CN114332634A - Method and device for determining position of electric power tower at risk, electronic equipment and storage medium - Google Patents
Method and device for determining position of electric power tower at risk, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114332634A CN114332634A CN202210205445.8A CN202210205445A CN114332634A CN 114332634 A CN114332634 A CN 114332634A CN 202210205445 A CN202210205445 A CN 202210205445A CN 114332634 A CN114332634 A CN 114332634A
- Authority
- CN
- China
- Prior art keywords
- electric tower
- electric
- tower
- map
- digital elevation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The present application relates to the field of data processing technologies, and in particular, to a method and an apparatus for determining a location of an electric power tower, an electronic device, and a storage medium, which are used to improve efficiency of determining a location of an electric power tower. The main technical scheme comprises: acquiring an electric tower position marking map corresponding to the digital elevation map; fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image; performing model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model; inputting the target image into an electric tower position identification model to obtain an electric tower position identification result; determining whether two other electric towers exist in a preset radius of each electric tower in the electric tower position identification result; and if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower.
Description
Technical Field
The present disclosure relates to the field of data processing technologies, and in particular, to a method and an apparatus for determining a position of a risk power tower, an electronic device, and a storage medium.
Background
Power transmission infrastructure, particularly high voltage transmission towers, plays a vital role in the proper operation of the power supply system. However, many towers are currently constructed along seas, rivers or forest areas, which are often susceptible to natural disasters such as wildfires, earthquakes and floods.
At present, the traditional mode generally adopts the manual work to patrol and examine the electric tower to confirm the position that the risk electric tower exists. However, because the towers are distributed, the efficiency of manually determining the location of the at-risk towers is low.
Disclosure of Invention
In view of the above, the present application provides a method and an apparatus for determining a location of an electric power tower, an electronic device, and a storage medium, which are used to improve efficiency of determining the location of the electric power tower.
In a first aspect, an embodiment of the present application provides a method for determining a position of an electric tower at risk, where the method includes:
acquiring an electric tower position marking map corresponding to the digital elevation map; the electric tower position mark map comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map;
filtering the electric tower mark points in the electric tower position mark diagram through the height corresponding to each pixel point in the digital elevation diagram, the normalized vegetation index and the electric tower height;
fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image;
performing model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model;
inputting the target image into the electric tower position identification model to obtain an electric tower position identification result, wherein the electric tower position identification result comprises position coordinates corresponding to a plurality of identified electric towers;
determining whether two other electric towers exist in a preset radius of each electric tower in the electric tower position identification result; the preset radius corresponds to the actual electric tower interval;
and if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower.
In a second aspect, an embodiment of the present application further provides a device for determining a location of an electric power pylon, where the device includes:
the acquisition module is used for acquiring an electric tower position marking map corresponding to the digital elevation map; the electric tower position mark map comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map;
the filtering module is used for filtering the electric tower mark points in the electric tower position mark diagram through the height corresponding to each pixel point in the digital elevation diagram, the normalized vegetation index and the electric tower height;
the fusion module is used for fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image;
the training module is used for carrying out model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model;
the identification module is used for inputting the target image into the electric tower position identification model to obtain an electric tower position identification result, and the electric tower position identification result comprises position coordinates corresponding to a plurality of identified electric towers;
the determining module is used for determining whether two other electric towers exist in the preset radius of each electric tower in the electric tower position identification result; the preset radius corresponds to the actual electric tower interval;
the determining module is further configured to determine the preset radius range of the electric tower as the location area of the risk electric tower if two other electric towers do not exist within the preset radius range.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, the memory storing processor-executable machine-readable instructions, the processor and the memory communicating via the bus when the electronic device is running, the machine-readable instructions being executed by the processor to perform the steps of the method for determining a location of a risky electrical tower of the first aspect.
In a fourth aspect, the present embodiment further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the method for determining a location of an electric risk tower in the first aspect.
According to the method and the device for determining the position of the electric tower at risk, the electronic equipment and the storage medium, an electric tower position mark diagram corresponding to a digital elevation diagram is obtained; fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image; performing model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model; inputting the target image into an electric tower position identification model to obtain an electric tower position identification result; determining whether two other electric towers exist in a preset radius of each electric tower in the electric tower position identification result; and if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower. Because the electric tower position identification model in the embodiment is obtained by training according to the three-channel information image and the filtered electric tower position label map, an electric tower position identification result can be obtained through the electric tower position identification model, so that the position area where the risk electric tower is located can be determined based on the electric tower position identification result, and the efficiency of determining the position of the risk electric tower is improved. In addition, in the embodiment, the electric tower position marking map is automatically marked through the digital elevation map, so that the work of manually marking the digital elevation map is saved, and the training efficiency of the electric tower position identification model is improved.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a flowchart illustrating a method for determining a location of an electric risk tower according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram illustrating a result of identifying a location of an electric tower according to an embodiment of the present application;
FIG. 3 is a schematic diagram illustrating another electric tower location identification result provided by an embodiment of the present application;
fig. 4 shows a block diagram of a position determining apparatus for an electric risk tower according to an embodiment of the present application.
