CN116416527A - Remote sensing image-based power transmission line corridor object identification method and system - Google Patents
Remote sensing image-based power transmission line corridor object identification method and system Download PDFInfo
- Publication number
- CN116416527A CN116416527A CN202310398286.2A CN202310398286A CN116416527A CN 116416527 A CN116416527 A CN 116416527A CN 202310398286 A CN202310398286 A CN 202310398286A CN 116416527 A CN116416527 A CN 116416527A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- corridor
- convolution
- remote sensing
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 27
- 238000013507 mapping Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000007246 mechanism Effects 0.000 claims abstract description 9
- 238000010586 diagram Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 239000013598 vector Substances 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 239000011800 void material Substances 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000013508 migration Methods 0.000 claims description 2
- 230000005012 migration Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 5
- 238000012549 training Methods 0.000 description 7
- 230000000875 corresponding effect Effects 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The disclosure provides a method and a system for identifying an object in a power transmission line corridor based on a remote sensing image, and relates to the technical field of image processing, wherein the method comprises the following steps: obtaining visible light data of a target area of a power transmission line, and preprocessing to obtain a digital orthophoto map; the digital orthographic image is input into a convolutional neural network model for corridor object identification after being cut through a sliding image window, in the convolutional neural network, the common convolution is replaced by the hole convolution in the convolutional layers, a certain number of the convolutional layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor; and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification. The method and the device can be used for efficiently identifying the power transmission line corridor object.
Description
Technical Field
The disclosure relates to the technical field of image processing, in particular to a method and a system for identifying an object in a power transmission line corridor based on a remote sensing image.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The power transmission line corridor is a strip-shaped area below a line extending to two sides by a specified width along a high-voltage overhead power line roadside wire. In the construction engineering of the transmission line engineering, the importance of the line corridor is increasingly highlighted, and the importance of the line corridor is improved to the same height as that of the line body.
In order to ensure the construction of the transmission line as expected and the safe and stable operation of the transmission line, the cleaning and protection of the transmission corridor are greatly relied on, and tall trees, buildings and the like in the corridor can have great influence on the safety of the transmission line engineering, so that the identification of objects existing in the corridor is particularly important.
At present, the method for identifying the transmission line is a traditional manual method, a method for identifying a helicopter with a person and the like. The manual method has the advantages of high labor intensity and low efficiency, is dangerous in high-voltage line inspection operation, and cannot be used for a power transmission line crossing barren mountain and wild mountains and deep-channel canyons. The identification of the manned helicopter is not influenced by the terrain environment factors, the inspection range is large, the precision is high, but the inspection cost is high due to the limitation of aviation control and weather conditions, and the unmanned helicopter cannot hover for a long time for fixed-point inspection.
Disclosure of Invention
In order to solve the problems, the present disclosure provides a method and a system for identifying an object in a power transmission line corridor based on a remote sensing image, wherein the method and the system divide a data set by a digital orthographic image of a target area, perform training learning by an improved convolutional neural network after data enhancement, and output a power transmission line corridor object identification result.
According to some embodiments, the present disclosure employs the following technical solutions:
the method for identifying the power transmission line corridor object based on the remote sensing image comprises the following steps:
obtaining visible light remote sensing image data of a target area of a power transmission line, and preprocessing to obtain a digital orthophoto map;
the digital orthographic image is input into a convolutional neural network model for corridor object identification after being cut through a sliding image window, in the convolutional neural network, the common convolution is replaced by the hole convolution in the convolutional layers, a certain number of the convolutional layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
According to some embodiments, the present disclosure employs the following technical solutions:
transmission line corridor object identification system based on remote sensing image, characterized by comprising:
the acquisition module is used for acquiring visible light remote sensing image data of a target area of the power transmission line, and acquiring a digital orthophoto map after preprocessing;
the identification classification module is used for cutting the digital orthophoto map through a sliding map window and inputting the digital orthophoto map into a convolution neural network model for identifying corridor objects, and in the convolution neural network, the common convolution is replaced by the hole convolution in the convolution layers, a certain number of the convolution layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
According to some embodiments, the present disclosure employs the following technical solutions:
a non-transitory computer readable storage medium for storing computer instructions which, when executed by a processor, implement a remote sensing image based transmission line corridor object identification method.
