according to an embodiment of the present invention;

FIG. 7 is a visual diagram of a tree height greater than 0 at a point coordinate corresponding to an electrical tower connection line according to an embodiment of the present invention;

FIG. 8 is a schematic diagram illustrating calculation of a point height on a tower connection according to an embodiment of the present invention;

FIG. 9 is a diagram of a visualized result of a hazard source on tree height data according to an embodiment of the present invention;

FIG. 10 is a diagram of a result of visualizing a hazard on an optical remote sensing image according to an embodiment of the present invention;

FIG. 11 is a schematic flow chart of another method for predicting high risk sources of a tower line tree according to an embodiment of the present invention;

FIG. 12 is a schematic structural diagram of a high risk source prediction device for a tower line tree according to an embodiment of the present invention;

fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described in conjunction with the embodiments, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

At present, the manual inspection mode is still adopted for inspecting the high-risk source of the electric tower line tree, and the problems of large inspection workload, high inspection cost, high inspection difficulty, low inspection efficiency and the like exist.

For the convenience of understanding the present embodiment, first, a method for predicting a high risk source of a tower line tree disclosed in the present embodiment will be described in detail, referring to a schematic flow chart of a method for predicting a high risk source of a tower line tree shown in fig. 1, the method mainly includes steps S102 to S108:

step S102, remote sensing image data and tree height image data of the area to be detected are obtained.

The remote sensing image data can adopt high-resolution optical remote sensing image data, and the tree height image data can adopt tree height optical image data.

Step S104, identifying a target tower detection frame contained in the to-be-detected area based on the remote sensing image data, and identifying tree height distribution data corresponding to the to-be-detected area based on the tree height image data.

The tree height distribution data are used for representing the distribution condition and the height information of trees growing in the to-be-detected area. In one embodiment, a tower detection model (such as YOLOv8 algorithm) and a sliding window detection algorithm can be used to detect the remote sensing image data of the area to be detected, so as to obtain detection frames of all towers on the remote sensing image data; and simultaneously, predicting the region to be detected by using a trained tree height prediction model (such as a SegFormer algorithm) to obtain tree height distribution data.

And S106, reasoning the tower lines in the area to be detected according to the target tower detection frame so as to determine a tower pair list corresponding to the area to be detected.

The electric tower pair list is used for recording coordinate information of a plurality of electric tower pairs, and an electric tower line is arranged between two electric towers contained in the electric tower pairs. In one embodiment, a designated point (such as a center point) of the target tower detection frame may be extracted, and the tower lines in the area to be detected are inferred, so as to obtain a connection line between the center points of the target tower detection frames in the area to be detected, where the connection line is used to characterize the tower lines, two towers with the connection line are used as tower pairs, and coordinate information of the two towers is stored in a tower pair list.

And S108, predicting whether a tree height dangerous source exists on a tower line in the area to be detected by using elevation information, tree height distribution data and a tower pair list corresponding to the area to be detected, so as to obtain a dangerous tree height position corresponding to the area to be detected.

In one embodiment, for each tower pair in the tower pair list, performing preliminary prediction on whether tree-height dangerous sources exist on two tower lines included in the tower pair according to tree-height distribution data, storing the tower pairs with the preliminary prediction on which the tree-height dangerous sources exist, and eliminating the tower pairs with the preliminary prediction on which the tree-height dangerous sources do not exist; the method comprises the steps of calculating the height of any point in a tower line between two towers by further utilizing elevation information and coordinate information of two towers in the stored tower pair, determining the actual tree height value at the point based on tree height distribution data and elevation information, comparing the actual point height with the actual tree height value, and judging whether a tree height dangerous source exists at the point based on a comparison result to further obtain the dangerous tree height position.

Optionally, because the tower line between the two towers is affected by gravity, the tower line will present a certain radian, so the actual point height can be further corrected, and the corrected actual point height is compared with the actual tree height value to obtain the dangerous tree height position. Illustratively, the correction process of the actual dot height is as follows: and fitting the electric tower line by using a Taylor formula to obtain an electric tower line expression, determining the coordinate of any point on the electric tower line based on the electric tower line expression, determining the other coordinate of the point by using a linear line segment expression between two electric towers, and correcting the height of the actual point by using the difference value between the two coordinates.