Detailed Description
The terms "first," "second," and "third," etc. in the description and claims of this application and the above-described drawings are used for distinguishing between different objects and not for limiting a particular order.
In the embodiments of the present application, words such as "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "e.g.," is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present relevant concepts in a concrete fashion for ease of understanding.
In the description of the present application, a "/" indicates a relationship in which the objects associated before and after are an "or", for example, a/B may indicate a or B; in the present application, "and/or" is only an association relationship describing an associated object, and means that there may be three relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. Also, in the description of the present application, "a plurality" means two or more than two unless otherwise specified. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
In the embodiments of the present application, at least one may also be described as one or more, and a plurality may be two, three, four or more, which is not limited in the present application.
As shown in fig. 1, an embodiment of the present application provides a method for determining a location of an electric power tower at risk, where the method for determining a location of an electric power tower at risk provided by the present application may include:
and S10, acquiring the electric tower position marked map corresponding to the digital elevation map.
The electric tower position mark graph comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark graph. Specifically, the preset region may be composed of N adjacent pixel points. The Digital elevation map is a Digital Surface Model (DSM), and includes heights of objects such as Surface buildings, bridges, trees, and the like, that is, each pixel point in the Digital elevation map has a corresponding height value, the height value can be represented by a brightness value of the pixel point, and the larger the brightness of the pixel point is, the higher the height value represented by the pixel point is.
After the digital elevation map is acquired, the electric tower position marker map can be determined by analyzing the brightness values of the pixel points in the digital elevation map. Specifically, position coordinates corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map are firstly obtained, then the obtained position coordinates of the pixel point are determined as the electric tower mark point, and then the electric tower position mark map is generated according to the determined electric tower mark point and the pixel point which is not marked in the digital elevation map. It should be noted that the electric tower position mark map in this embodiment is a binary map composed of 0 and 1, where 1 represents an electric tower mark point, and 0 represents an unmarked pixel point.
In an alternative embodiment, obtaining a map of electric tower location markers corresponding to the digital elevation map comprises:
s101, performing gray scale morphological erosion on each digital elevation map by using disc-shaped structural elements to generate a marked image; the disc-shaped structural elements are determined according to the size of the electric tower.
The marked image is a binary image represented by gray values, and the disc-shaped structural elements are determined according to the actual size of the electric tower.
And S102, generating a morphological reconstruction mask from the marked image through an iterative process.
Wherein the morphological reconstruction mask is a result of morphological filtering of the digital elevation map.
S103, marking the position coordinates corresponding to the highest brightness pixel points in each disc-shaped structural element in the electric tower position mark map by subtracting the form reconstruction mask from the digital elevation map to obtain the electric tower position mark map.
In this embodiment, the conical-like landscape is presented in the digital elevation map due to most electric towers. Therefore, the implementation can detect the position of the electric tower in the digital elevation map by the tophat reconstruction method. Gray scale morphological roof fall-offs by reconstruction operations are an effective method of blob-like shape detection compared to many other local maximum detectors and are less sensitive to filter size. The morphological top-level cap is defined as the peak of the image grid calculated by the morphological operation, and several of the most advanced protrusion detection methods have been successfully detected using this morphological operation.
Specifically, tophat detection uses disk-shaped structuring elements (e) to perform grey scale morphological erosion on DSM to generate a marking image. Then, a morphological reconstruction mask is generated from the marker image by an iterative process. Finally, by subtracting the morphological reconstruction mask from the DSM, the position coordinates corresponding to the highest brightness pixel points within each disc-shaped structuring element on the DSM, i.e. the detected electric towers, can be extracted. Regarding the differentiation between the electric tower and other similar objects in the digital elevation map in the embodiment, for example, buildings, wind power generators, etc., different objects can be differentiated by trying the most appropriate size of the disc structural elements, that is, the corresponding disc structural elements are set according to the size of the electric tower, so as to realize the identification of the position of the electric tower in the digital elevation map.
In another alternative embodiment, the acquiring the electric tower position marked map corresponding to the digital elevation map includes: and sliding a fixed window for each digital elevation map, and obtaining an electric tower position mark map according to the position coordinates corresponding to the highest brightness pixel points in each fixed window in the digital elevation maps.
In this embodiment, as a comparison using different unsupervised detectors, a local maximum filter with a fixed window may also be utilized on the DSM to detect the electric tower. In a certain area (in a fixed window) with a fixed size, if an obvious pixel point with the highest brightness exists, the position coordinate corresponding to the pixel point with the highest brightness in the fixed window can be regarded as the potential position of the electric tower. Optionally, the fixed window in this embodiment may be 7 continuous pixels, that is, the sliding window size with 7 pixels (corresponding to 2.1 meters) is selected in this embodiment, because it has the best average performance in the test scheme, and at the same time, confusable objects such as buildings and the like can be excluded.
Further, in order to avoid the situation that the electric tower is not identified, the embodiment may superimpose the electric tower position mark maps obtained in the two manners, and then perform processing based on the superimposed electric tower position mark map, that is, after obtaining the superimposed electric tower position mark map, jump to S20 to continue execution.