According to some embodiments, the present disclosure employs the following technical solutions:
an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected with the memory, and the computer program is stored in the memory, and when the electronic equipment runs, the processor executes the computer program stored in the memory, so that the electronic equipment executes the power transmission line corridor object identification method based on the remote sensing image.
Compared with the prior art, the beneficial effects of the present disclosure are:
the power transmission line corridor object identification method has the advantages of high wide area census efficiency, low cost, rich image textures, no influence of terrain, flexibility and high efficiency by acquiring the image data of the target area through the unmanned aerial vehicle. The identification of the power transmission line corridor objects has important significance for lean operation and maintenance of a power grid, acceptance of the power transmission corridor and cleaning of the power transmission corridor. The unmanned aerial vehicle is used for acquiring the image data of the target area, and has the advantages of high wide area census efficiency, low cost, rich image textures, no influence of terrain, flexibility and high efficiency. The identification of the power transmission line corridor objects has important significance for lean operation and maintenance of a power grid, acceptance of the power transmission corridor and cleaning of the power transmission corridor.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the exemplary embodiments of the disclosure and together with the description serve to explain the disclosure, and do not constitute an undue limitation on the disclosure.
FIG. 1 is a schematic flow diagram of a method in accordance with an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a convolutional neural network of an embodiment of the present disclosure;
FIG. 3 is a normal convolution schematic diagram of an embodiment of the present disclosure;
fig. 4 is a schematic diagram of hole convolution according to an embodiment of the present disclosure.
The specific embodiment is as follows:
the disclosure is further described below with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present disclosure. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
Example 1
An embodiment of the present disclosure provides a method for identifying an object in a power transmission line corridor based on a remote sensing image, including:
step one: obtaining visible light data of a target area of a power transmission line, and preprocessing to obtain a digital orthophoto map;
step two: the digital orthographic image is input into a convolutional neural network model for corridor object identification after being cut through a sliding image window, in the convolutional neural network, the common convolution is replaced by the hole convolution in the convolutional layers, a certain number of the convolutional layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
As an embodiment, the method flow of the remote sensing image-based power transmission line corridor identification of the present disclosure is as follows:
step 1: firstly, collecting the whole actual data of a power transmission line, and collecting the data by adopting an unmanned aerial vehicle; and shooting an aerial remote sensing image of the target area by using the unmanned aerial vehicle, and acquiring visible light remote sensing image data of the target area of the power transmission line.
Step 2: and (5) preprocessing data. The existing intelligent map software in Xinjiang is adopted to process remote sensing images, corresponding digital orthophoto DOM is output, meanwhile, generated DOM is cut, the digital orthophoto map is cut to remove the background, and only the digital orthophoto map DOM of a target area is reserved.
Step 3: and cutting the digital orthographic image of the target area into a plurality of images by a sliding image window method, and inputting the images into a convolutional neural network model for corridor object recognition for recognition.
Because the image data of the target area is very large in size, the overflow of the computer memory can be caused, a plurality of images are generated by cutting through a sliding window, and then the images are input into a convolutional neural network model for identifying corridor objects for identification and classification, so that the overflow of the computer memory can be effectively relieved.
The training process of the convolutional neural network for corridor object recognition is as follows:
acquiring training data: the training data is based on ground survey samples, including images cut according to a sliding map window, including all trees, houses, tombs, etc. in the category. Because the illumination intensity can influence the recognition rate of the network, in order to make the training sample more feasible, the sample diversity is ensured, and various types of data contain different time and different weather as far as possible. The test data and the training data are independent of each other.
Data enhancement: when the convolutional neural network is used for identification, sufficient data are needed to ensure that the network has a good identification effect, if the sample is insufficient, the problems of low network identification precision, network overfitting and the like can be caused, and the existing image data are continuously generated into new sample data by random distortion, deformation overturning and the like.
The convolutional neural network performs training learning: the convolutional neural network adopts a mechanism of local connection and weight sharing to extract the characteristic value of the remote sensing image, and the characteristic value has scale and translation invariance. Convolutional neural networks generally consist of an input layer, a convolutional layer, a pooling layer, and a fully connected layer.