According to the method for predicting the tree height dangerous source of the electric tower line, the remote sensing image data and the tree height image data are utilized to respectively identify the target electric tower detection frame and the tree height distribution data, then the electric tower line is inferred on the basis of the target electric tower detection frame to obtain a corresponding electric tower pair list, so that the tree height dangerous source on the electric tower line is predicted by combining the tree height distribution data and the height data.

Along with the rapid development of aerospace technology, remote sensing data are increasingly widely applied in various fields. In the power industry, the remote sensing data is of great significance in predicting the tree height dangerous sources on the electric tower line.

Firstly, the security of the electric power facilities can be improved by predicting the tree height dangerous sources on the electric tower line by using the remote sensing data. The power line passes through a plurality of areas with dense trees, if the heights of the trees are too high, the power line can be contacted with the electric wires, so that the electric wires are broken, short-circuited and the like, and further the damage of power facilities is caused, and even safety accidents such as fire disaster and the like are triggered. Through predictive analysis of remote sensing data, a power grid company can make judgment and prevention in advance, and safety of electric power facilities is guaranteed.

Second, predicting high risk trees helps to improve the reliability of the power supply. The utility company can trim according to remote sensing data more pertinently, prevent that the tree height from stretching into the electric wire to improve the reliability and the stability of power supply line. The application of remote sensing data also helps to save operational costs. If the potential risk of the power line can be formed by the tree heights of the areas in advance, maintenance work such as tree pruning can be effectively planned and executed, emergency repair after the line is in a problem is avoided, and the maintenance cost is greatly saved. The remote sensing data is used for predicting the tree height dangerous source on the electric tower line, so that the environment is protected. Through accurate prediction and subsequent targeted tree pruning, excessive pruning of non-dangerous trees can be avoided, and the ecological environment is protected.

Based on the above, the embodiment of the invention provides a specific implementation mode of a high risk source prediction method for a tower line tree.

For the foregoing step S104, when the step of identifying the target tower detection frame included in the region to be detected based on the remote sensing image data is performed, the following steps A1 to A3 may be referred to:

and A1, intercepting a plurality of images to be detected from the remote sensing image data according to a preset overlapping rate and a preset image size.

In one embodiment, the sliding window size and the sliding step length may be set so as to intercept a plurality of images to be detected from the remote sensing image data by using a sliding window detection algorithm, wherein the preset image size is related to the sliding window size, and the preset overlapping rate is related to the sliding step length.

By way of example, assuming that the overlap ratio is set to 20% and the image size is set to 1024 in both width and height, a data source can be constructed from image blocks (i.e., images to be detected) of 1024 in both width and height, sequentially taken from the high resolution image at 20% overlap ratio.

And step A2, performing electric tower detection on the image to be detected through an electric tower detection model obtained through pre-training so as to obtain an initial electric tower detection frame contained in the image to be detected and a confidence corresponding to the initial electric tower detection frame.

The electric tower detection model can adopt a YOLOv8 electric tower detection model. In one embodiment, the images to be detected may be sequentially obtained from the data source, and input into the YOLOv8 electric tower detection model to perform electric tower detection, so as to obtain an initial electric tower detection frame and a confidence coefficient corresponding to the initial electric tower detection frame included in each image to be detected.

And step A3, screening the initial electric tower detection frame based on the confidence coefficient to obtain a target electric tower detection frame contained in the region to be detected.

In one embodiment, the initial tower detection frames with lower confidence and higher overlapping rate can be filtered out by using a confidence threshold and NMS (non maximum suppression) algorithm, and the initial tower detection frames which are not removed are target tower detection frames.

Specifically, the target tower detection frame included in the to-be-detected area may be determined according to the steps shown in the following first to second modes:

mode one: and if the confidence coefficient corresponding to the initial electric tower detection frame is larger than a preset confidence coefficient threshold value, determining the initial electric tower detection frame as a target electric tower detection frame.

In one example, for each initial tower detection frame, determining whether a confidence threshold corresponding to the initial tower detection frame is greater than a preset confidence threshold, if so, reserving the initial tower detection frame, determining the initial tower detection frame as a target tower detection frame, and if not, rejecting the initial tower detection frame.