And S20, filtering the electric tower mark points in the electric tower position mark diagram through the height corresponding to each pixel point in the digital elevation diagram, the normalized vegetation index and the electric tower height.
Wherein, the height of the electric tower in the embodiment is the actual height of the electric tower. Normalized Difference Vegetation Index (NDVI) Vegetation is quantified by measuring the Difference between near infrared (Vegetation strong reflection) and red light (Vegetation absorption). The range of NDVI is always-1 to + 1. But each type of land cover has no clear limits. For example, when the value is negative, it is likely to be water. On the other hand, if the NDVI value is close to +1, then there is a high probability of a bushy green leaf. However, when NDVI is close to zero, there are no green leaves, and even possibly a urbanized area. Specifically, the normalized vegetation index: NDVI = (NIR-R)/(NIR + R), NIR is a reflection value of a near infrared band, and R is a reflection value of a red light band.
Specifically, the filtering the electric tower mark points in the electric tower position mark map through the height corresponding to each pixel point in the digital elevation map, the normalized vegetation index and the electric tower height includes:
1. and filtering the electric tower mark points belonging to the vegetation in the electric tower position mark map by adopting the normalized difference vegetation index.
Because the larger the NDVI value is, the more vegetation coverage is indicated, so that the vegetation coverage area can be detected by setting a reasonable threshold value in this embodiment, and if the NDVI corresponding to a certain electric tower marking point in the electric tower position marking map is in a larger area, the electric tower marking point is indicated as a vegetation point (such as a tree), that is, the electric tower marking point is inaccurate, and therefore the electric tower marking point needs to be removed from the electric tower position marking map.
2. And filtering the electric tower mark points which do not belong to the height of the electric tower in the electric tower position mark diagram.
Since the digital elevation map in this embodiment has a height value corresponding to each pixel point, in this embodiment, after the electric tower position marker map is obtained, the electric tower marker points that do not belong to the electric tower height are filtered according to the height values of the electric tower marker points corresponding to the digital elevation map. For example, if the height of the electric tower is 20 meters, the corresponding height value of the electric tower mark points in the digital elevation map is determined, and then the height value is filtered for the electric tower mark points which are significantly higher than 20 meters or 20 meters, and specifically, the electric tower mark points whose height value is not 18-22 meters can be filtered.
3. And filtering a plurality of electric tower mark points existing in a preset range in the electric tower position mark diagram according to the average value of the height of the electric tower.
Wherein, the preset range can be determined according to the area occupied by a single electric tower. If only one electric tower can exist at the radius of 50 meters, the preset range can be determined according to the corresponding range of the electric tower position mark map at the radius of 50 meters. For example, a plurality of electric tower mark points exist in a preset range in the electric tower position mark map, and the plurality of electric tower mark points in the preset range are filtered according to the average value of the electric tower heights, namely, the electric tower mark point with the height value closest to the average value of the electric tower heights in the preset range is reserved.
It should be noted that, the order of filtering the electric tower mark points is not limited in this embodiment, that is, the electric tower mark points may be filtered according to the normalized vegetation index first, or the electric tower mark points may be filtered according to the height corresponding to the pixel point first, which is not specifically limited in this embodiment.
And S30, fusing the digital elevation map and the multiband ortho-image or RGB image corresponding to the digital elevation map into a three-channel information image.
The three-channel information image comprises three-dimensional information, namely the three-channel information image comprises a red waveband, a normalized vegetation index and a height.
Specifically, the present embodiment can obtain a digital elevation map, a multiband ortho-image, and an RGB image through photogrammetric processing of a multiview image. The digital elevation map and the multiband orthogonal images corresponding to the digital elevation map can be fused to obtain a three-channel information image, wherein the red waveband and the normalized vegetation index of each pixel point are calculated according to the multiband orthogonal images, the height of each pixel point is determined according to the digital elevation map, and then the results of the two images are fused into the three-channel information image.
In addition, the embodiment may further fuse the digital elevation map and the RGB images corresponding to the digital elevation map to obtain a three-channel information image. The height of each pixel point is determined according to the digital elevation map, and the red wave band and the normalized vegetation index of each pixel point are calculated according to the RGB image. Specifically, the normalized vegetation index is calculated by the following formula:
NDVI=(2*G'-R'-B')–(1.4*R'-G')；
wherein G ' = G/(R + G + B), R ' = R/(R + G + B), and B ' = B/(R + G + B).
And S40, performing model training according to the three-channel information image and the filtered electric tower position label diagram to obtain an electric tower position identification model.
In an embodiment provided by the present invention, the performing model training according to the three-channel information image and the filtered electric tower position marker map to obtain an electric tower position identification model includes: performing full convolution network model training according to the three-channel information image and the filtered electric tower position mark image to obtain an electric tower position identification model; the three-channel information image comprises a red waveband, a normalized vegetation index and a height, wherein each pixel point is respectively and correspondingly normalized to be 0-1;
it should be noted that, in this embodiment, the three-channel information image is a sample image, the filtered electric tower position marker map is a sample label, and a full convolution network model training is performed on the three-channel information image and the filtered electric tower position marker map, so that an electric tower position identification model can be obtained. The electric tower position recognition result corresponding to the input target image can be recognized through the electric tower position recognition model.