The input layer reads the image data and generates vectors with specific sizes as inputs of the convolutional neural network. The convolution layer is composed of a plurality of feature graphs, is obtained through image convolution operation and nonlinear mapping conversion by utilizing a trainable convolution kernel, and the convolution kernel is used for influencing feature extraction, and has good and bad effects on the feature extraction performance of the convolution neural network. The pooling layer is a downsampling layer, reduces the dimension of the characteristic diagram of the convolution layer, keeps the original characteristic diagram information to the greatest extent, enhances the characteristic translation invariance, reduces the input size and the participation of the characteristic diagram of the next layer, reduces the complexity of a model and reduces the risk of overfitting. The multi-neuron of the full-connection layer is connected with the characteristic node of the upper layer, and the characteristic graphs obtained by convolution, pooling and nonlinear operation in the network are integrated and converted into one-dimensional characteristic vectors representing the global information of the image for image classification.
In general, network model parameters in a convolutional neural network are mainly concentrated on a convolutional kernel of a convolutional layer and connection weights of a full-connection layer, wherein the convolutional layer extracts features of an image, and the full-connection layer is used for integrating and classifying the features.
For the convolutional layer, the present disclosure replaces the normal convolution with the hole convolution, which has the following advantages over the normal convolution:
1. expanding receptive field: conventional downsampling increases receptive field but reduces spatial resolution. And the cavity convolution can ensure resolution while expanding receptive fields. The method is very suitable for detection and segmentation tasks, large targets can be detected and segmented by increasing receptive fields, and the targets can be accurately positioned by high resolution.
2. Capturing multi-scale context information: the parameter condition rate in the hole convolution indicates that (condition rate-1) 0 s are filled in the convolution kernel. Different conditions are set to bring different receptive fields to the network, namely, multi-scale information is acquired.
Meanwhile, hole convolution is called gridding, that is, a gridding effect/checkerboard problem. Because adjacent pixels are convolved from mutually independent subsets in the result of a layer obtained by hole convolution, there is a lack of dependence on each other.
The hole convolution may have local information loss: because the calculation mode of the cavity convolution is similar to a chessboard format, the convolution result obtained by a certain layer is independent from the independent set of the previous layer, and the independent sets are not interdependent, so that the convolution results of the layer are not correlated, namely local information is lost; there is also a problem that information obtained remotely has no correlation, and because of the sparse sampling input signal of the cavity convolution, information obtained by remote convolution has no correlation, and a classification result is affected.
The present disclosure differs from the deeplab scheme that employs the same void fraction, which forms a certain number of convolutional layers (layers) into one group, and then each group uses a continuously increasing void fraction, with the other groups repeating. According to the scheme, information can be acquired from a wider pixel range, and the grid problem is avoided. Meanwhile, the scheme can also adjust the receptive field at will by modifying the rate.
In order to prevent the network from generating over fitting, the convolutional neural network is improved to a certain extent, and a Dropout layer is added after the pooling layer, so that the Dropout layer can prevent the network from generating over fitting, and the generalization capability of the network is stronger.
The original image before generating the digital orthophoto image DOM has a plurality of similar places with the DOM, and only has a small gap, so that the weight parameters trained on the original data by the convolutional neural network can be finely adjusted on the data set of the DOM by adopting a model migration mode.
Firstly, preprocessing such as removing average value and normalizing an input image through an input layer; then the convolution layer greatly reduces the calculation parameters of the model by using a plurality of cavity convolution kernels to perform local perception; the convolution layer is followed by downsampling through the pooling layer, and the characteristic values are extracted, so that the characteristic diagram is smaller, the network computing complexity is reduced, and the algorithm computing time is shortened; after the pooling layer is added with a Dropout layer, the Dropout layer can prevent the network from being fitted, so that the generalization capability of the network is stronger; and finally classifying the image features through the full connection layer.
Step 4: and outputting a result of identifying the transmission line corridor, and mapping the result from the image coordinate system to the real coordinate system.
GeoTIFF, geoTIFF is a common domain metadata standard for orthographic image formats that has geographic registration information embedded in the image file. The geographic registration information is contained by tif tags that contain spatial information about the image file, such as map projections, coordinate systems, ellipsoids, fiducials. In this way, the recognized target of the digital orthographic image can be converted to obtain the corresponding real coordinate.
