CN113392803A - Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium - Google Patents
Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium Download PDFInfo
- Publication number
- CN113392803A CN113392803A CN202110742552.XA CN202110742552A CN113392803A CN 113392803 A CN113392803 A CN 113392803A CN 202110742552 A CN202110742552 A CN 202110742552A CN 113392803 A CN113392803 A CN 113392803A
- Authority
- CN
- China
- Prior art keywords
- image data
- remote sensing
- sensing image
- transmission line
- power transmission
- 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 41
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 49
- 238000012549 training Methods 0.000 claims abstract description 35
- 230000009467 reduction Effects 0.000 claims abstract description 26
- 238000012360 testing method Methods 0.000 claims abstract description 21
- 238000001914 filtration Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 230000004927 fusion Effects 0.000 claims description 14
- 230000009466 transformation Effects 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000009792 diffusion process Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 230000002776 aggregation Effects 0.000 claims description 7
- 238000004220 aggregation Methods 0.000 claims description 7
- 230000008602 contraction Effects 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 5
- 238000007689 inspection Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 235000005770 birds nest Nutrition 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002401 inhibitory effect Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 235000005765 wild carrot Nutrition 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device, a terminal and a storage medium for identifying suspended foreign matters of a power transmission line, wherein the method comprises the following steps: acquiring historical remote sensing image data of a target area, and performing noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data; dividing the second historical remote sensing image data into a training image data set and a test image data set; respectively inputting the training image data set and the test image data set into a preset Faster-RCNN model for training to obtain a second Faster-RCNN model; and acquiring real-time remote sensing image data of a target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after noise reduction processing to obtain a recognition result of the suspended foreign matter of the power transmission line. The method and the device can improve the speed of identifying the suspended foreign matters of the power transmission line and the accuracy of identification.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method, a device, a terminal and a storage medium for recognizing suspended foreign matters of a power transmission line.
Background
The high-voltage transmission line mainly plays a role in transmitting electric energy in daily life, and the importance of maintaining the safe and stable operation of the high-voltage transmission line is self-evident. Because the transmission lines are mainly distributed in the outdoor environment, potential safety hazards are inevitably generated due to the influence of external foreign matters (plastic bags, branches, bird nests, kites and the like), and the environment around the transmission lines needs to be regularly inspected. The manual inspection of the foreign matters consumes a great amount of manpower, and due to negligence of workers, the conditions of lack of inspection and omission of inspection often occur. At present, the remote sensing mode is mainly adopted domestically to shoot the images of the suspended foreign matters on the wires, so that the inspection efficiency is guaranteed, and the line maintenance speed is increased. In addition, due to the influence of weather factors such as crosswind and light, the obtained image is often not clear enough, which requires the image to be subjected to noise reduction before application.
At present, the existing identification method has the following problems:
1. when the remote sensing image is obtained, the background of the image is complex and changeable due to the movement of the suspension object of the power transmission line, and the image is not clear enough, so that some important characteristic information in the image is often lost when the existing method is used for carrying out noise reduction processing on the image.
2. Due to the fact that the states of foreign matters such as bird nests, branches and plastic bags in the acquired images are complex and changeable and the background distribution of the images is complex, the identification accuracy rate of the prior art scheme is low.
Disclosure of Invention
The purpose of the invention is: the invention provides a method, a device, a terminal and a storage medium for identifying suspended foreign matters of a power transmission line.
In order to achieve the above object, the present invention provides a method for identifying a suspended foreign object in a power transmission line, comprising:
acquiring historical remote sensing image data of a target area, and performing noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data;
dividing the second historical remote sensing image data into a training image data set and a test image data set;
respectively inputting the training image data set and the test image data set into a preset Faster-RCNN model for training to obtain a second Faster-RCNN model;
and acquiring real-time remote sensing image data of a target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after noise reduction processing to obtain a recognition result of the suspended foreign matter of the power transmission line.