In an optional embodiment, in order to improve the training efficiency of the electric tower position model, the internal structure of the full convolutional network is modified, specifically, the full convolutional network in this application includes 3 modules and an upsampling layer, that is, a first module, a second module, a third module and an upsampling layer, where each module includes a convolutional layer and a max-pooling layer. Preferably, the input size of the three-channel information image is set to 48 × 48. Because the input size of the three-channel information image is small (48 x 48) and the modeling task is simpler (binary), after the largest pooling layer of three modules (pooling layer in one module shrinks the input data by a factor of 2, three times, or 8 times), the size of the three-channel information image will be reduced to 6 x 6, and the following two fully convolutional layers will generate two classes (electric tower and background) of predictions with reduced sampling resolution (6 x 6). The last layer is an upsampling layer for resizing the output image as large as the input image, and in order for the final output data to remain the same size as the input data, all data needs to be enlarged by a factor of 8 at the last upsampling layer.
Since the towers detected by the tower location identification model are usually represented as regions, the invention needs to further locate their positions at the pixel level by finding the highest point in each region, i.e. after the tower location identification result obtained by the tower location identification model, further verification of the identification result is needed. Specifically, in this embodiment, after the target image is input to the electric tower position identification model to obtain an electric tower position identification result, the method further includes:
s401, marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the normalized vegetation index which is respectively corresponding to each pixel point in the three-channel information image.
S402, marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the heights corresponding to the pixel points in the three-channel information image and the average value of the electric tower heights.
It should be noted that, in this embodiment, the confirmation method for misidentifying the electric tower in S401, S402 and S20 is the same, and the description of this embodiment is omitted.
And S403, performing model training according to the marked electric tower position identification result and the corresponding target image, and updating the electric tower position identification model.
In this embodiment, after the electric towers which are mistakenly identified in the electric tower position identification result are marked through S401 and S402, four training samples are randomly generated around each electric tower which is mistakenly identified, and are merged into training data as negative training samples, so that the electric tower position identification model is trained again according to the training data, the purpose of updating the electric tower position identification model is achieved, and the accuracy of the electric tower position identification result obtained through the electric tower position identification model is further ensured.
And S50, inputting the target image into the electric tower position identification model to obtain an electric tower position identification result, wherein the electric tower position identification result comprises position coordinates corresponding to a plurality of identified electric towers.
The target image may be a three-way information map obtained by fusing a digital elevation map and a multiband ortho-image corresponding to the digital elevation map, may also be a three-channel information map obtained by fusing a digital elevation map and an RGB image corresponding to the digital elevation map, and may also be an RGB image, a digital elevation map, a multiband ortho-image, and the like, which is not specifically limited in this embodiment. Preferably, the target image in this embodiment is a fused three-channel information map.
The size of the target image input to the electric tower position recognition model may be set arbitrarily, and the size of the target image is not limited in this embodiment, for example, the size of the target image may be 256 × 256 or 1024 × 1024. In this embodiment, the electric tower position recognition result obtained by the electric tower position recognition model can be as shown in fig. 2, where 1 represents the recognized electric tower and 0 represents other objects except the electric tower.
And S60, determining whether two other electric towers exist in the preset radius of each electric tower in the electric tower position identification result.
Wherein the preset radius corresponds to an average interval of two adjacent electric towers in an actual situation. For example, if the average distance between two actual electric towers is 50 meters, after the electric tower position identification result is obtained through the electric tower position identification model, it is determined whether two other electric towers exist within the preset radius of each electric tower in the electric tower position identification result, for example, if the distance represented by 3 pixels in fig. 2 is 50 meters, two other identified electric towers do not exist within the 50 meter range respectively corresponding to the electric tower a and the electric tower C in fig. 2, which indicates that one unidentified electric tower exists between the electric tower a and the electric tower C.
And S70, if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower.
It should be noted that, in the present embodiment, the electric towers in the normal state may be determined by the electric tower position identification model, and based on the spacing distance between the electric towers in the normal state and the identified position of the electric tower in the normal state, the unidentified position area of the electric tower may be determined, that is, if there are no two other electric towers in the preset radius of a certain electric tower, the preset radius range of the electric tower is determined as the position area of the electric tower in risk.
In the method for determining the location of the electric tower at risk provided by this embodiment, after the preset radius range of the electric tower is determined as the location area of the electric tower at risk, the specific location of the electric tower at risk in the location area may be further determined, and the specific process may be as follows: and if only one electric tower exists in a preset radius of a certain electric tower in the electric tower identification result, acquiring the electric tower which is the second closest to the electric tower, and determining the middle point between the electric tower and the second closest electric tower as the position of the risk electric tower. As shown in fig. 2, there is only one electric tower E in the preset radius of the electric tower a, and adjacent to the electric tower a are an electric tower E and an electric tower C. If the electric tower C is the electric tower next to the electric tower a, the intermediate point of the connection line between the electric tower a and the electric tower C can be determined as the position of the electric tower at risk (not recognized due to the inclination or collapse), and thus the position of the electric tower at risk can be obtained as the position corresponding to the point B.