Wherein: x' is the geographic X coordinate (distance or longitude) corresponding to the pixel; y' is the geographic Y coordinate (distance or latitude) corresponding to the pixel; x is the pixel coordinate column number; y is the pixel coordinate line number; a is the pixel resolution in the X direction; d is the rotation coefficient in the X direction; b is the rotation coefficient in the Y direction; e is the pixel resolution in the Y direction; c is the center X coordinate (distance or longitude) of the pixel at the upper left corner of the grid map; f is the center Y coordinate (distance or latitude) of the upper left corner pixel of the grid map.
Example 2
In one embodiment of the present disclosure, a power transmission line corridor object recognition system based on a remote sensing image is provided, which is characterized in that the system includes:
the acquisition module is used for acquiring visible light data of a target area of the power transmission line, and acquiring a digital orthophoto map after preprocessing;
the identification classification module is used for cutting the digital orthophoto map through a sliding map window and inputting the digital orthophoto map into a convolution neural network model for identifying corridor objects, and in the convolution neural network, the common convolution is replaced by the hole convolution in the convolution layers, a certain number of the convolution layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
Example 3
In one embodiment of the disclosure, a non-transitory computer readable storage medium is provided, where the non-transitory computer readable storage medium is configured to store computer instructions, where the computer instructions, when executed by a processor, implement the steps of the method for identifying an object in a power transmission line corridor based on a remote sensing image.
Example 4
In one embodiment of the present disclosure, there is provided an electronic device including: a processor, a memory, and a computer program; the processor is connected with the memory, the computer program is stored in the memory, and when the electronic equipment runs, the processor executes the computer program stored in the memory so as to enable the electronic equipment to execute the steps of the method for identifying the object of the transmission line corridor based on the remote sensing image.
Example 2, example 3 and example 4 specifically perform the method steps referred to in example 1.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.
Claims (10)
1. The method for identifying the object in the transmission line corridor based on the remote sensing image is characterized by comprising the following steps of:
obtaining visible light remote sensing image data of a target area of a power transmission line, and preprocessing to obtain a digital orthophoto map;
the digital orthographic image is input into a convolutional neural network model for corridor object identification after being cut through a sliding image window, in the convolutional neural network, the common convolution is replaced by the hole convolution in the convolutional layers, a certain number of the convolutional layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
2. The method for identifying the object in the transmission line corridor based on the remote sensing image according to claim 1, wherein the visible light data of the target area is shot by the unmanned aerial vehicle, the corresponding digital orthophoto is output after preprocessing, the digital orthophoto map is cut to remove the background, and only the digital orthophoto map of the target area is reserved.
3. The method for identifying the corridor object of the power transmission line based on the remote sensing image according to claim 1, wherein the digital orthographic image of the target area is cut into a plurality of images by a sliding image window method and then is input into a convolutional neural network model for identifying the corridor object for identification.
4. The remote sensing image-based power transmission line corridor object identification method as claimed in claim 1, wherein the convolutional neural network adopts a mechanism of local connection and weight sharing to extract characteristic values of images, and the characteristic values have scale and translation invariance; the convolutional neural network consists of an input layer, a convolutional layer, a pooling layer and a full connection layer.
5. The remote sensing image-based power transmission line corridor object identification method as claimed in claim 4, wherein the input layer of the convolutional neural network reads image data, generates vectors with specific sizes as the input of the convolutional neural network, and the convolutional layer is formed by a plurality of feature maps, is obtained by image convolution operation and nonlinear mapping conversion by using a trainable convolution kernel; the pooling layer reduces the dimension of the characteristic diagram of the convolution layer, retains the original characteristic diagram information, connects the multi-neuron of the full-connection layer with the characteristic node of the upper layer, integrates the characteristic diagram obtained by convolution, pooling and nonlinear operation in the network, and converts the characteristic diagram into a one-dimensional characteristic vector representing the global information of the image to output the identification classification result.
6. The method for identifying objects in a power transmission line corridor based on a remote sensing image according to claim 5, wherein the convolution layers replace a normal convolution with a hole convolution, capture multi-scale context information and expand a receptive field, a certain number of convolution layers are formed into one group, then each group uses a continuously increasing void ratio, and the other groups repeat.