Further, the acquiring historical remote sensing image data of the target area, and performing noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data includes:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
Further, the preset fast-RCNN model comprises: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
Further, the feature extraction unit adopts the following calculation formula:
wherein p isnIs the conventional convolution output position p0Corresponding to an integer offset, ω (p)n) For corresponding weights, Δ pnIndicating that an offset, p, is added to the conventional convolution0+pn+ΔpnRepresenting possible spatial positions in the feature χ, y (p)0) Is shown in position p0The output characteristic map of (1).
Further, the feature fusion unit adopts the following calculation formula:
where γ and b are two constants, set to 2 and 1, | toodRepresenting the odd number closest to t, there is a mapping ψ between the convolution kernel size k and the number of channels C, the value of k being determined adaptively.
The invention also provides a device for identifying the suspended foreign matter of the power transmission line, which comprises: a data processing module, a splitting module, a training module and an identification module, wherein,
the data processing module is used for acquiring historical remote sensing image data of a target area, and performing noise reduction processing on the historical remote sensing image data to acquire second historical remote sensing image data;
the splitting module is used for splitting the second historical remote sensing image data into a training image data set and a test image data set;
the training module is used for inputting the training image data set and the testing image data set into a preset Faster-RCNN model respectively for training to obtain a second Faster-RCNN model;
the identification module is used for acquiring real-time remote sensing image data of a target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after noise reduction processing to obtain an identification result of the power transmission line suspended foreign matter.
Further, the data acquisition module is specifically configured to:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
Further, the preset fast-RCNN model comprises: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
The present invention also provides a computer terminal device, comprising: one or more processors; a memory coupled to the processor for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method for identifying the power transmission line hanging foreign object as described in any one of the above.
The invention also provides a computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for identifying a hanging foreign object on a power transmission line according to any one of the above.
Compared with the prior art, the method, the device, the terminal equipment and the computer readable storage medium for identifying the suspended foreign matter of the power transmission line have the advantages that:
1. the invention adopts an improved BM3D image noise reduction method, firstly carries out anisotropic diffusion filtering on a noise image, can retain the edge information of the image while inhibiting noise, and then uses a method of searching similar blocks along the edge direction instead of the horizontal direction, and the similar blocks obtained at the edge of the image can well retain the detail information of the image.
2. The invention improves the fast-RCNN model, and particularly shows that a ResNet-101 network is selected as a basic network for characteristic extraction, and a DCN structure is added to the last three stages of the ResNet network so as to enhance the transformation modeling capability of CNN and expand the receptive field of a high-level network. And in the feature fusion stage, the ECA high-efficiency channel attention module is combined to promote the fusion of the network to the multi-level features, at the moment, the feature pyramid can pay more attention to the interactive information of the adjacent channels, and the identifiability of the FPN structure is improved. In order to improve the performance of the model, the ROI-Align algorithm is used for replacing the ROI-Powing algorithm in the target detection network, and the improvement enhances the recognition capability of the fast-RCNN model on small target objects.
Drawings
Fig. 1 is a schematic flow chart of a method for identifying a suspended foreign object in a power transmission line provided by the invention;
FIG. 2 is a schematic flow chart of the improved BM3D denoising algorithm provided by the present invention;
FIG. 3 is a schematic diagram of the structure of a network of the improved Faster-RCNN model provided by the present invention;
FIG. 4 is a schematic diagram of a conventional convolution structure provided by the present invention;
FIG. 5 is a schematic diagram of an ECA high efficiency channel attention module configuration provided by the present invention;
fig. 6 is a schematic structural diagram of a device for identifying a suspended foreign object in a power transmission line provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be understood that the step numbers used herein are for convenience of description only and are not intended as limitations on the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of the described features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to and includes any and all possible combinations of one or more of the associated listed items.
As shown in fig. 1, a method for identifying a suspended foreign object in a power transmission line according to an embodiment of the present invention at least includes the following steps:
s1, obtaining historical remote sensing image data of a target area, and carrying out noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data;
specifically, after a remote sensing image is obtained, anisotropic diffusion filtering is carried out on a noise image, then an edge region and a smooth region of the image are obtained by utilizing a Sobel operator, similar blocks are searched in the smooth region along the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, similar blocks are searched in the edge region along the vertical direction and the edge direction, and finally a de-noised image is obtained through grouping, 3D conversion, coefficient contraction, inverse 3D conversion, block estimation and aggregation. The flow chart of the improved BM3D algorithm is shown in fig. 2.
S2, dividing the second historical remote sensing image data into a training image data set and a test image data set;
specifically, dividing the second historical remote sensing image data into a training image data set and a test image data set;
s3, inputting the training image data set and the testing image data set into a preset Faster-RCNN model respectively for training to obtain a second Faster-RCNN model;
specifically, the structure of the fast-RCNN model network is shown in FIG. 3. ResNet-101 is adopted in the backbone network of the model, and the low-level semantic information of pictures is mainly learned in the shallow layer of the backbone network, so that the DCN structure is introduced only in the last 3 stages of the network. The conventional convolution structure (see (a) diagram in fig. 4) is calculated as follows:
in comparison, the deformable convolution is added with a small offset Δ p based on the conventional convolutionn，ΔpnAnother convolution calculation can be used to obtain the updated deformable convolution calculation method as follows:
in the training process, firstly, feature extraction is carried out on an input image through a ResNet-101 network, feature fusion is carried out on obtained feature maps containing different semantic information from top to bottom at the output ends of different Bottleneck structures, an ECA efficient channel attention module is introduced into the highest layer of an FPN to enhance the correlation among channels, and then generation of a feature pyramid is guided in the layer, so that the feature maps containing low-level and high-level semantic information are obtained.
The model structure of the ECA efficient channel attention module is shown in fig. 5. The module can perform global average pooling on feature graphs χ input into all channels under the condition of not reducing dimensionality, learn through one-dimensional convolution which can share weight values, capture interaction among cross-channels by each channel and k neighbors thereof in the learning process, wherein k represents the size of a one-dimensional convolution kernel, and the following calculation formula can be obtained through the proportional relation between k and a channel dimensionality C:
wherein, the values of γ ═ 2, b ═ 1, and k can be determined adaptively.
For the feature map output by the backbone network, a generation recommendation area can be obtained by outputting some target frames through the area nomination network, and then a Softmax classifier judges whether the anchor point belongs to the background or the foreground, so that accurate proposals can be obtained finally. When pooling operation is carried out, the collected feature maps and the propusals are integrated, ROI-Align algorithm is adopted to pool the ROIs with different sizes into the feature maps with the same size, the ROI mapping method is set as a nearest neighbor interpolation method, and finally prediction results of the category and position information of the target can be obtained.
And S4, acquiring real-time remote sensing image data of the target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after denoising processing to obtain the recognition result of the power transmission line suspended foreign matter.
Specifically, real-time remote sensing image data of a target area are obtained, and the real-time remote sensing image data are input into the second Faster-RCNN model after noise reduction processing, so that a recognition result of the power transmission line suspended foreign matter is obtained.
In an embodiment of the present invention, the obtaining of the historical remote sensing image data of the target area and the denoising of the historical remote sensing image data to obtain the second historical remote sensing image data includes:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
In one embodiment of the present invention, the preset fast-RCNN model includes: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
In an embodiment of the present invention, the feature extraction unit adopts the following calculation formula:
wherein p isnIs the conventional convolution output position p0Corresponding to an integer offset, ω (p)n) For corresponding weights, Δ pnIndicating that an offset, p, is added to the conventional convolution0+pn+ΔpnRepresenting possible spatial positions in the feature χ, y (p)0) Is shown in position p0The output characteristic map of (1).
In an embodiment of the present invention, the feature fusion unit adopts the following calculation formula:
where γ and b are two constants, set to 2 and 1, | toodRepresenting the odd number closest to t, there is a mapping ψ between the convolution kernel size k and the number of channels C, the value of k being determined adaptively.
Compared with the prior art, the method for identifying the suspended foreign matter of the power transmission line has the beneficial effects that:
1. the invention adopts an improved BM3D image noise reduction method, firstly carries out anisotropic diffusion filtering on a noise image, can retain the edge information of the image while inhibiting noise, and then uses a method of searching similar blocks along the edge direction instead of the horizontal direction, and the similar blocks obtained at the edge of the image can well retain the detail information of the image.
2. The invention improves the fast-RCNN model, and particularly shows that a ResNet-101 network is selected as a basic network for characteristic extraction, and a DCN structure is added to the last three stages of the ResNet network so as to enhance the transformation modeling capability of CNN and expand the receptive field of a high-level network. And in the feature fusion stage, the ECA high-efficiency channel attention module is combined to promote the fusion of the network to the multi-level features, at the moment, the feature pyramid can pay more attention to the interactive information of the adjacent channels, and the identifiability of the FPN structure is improved. In order to improve the performance of the model, the ROI-Align algorithm is used for replacing the ROI-Powing algorithm in the target detection network, and the improvement enhances the recognition capability of the fast-RCNN model on small target objects.
As shown in fig. 5, the present invention further provides an apparatus 200 for identifying a foreign object suspended on a power transmission line, comprising: a data processing module 201, a splitting module 202, a training module 203, and a recognition module 204, wherein,
the data processing module 201 is configured to obtain historical remote sensing image data of a target area, and perform noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data;
the splitting module 202 is configured to split the second historical remote sensing image data into a training image data set and a test image data set;
the training module 203 is configured to input the training image data set and the test image data set to a preset fast-RCNN model for training, so as to obtain a second fast-RCNN model;
the identification module 204 is configured to obtain real-time remote sensing image data of a target area, perform noise reduction processing on the real-time remote sensing image data, and input the real-time remote sensing image data into the second fast-RCNN model to obtain an identification result of a power transmission line suspended foreign matter.
In an embodiment of the present invention, the data obtaining module is specifically configured to:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
In one embodiment of the present invention, the preset fast-RCNN model includes: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
The present invention also provides a computer terminal device, comprising: one or more processors; a memory coupled to the processor for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method for identifying the power transmission line hanging foreign object as described in any one of the above.
It should be noted that the processor may be a Central Processing Unit (CPU), other general-purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an application-specific programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., the general-purpose processor may be a microprocessor, or the processor may be any conventional processor, the processor is a control center of the terminal device, and various interfaces and lines are used to connect various parts of the terminal device.
The memory mainly includes a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function, and the like, and the data storage area may store related data and the like. In addition, the memory may be a high speed random access memory, may also be a non-volatile memory, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (FlashCard), and the like, or may also be other volatile solid state memory devices.
It should be noted that the terminal device may include, but is not limited to, a processor and a memory, and those skilled in the art will understand that the terminal device is only an example and does not constitute a limitation of the terminal device, and may include more or less components, or combine some components, or different components.
The invention also provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for identifying a hanging foreign object on a power transmission line according to any one of the above.
It should be noted that the computer program may be divided into one or more modules/units (e.g., computer program), and the one or more modules/units are stored in the memory and executed by the processor to implement the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the terminal device.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (10)
1. A method for identifying a suspended foreign matter in a power transmission line is characterized by comprising the following steps:
acquiring historical remote sensing image data of a target area, and performing noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data;
dividing the second historical remote sensing image data into a training image data set and a test image data set;
respectively inputting the training image data set and the test image data set into a preset Faster-RCNN model for training to obtain a second Faster-RCNN model;
and acquiring real-time remote sensing image data of a target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after noise reduction processing to obtain a recognition result of the suspended foreign matter of the power transmission line.