In another method for determining the location of a risk power tower provided in this embodiment, due to reasons such as terrain, power towers distributed in the field may not be in a straight line, that is, the distribution of the power towers is irregular, and in order to solve the above problem, after determining the preset radius range of the power tower as the location area of the risk power tower, the present invention may further determine the specific location of the risk power tower in the location area, and the specific process may be as follows: and if only one electric tower exists in a preset radius of a certain electric tower in the electric tower identification result, acquiring two electric towers adjacent to the electric tower, taking the electric tower which is the second nearest to the electric tower as a target electric tower, respectively drawing two circles by taking the electric tower and the target electric tower as the center of a circle, and determining the intersection point of the two circles as the position of the risk electric tower. Wherein the corresponding radius of the elevator and the target electric tower is the distance between the electric tower which is the nearest to the elevator and the target electric tower. As shown in fig. 2, the electric tower a has only one electric tower E within a preset radius, the electric tower C has only one electric tower F within a preset radius, and the electric towers a and C are adjacent electric towers. And drawing circles respectively by taking the electric tower A and the electric tower E as the circle centers and taking the intersection point of the two circles as the position of the risk electric tower. The radius corresponding to the electric tower A is the distance between the electric tower A and the electric tower E, and the radius corresponding to the electric tower C is the distance between the electric tower C and the electric tower F.
Further, if there are a plurality of consecutive electric towers that cannot be identified in the obtained electric tower location identification result, the two ways for determining the location of the electric tower at risk cannot be used to determine the specific location of the electric tower at risk, that is, the two ways for determining the location of the electric tower at risk are the case where it is determined that one electric tower cannot be identified between only two electric towers, that is, only one electric tower that cannot be identified (an electric tower that has a tilt or collapse) may exist between the electric tower a and the electric tower C in fig. 2, and whether the distance between the specific electric tower a and the electric tower C is far greater than 2 times of the preset radius may be used, if so, it may be determined that there are a plurality of electric towers that have no risk and that have only one electric tower that cannot be identified between the electric tower a and the electric tower C.
In this embodiment, in order to solve the situation of the existing multiple consecutive unidentifiable risk electric towers, the present embodiment may identify all the electric towers in the image in the normal state by calling the image in the normal state corresponding to the target image, and determine the locations of the multiple consecutive unidentifiable risk electric towers by comparing the target image with the image in the normal state.
In an optional embodiment, if there are a plurality of other electric towers in a preset range of a certain electric tower in this embodiment, the situation may also be determined as a risk electric tower, where the situation may correspond to the fact that other interfering objects (such as trees, etc.) are present around the electric tower, and the preset range is a range smaller than the preset radius. As shown in fig. 3, if an electric tower D, an electric tower C, and an electric tower F exist in an electric tower a in the preset range in the drawing, it is described that positions of the electric tower a, the electric tower D, the electric tower C, and the electric tower F have risks, that is, only one point is an identified electric tower, and the other three points are misidentified electric towers, and in order to accurately obtain the position of the electric tower with risks, the present embodiment may determine mark points belonging to vegetation in the electric tower a, the electric tower D, the electric tower C, and the electric tower F by normalizing the difference vegetation index; or filtering the electric tower mark points which do not belong to the height of the electric tower in the electric tower A, the electric tower D, the electric tower C and the electric tower F; or filtering a plurality of mark points existing in the preset ranges in the electric towers A, D, C and F according to the average value of the heights of the electric towers. The actual electric tower positions in the electric towers A, D, C and F are obtained through the method.
Further, when the electric tower a, the electric tower D, the electric tower C and the electric tower F are filtered according to the three modes, if two electric towers still exist after filtering, the real position of the electric tower can be determined according to the preset radius. For example, in fig. 3, the electric tower D and the electric tower C may be filtered out in the above manner, and there are two electric towers a and F, at this time, the true electric tower positions in the electric towers a and F are determined according to the preset radius, and if the preset radius corresponds to the three pixel points in fig. 3, it may be finally determined that the electric tower a is the true position of the electric tower.
According to the method for determining the position of the electric tower at risk, which is provided by the embodiment of the application, an electric tower position mark diagram corresponding to a digital elevation diagram is obtained; fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image; performing model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model; inputting the target image into an electric tower position identification model to obtain an electric tower position identification result; determining whether two other electric towers exist in a preset radius of each electric tower in the electric tower position identification result; and if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower. Because the electric tower position identification model in the embodiment is obtained by training according to the three-channel information image and the filtered electric tower position label map, an electric tower position identification result can be obtained through the electric tower position identification model, so that the position area where the risk electric tower is located can be determined based on the electric tower position identification result, and the efficiency of determining the position of the risk electric tower is improved. In addition, in the embodiment, the electric tower position marking map is automatically marked through the digital elevation map, so that the work of manually marking the digital elevation map is saved, and the training efficiency of the electric tower position identification model is improved.