7. The remote sensing image based transmission line corridor object identification method of claim 5, wherein a Dropout layer is added after the pooling layer; and (3) adopting a model migration mode to finely adjust weight parameters trained on the original data by the convolutional neural network on the data set of the digital orthographic image.
8. Transmission line corridor object identification system based on remote sensing image, characterized by comprising:
the acquisition module is used for acquiring visible light remote sensing image data of a target area of the power transmission line, and acquiring a digital orthophoto map after preprocessing;
the identification classification module is used for cutting the digital orthophoto map through a sliding map window and inputting the digital orthophoto map into a convolution neural network model for identifying corridor objects, and in the convolution neural network, the common convolution is replaced by the hole convolution in the convolution layers, a certain number of the convolution layers are formed into a group, and each group uses continuously increased hole rate; extracting a characteristic value of the remote sensing image by adopting a mechanism of local connection and weight sharing, and carrying out identification, classification and output on a power transmission line corridor;
and mapping the identification classification result from the image coordinate system to a real coordinate system, and acquiring the real position of the corridor object identification.
9. A non-transitory computer readable storage medium storing computer instructions which, when executed by a processor, implement the remote sensing image based transmission line corridor object identification method of any one of claims 1-7.
10. An electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the method for identifying the object of the transmission line corridor based on the remote sensing image according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310398286.2A CN116416527A (en) | 2023-04-10 | 2023-04-10 | Remote sensing image-based power transmission line corridor object identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310398286.2A CN116416527A (en) | 2023-04-10 | 2023-04-10 | Remote sensing image-based power transmission line corridor object identification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116416527A true CN116416527A (en) | 2023-07-11 |
Family
ID=87056066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310398286.2A Pending CN116416527A (en) | 2023-04-10 | 2023-04-10 | Remote sensing image-based power transmission line corridor object identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116416527A (en) |
-
2023
- 2023-04-10 CN CN202310398286.2A patent/CN116416527A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583369B (en) | Target identification method and device based on target area segmentation network | |
| CN108573276B (en) | Change detection method based on high-resolution remote sensing image | |
| Peng et al. | Wild animal survey using UAS imagery and deep learning: modified Faster R-CNN for kiang detection in Tibetan Plateau | |
| Chen et al. | Vehicle detection in high-resolution aerial images via sparse representation and superpixels | |
| Chen et al. | Vehicle detection in high-resolution aerial images based on fast sparse representation classification and multiorder feature | |
| CN107918776B (en) | Land planning method and system based on machine vision and electronic equipment | |
| CN111046880A (en) | Infrared target image segmentation method and system, electronic device and storage medium | |
| CN111797712B (en) | Remote sensing image cloud and cloud shadow detection method based on multi-scale feature fusion network | |
| CN114092833B (en) | Remote sensing image classification method and device, computer equipment and storage medium | |
| CN112801158A (en) | Deep learning small target detection method and device based on cascade fusion and attention mechanism | |
| CN111639530B (en) | Method and system for detecting and identifying power transmission tower and insulator of power transmission line | |
| CN112419333B (en) | Remote sensing image self-adaptive feature selection segmentation method and system | |
| CN114943893B (en) | Feature enhancement method for land coverage classification | |
| CN113537180B (en) | Tree obstacle identification method and device, computer equipment and storage medium | |
| CN113515655A (en) | Fault identification method and device based on image classification | |
| CN110852327A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN114972646B (en) | Method and system for extracting and modifying independent ground objects of live-action three-dimensional model | |
| CN116206223A (en) | Fire detection method and system based on unmanned aerial vehicle edge calculation | |
| CN115619719A (en) | Pine wood nematode infected wood detection method based on improved Yolo v3 network model | |
| Ataş | Performance Evaluation of Jaccard-Dice Coefficient on Building Segmentation from High Resolution Satellite Images | |
| CN109284752A (en) | A kind of rapid detection method of vehicle | |
| CN113033386A (en) | High-resolution remote sensing image-based transmission line channel hidden danger identification method and system | |
| CN111931577A (en) | Intelligent inspection method for specific foreign matters of power grid line | |
| Wu et al. | Research on asphalt pavement disease detection based on improved YOLOv5s | |
| CN115861756A (en) | Earth background small target identification method based on cascade combination network |
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 |