2. The method for identifying the power transmission line suspended foreign matter according to claim 1, wherein the step of obtaining historical remote sensing image data of a target area and performing noise reduction processing on the historical remote sensing image data to obtain second historical remote sensing image data comprises the steps of:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
3. The method for identifying the power transmission line hanging foreign matter according to claim 1, wherein the preset fast-RCNN model comprises: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
4. The method for identifying the power transmission line suspended foreign matter according to claim 3, wherein the feature extraction unit adopts the following calculation formula:
wherein p isnIs the conventional convolution output position p0Corresponding to an integer offset, ω (p)n) For corresponding weights, Δ pnIndicating that an offset, p, is added to the conventional convolution0+n+ΔpnTo representPossible spatial positions in the feature map x, y (p)0) Is shown in position p0The output characteristic map of (1).
5. The method for identifying the power transmission line suspended foreign matter according to claim 3, wherein the feature fusion unit adopts the following calculation formula:
where γ and b are two constants, set to 2 and 1, | toodThe value of k is adaptively determined, indicating the odd number closest to t, and there is a mapping ψ between the convolution kernel size and the number of channels C.
6. The utility model provides a transmission line hangs recognition device of foreign matter which characterized in that includes: a data processing module, a splitting module, a training module and an identification module, wherein,
the data processing module is used for acquiring historical remote sensing image data of a target area, and performing noise reduction processing on the historical remote sensing image data to acquire second historical remote sensing image data;
the splitting module is used for splitting the second historical remote sensing image data into a training image data set and a test image data set;
the training module is used for inputting the training image data set and the testing image data set into a preset Faster-RCNN model respectively for training to obtain a second Faster-RCNN model;
the identification module is used for acquiring real-time remote sensing image data of a target area, and inputting the real-time remote sensing image data into the second Faster-RCNN model after noise reduction processing to obtain an identification result of the power transmission line suspended foreign matter.
7. The device for identifying the foreign object hung on the power transmission line according to claim 6, wherein the data acquisition module is specifically configured to:
carrying out anisotropic diffusion filtering processing on the historical remote sensing image data to obtain the training data set and the test data set after filtering processing;
calculating to obtain an edge area and a smooth area of the historical remote sensing image data by adopting a Sobel operator, wherein the smooth area searches similar blocks in the horizontal direction and the vertical direction by adopting a traditional BM3D algorithm, and the edge area searches the similar blocks in the vertical direction and the edge direction;
and grouping, 3D transformation, coefficient contraction, inverse 3D transformation, block estimation and aggregation processing are carried out on the edge area and the smooth area, and historical remote sensing image data subjected to noise reduction processing are obtained.
8. The apparatus for identifying the power transmission line hanging foreign object as claimed in claim 6, wherein the preset fast-RCNN model comprises: the device comprises a feature extraction unit, a feature fusion unit and a prediction unit.
