Power facility collapse assessment method based on SAR data of remote sensing images

Abstract

The present invention discloses a method for evaluating the collapse of power facilities based on SAR data of remote sensing images, belonging to the technical field of image processing, including: Step S1: Obtain the analysis point data in the power facilities through remote sensing images and generate the distribution maps of each power facility; Step S2: Extract four adjacent analysis point data in the power facilities and combine and connect them to generate a plurality of second combined surfaces; Step S3: Obtain the second vector of the power facilities, calculate the included angle between the first vector and the second vector of each second combined surface in the power facilities based on the time interval to obtain the corresponding second included angle set; Step S4: Generate a prediction model based on the change rate of the angle change sequence and output a prediction result set, and evaluate the danger index of the power facilities and the safety range between each of the power facilities based on the distribution map. Through the present invention, the collapse degree of power facilities in different regions can be evaluated and monitored, thereby avoiding the problem of untimely maintenance of power facilities.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and particularly relates to a method for evaluating the collapse of power facilities based on SAR data of remote sensing images. Background Art

[0002] Power facilities are a general term for a series of facilities such as electrical equipment, metal structures, and auxiliary buildings used for power generation, transmission, transformation, and distribution. They are widely distributed with diverse distribution environments. Monitoring the safety of each power facility through SAR data of remote sensing images not only saves labor costs but also can accurately identify the degree of collapse and damage and the affected range of each power facility, effectively monitoring each power facility.

[0003] For example, Chinese Patent Application "CN108594226B" discloses a method for detecting transmission tower frames in mountainous areas using SAR images considering terrain, including the following steps: for the original single-polarization SAR image set, calculate its corresponding average radar power image; at the same time, with the help of external elevation data, generate a range-direction slope angle image; estimate the quantitative relationship between the average radar power and the range-direction slope angle, and use the obtained relationship to simulate the terrain-related power image; subtract the simulated power image from the real power image to obtain a radar power image with reduced terrain influence; perform dual-parameter constant false alarm detection on the target in the radar power image with reduced terrain influence to obtain a binary image; perform target pixel clustering analysis on the binary image and extract according to the linear arrangement feature. This invention effectively eliminates the influence of terrain in single-polarization SAR images by simulating the amplitude image with elevation data, highlighting the ground object targets, and improving the detection accuracy using linear extraction and clustering algorithms. Another example is Chinese Patent Application "CN110488151B" which discloses a vegetation risk warning system and method for transmission lines based on remote sensing technology. The system includes: a vegetation growth prediction module, a vegetation height extraction module, a power line elevation extraction module, a relative distance calculation module, and a vegetation risk warning module. This invention combines the advantages of spaceborne remote sensing technology and airborne remote sensing technology and can realize the safety distance assessment of vegetation risk sections. According to the assessment results, it is possible to timely cut down overgrown vegetation, and at the same time formulate a more reasonable inspection plan, reduce the number of manual inspections, and reduce the consumption of human and material resources while ensuring the safety of the transmission line.

[0004] However, in the above-mentioned prior art, only mountainous power facilities and transmission lines are detected by using radar remote sensing data. In actual situations, power facilities with their bottoms covered by plants may collapse due to gravity and geology, and when the degree of collapse of power facilities is large, it will affect other connected power facilities. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method, system and storage medium for evaluating the collapse of power facilities based on remote sensing image SAR data to solve the problems in the prior art.

[0006] To achieve the above invention purpose, the present invention proposes a method for evaluating the collapse of power facilities based on remote sensing image SAR data, including:

[0007] Step S1: Obtain an image data set containing power facilities through remote sensing images, identify each feature point data in the image data set, obtain the feature coordinates of the feature point data, take the first distance as the target, and cluster each feature point data in combination with the feature coordinates to generate a distribution map of each power facility, and define the feature point data located in the same power facility as analysis point data;

[0008] Step S2: Extract four adjacent analysis point data in the power facility and combine them to generate a first combined surface. The first combined surface includes multiple sub - surfaces. Obtain the first vectors of the sub - surfaces in the first combined surface, and determine whether all the first vectors are parallel to each other. If so, merge the sub - surfaces into the first combined surface, and then perform the combined connection of the first combined surface with adjacent analysis point data to generate a second combined surface. If not, perform the combined connection of the sub - surfaces with adjacent analysis point data to generate the second combined surface. Respectively judge each first vector in the second combined surface, and repeat this step until the analysis point data in the same power facility is expanded;

[0009] Step S3: Obtain the second vector of the power facility, calculate the included angle between the first vector of each second combined surface in the power facility and the second vector, and define it as a first included - angle set. Set a time interval, and execute Step S1 and Step S2 based on the time interval to correspondingly generate the first vectors of each second combined surface and define them as updated vectors. Obtain the included angle between the updated vector and the second vector, and define it as a second included - angle set;

[0010] Step S4: Obtain an angle change sequence based on the second included - angle set and the time interval, generate a prediction model based on the change rate of the angle change sequence, predict the change rate based on the prediction model to output a prediction result set, and evaluate the danger index of the power facility and the safety range between each power facility based on the prediction result set and the distribution map.