In the case of dividing each function module according to each function, fig. 4 shows a schematic diagram of a possible composition of the location determining apparatus of the electric risk tower mentioned above and in the embodiment, as shown in fig. 4, the location determining apparatus of the electric risk tower may include:
an obtaining module 41, configured to obtain an electric tower position marker map corresponding to the digital elevation map; the electric tower position mark map comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map;
the filtering module 42 is configured to filter the electric tower mark points in the electric tower position mark map through the heights corresponding to the pixel points in the digital elevation map, the normalized vegetation indexes, and the electric tower heights;
a fusion module 43, configured to fuse the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image;
the training module 44 is used for performing model training according to the three-channel information image and the filtered electric tower position label diagram to obtain an electric tower position identification model;
the identification module 45 is configured to input the target image to the electric tower position identification model to obtain an electric tower position identification result, where the electric tower position identification result includes position coordinates corresponding to a plurality of identified electric towers;
a determining module 46, configured to determine, for each of the electric towers in the electric tower position identification result, whether two other electric towers exist within a preset radius of the electric tower; the preset radius corresponds to the actual electric tower interval;
the determining module 46 is further configured to determine the preset radius range of the electric tower as the location area of the risk electric tower if two other electric towers do not exist within the preset radius range.
In an optional embodiment, the obtaining module 41 is specifically configured to:
performing a gray scale morphological erosion on each digital elevation map using disk-shaped structuring elements to generate a marker image; the disc-shaped structural elements are determined according to the size of the electric tower;
generating a morphological reconstruction mask from the marker image by an iterative process;
and marking the position coordinates corresponding to the highest brightness pixel points in each disc-shaped structural element in the electric tower position mark map by subtracting the form reconstruction mask from the digital elevation map to obtain the electric tower position mark map.
In an optional embodiment, the obtaining module 41 is configured to perform sliding on a fixed window for each digital elevation map, and obtain the electric tower position marker map according to a position coordinate corresponding to a highest-brightness pixel point in each fixed window in the digital elevation map.
In an alternative embodiment, the filtering module 42 is specifically configured to:
filtering the electric tower mark points belonging to the vegetation in the electric tower position mark map by adopting a normalized difference vegetation index;
filtering the electric tower mark points which do not belong to the height of the electric tower in the electric tower position mark diagram;
and filtering a plurality of electric tower mark points existing in a preset range in the electric tower position mark diagram according to the average value of the electric tower heights.
In an alternative embodiment, training module 44 is specifically configured to:
performing full convolution network model training according to the three-channel information image and the filtered electric tower position mark image to obtain an electric tower position identification model; the three-channel information image comprises a red waveband, a normalized vegetation index and a height, wherein each pixel point is respectively and correspondingly normalized to be 0-1;
the full convolutional network comprises 3 modules and an up-sampling layer, wherein each module comprises a convolutional layer and an up-sampling layer.
In an optional embodiment, the size of the three-channel information image is 48 × 48, the modules include a first module, a second module and a third module, and the three-channel information image is sequentially reduced by 2 times, 4 times and 8 times through the first module, the second module and the third module; the size of the up-sampling layer is 8 times.
In an alternative embodiment, training module 44 is specifically configured to:
marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the normalized vegetation index which corresponds to each pixel point in the three-channel information image;
marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the heights which correspond to the pixel points in the three-channel information image respectively and the average value of the electric tower heights;
and the system is also used for carrying out model training according to the marked electric tower position identification result and the corresponding target image and updating the electric tower position identification model.
Based on the same application concept, the embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the method for determining the position of an electric risk tower provided by the above embodiment.
Specifically, the storage medium can be a general-purpose storage medium, such as a removable disk, a hard disk, and the like, when a computer program on the storage medium is executed, the method for determining the position of the electric risk tower can be executed, and by operating the mark control by a user, the method can realize that the control mark icon moves on the surface of the virtual object in the virtual game environment, and then marks the specific position of the virtual object, so that the accuracy of marking the virtual object can be improved by the method and the device.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
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 non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for determining the location of an at-risk power tower, the method comprising:
acquiring an electric tower position marking map corresponding to the digital elevation map; the electric tower position mark map comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map;
filtering the electric tower mark points in the electric tower position mark diagram through the height corresponding to each pixel point in the digital elevation diagram, the normalized vegetation index and the electric tower height;
fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image;
performing model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model;
inputting the target image into the electric tower position identification model to obtain an electric tower position identification result, wherein the electric tower position identification result comprises position coordinates corresponding to a plurality of identified electric towers;
determining whether two other electric towers exist in a preset radius of each electric tower in the electric tower position identification result; the preset radius corresponds to the actual electric tower interval;
and if two other electric towers do not exist in the preset radius, determining the preset radius range of the electric tower as the position area of the risk electric tower.
2. The method of claim 1, wherein obtaining the electrical tower position marker maps corresponding to the digital elevation maps comprises:
performing a gray scale morphological erosion on each digital elevation map using disk-shaped structuring elements to generate a marker image; the disc-shaped structural elements are determined according to the size of the electric tower;
generating a morphological reconstruction mask from the marker image by an iterative process;
and marking the position coordinates corresponding to the highest brightness pixel points in each disc-shaped structural element in the electric tower position mark map by subtracting the form reconstruction mask from the digital elevation map to obtain the electric tower position mark map.