9. A computer terminal device, comprising:
one or more processors;
a memory coupled to the processor for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of identifying a hanging foreign object on a power transmission line according to any one of claims 1 to 5.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements a method for identifying a hanging foreign object on a power transmission line according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110742552.XA CN113392803A (en) | 2021-06-30 | 2021-06-30 | Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110742552.XA CN113392803A (en) | 2021-06-30 | 2021-06-30 | Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113392803A true CN113392803A (en) | 2021-09-14 |
Family
ID=77624835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110742552.XA Pending CN113392803A (en) | 2021-06-30 | 2021-06-30 | Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113392803A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738307A (en) * | 2020-06-02 | 2020-10-02 | 国网宁夏电力有限公司检修公司 | Foreign matter identification method and system in power transmission line environment based on fast RCNN and computer readable storage medium |
| CN111950488A (en) * | 2020-08-18 | 2020-11-17 | 山西大学 | Improved fast-RCNN remote sensing image target detection method |
| CN112651337A (en) * | 2020-12-25 | 2021-04-13 | 国网黑龙江省电力有限公司电力科学研究院 | Sample set construction method applied to training line foreign object target detection model |
-
2021
- 2021-06-30 CN CN202110742552.XA patent/CN113392803A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738307A (en) * | 2020-06-02 | 2020-10-02 | 国网宁夏电力有限公司检修公司 | Foreign matter identification method and system in power transmission line environment based on fast RCNN and computer readable storage medium |
| CN111950488A (en) * | 2020-08-18 | 2020-11-17 | 山西大学 | Improved fast-RCNN remote sensing image target detection method |
| CN112651337A (en) * | 2020-12-25 | 2021-04-13 | 国网黑龙江省电力有限公司电力科学研究院 | Sample set construction method applied to training line foreign object target detection model |
Non-Patent Citations (4)
| Title |
|---|
| QILONG WANG ET AL.: "ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks", 2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, 5 August 2020 (2020-08-05), pages 1 - 9 * |
| 刘黎等: "改进的Faster-RCNN目标检测方法在变电站悬挂异物检测中的应用", 《电测与仪表》, vol. 58, no. 1, 15 January 2021 (2021-01-15), pages 142 - 146 * |
| 肖佳: "基于改进三维块匹配算法的图像去噪研究", 《中国优秀硕士学位论文全文数据库信息科技辑》, no. 2020, 15 March 2020 (2020-03-15), pages 138 - 759 * |
| 董洪义: "《深度学习之PyTorch物体检测实战》", vol. 2020, 31 March 2020, 机械工业出版社, pages: 135 - 138 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11429818B2 (en) | Method, system and device for multi-label object detection based on an object detection network | |
| CN106951830B (en) | Image scene multi-object marking method based on prior condition constraint | |
| CN109801265B (en) | Real-time transmission equipment foreign matter detection system based on convolutional neural network | |
| CN110555420B (en) | Fusion model network and method based on pedestrian regional feature extraction and re-identification | |
| CN112132145B (en) | Image classification method and system based on model extended convolutional neural network | |
| CN112419202B (en) | Automatic wild animal image recognition system based on big data and deep learning | |
| CN103853724A (en) | Multimedia data sorting method and device | |
| US20230326173A1 (en) | Image processing method and apparatus, and computer-readable storage medium | |
| CN112580662A (en) | Method and system for recognizing fish body direction based on image features | |
| CN110633640A (en) | Method for identifying complex scene by optimizing PointNet | |
| CN114444565A (en) | Image tampering detection method, terminal device and storage medium | |
| CN105354547A (en) | Pedestrian detection method in combination of texture and color features | |
| CN114782919A (en) | Road grid map construction method and system with real and simulation data enhanced | |
| CN111223125B (en) | Target motion video tracking method based on Python environment | |
| CN110633630B (en) | Behavior identification method and device and terminal equipment | |
| CN113392803A (en) | Method and device for identifying suspended foreign matters of power transmission line, terminal and storage medium | |
| CN116363583A (en) | Human body identification method, device, equipment and medium for top view angle | |
| CN113284221B (en) | Target detection method and device and electronic equipment | |
| CN112446292B (en) | 2D image salient object detection method and system | |
| CN111950586B (en) | Target detection method for introducing bidirectional attention | |
| CN115272741A (en) | Detection method of slender flexible object, terminal equipment and storage medium | |
| CN113343749A (en) | Fruit identification method and system based on D2Det model | |
| CN111950475A (en) | Yalhe histogram enhancement type target recognition algorithm based on yoloV3 | |
| CN111950451A (en) | Multi-class target identification method based on multi-scale prediction CNN and Loongson chip | |
| CN115984583B (en) | Data processing method, apparatus, computer device, storage medium, and program product |
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 |