Mode two: for any two initial electric tower detection frames, detecting an intersection area between the two initial electric tower detection frames to determine a detection frame overlapping rate or a detection frame similarity between the two initial electric tower detection frames based on the intersection area; if the overlapping rate of the detection frames is larger than a preset overlapping rate threshold value, or if the similarity of the detection frames is larger than a preset similarity threshold value, determining the initial electric tower detection frame corresponding to the maximum confidence as a target electric tower detection frame.

The two initial electric tower detection frames are respectively marked as a first detection frame and a second detection frame, an intersection area between the first detection frame and the second detection frame is detected, the ratio between the area of the intersection area and the area of any electric tower detection frame is calculated, the ratio is used for representing the overlapping rate of the detection frames, whether the overlapping rate of the detection frames is larger than a preset overlapping threshold value is judged, and if not, the first detection frame and the second detection frame are reserved; if so, eliminating one detection frame from the first detection frame and the second detection frame.

Further, when the detection frame is removed, the following steps may be adopted: if the confidence coefficient corresponding to the first detection frame is greater than that corresponding to the second detection frame, reserving the first detection frame and eliminating the second detection frame, namely, the first detection frame is the target electric tower detection frame; otherwise, the second detection frame is the target electric tower detection frame.

In practical applications, the initial tower detection frame may be selected by any one of the first mode, the second mode or a combination of the two modes, which is not limited in the embodiment of the present invention. Illustratively, after screening the initial tower test frames in the first and second modes, a tower test overall result diagram shown in fig. 2 and a tower test partial result diagram shown in fig. 3 are obtained, where points in fig. 2 are used to characterize the target tower test frame, and boxes in fig. 3 are used to characterize the target tower test frame.

Alternatively, after the target tower detection box is determined, the center point (i.e., the designated point) of the target tower detection box may be extracted and saved to the list.

For the foregoing step S104, when the step of identifying the tree height distribution data corresponding to the region to be detected based on the tree height image data is performed, the following steps may be referred to: and predicting the tree height of the region to be detected based on the tree height image data through a pre-trained tree height prediction model so as to obtain tree height distribution data corresponding to the region to be detected.

In one embodiment, referring to a schematic diagram of tree height image data shown in fig. 4, the tree height image data may be input into a trained SegFormer tree height prediction model, the tree height distribution of the area to be detected is predicted by the SegFormer tree height prediction model, and the prediction result is stored in a single-channel image, where the tree height size is the pixel value of the single-channel image, and the single-channel image is the tree height distribution data, such as the schematic diagram of a tree height distribution data shown in fig. 5.

For the step S106, when the step of reasoning the electric tower lines in the area to be detected according to the target electric tower detection frame to determine the electric tower pair list corresponding to the area to be detected is performed, the electric tower center point connection line can be obtained by using the center point of the target electric tower detection frame and the shortest path search algorithm, and the black connection line, that is, the electric tower center point connection line, can be referred to as a schematic diagram of the electric tower center point connection line shown in fig. 6, so that the electric tower center point connection line data is utilized to determine whether the electric tower lines between two electric towers need to be subjected to tree height dangerous source determination, and two electric tower coordinates needing to be determined are reserved in pairs in the list to form the electric tower pair list, thereby reducing the time consumption of carrying out tree height dangerous source prediction on the whole electric tower connection line in the area to be detected.

In specific implementation, the central point of the electric tower detection frame and a shortest path search algorithm are used for completing reasoning of the electric tower line trend, and coordinates of all points from the beginning to the end of the electric tower central point connecting line in the current area are sequentially stored by using a list to form an electric tower pair list. Wherein the shortest path search algorithm comprises the following steps B1 to B6:

and B1, determining the designated point of the electric tower detection frame. The designated point adopts a central point, namely, the central point of the electric tower detection frame is determined.

And B2, creating a first set and a second set. Wherein the first set is used for storing the specified points of the determined shortest paths and the second set is used for storing the specified points of the undetermined shortest paths.

Illustratively, the distances of all center points are set to infinity, and the distances of the starting points are set to 0. Creating an empty set S (i.e. a first set) for storing the center points for which the shortest paths have been determined; a set Q (i.e., a second set) is created containing all the center points for the undetermined shortest paths.

And B3, selecting a current target specified point closest to the starting point from the second set, and storing the current target specified point into the first set.

Illustratively, a center point u with the smallest distance, that is, the above-mentioned current target specific point, is selected from the set Q, and added to the set S. If u is the endpoint, the algorithm ends; otherwise, the step B4 is continued.

And B4, determining a current adjacent designated point corresponding to the current target designated point from the second set, and updating the second distance when the first distance between the current target designated point and the current adjacent designated point is smaller than the second distance between the starting point and the current adjacent designated point.

Illustratively, all center points v adjacent to the center point u, i.e., the currently neighboring designated point, are traversed, and the distance from the start point to the center point v is updated if the distance from the center point u to the center point v is less than the currently known distance from the start point to the center point v.

Step B5, determining a next target designated point from the current adjacent designated points based on the updated second distance, storing the next target designated point into the first set, and continuously determining the next adjacent designated point corresponding to the next target designated point from the second set until the second set is empty, so as to obtain electric tower connection data in the region to be detected; wherein the tower wiring data is used to characterize the tower wiring.

For example, from among all the center points v, the center point v having the shortest distance from the start point may be determined as the next target specification point and stored in the set S, and this process may be repeated until the set Q is empty or the end point is found.

And step B6, storing coordinate information of two towers corresponding to the tower connection data as a tower pair list.

For the step S106, when the step of predicting whether the tree height dangerous source exists on the tower line in the area to be detected by using the elevation information, the tree height distribution data and the tower pair list corresponding to the area to be detected to obtain the dangerous tree height position corresponding to the area to be detected is performed, all coordinate points on the connection line between the paired towers can be obtained by using the tower pair list by using a linear equation, the tree height value is obtained by using the coordinates of the points and the tree height distribution diagram, and all points with the tree height greater than 0 on the line are saved. And judging all dangerous tree height positions on the power tower line by using the actual tree height values of each point and the elevation (DEM) data of the area to be detected and the height of the power tower.

In a specific implementation, the following steps C1 to C4 can be referred to:

and step C1, adjusting the electric tower pair list based on the tree height distribution data to obtain a target electric tower pair list. Specifically, the method comprises the following steps C1-1 to C1-3:

and C1-1, constructing a linear equation of each electric tower pair in the electric tower pair list, and selecting a plurality of reference points from the linear equation.

In one example, tower pairs are sequentially taken from a list of all tower pairs, a linear equation between 2 points (tower coordinates) in the tower pairs is fitted, and 10 points are uniformly taken from the linear equation as reference points.

And C1-2, extracting the tree height value corresponding to each reference point from the tree height distribution data.

In one example, the coordinate information of the reference point may be determined using the above linear equation, and then the corresponding tree height value may be extracted from the tree height distribution data based on the coordinate information.

Step C1-3, if the tree height value corresponding to any reference point is greater than 0, reserving a tower pair; and if the tree height value corresponding to each reference point is equal to 0, eliminating the tower pairs to obtain a target tower pair list.

In one example, if there are points with tree height values greater than 0, the coordinates of the 2 towers are saved, so as to form a target tower pair list, see a visual chart of tree height greater than 0 at the point coordinates corresponding to one tower connection line shown in fig. 7, the points shown in fig. 7 are all tower coordinates, and the black line segment is a point with tree height value greater than 0 on the connection line between the coordinates.

And C2, for each tower pair in the target tower pair list, determining the actual point height corresponding to any point on a line segment between the first tower and the second tower in the tower pair according to the elevation information corresponding to the area to be detected.

In specific implementation, the coordinates of each point on the line segment of the tower pair are obtained according to a calculation method (y=kx+b) related to the slope of the straight line. According to the coordinates of a certain point, the position coordinates of the two towers and DEM data, the height of the point on the tower connecting line can be calculated.

For example, referring to a schematic diagram of calculation of the height of a point on a tower connection shown in fig. 8, on the basis of fig. 8, the specific steps of calculation are as follows:

firstly, a certain Point (x, y) on a power tower connecting line is downwards shifted by one power tower height, the power tower height is mapped to a connecting line position p (x, y) between the bottoms of 2 power towers, the bottom of a second power tower p2 is used as a horizontal plane, according to DEM data, the terrain heights of a first power tower p1 and a second power tower p2 can be obtained according to pixel Point coordinate values, and H is the height difference between the first power tower p1 and the second power tower p 2. Using the rule of similar triangle correlation, there is formula (1):

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>、/>Coordinates of the first electric tower p1 and the second electric tower p2 are respectively +. >Is the coordinates of point p (x, y), -, for example>Is the height of point p (x, y).

Since the height difference between the two towers can be directly obtained from DEM data, as shown in formula (2):

the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>、/>The elevations of the first electric tower p1 and the second electric tower p2 are respectively.

Can be obtained according to the formula (1) and the formula (2)Since the level of the topography of the location of the electric tower p2 is here taken as the level, the electric tower height +.>Let =40 is known, so the actual dot height of the dot +.>The method comprises the following steps:

。

and C3, extracting a tree height value corresponding to the point from the tree height distribution data, and taking the sum value of the elevation information corresponding to the point and the tree height value as an actual tree height value corresponding to the point.

In one example, the tree height value and the terrain height of the point can be obtained by combining the tree height distribution data and the DEM data according to the position of the point, and the actual tree height value of the tree height of the point can be obtained by addition operationBy calculation ofAnd->And (3) obtaining whether the tree height of the point belongs to a dangerous source or not.

And C4, if the difference value between the actual point height corresponding to the point and the actual tree height value is smaller than a preset difference value threshold, determining that a tree height dangerous source exists at the position of the point, and determining the position of the point as a dangerous tree height position.

For example, assuming that the preset difference threshold is 5m, if the difference between the actual point height corresponding to the point and the actual tree height value is greater than 5m, the tree height dangerous source is considered to be absent; otherwise, if the difference between the actual point height corresponding to the point and the actual tree height value is smaller than 5 meters, determining the position of the point as the dangerous tree height position.

Further, the tree high risk level can be divided: determining a dangerous level corresponding to the dangerous tree height position according to a threshold interval in which a difference value between the actual point height corresponding to the dangerous tree height position and the actual tree height value is positioned; wherein, there is mapping relation between threshold interval and dangerous level. By way of example, by setting a threshold value for the difference between the actual point height and the actual tree height value and corresponding to different risk levels, such as a serious risk within 2m, a risk within 2 m-5 m, a safety outside 5m, etc., different measures may be taken according to the different levels.

In summary, the embodiment of the invention combines the multi-mode remote sensing data and the artificial intelligent algorithm to predict the high risk source of the electric tower line tree for the first time, verifies the actual data, and can achieve the purpose of accurately predicting the risk source of the electric tower line within the error allowable range. Compared with the traditional manual inspection method, the method has the following advantages: the electric tower line inspection device has the advantages that a large amount of manpower and material resources can be saved, and the electric tower line which is difficult to reach by manpower can be inspected, so that the electric tower line inspection device is not influenced by environment and weather. The method has the advantages that the artificial intelligence is adopted, the automatic inspection speed is higher, and the prediction of the high risk source of the electric tower line tree in a certain area can be completed in a short time. The method has the advantages that the adaptability is strong, and for the areas with large environmental differences, only a small amount of local optical remote sensing images and tree height image fine tuning tower detection and tree height prediction models are required to be obtained, so that the method can be applied to the whole area. For example, referring to a visualized result diagram of a hazard source on tree height data shown in fig. 9 and a visualized result diagram of a hazard source on an optical remote sensing image shown in fig. 10, fig. 9 and 10 are respectively visualized result diagrams of a tree height hazard source on tree height data and optical remote sensing data when a difference between a tree height and a line height is less than 5m, and black line segments in the diagrams represent positions of the hazard tree height.

In order to facilitate understanding, the embodiment of the invention provides an application example of a method for predicting a high risk source of a power tower line tree by taking a tin-free city as an example, referring to a flow diagram of another method for predicting a high risk source of a power tower line tree shown in fig. 11, firstly, optical remote sensing data of the tin-free city are obtained, all power tower positions and power tower central point connecting lines are determined through a power tower detection model, meanwhile, tree height image data of the tin-free city are obtained, tree height distribution data is predicted through a tree height prediction model, and the risk source judgment is performed by combining the power tower central point connecting lines, the tree height distribution data and DEM data of the tin-free city.

Aiming at the problems that the existing remote sensing data and an artificial intelligent algorithm are used for predicting the tree high risk sources on the electric tower line, the work is less, and the efficiency of manually inspecting the tree high risk sources on the electric tower line is low, the embodiment of the invention provides a method for predicting the tree high risk sources on the electric tower line by combining multi-mode remote sensing data and an artificial intelligent technology. The method combines the advantages of an optical remote sensing image, tree height data, DEM data and YOLOv8 and SegFormer artificial intelligent algorithm, can rapidly complete detection of the electric tower in a certain area, predicts tree height and predicts dangerous tree height positions on an electric tower line, greatly reduces cost of manpower and material resources, and can be widely applied to inspection of the electric tower line in the power industry.

On the basis of the foregoing embodiments, the embodiment of the present invention provides a device for predicting a high risk source of a tower line tree, referring to a schematic structural diagram of the device for predicting a high risk source of a tower line tree shown in fig. 12, where the device mainly includes the following parts:

the data acquisition module 1202 is configured to acquire remote sensing image data and tree height image data of an area to be detected;

the tower and tree height identification module 1204 is configured to identify a target tower detection frame included in the to-be-detected area based on the remote sensing image data, and identify tree height distribution data corresponding to the to-be-detected area based on the tree height image data;

the tower pair determining module 1206 is configured to infer a tower line in the area to be detected according to the target tower detection frame, so as to determine a tower pair list corresponding to the area to be detected; the electric tower pair list is used for recording coordinate information of a plurality of electric tower pairs, and an electric tower circuit is arranged between two electric towers contained in the electric tower pairs;

the dangerous source prediction module 1208 is configured to predict whether a tree-height dangerous source exists on a tower line in the area to be detected by using elevation information, tree-height distribution data and a tower pair list corresponding to the area to be detected, so as to obtain a dangerous tree-height position corresponding to the area to be detected.

According to the high-risk source prediction device for the electric tower line, the remote sensing image data and the high-risk source image data are utilized to respectively identify the target electric tower detection frame and the high-risk source distribution data, then the electric tower line is inferred on the basis of the target electric tower detection frame to obtain a corresponding electric tower pair list, so that the high-risk source on the electric tower line is predicted by combining the high-risk source distribution data and the high-risk data.

In one embodiment, the tower and tree height identification module 1204 is further configured to:

Intercepting a plurality of images to be detected from remote sensing image data according to a preset overlapping rate and a preset image size;

performing electric tower detection on an image to be detected through an electric tower detection model obtained through pre-training, so as to obtain an initial electric tower detection frame contained in the image to be detected and a confidence corresponding to the initial electric tower detection frame;

the tower and tree height identification module 1204 is further configured to:

and/or, for any two initial tower detection frames, detecting an intersection region between the two initial tower detection frames to determine a detection frame overlap rate or a detection frame similarity between the two initial tower detection frames based on the intersection region;

If the overlapping rate of the detection frames is larger than a preset overlapping rate threshold value, or if the similarity of the detection frames is larger than a preset similarity threshold value, determining the initial electric tower detection frame corresponding to the maximum confidence as a target electric tower detection frame.

In one embodiment, the tower pair determination module 1206 is further configured to:

determining a designated point of a tower detection frame;

creating a first set and a second set; wherein the first set is used for storing the designated points of the determined shortest paths, and the second set is used for storing the designated points of the undetermined shortest paths;

selecting a current target designated point closest to the starting point from the second set, and storing the current target designated point into the first set;

determining a current adjacent designated point corresponding to the current target designated point from the second set, and updating the second distance when the first distance between the current target designated point and the current adjacent designated point is smaller than the second distance between the starting point and the current adjacent designated point;

determining a next target designated point from the current adjacent designated points based on the updated second distance, storing the next target designated point to the first set, and continuously determining the next adjacent designated point corresponding to the next target designated point from the second set until the second set is empty, so as to obtain electric tower connection data in the area to be detected; the electric tower connection data are used for representing electric tower lines;

In one embodiment, the hazard source prediction module 1208 is further configured to:

based on the tree height distribution data, adjusting the electric tower pair list to obtain a target electric tower pair list;

for each electric tower pair in the target electric tower pair list, determining the actual point height corresponding to any point on a line segment between a first electric tower and a second electric tower in the electric tower pair according to the elevation information corresponding to the area to be detected;

if the difference value between the actual point height corresponding to the point and the actual tree height value is smaller than a preset difference value threshold, determining that a tree height dangerous source exists at the position of the point, and determining the position of the point as a dangerous tree height position.

If the tree height value corresponding to any reference point is greater than 0, reserving a tower pair; and if the tree height value corresponding to each reference point is equal to 0, eliminating the tower pairs to obtain a target tower pair list.

determining a dangerous level corresponding to the dangerous tree height position according to a threshold interval in which a difference value between the actual point height corresponding to the dangerous tree height position and the actual tree height value is positioned;

wherein, there is mapping relation between threshold interval and dangerous level.

The device provided by the embodiment of the present invention has the same implementation principle and technical effects as those of the foregoing method embodiment, and for the sake of brevity, reference may be made to the corresponding content in the foregoing method embodiment where the device embodiment is not mentioned.

The embodiment of the invention provides electronic equipment, which comprises a processor and a storage device; the storage means has stored thereon a computer program which, when executed by the processor, performs the method of any of the embodiments described above.

Fig. 13 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, where the electronic device 100 includes: a processor 130, a memory 131, a bus 132 and a communication interface 133, the processor 130, the communication interface 133 and the memory 131 being connected by the bus 132; the processor 130 is arranged to execute executable modules, such as computer programs, stored in the memory 131.

The memory 131 may include a high-speed random access memory (RAM, random Access Memory), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is implemented via at least one communication interface 133 (which may be wired or wireless), and may use the internet, a wide area network, a local network, a metropolitan area network, etc.

Bus 132 may be an ISA bus, a PCI bus, an EISA bus, or the like. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 13, but not only one bus or type of bus.

The memory 131 is configured to store a program, and the processor 130 executes the program after receiving an execution instruction, and the method executed by the apparatus for flow defining disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 130 or implemented by the processor 130.

The processor 130 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuitry in hardware or instructions in software in processor 130. The processor 130 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a digital signal processor (Digital Signal Processing, DSP for short), application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in the memory 131, and the processor 130 reads the information in the memory 131, and in combination with the hardware, performs the steps of the above method.

The computer program product of the readable storage medium provided by the embodiment of the present invention includes a computer readable storage medium storing a program code, where the program code includes instructions for executing the method described in the foregoing method embodiment, and the specific implementation may refer to the foregoing method embodiment and will not be described herein.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. The method for predicting the high-risk source of the electric tower line tree is characterized by comprising the following steps of:

2. The method for predicting a high risk source of a tower route tree according to claim 1, wherein the step of identifying a target tower detection frame included in the region to be detected based on the remote sensing image data comprises:

3. The method for predicting a high risk source of a tower wire tree according to claim 2, wherein the step of screening the initial tower detection frame based on the confidence level to obtain a target tower detection frame contained in the to-be-detected area comprises the steps of:

and/or, for any two initial electric tower detection frames, detecting an intersection area between the two initial electric tower detection frames to determine a detection frame overlapping rate or a detection frame similarity between the two initial electric tower detection frames based on the intersection area;

4. The method for predicting a high risk source of a tower line tree according to claim 1, wherein the step of reasoning the tower line in the to-be-detected area according to the target tower detection frame to determine a tower pair list corresponding to the to-be-detected area comprises the steps of:

determining a designated point of the electric tower detection frame;

5. The method for predicting a tree height hazard source of a tower line according to claim 1, wherein the step of predicting whether a tree height hazard source exists on the tower line in the area to be detected by using elevation information corresponding to the area to be detected, the tree height distribution data and the tower pair list to obtain a hazard tree height position corresponding to the area to be detected comprises the steps of:

6. The method for predicting a high risk source of a tower line tree according to claim 5, wherein the step of adjusting the tower pair list based on the tree height distribution data to obtain a target tower pair list comprises:

7. The method for predicting a tree height risk source of a tower line according to claim 5, wherein after predicting whether a tree height risk source exists on the tower line in the area to be detected by using elevation information corresponding to the area to be detected, the tree height distribution data and the tower pair list, the method further comprises:

8. The utility model provides a high dangerous source prediction device of electricity tower circuit tree which characterized in that includes:

9. An electronic device comprising a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer readable storage medium storing computer executable instructions which, when invoked and executed by a processor, cause the processor to implement the method of any one of claims 1 to 7.