3. The method of claim 1, wherein obtaining the electrical tower position marker maps corresponding to the digital elevation maps comprises:
and sliding a fixed window for each digital elevation map, and obtaining the electric tower position mark map according to the position coordinates corresponding to the highest brightness pixel points in each fixed window in the digital elevation maps.
4. The method of any one of claims 1-3, wherein filtering the electrical tower location marker points in the electrical tower location marker map by the height corresponding to each pixel point in the digital elevation map, the normalized vegetation index, and the electrical tower height comprises:
filtering the electric tower mark points belonging to the vegetation in the electric tower position mark map by adopting a normalized difference vegetation index;
filtering the electric tower mark points which do not belong to the height of the electric tower in the electric tower position mark diagram;
and filtering a plurality of electric tower mark points existing in a preset range in the electric tower position mark diagram according to the average value of the electric tower heights.
5. The method of claim 4, wherein the model training according to the three-channel information image and the filtered electric tower position label map to obtain an electric tower position identification model comprises:
performing full convolution network model training according to the three-channel information image and the filtered electric tower position mark image to obtain an electric tower position identification model; the three-channel information image comprises a red waveband, a normalized vegetation index and a height, wherein each pixel point is respectively and correspondingly normalized to be 0-1;
the full convolutional network comprises 3 modules and an up-sampling layer, wherein each module comprises a convolutional layer and an up-sampling layer.
6. The method of claim 5, wherein the three channel information image is 48 x 48 in size, the modules include a first module, a second module, and a third module, and the three channel information image is sequentially reduced by 2 times, 4 times, and 8 times by the first module, the second module, and the third module; the size of the up-sampling layer is 8 times.
7. The method of claim 4, wherein after inputting the target image into the electric tower position identification model and obtaining the electric tower position identification result, the method further comprises:
marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the normalized vegetation index which corresponds to each pixel point in the three-channel information image;
marking the electric tower which is identified by mistake in the obtained electric tower position identification result according to the heights which correspond to the pixel points in the three-channel information image respectively and the average value of the electric tower heights;
and performing model training according to the marked electric tower position identification result and the corresponding target image, and updating the electric tower position identification model.
8. An apparatus for determining a location of an at-risk power tower, the apparatus comprising:
the acquisition module is used for acquiring an electric tower position marking map corresponding to the digital elevation map; the electric tower position mark map comprises a plurality of electric tower mark points, and each electric tower mark point is a position coordinate corresponding to a pixel point with the highest brightness in a preset area in the electric tower position mark map;
the filtering module is used for filtering the electric tower mark points in the electric tower position mark diagram through the height corresponding to each pixel point in the digital elevation diagram, the normalized vegetation index and the electric tower height;
the fusion module is used for fusing the digital elevation map and the multiband orthogonal image or RGB image corresponding to the digital elevation map into a three-channel information image;
the training module is used for carrying out model training according to the three-channel information image and the filtered electric tower position mark diagram to obtain an electric tower position identification model;
the identification module is used for inputting the target image into the electric tower position identification model to obtain an electric tower position identification result, and the electric tower position identification result comprises position coordinates corresponding to a plurality of identified electric towers;
the determining module is used for determining whether two other electric towers exist in the preset radius of each electric tower in the electric tower position identification result; the preset radius corresponds to the actual electric tower interval;
the determining module is further configured to determine the preset radius range of the electric tower as the location area of the risk electric tower if two other electric towers do not exist within the preset radius range.
9. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine readable instructions executable by the processor, the processor and the memory communicating over the bus when an electronic device is running, the machine readable instructions when executed by the processor performing the steps of the method of determining the position of a jeopardy tower according to any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, performs the steps of the method for determining a location of a risky electrical tower according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210205445.8A CN114332634B (en) | 2022-03-04 | 2022-03-04 | Method and device for determining position of risk electric tower, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210205445.8A CN114332634B (en) | 2022-03-04 | 2022-03-04 | Method and device for determining position of risk electric tower, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114332634A true CN114332634A (en) | 2022-04-12 |
| CN114332634B CN114332634B (en) | 2022-06-10 |
Family
ID=81030679