[0011] Further, in Step S1, each feature point data is clustered to generate a distribution map of each power facility based on the following steps:

[0012] Calculate the coordinate distance between two pieces of the feature point data based on the feature coordinates, extract one piece of the feature point data and set it as the first type of labeled point, obtain the feature point data whose coordinate distance from the first type of labeled point is less than or equal to the first distance and set it as the first type of labeled point, and repeat this step to obtain all the first type of labeled points among all the feature point data. If the coordinate distance between the feature point data and the first type of labeled point is greater than the first distance, then set the feature point data as the second type of labeled point, obtain the feature point data whose coordinate distance from the second type of labeled point is less than or equal to the first distance and set it as the second type of labeled point, and repeat this step to cluster all the feature point data in the image dataset into multiple label types;

[0013] Count the number of label types of the feature point data corresponding to each power facility, and set the number of label types as the distribution quantity. Based on the distribution quantity, extract one piece of the feature point data from both power facilities and calculate the coordinate distance between the two pieces of the feature point data. Set the minimum value of the coordinate distance as the distribution distance between the corresponding two power facilities. Extract two pieces of the analysis point data from the power facilities to calculate the coordinate distance, and set the maximum value of the coordinate distance as the height value corresponding to the power facility. Generate the distribution map of each power facility based on the distribution quantity, the distribution distance, the height value, and the feature coordinates.

[0014] Further, in step S3, obtain the second vector based on the following steps and set the time interval:

[0015] Calculate the second vector based on the first formula , and the first formula is: , where and are the first vectors corresponding to two adjacent second combined faces, is the vector cross product operation, is the first vector is the variance value of each analysis point data included in the second combined face corresponding to the first vector is the first vector is the variance value of each analysis point data included in the second combined face corresponding to the first vector, and m is the number of permutations and combinations of two adjacent second combined faces of the power facility;

[0016] Set the time interval based on the second formula , and the second formula is: , where is the first angle set, is the set of the second included angles corresponding to the i-th time interval, and are respectively the first and the second randomly preset time intervals.

[0017] Further, in the step S3, in the step S4, evaluating the risk index of the power facilities based on the prediction result set and the distribution map includes the following steps:

[0018] Obtaining the change rate of the angle change sequence based on the third formula , and the third formula is , obtaining the predicted change rate of the angle change sequence based on the fourth formula , and the fourth formula is: , where , is the change rate corresponding to the i-th time interval, is the set of the second included angles corresponding to the i-th time interval, and obtaining the prediction result set based on the fifth formula , and the fifth formula is: , obtaining the risk index W based on the sixth formula, and the sixth formula is: , where is the standard normal distribution function, is the average value of the respective change rates, is the standard deviation of the respective change rates.

[0019] Further, in the step S4, obtaining the safety range between the respective power facilities based on the following steps:

[0020] Calculating the safety range S of the respective power facilities based on the seventh formula, and the seventh formula is: , where is the height value of the power facility, is the distribution distance between two connected power facilities in the distribution map. When the risk index is greater than or equal to a preset evaluation threshold, obtaining the second combined surface corresponding to the subset with the largest change rate in the prediction result set in the power facilities as the collapse surface, obtaining the analysis point data included in the collapse surface as the collapse point data, mapping the collapse point data based on the second vector, and determining whether there is collapse point data mapped within the safety range of each power facility. If so, maintaining the power facility; if not, maintaining the power facility and evaluating and monitoring other connected power facilities.

[0021] The present invention also provides a power facility collapse assessment system based on remote sensing image SAR data, which is used to implement the power facility collapse assessment method based on remote sensing image SAR data described above. The system mainly includes:

[0022] An image extraction module, which is used to obtain an image data set containing power facilities in a remote sensing image, identify each feature point data in the image data set, obtain the feature coordinates of the feature point data, take the first distance as the target, and cluster each feature point data based on the feature coordinates to generate a distribution map of each power facility, and define the feature point data located in the same power facility as analysis point data;

[0023] A three-dimensional construction module, which is used to extract four adjacent analysis point data in the power facility and combine them to generate a first combined surface. The first combined surface includes multiple sub-planes. Obtain the first vector of the sub-planes in the first combined surface, and determine whether all the first vectors are parallel to each other. If so, merge the sub-planes into the first combined surface, and then perform the combined connection of the first combined surface and the adjacent analysis point data to generate the second combined surface. If not, perform the combined connection of the sub-planes and the adjacent analysis point data to generate the second combined surface, and respectively judge each first vector in the second combined surface, and repeat this step until the analysis point data in the same power facility is expanded and completed;

[0024] An evaluation and monitoring module, which is used to obtain the second vector of the power facility, calculate the included angle between the first vector of each second combined surface in the power facility and the second vector, and define it as the first included angle set. Set a time interval, and perform step S1 and step S2 based on the time interval to correspondingly generate the first vector of each second combined surface and define it as an updated vector, and obtain the included angle between the updated vector and the second vector, and define it as the second included angle set;

[0025] A prediction and analysis module, which obtains an angle change sequence based on the second included angle set and the time interval, generates a prediction model based on the change rate of the angle change sequence, predicts the change rate based on the prediction model and outputs a prediction result set, and evaluates the danger index of the power facility and the safety range between each power facility based on the prediction result set and the distribution map.

[0026] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0027] First, the present invention extracts feature point data from an image dataset containing various power facilities by using remote sensing images. Through clustering, the feature point data belonging to each power facility in the image dataset can be classified and extracted. The distribution map between each power facility is constructed from the characteristic coordinates of the analysis point data in each power facility, which is convenient for analyzing the safety between each power facility. Then, the analysis point data contained in the power facilities are combined and connected, and the generated second combined surfaces and corresponding first vectors are judged. Further, a three-dimensional construction of the power facilities is carried out. Finally, the set of included angles between each first vector and the second vector in the power facilities is obtained. By setting a time interval and repeating the above steps to generate a corresponding second included angle set, the update vectors of each surface in the power facilities can be monitored.

[0028] The present invention also constructs a prediction model by using an angle change sequence, outputs a prediction result set to obtain the set of included angles between the update vectors of each surface and the second vector in each power facility in the next time interval, and calculates a danger index to judge the collapse degree of the power facilities. The safety range of each power facility is calculated through the distribution map of each power facility. Not only can the collapsed power facilities be maintained in time, but also it can be judged whether there is a safety impact on other connected power facilities. If there is a safety impact, the same steps of monitoring are carried out on other connected power facilities in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of the steps of the power facility collapse assessment method based on remote sensing image SAR data of the present invention;

[0030] Figure 2 It is a flowchart of the steps of generating each second combined surface in the power facilities of the present invention;

[0031] Figure 3 It is a distribution map between each power facility of the present invention;

[0032] Figure 4 It is a structural diagram of the power facility collapse assessment system based on remote sensing image SAR data of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0034] It can be understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish the first element from another element. For example, without departing from the scope of this application, the first xx script may be referred to as the second xx script, and similarly, the second xx script may be referred to as the first xx script.

[0035] As Figure 1 shown, a power facility collapse assessment method based on remote sensing image SAR data includes:

[0036] Step S1: Obtain an image data set containing power facilities through remote sensing images, identify the respective feature point data in the image data set, obtain the feature coordinates of the feature point data, and use the first distance as the target. Combine the feature coordinates to cluster the respective feature point data to generate a distribution map of each power facility, and define the feature point data located in the same power facility as analysis point data.

[0037] Specifically, SAR radar remote sensing has strong penetration ability for clouds and dark light, and can provide rich surface information to obtain clear and accurate SAR image data. Power facilities refer to any power buildings that need to be monitored, including but not limited to power towers, power stations, and building structures, etc. In this embodiment, the remote sensing image refers to a surface area image containing multiple power facilities generated by the SAR remote sensing running in a circular motion with the longitude and latitude coordinates of the power facility as the center point. The surface area image is intercepted according to the set image frame rate, so that there is an overlapping area containing each power facility in two images with adjacent image frame rates. The image data set is composed of each image frame rate. The feature point data in each image data set is extracted by the SURF accelerated robust feature algorithm of image processing technology. The feature point data refers to the feature data set combined by the local features of each image in the image data set. For example, the pixel protruding points, corner points, and edge points of each surface of the building, etc. The corresponding feature coordinates are obtained from the feature point data included in each overlapping area of the image data set. The feature coordinates are the position information used to describe the feature point data based on the image coordinates. For example, the feature point data . The feature point data of multiple power facilities and other buildings are included in the image dataset. By setting the first distance, the feature point data of different building facilities can be clustered and distinguished. For example, the shortest actual distance between connected power facilities is 10 meters, which is represented by 5 pixel points in the remote sensing image. Then the first distance is set to 5. Based on the feature point data of the power facilities, clustering is performed with the first distance of 5 to extract all the feature point data included in each power facility. The specific clustering method will be explained in detail later. Based on the number of results after clustering, the number of power facilities included in the image dataset is judged. According to the characteristic coordinates of the feature point data in each power facility, the distribution distance between each power facility and the height value corresponding to the power facility are calculated. Thus, the distribution map of each power facility in the image dataset is constructed, which is convenient for analyzing and judging the safety impact degree of each adjacent power facility. Among them, the analysis point data refers to the feature point data included in each power facility.

[0038] Step S2: Extract four adjacent analysis point data in the power facility and combine and connect them to generate a first combined surface. The first combined surface includes multiple sub-planes. Obtain the first vector of the sub-planes in the first combined surface, and judge whether all the first vectors are parallel to each other. If so, merge the sub-planes into the first combined surface, and then combine and connect the first combined surface with adjacent analysis point data to generate a second combined surface. If not, combine and connect the sub-planes with adjacent analysis point data to generate a second combined surface. Judge each first vector in the second combined surface respectively, and repeat this step until the analysis point data in the same power facility is expanded and completed.

[0039] Specifically, in this embodiment, in order to accurately analyze the collapse degree of each power facility, it is necessary to perform three-dimensional construction on the analysis point data included in the power facility based on the corresponding characteristic coordinates to restore the structure diagram of the power facility. As Figure 2 shown, the combined connection means connecting the included analysis point data crosswise in pairs to form a line, and further generating a first combined surface , which includes multiple sub-planes , and . The first vector refers to the normal vector of any plane. The normal vector is perpendicular to the plane and can be used to describe the inner direction of any plane in a three-dimensional space. If all the first vectors are parallel, it means that the corresponding planes are parallel or belong to the same plane. For example, for the first vectors f1, f2, and f3 of the sub-planes in the first combined surface , where the first vectors f1, f2, and f3 are not parallel to each other, then connect each sub-plane , and in the first combined plane with other adjacent analysis point data to form a second combined surface. For example, the sub-plane With the analysis point data Expand to form a second combined surface , and perform the same judgment steps on it. Connect all the analysis point data in the power facility through the above combination. Each second combined surface is all the planes of the power facility, and the first vector corresponding to each second combined plane is the connection structure of all the planes in the power facility. Thus, the three-dimensional construction of the corresponding power facility is completed.

[0040] Step S3: Obtain the second vector of the power facility, calculate the angle between the first vector of each second combined surface in the power facility and the second vector, and define it as the first angle set. Set a time interval, and based on the time interval, execute Step S1 and Step S2 to correspondingly generate the first vector of each second combined surface and define it as the updated vector. Obtain the angle between the updated vector and the second vector, and define it as the second angle set.

[0041] Specifically, in this embodiment, the second vector refers to the gravity vector of the power facility in the earth coordinate system, and its specific acquisition method will be explained in detail later. The first angle set refers to the angle set generated by the angles between the first vectors corresponding to each second combined surface and the second vector when the above operations are first performed on the power facility. For example, the first angle set A0 is . The time interval refers to the monitoring time interval of the power facility, and its specific setting method will be explained in detail later. Through the time interval, obtain the updated vector of the first vector corresponding to each surface of the power facility, and at the same time obtain the angle between the updated vector of each surface and the second vector to generate the second angle set. For example, the first time interval is set to 1 hour, then after an interval of 1 hour, execute Step S1 and Step S2 to generate the updated vectors of each second combined surface in the power facility M1. The second angle set A1 between the updated vectors of the power facility M1 and the second vector is .

[0042] Step S4: Obtain the angle change sequence based on the second angle set and the time interval, generate a prediction model based on the change rate of the angle change sequence, predict the change rate based on the prediction model to output a prediction result set, and evaluate the danger index of the power facility and the safety range between each power facility based on the prediction result set and the distribution map.

[0043] Specifically, in this embodiment, the angle change sequence refers to the angle data set of the second included angle set corresponding to the power facility changing with the time interval. The prediction model analyzes and predicts the change rate of the next time interval by using the change rate between the angle change sequences obtained for each adjacent time interval. By obtaining the change rate of the next time interval, a prediction result set can be output. The prediction result set refers to the included angle size between the update vector of each second combined surface and the second vector of the power facility in the next time interval. The danger index intuitively evaluates the degree of collapse of the power facility. If the danger index of the power facility is greater than the evaluation threshold, not only does the power facility need to be maintained, but it will also affect the safety of other power facilities connected to the power facility. The influence range refers to the safety range between each connected power facility. Based on the safety range of this power facility, the positions of other power facilities in the distribution map that will be affected by safety can be determined, and further real-time monitoring of other power facilities can be carried out. For example, as Figure 3 shown, if the danger index of the power facility M1 is too large and it collapses, it is determined whether the other power facilities M2 and M3 connected to the power facility M1 are affected by the collapse angle of the power facility M1. If the collapse angle of the power facility M1 is within the safety range of the power facility M2 but outside the safety range of the power facility M3, due to the influence of the force on the transmission line between the power facilities, only the power facility M3 needs to be continuously monitored.

[0044] The present invention first extracts the feature point data from the image data set containing each power facility by using remote sensing images. Through clustering, the feature point data belonging to each power facility in the image data set can be classified and extracted. The distribution map between each power facility is constructed from the feature coordinates of the analysis point data in each power facility, which is convenient for analyzing the safety between each power facility. Then, the analysis point data contained in the power facility are combined and connected, and each second combined surface and the corresponding first vector are generated and judged. Further, a three-dimensional construction of the power facility is carried out. Finally, the included angle set between each first vector and the second vector in the power facility is obtained. By setting the time interval, the above steps are repeatedly executed to generate the corresponding second included angle set, so as to monitor the update vectors of each surface in the power facility.

[0045] The present invention also constructs a prediction model by using the angle change sequence, outputs a prediction result set to obtain the included angle set between the update vector of each surface and the second vector of the power facility in the next time interval, and calculates the danger index to judge the degree of collapse of the power facility. The safety range of each power facility is calculated from the distribution map of each power facility. Not only can the collapsed power facility be maintained in time, but it can also be judged whether there is a safety impact on other connected power facilities. If there is a safety impact, the same steps of monitoring are carried out on other connected power facilities in time.

[0046] Particularly noteworthy is that through the present invention, it is possible to conduct a collapse assessment of power facilities and judge the safety of other connected power facilities, thereby solving the problems of untimely maintenance of power facilities and damage to adjacent power facilities.

[0047] In step S1, the following steps are used to cluster the data of each feature point to generate a distribution map of each power facility:

[0048] Calculate the coordinate distance between two feature point data based on the feature coordinates, extract a feature point data and set it as the first type of label point, obtain the feature point data whose coordinate distance from the first type of label point is less than or equal to the first distance and set it as the first type of label point, and repeat this step to obtain all the first type of label points in the data of each feature point. If the coordinate distance between the feature point data and the first type of label point is greater than the first distance, set the feature point data as the second type of label point, obtain the feature point data whose coordinate distance from the second type of label point is less than or equal to the first distance and set it as the second type of label point, and repeat this step to cluster all the feature point data in the image dataset into multiple label types;

[0049] Count the number of label types of the feature point data corresponding to each power facility, and set the number of label types as the distribution quantity. Based on the distribution quantity, extract a feature point data from both power facilities and calculate the coordinate distance between the two feature point data. Set the minimum value of the coordinate distance as the distribution distance between the corresponding two power facilities. Extract two analysis point data from the power facility and calculate the coordinate distance, and set the maximum value of the coordinate distance as the height value of the corresponding power facility. Generate a distribution map of each power facility based on the distribution quantity, distribution distance, height value, and feature coordinates.

[0050] Specifically, in this embodiment, the coordinate distance between any two feature point data is calculated by the three-dimensional space distance formula. Extract a feature point data belonging to a power facility from the pixel point features of the power facility in the image dataset and set it as the first type of label point. It is possible to cluster through the label to identify the feature point data belonging to different buildings. For example, with the first label point as the center and a radius of 5 for the first distance, form a circular area, and set all the feature point data within the circular area as the first type of label point. Then, execute this step with a radius of 5 for the first distance for each first type of label point to cluster all the first type of label points in the image dataset. Extract a feature point data from the remaining feature point data in the image dataset and set it as the second type of label point, and repeat the above steps to cluster the feature point data of each building;

[0051] Set the number of label types belonging to each power facility to the number of power facilities in the image dataset. The distribution distance refers to the interval distance between any two connected power facilities, and the height value refers to the straight-line distance between the highest point and the lowest point of the power facility. For example, after clustering, there are the first type of label points, the second type of label points, and the third type of label points belonging to the power facility, then there are three connected power facilities in the image dataset. The minimum value of the coordinate distance between each sub-label point in the first type of label points and the second type of label points is 5. According to the characteristic coordinates of the two sub-label points, the distribution directions of the corresponding first type of label points and the second type of label points can be obtained. The maximum value of the coordinate distance between any two sub-label points in the first type of label points is 100. By the same method, the distribution diagrams of each power facility are constructed based on the distribution distance, distribution direction, and height value of each power facility.

[0052] In step S3, obtain the second vector based on the following steps and set the time interval:

[0053] Calculate the second vector based on the first formula , the first formula is: , where and are the first vectors corresponding to two adjacent second combined faces, is the vector cross product operation, is the first vector the variance value of the data of each analysis point included in the second combined face corresponding to is the first vector the variance value of the data of each analysis point included in the second combined face corresponding to, m is the number of permutations and combinations of two adjacent second combined faces of the power facility;

[0054] Set the time interval based on the second formula , the second formula is: , where is the first set of included angles, is the second set of included angles corresponding to the i-th time interval, and are the first time interval and the second time interval randomly preset respectively.

[0055] Specifically, in this embodiment, since each power facility is constructed based on the gravitational direction of the earth coordinates, the plane vectors composed of the characteristic point data on the ground cannot be fully used as reference coordinates. It is necessary to obtain the gravitational vector of the power facility in the earth coordinates. In the first formula, by performing an outer product operation on the first vectors in any two adjacent second combined surfaces of the power facility and then multiplying by a set coefficient value, the combined vector of these two second combined surfaces can be generated. After performing this operation on each adjacent second combined surface and then calculating the average vector, the gravitational direction of the power facility can be generated. For example, by substituting the first vectors of each surface in the power facility obtained from the first monitoring into the first formula to calculate the second vector, the gravitational direction of the power facility can be accurately described in a three-dimensional construction without a reference object.

[0056] Setting the time interval can not only effectively monitor the update vectors of each surface in the power facility, but also reduce the device operation load during the image processing process. For example, the first time interval is set to 1 hour, and the subsequent time intervals are sequentially set according to the change rate of the second angle set.

[0057] In step S4, evaluating the risk index of the power facility based on the prediction result set and the distribution map includes the following steps:

[0058] Obtaining the change rate of the angle change sequence based on the third formula , and the third formula is , obtaining the predicted change rate of the angle change sequence based on the fourth formula , and the fourth formula is: , where , is the change rate corresponding to the i-th time interval, is the second angle set corresponding to the i-th time interval, obtaining the prediction result set based on the fifth formula , and the fifth formula is: , obtaining the risk index W based on the sixth formula, and the sixth formula is: , where is the standard normal distribution function, is the average value of each change rate, is the standard deviation of each change rate.

[0059] Specifically, in this embodiment, the change rate refers to the change speed of the included angle between the update vectors of each surface in the power facility and the second gravity. The change rate of the previous time interval can be generated by the ratio of the difference between the second included angle sets monitored in any two adjacent time intervals. By predicting the change rate, the values of each subset in the second included angle set of the power facility after a future time interval can be obtained. In the sixth formula, specific values can be output by using the standard normal distribution function. For example, the second included angle set A2 obtained by monitoring in the second time interval is , and the second included angle set A1 corresponding to the first time interval are used to calculate the change rate of the second time. is (0, 10, 0, -10), and its risk index is 0.8, which is greater than the evaluation threshold of 0.3. This indicates that the collapse degree of the power facility is relatively large. Further, the collapse angle is judged whether it is within the safety range of the connected power facilities based on the collapse point data of the power facility.

[0060] In step S4, the safety range between each power facility is obtained based on the following steps:

[0061] Calculate the safety range S of each power facility based on the seventh formula. The seventh formula is: , where is the height value of the power facility, is the distribution distance between two connected power facilities in the distribution map. When the risk index is greater than or equal to the preset evaluation threshold, the second combined surface corresponding to the subset with the largest change rate in the prediction result set in the power facility is defined as the collapse surface, and the analysis point data included in the collapse surface is defined as the collapse point data. The collapse point data is mapped based on the second vector, and it is judged whether there is collapse point data mapped within the safety range of each power facility. If so, the power facility is maintained; if not, the power facility is maintained and the other connected power facilities are evaluated and monitored.

[0062] Specifically, in this embodiment, as Figure 3As shown, the safety range is dynamically set relative to the connected power facilities. Specifically, it is the safety range generated with the connected power facilities as the center. In the seventh formula, in order to avoid damage to the transmission lines connected between power facilities due to the pulling caused by the collapse of power facilities, the distribution distance and the height value of the connected power facilities are substituted into the seventh formula to calculate the safety range of the connected power facilities. Through the safety range, it can be judged whether any connected power facilities will be affected by the connected transmission lines if they collapse, and whether the transmission lines between power facilities will break and cause safety problems. If within the safety range, the connected power facilities will not be affected and the lines will not break. If outside the safety range, not only the power facilities need to be maintained in time, but also other connected power facilities need to be monitored in time, and the transmission lines between them need to be repaired in time. For example, if power facility M1 collapses and the corresponding danger index is greater than the evaluation threshold, then according to the prediction result set, the collapse direction of power facility M1 can be judged by the analysis point data it contains. If the highest analysis point data of power facility M1 is mapped within the safety range of power facility M2, it means that power facility M1 will not cause safety impact on power facility M2, and then continue to judge whether the highest analysis point data of power facility M1 is mapped within the safety range of another connected power facility M3.

[0063] As Figure 4 shown, the present invention also provides a power facility collapse assessment system based on remote sensing image SAR data. This system is used to implement the above-mentioned power facility collapse assessment method based on remote sensing image SAR data. The system mainly includes:

[0064] An image extraction module, which is used to obtain an image data set containing power facilities in the remote sensing image, identify each feature point data in the image data set, obtain the feature coordinates of the feature point data, and use the first distance as the target to cluster each feature point data in combination with the feature coordinates to generate a distribution map of each power facility, and define the feature point data located in the same power facility as analysis point data;

[0065] A three-dimensional construction module, which is used to extract four adjacent analysis point data in the power facility and combine and connect them to generate a first combined surface. The first combined surface includes multiple sub-planes. Obtain the first vectors of the sub-planes in the first combined surface and judge whether all the first vectors are parallel to each other. If so, merge the sub-planes into the first combined surface, and then combine and connect the first combined surface with adjacent analysis point data to generate a second combined surface. If not, combine and connect the sub-planes with adjacent analysis point data to generate a second combined surface, and judge each first vector in the second combined surface respectively, and repeat this step until the analysis point data in the same power facility is expanded and completed;

[0066] An evaluation and monitoring module, configured to obtain a second vector of a power facility, calculate the angle between the first vector of each second combined surface in the power facility and the second vector, and define it as a first angle set, set a time interval, and execute step S1 and step S2 based on the time interval to correspondingly generate the first vector of each second combined surface and define it as an updated vector, obtain the angle between the updated vector and the second vector, and define it as a second angle set;

[0067] A prediction and analysis module, configured to obtain an angle change sequence based on the second angle set and the time interval, generate a prediction model based on the change rate of the angle change sequence, predict the change rate based on the prediction model to output a prediction result set, and evaluate the hazard index of the power facility and the safety range between each power facility based on the prediction result set and the distribution map.

[0068] The present invention also provides a computer storage medium, which stores program instructions. When the program instructions run, the device where the computer storage medium is located is controlled to execute the above-mentioned power facility collapse evaluation method based on remote sensing image SAR data.

[0069] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages do not necessarily need to be executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages does not necessarily need to be sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0070] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The above program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0071] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0072] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be subject to the appended claims.

[0073] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for assessing the collapse of power facilities based on remote sensing image SAR data, characterized in that: The method comprises the following steps: Step S1: acquiring an image data set including electric power facilities through remote sensing images, identifying each feature point data in the image data set, acquiring feature coordinates of the feature point data, taking a first distance as a target, clustering each feature point data in combination with the feature coordinates to generate a distribution map of each electric power facility, and defining the feature point data located in the same electric power facility as analysis point data; Step S2: extract four adjacent analysis point data in the power facility and combine and connect them to generate a first combined surface, the first combined surface includes multiple sub-planes, obtain the first vectors of the sub-planes in the first combined surface, the first vector refers to the normal vector of the sub-plane, and judge whether the first vectors are parallel to each other. If yes, merge the sub-planes into the first combined surface, and then combine and connect the first combined surface with the adjacent analysis point data to generate a second combined surface. If no, combine the sub-plane with the adjacent analysis point data to generate the second combined surface, judge each of the first vectors in the second combined surface respectively, and repeat this step until the expansion of the analysis point data in the same power facility is completed; Step S3: obtaining the second vector of the power facility, calculating the angle between the first vector and the second vector of each second combination surface in the power facility, and defining it as a first angle set, setting a time interval, executing the step S1 and the step S2 based on the time interval to correspondingly generate the first vector of each second combination surface and define it as an update vector, obtaining the angle between the update vector and the second vector, and defining it as a second angle set; The second vector is obtained and the time interval is set based on the following steps: Calculate the second vector based on the first formula , the first formula is: ,in and are the first vectors corresponding to two adjacent second combined surfaces, is the vector outer product operation, The first vector The corresponding variance value of each analysis point data contained in the second combined surface, The first vector The variance value of each of the analysis point data contained in the corresponding second combination surface, m is the number of permutations and combinations of two adjacent second combination surfaces of the power facilities; The time interval is set based on a second formula , the second formula is: ,in is the first angle set, is the second angle set corresponding to the i-th time interval, and The first time interval and the second time interval are respectively randomly preset; Step S4: Acquire an angle change sequence based on the second angle set and the time interval, generate a prediction model based on the change rate of the angle change sequence, predict the change rate based on the prediction model and output a prediction result set, and evaluate the danger index of the power facilities and the safety range between each of the power facilities based on the prediction result set and the distribution map.

2. The method for assessing the collapse of electric power facilities based on remote sensing image SAR data according to claim 1, characterized in that: In step S1, each of the feature point data is clustered based on the following steps to generate a distribution map of each of the power facilities: Calculate the coordinate distance between two feature point data based on the feature coordinates, extract one feature point data and set it as a first-category label point, obtain the feature point data whose coordinate distance with the first-category label point is less than or equal to the first distance and set it as a first-category label point, repeat this step to obtain all the first-category label points in each feature point data, if the coordinate distance between the feature point data and the first-category label point is greater than the first distance, set the feature point data as a second-category label point, obtain the feature point data whose coordinate distance with the second-category label point is less than or equal to the first distance and set it as a second-category label point, repeat this step to cluster all the feature point data in the image data set into multiple label types; The number of label types corresponding to the feature point data of each of the power facilities is counted, and the number of label types is set as the distribution number. Based on the distribution number, one feature point data is extracted from each of the two power facilities and the coordinate distance of the two feature point data is calculated. The minimum value of the coordinate distance is set as the corresponding distribution distance of the two power facilities. Two analysis point data are extracted from the power facilities to calculate the coordinate distance. The maximum value of the coordinate distance is set as the height value corresponding to the power facility. The distribution map of each of the power facilities is generated based on the distribution number, the distribution distance, the height value and the feature coordinates.

3. The method for assessing the collapse of electric power facilities based on remote sensing image SAR data according to claim 1, characterized in that: In step S4, evaluating the danger index of the power facility based on the prediction result set and the distribution map includes the following steps: The change rate of the angle change sequence is obtained based on the third formula , the third formula is , based on the fourth formula, obtain the predicted change rate of the angle change sequence , the fourth formula is: ,in, , is the rate of change corresponding to the i-th time interval, The second angle set corresponding to the i-th time interval is obtained based on the fifth formula to obtain the prediction result set , the fifth formula is: , the risk index W is obtained based on the sixth formula, the sixth formula is: ,in, is the standard normal distribution function, is the average value of each of the change rates, is the standard deviation of each of the above mentioned change rates.

4. The method for assessing the collapse of electric power facilities based on remote sensing image SAR data according to claim 1, characterized in that: In step S4, the safety range between the power facilities is obtained based on the following steps: The safety range S of each of the power facilities is calculated based on the seventh formula, which is: ,in, is the height of the power facility, The distribution distance between the two connected power facilities in the distribution map is determined. When the hazard index is greater than or equal to a preset evaluation threshold, the second combined surface corresponding to the subset with the largest change rate in the prediction result set in the power facility is obtained and defined as a collapse surface. The analysis point data contained in the collapse surface is obtained and defined as collapse point data. The collapse point data is mapped based on the second vector to determine whether the collapse point data mapping exists within the safety range of each power facility. If so, the power facility is maintained. If not, the power facility is maintained and other connected power facilities are evaluated and monitored.

5. A power facility collapse assessment system based on remote sensing image SAR data, used to implement the power facility collapse assessment method based on remote sensing image SAR data as described in any one of claims 1 to 4, characterized in that: The system includes the following modules: An image extraction module is used to obtain an image data set containing electric power facilities in a remote sensing image, identify each feature point data in the image data set, obtain feature coordinates of the feature point data, cluster each feature point data based on the feature coordinates to generate a distribution map of each electric power facility with a first distance as a target, and define the feature point data located in the same electric power facility as analysis point data; A three-dimensional construction module, used for extracting four adjacent analysis point data in the power facility and combining and connecting them to generate a first combination surface, the first combination surface includes multiple sub-planes, obtaining the first vectors of the sub-planes in the first combination surface, the first vector refers to the normal vector of the sub-plane, and judging whether the first vectors are parallel to each other. If yes, the sub-planes are merged into the first combination surface, and then the first combination surface and the adjacent analysis point data are combined and connected to generate the second combination surface. If no, the sub-planes and the adjacent analysis point data are combined and connected to generate the second combination surface, and the first vectors in the second combination surface are judged respectively, and this step is repeated until the expansion of the analysis point data in the same power facility is completed; An evaluation and monitoring module is used to obtain the second vector of the power facility, calculate the angle between the first vector and the second vector of each second combination surface in the power facility, and define it as a first angle set, set a time interval, execute the steps S1 and S2 based on the time interval to correspondingly generate the first vector of each second combination surface and define it as an update vector, obtain the angle between the update vector and the second vector, and define it as a second angle set; calculate the second vector based on the first formula , the first formula is: ,in and are the first vectors corresponding to two adjacent second combined surfaces, is the vector outer product operation, The first vector The corresponding variance value of each analysis point data contained in the second combined surface, The first vector The variance value of each of the analysis point data contained in the corresponding second combination surface, m is the number of permutations and combinations of two adjacent second combination surfaces of the power facilities; The time interval is set based on a second formula , the second formula is: ,in is the first angle set, is the second angle set corresponding to the i-th time interval, and The first time interval and the second time interval are respectively randomly preset; A prediction and analysis module obtains an angle change sequence based on the second angle set and the time interval, generates a prediction model based on the change rate of the angle change sequence, predicts the change rate based on the prediction model and outputs a prediction result set, and evaluates the danger index of the power facilities and the safety range between each of the power facilities based on the prediction result set and the distribution map.

Images