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210205445.8A Active CN114332634B (en) | 2022-03-04 | 2022-03-04 | Method and device for determining position of risk electric tower, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114332634B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115115221A (en) * | 2022-06-25 | 2022-09-27 | 国网安徽省电力有限公司经济技术研究院 | Narrow-base steel tower construction analysis method based on cloud computing |
| CN117095359A (en) * | 2023-10-17 | 2023-11-21 | 山东和兑智能科技有限公司 | Transmission line safety monitoring and early warning method and system based on artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120257792A1 (en) * | 2009-12-16 | 2012-10-11 | Thales | Method for Geo-Referencing An Imaged Area |
| CN106526610A (en) * | 2016-11-04 | 2017-03-22 | 广东电网有限责任公司电力科学研究院 | Power tower automatic positioning method and apparatus based on unmanned aerial vehicle laser point cloud |
| CN109671174A (en) * | 2018-12-20 | 2019-04-23 | 北京中飞艾维航空科技有限公司 | A kind of pylon method for inspecting and device |
| WO2019232830A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for detecting foreign object debris at airport, computer apparatus, and storage medium |
| CN111860488A (en) * | 2020-07-31 | 2020-10-30 | 广东电网有限责任公司 | Method, device, equipment and medium for detecting and identifying bird nest of tower |
| CN112132795A (en) * | 2020-09-14 | 2020-12-25 | 中山大学 | Electric tower disaster risk assessment method and system based on LiDAR point cloud |
| CN113888514A (en) * | 2021-10-09 | 2022-01-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Method and device for detecting defects of ground wire, edge computing equipment and storage medium |
-
2022
- 2022-03-04 CN CN202210205445.8A patent/CN114332634B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120257792A1 (en) * | 2009-12-16 | 2012-10-11 | Thales | Method for Geo-Referencing An Imaged Area |
| CN106526610A (en) * | 2016-11-04 | 2017-03-22 | 广东电网有限责任公司电力科学研究院 | Power tower automatic positioning method and apparatus based on unmanned aerial vehicle laser point cloud |
| WO2019232830A1 (en) * | 2018-06-06 | 2019-12-12 | 平安科技（深圳）有限公司 | Method and device for detecting foreign object debris at airport, computer apparatus, and storage medium |
| CN109671174A (en) * | 2018-12-20 | 2019-04-23 | 北京中飞艾维航空科技有限公司 | A kind of pylon method for inspecting and device |
| CN111860488A (en) * | 2020-07-31 | 2020-10-30 | 广东电网有限责任公司 | Method, device, equipment and medium for detecting and identifying bird nest of tower |
| CN112132795A (en) * | 2020-09-14 | 2020-12-25 | 中山大学 | Electric tower disaster risk assessment method and system based on LiDAR point cloud |
| CN113888514A (en) * | 2021-10-09 | 2022-01-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Method and device for detecting defects of ground wire, edge computing equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| JIANPING CHENG 等: "Research on Corrosion Problem and Residual Bearing Capacity of Transmission Tower Engineering Application Process", 《IEEE》 * |
| 黄陆君等: "电力巡线LiDAR点云电塔自动定位和提取算法", 《测绘通报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115115221A (en) * | 2022-06-25 | 2022-09-27 | 国网安徽省电力有限公司经济技术研究院 | Narrow-base steel tower construction analysis method based on cloud computing |
| CN115115221B (en) * | 2022-06-25 | 2023-08-08 | 国网安徽省电力有限公司经济技术研究院 | Narrow-base steel tower construction analysis method based on cloud computing |
| CN117095359A (en) * | 2023-10-17 | 2023-11-21 | 山东和兑智能科技有限公司 | Transmission line safety monitoring and early warning method and system based on artificial intelligence |
| CN117095359B (en) * | 2023-10-17 | 2024-02-02 | 山东和兑智能科技有限公司 | Transmission line safety monitoring and early warning method and system based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114332634B (en) | 2022-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sylvain et al. | Mapping dead forest cover using a deep convolutional neural network and digital aerial photography | |
| CN114332634B (en) | Method and device for determining position of risk electric tower, electronic equipment and storage medium | |
| Hamraz et al. | A robust approach for tree segmentation in deciduous forests using small-footprint airborne LiDAR data | |
| US9947109B2 (en) | Method and apparatus for direct detection, location, analysis, identification, and reporting of vegetation clearance violations | |
| Awrangjeb et al. | Building detection in complex scenes thorough effective separation of buildings from trees | |
| CA2840436C (en) | System for mapping and identification of plants using digital image processing and route generation | |
| Grigillo et al. | Urban object extraction from digital surface model and digital aerial images | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| Chen et al. | Rapid urban roadside tree inventory using a mobile laser scanning system | |
| US8855427B2 (en) | Systems and methods for efficiently and accurately detecting changes in spatial feature data | |
| CN109583345A (en) | Roads recognition method, device, computer installation and computer readable storage medium | |
| Grigillo et al. | Automated building extraction from IKONOS images in suburban areas | |
| CN114519712A (en) | Point cloud data processing method and device, terminal equipment and storage medium | |
| Xiao et al. | Individual tree detection and crown delineation with 3D information from multi-view satellite images | |
| Xiao et al. | Individual tree detection from multi-view satellite images | |
| CN108109125A (en) | Information extracting method and device based on remote sensing images | |
| Albuquerque et al. | Remotely piloted aircraft imagery for automatic tree counting in forest restoration areas: a case study in the Amazon | |
| JP6434834B2 (en) | Inspection object extraction device and inspection object extraction method | |
| Butenko et al. | Geospatial data processing characteristics for environmental monitoring tasks | |
| CN113111793A (en) | Tree identification method and device | |
| JP6734213B2 (en) | Information processing device and program | |
| Lertaramsaeng et al. | An Appropriate Interpretation Approach for Land Use/Land Cover Map in Chomthong, Phitsanulok, Thailand | |
| Watanakij et al. | Improving the Accuracy of Mangrove Species Discrimination using Object Based and High Spatial Resolution Imagery: A Case Study in Pak Phanang, Thailand. | |
| Münzinger et al. | Mapping the Urban Forest in Detail: From LiDAR Point Clouds to 3D Tree Models | |
| Höhle | From classification results to topographic maps |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |