CN113392730A - Power distribution network equipment image identification method and computer readable storage medium - Google Patents
Power distribution network equipment image identification method and computer readable storage medium Download PDFInfo
- Publication number
- CN113392730A CN113392730A CN202110598038.3A CN202110598038A CN113392730A CN 113392730 A CN113392730 A CN 113392730A CN 202110598038 A CN202110598038 A CN 202110598038A CN 113392730 A CN113392730 A CN 113392730A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- power distribution
- network equipment
- image
- picture
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 33
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 238000003709 image segmentation Methods 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 description 8
- 239000013604 expression vector Substances 0.000 description 7
- 230000009467 reduction Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000008520 organization Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002146 bilateral effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000003707 image sharpening Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000007619 statistical method Methods 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
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Economics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a power distribution network equipment image identification method and a computer readable storage medium, wherein the method comprises the following steps: establishing a power distribution network equipment image recognition training library; acquiring an image of the power distribution network equipment, and preprocessing the image; obtaining the area of the power distribution network equipment in the preprocessed power distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified; extracting the characteristics of the power distribution network equipment; according to the characteristics, determining the equipment subtype of the power distribution network equipment, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library; and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized. The invention can improve the identification efficiency of the images of the power distribution network equipment.
Description
Technical Field
The invention relates to the field of power grid image recognition, in particular to a power distribution network equipment image recognition method and a computer readable storage medium.
Background
At present, image recognition technology is mature, and many researches are made on electric power equipment, such as '*** lenet inclusion-V3 model-based electric power equipment image recognition' by yogkai et al, 'migration learning and convolutional neural network electric power equipment image recognition method' by king xin et al, and 'deep learning-based electric power equipment image recognition model construction' by tangy et al.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the image recognition method for the power distribution network equipment and the computer readable storage medium can improve the recognition efficiency of the power distribution network equipment image.
In order to solve the technical problems, the invention adopts the technical scheme that: an image identification method for power distribution network equipment comprises the following steps:
establishing a power distribution network equipment image recognition training library, wherein sample pictures corresponding to the subtype of each equipment are stored in the power distribution network equipment image recognition training library, and the corresponding equipment types are marked on the sample pictures;
acquiring a power distribution network equipment image, and preprocessing the power distribution network equipment image;
obtaining the area of the distribution network equipment in the preprocessed distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified;
extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics;
determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library;
and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
The invention also proposes a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, realizes the following steps.
The invention has the beneficial effects that: the image of the power distribution network equipment is preprocessed, and the area where the power distribution network equipment is located in the image of the power distribution network equipment is obtained and serves as a picture to be identified, so that the image identification range is reduced, and the identification efficiency is improved; by establishing the power distribution network equipment image recognition training library, classifying the sample pictures according to the equipment subtypes, and subsequently recognizing the pictures to be recognized and the sample pictures corresponding to the same equipment subtypes, the recognition efficiency can be further improved, and the recognition accuracy can be improved.
Drawings
Fig. 1 is a flowchart of an image recognition method for power distribution network equipment according to a first embodiment of the present invention.
Detailed Description
In order to explain technical contents, objects and effects of the present invention in detail, the following detailed description is given with reference to the accompanying drawings in conjunction with the embodiments.
Referring to fig. 1, an image recognition method for power distribution network equipment includes:
establishing a power distribution network equipment image recognition training library, wherein sample pictures corresponding to the subtype of each equipment are stored in the power distribution network equipment image recognition training library, and the corresponding equipment types are marked on the sample pictures;
acquiring a power distribution network equipment image, and preprocessing the power distribution network equipment image;
obtaining the area of the distribution network equipment in the preprocessed distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified;
extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics;
determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library;
and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
From the above description, the beneficial effects of the present invention are: the identification efficiency of the images of the power distribution network equipment can be improved.
Further, the establishing of the power distribution network equipment image recognition training library specifically includes:
acquiring line information of a line to be acquired from a power grid GIS platform, wherein the line information comprises an equipment subtype of power distribution network equipment in the line to be acquired;
creating a data set according to the device subtype;
obtaining a sample picture, and labeling the sample picture according to the equipment type of the power distribution network equipment in the sample picture;
and storing the sample picture into a corresponding data set according to the equipment subtype of the power distribution network equipment in the sample picture.
According to the description, data organization and management are carried out on the image recognition training library of the power distribution network equipment based on the line information of the power grid GIS platform, and the retrieval performance and the management level of the image recognition training library of the power distribution network equipment are improved by taking all sample pictures of the equipment subtype as a verification model.
Further, the acquiring the power distribution network equipment image and the preprocessing the power distribution network equipment image specifically include:
acquiring an orthographic picture of power distribution network equipment on a line to be acquired to obtain an image of the power distribution network equipment;
and carrying out graying processing, denoising processing, sharpening processing and image segmentation processing on the power distribution network equipment image.
According to the above description, the efficiency and accuracy of the subsequent image recognition are improved by preprocessing the image.
Further, the image recognition of the picture to be recognized and each sample picture corresponding to the device subtype is performed to obtain a recognition result of the picture to be recognized, specifically:
obtaining the matching reliability of the picture to be identified and each sample picture corresponding to the equipment subtype by a neural network matching reliability judging method;
and if the matching reliability of the picture to be recognized and the sample picture exceeds a preset threshold value, taking the equipment type marked by the sample picture as the recognition result of the picture to be recognized.
According to the description, the nonlinear dimension reduction image identification technology is adopted, so that the high-dimensional image identification problem is converted into the characteristic expression vector identification problem, the calculation complexity is reduced, and the high efficiency of the power distribution network equipment identification technology is improved.
Further, after the image recognition is performed on the picture to be recognized and each sample picture corresponding to the device subtype, the method further includes:
and if no identification result exists, storing the picture to be identified as a sample picture in the power distribution network equipment image identification training library.
According to the description, the picture to be recognized, which cannot be recognized by the corresponding equipment, is taken as the sample picture and is brought into the image recognition training library of the power distribution network equipment, so that the image recognition training library of the power distribution network equipment can be continuously improved, and the reliability of the image recognition result is improved.
The invention also proposes a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, realizes the following steps.
Example one
Referring to fig. 1, a first embodiment of the present invention is: a power distribution network equipment image identification method is based on a power grid GIS platform and can be applied to a power distribution network equipment image identification application terminal based on the power grid GIS platform.
As shown in fig. 1, the method comprises the following steps:
s1: and establishing an image recognition training library of the power distribution network equipment.
Specifically, the name of a distribution network line to be collected is determined, and then line information of the line to be collected is obtained from a power grid GIS platform according to the line name, wherein the line information comprises the name of a local city, an operation and maintenance unit, the equipment subtype of distribution network equipment in the line to be collected and the like.
Then, according to the type of the equipment, a training model is established, and a model label is set, wherein the model label comprises information such as a model ID and a model name; depending on the device subtype, a data set is created for which a data set ID and a data set name may be defined.
And then uploading a sample picture, wherein the sample picture can be collected or acquired according to the device subtype, and is processed (a series of image technology processing including graying, image denoising, image sharpening, image segmentation, difference and image accumulation and the like), selecting a model label, and labeling the sample picture (namely, a sample of which device), namely labeling the sample picture according to the device type of the power distribution network device in the sample picture.
And finally, storing the sample picture into a corresponding data set according to the equipment subtype of the power distribution network equipment in the sample picture.
S2: acquiring an image of the power distribution network equipment, and preprocessing the image of the power distribution network equipment.
Specifically, acquiring an orthographic high-definition picture of power distribution network equipment on a line to be acquired to obtain an image of the power distribution network equipment; and then carrying out graying processing, denoising processing, sharpening processing and image segmentation processing on the power distribution network equipment image.
The method is characterized in that the potential of bilateral filtering is fully excavated on the basis of analyzing the internal characteristics of the Bilateral Filtering (BF), and algorithm improvement is carried out by combining wavelet analysis, Principal Component Analysis (PCA) technology, graph theory and the like, so that the method is suitable for removing noise of different types of power distribution network equipment images and noise of different types of noise and sharpening images of multimode power distribution network equipment images; the PCNN algorithm (coupled neural network algorithm) is combined with the intra-class absolute difference method, and the distribution network equipment image is segmented by improving the parameter selection and threshold optimization modes.
S3: and obtaining the area where the power distribution network equipment is located in the preprocessed power distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain the picture to be identified.
Specifically, performing wavelet decomposition on a color, brightness and direction characteristic diagram of an image to construct a wavelet multi-scale pyramid of each characteristic; then, processing the feature pyramid of the image by using a Center-periphery (Center-periphery) operator to obtain a focus Map (convergence Map) with 3 features, respectively performing local iterative processing on the focus maps with different scales to further enhance a significant region and inhibit an insignificant region, and segmenting the power distribution network equipment image by applying a winner total network and a region growing method to obtain an interested region. And then, carrying out difference and accumulated image processing on the region based on a method of accumulating the sequence difference images, and outputting a result picture to be subjected to power distribution network equipment identification.
S4: and extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics.
Specifically, color features of the power distribution network equipment are extracted by utilizing a color feature extraction technology comprising a color histogram, a color moment, a dominant hue method and the like; extracting the texture features of the power distribution network equipment by using texture feature extraction technologies including a statistical method, a structural method, a model method, a frequency spectrum method and the like; and extracting the shape characteristics of the power distribution network equipment by utilizing a shape characteristic extraction technology, including methods based on boundary characteristics, region characteristics and the like.
S5: and determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library.
Specifically, the sub-type of the distribution network equipment is obtained according to the color feature, the texture feature and the shape feature of the distribution network equipment, and then the corresponding verification model, namely the corresponding data set is obtained from the distribution network equipment image recognition training library according to the sub-type of the distribution network equipment, and the sample picture corresponding to the sub-type of the equipment is stored in the data set.
S6: and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
Specifically, image recognition is carried out by adopting a nonlinear dimension reduction image recognition technology, and a matching reliability judging method based on a neural network is adopted, namely, a matching experiment result of a reference image and a plurality of real-time images is used as sample data for training, then, a BP network is used for training, and the trained network is used for judging matching reliability to obtain the matching reliability.
And if the matching reliability of the picture to be recognized and the sample picture exceeds a preset threshold (such as 80%), judging that the corresponding equipment is recognized, and outputting an image recognition result (output in an excel table format) of the power distribution network equipment, namely, taking the equipment type marked by the sample picture as the recognition result of the picture to be recognized. Further, after the device type is identified, the relevant information of the power distribution network device can be acquired through the association relationship between the power grid GIS platform and the PMS.
Further, if the corresponding equipment is not identified, namely the matching reliability of the picture to be identified and each sample picture corresponding to the equipment subtype does not exceed a preset threshold value, the picture to be identified is taken as the sample picture through manual identification and classification, and the sample picture is marked and then stored into a corresponding data set in an image identification training library of the power distribution network equipment.
The image data of the densely distributed power distribution equipment is converted into a data set in a high-dimensional space through high dimensionality reduction, a low-dimensional expression vector is sought and used as a feature expression vector of the image data, so that the high-dimensional image recognition problem is converted into the recognition problem of the feature expression vector, the complexity of calculation is greatly reduced, the recognition error caused by redundant information is reduced, and the high efficiency of the power distribution network equipment recognition technology is improved.
In the embodiment, data organization and management are performed on the power distribution network equipment image recognition training library based on the power grid GIS platform line information, so that the result data recognized by the images can be well matched with the power grid GIS platform data, and the application of the associated PMS accounts is supported (through the association relationship between the power grid GIS platform and the PMS). The automation level of the whole process is improved, and the maintenance cost of data operation is reduced.
Collecting or acquiring an orthographic high-definition picture of the power distribution network equipment according to the equipment subtype, processing the orthographic high-definition picture through a series of image technologies such as graying, image denoising, image sharpening, image segmentation, difference and accumulated image and the like to form a sample picture, and bringing the sample picture into an image recognition training library of the power distribution network equipment. In the implementation of business application, high-definition pictures can be orthographically acquired from a specific acquisition line to a field acquisition power distribution network device, after image processing, the orthographically acquired high-definition pictures are compared with all sample pictures of the device subtype acquired from a power distribution network device image identification training library through the identified device subtype to be used as a verification model, specific devices are identified, relevant information output results are acquired, and if the corresponding devices cannot be identified, the sample pictures are used as the sample pictures and are included in the power distribution network device image identification training library, so that the power distribution network device image identification training library can be continuously improved, and the reliability of the image identification results is improved.
By carrying out nonlinear dimensionality reduction on the image, seeking a low-dimensional expression vector of the image, and using the low-dimensional expression vector as a feature expression vector of image data, the identification error caused by redundant information is reduced, and the identification rate of the power distribution network equipment is improved。
Example two
The present embodiment is a computer-readable storage medium corresponding to the above-mentioned embodiments, on which a computer program is stored, the program, when executed by a processor, implementing the steps of:
establishing a power distribution network equipment image recognition training library, wherein sample pictures corresponding to the subtype of each equipment are stored in the power distribution network equipment image recognition training library, and the corresponding equipment types are marked on the sample pictures;
acquiring a power distribution network equipment image, and preprocessing the power distribution network equipment image;
obtaining the area of the distribution network equipment in the preprocessed distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified;
extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics;
determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library;
and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
Further, the establishing of the power distribution network equipment image recognition training library specifically includes:
acquiring line information of a line to be acquired from a power grid GIS platform, wherein the line information comprises an equipment subtype of power distribution network equipment in the line to be acquired;
creating a data set according to the device subtype;
obtaining a sample picture, and labeling the sample picture according to the equipment type of the power distribution network equipment in the sample picture;
and storing the sample picture into a corresponding data set according to the equipment subtype of the power distribution network equipment in the sample picture.
Further, the acquiring the power distribution network equipment image and the preprocessing the power distribution network equipment image specifically include:
acquiring an orthographic picture of power distribution network equipment on a line to be acquired to obtain an image of the power distribution network equipment;
and carrying out graying processing, denoising processing, sharpening processing and image segmentation processing on the power distribution network equipment image.
Further, the image recognition of the picture to be recognized and each sample picture corresponding to the device subtype is performed to obtain a recognition result of the picture to be recognized, specifically:
obtaining the matching reliability of the picture to be identified and each sample picture corresponding to the equipment subtype by a neural network matching reliability judging method;
and if the matching reliability of the picture to be recognized and the sample picture exceeds a preset threshold value, taking the equipment type marked by the sample picture as the recognition result of the picture to be recognized.
Further, after the image recognition is performed on the picture to be recognized and each sample picture corresponding to the device subtype, the method further includes:
and if no identification result exists, storing the picture to be identified as a sample picture in the power distribution network equipment image identification training library.
In summary, according to the power distribution network device image identification method and the computer readable storage medium provided by the invention, the power distribution network device image is preprocessed, and the region where the power distribution network device is located in the power distribution network device image is obtained as the picture to be identified, so that the image identification range is reduced, and the identification efficiency is improved; by establishing the power distribution network equipment image recognition training library, classifying the sample pictures according to the equipment subtypes, and subsequently recognizing the pictures to be recognized and the sample pictures corresponding to the same equipment subtypes, the recognition efficiency can be further improved, and the recognition accuracy can be improved. Data organization and management are carried out on the power distribution network equipment image recognition training library based on power grid GIS platform line information, so that result data recognized by images can be well matched with power grid GIS platform data, relevant PMS account application is supported, the automation level of the whole process is improved, and the maintenance cost of data operation is reduced.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (10)
1. An image identification method for power distribution network equipment is characterized by comprising the following steps:
establishing a power distribution network equipment image recognition training library, wherein sample pictures corresponding to the subtype of each equipment are stored in the power distribution network equipment image recognition training library, and the corresponding equipment types are marked on the sample pictures;
acquiring a power distribution network equipment image, and preprocessing the power distribution network equipment image;
obtaining the area of the distribution network equipment in the preprocessed distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified;
extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics;
determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library;
and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
2. The power distribution network equipment image recognition method according to claim 1, wherein the establishing of the power distribution network equipment image recognition training library specifically comprises:
acquiring line information of a line to be acquired from a power grid GIS platform, wherein the line information comprises an equipment subtype of power distribution network equipment in the line to be acquired;
creating a data set according to the device subtype;
obtaining a sample picture, and labeling the sample picture according to the equipment type of the power distribution network equipment in the sample picture;
and storing the sample picture into a corresponding data set according to the equipment subtype of the power distribution network equipment in the sample picture.
3. The power distribution network equipment image recognition method according to claim 1, wherein the acquiring of the power distribution network equipment image and the preprocessing of the power distribution network equipment image specifically comprise:
acquiring an orthographic picture of power distribution network equipment on a line to be acquired to obtain an image of the power distribution network equipment;
and carrying out graying processing, denoising processing, sharpening processing and image segmentation processing on the power distribution network equipment image.
4. The power distribution network equipment image identification method according to claim 1, wherein the image identification of the picture to be identified and each sample picture corresponding to the equipment subtype is performed, and the identification result of the picture to be identified is specifically obtained as follows:
obtaining the matching reliability of the picture to be identified and each sample picture corresponding to the equipment subtype by a neural network matching reliability judging method;
and if the matching reliability of the picture to be recognized and the sample picture exceeds a preset threshold value, taking the equipment type marked by the sample picture as the recognition result of the picture to be recognized.
5. The power distribution network equipment image recognition method according to claim 1, wherein after the image recognition is performed on the picture to be recognized and each sample picture corresponding to the equipment subtype, the method further comprises the following steps:
and if no identification result exists, storing the picture to be identified as a sample picture in the power distribution network equipment image identification training library.
6. A computer-readable storage medium, on which a computer program is stored, said program, when executed by a processor, performing the steps of:
establishing a power distribution network equipment image recognition training library, wherein sample pictures corresponding to the subtype of each equipment are stored in the power distribution network equipment image recognition training library, and the corresponding equipment types are marked on the sample pictures;
acquiring a power distribution network equipment image, and preprocessing the power distribution network equipment image;
obtaining the area of the distribution network equipment in the preprocessed distribution network equipment image, and carrying out difference and accumulated image processing on the area by an accumulated sequence difference image method to obtain a picture to be identified;
extracting the characteristics of the power distribution network equipment according to the power distribution network equipment image, wherein the characteristics comprise color characteristics, texture characteristics and shape characteristics;
determining the equipment subtype of the power distribution network equipment according to the characteristics, and acquiring each sample picture corresponding to the equipment subtype of the power distribution network equipment from a power distribution network equipment image recognition training library;
and carrying out image recognition on the picture to be recognized and each sample picture corresponding to the equipment subtype to obtain a recognition result of the picture to be recognized.
7. The computer-readable storage medium according to claim 6, wherein the establishing of the power distribution network equipment image recognition training library specifically includes:
acquiring line information of a line to be acquired from a power grid GIS platform, wherein the line information comprises an equipment subtype of power distribution network equipment in the line to be acquired;
creating a data set according to the device subtype;
obtaining a sample picture, and labeling the sample picture according to the equipment type of the power distribution network equipment in the sample picture;
and storing the sample picture into a corresponding data set according to the equipment subtype of the power distribution network equipment in the sample picture.
8. The computer-readable storage medium according to claim 6, wherein the acquiring the power distribution network device image and the preprocessing the power distribution network device image specifically include:
acquiring an orthographic picture of power distribution network equipment on a line to be acquired to obtain an image of the power distribution network equipment;
and carrying out graying processing, denoising processing, sharpening processing and image segmentation processing on the power distribution network equipment image.
9. The computer-readable storage medium according to claim 6, wherein the performing image recognition on the to-be-recognized picture and each sample picture corresponding to the device sub-type obtains a recognition result of the to-be-recognized picture specifically includes:
obtaining the matching reliability of the picture to be identified and each sample picture corresponding to the equipment subtype by a neural network matching reliability judging method;
and if the matching reliability of the picture to be recognized and the sample picture exceeds a preset threshold value, taking the equipment type marked by the sample picture as the recognition result of the picture to be recognized.
10. The computer-readable storage medium according to claim 6, wherein after the image recognition of each sample picture corresponding to the to-be-recognized picture and the device subtype, the method further comprises:
and if no identification result exists, storing the picture to be identified as a sample picture in the power distribution network equipment image identification training library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110598038.3A CN113392730A (en) | 2021-05-31 | 2021-05-31 | Power distribution network equipment image identification method and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110598038.3A CN113392730A (en) | 2021-05-31 | 2021-05-31 | Power distribution network equipment image identification method and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113392730A true CN113392730A (en) | 2021-09-14 |
Family
ID=77619529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110598038.3A Pending CN113392730A (en) | 2021-05-31 | 2021-05-31 | Power distribution network equipment image identification method and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113392730A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104281843A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Image recognition method and system based on self-adaptive features and classification model selection |
| CN105678344A (en) * | 2016-02-29 | 2016-06-15 | 浙江群力电气有限公司 | Intelligent classification method for power instrument equipment |
| US20170011257A1 (en) * | 2015-07-08 | 2017-01-12 | University Of Central Florida Research Foundation, Inc. | Supervised facial recognition system and method |
| CN109389180A (en) * | 2018-10-30 | 2019-02-26 | 国网四川省电力公司广元供电公司 | A power equipment image-recognizing method and inspection robot based on deep learning |
| CN109685075A (en) * | 2018-11-27 | 2019-04-26 | 山东鲁能软件技术有限公司 | A kind of power equipment recognition methods based on image, apparatus and system |
| CN110110726A (en) * | 2019-05-15 | 2019-08-09 | 深圳供电局有限公司 | The recognition methods of power equipment nameplate, device, computer equipment and storage medium |
| CN110674684A (en) * | 2019-08-15 | 2020-01-10 | 深圳壹账通智能科技有限公司 | Micro-expression classification model generation method, micro-expression classification model generation device, micro-expression classification model image recognition method, micro-expression classification model image recognition device, micro-expression classification model image recognition equipment and micro-expression classification model image recognition medium |
| CN111079792A (en) * | 2019-11-19 | 2020-04-28 | 全球能源互联网研究院有限公司 | Power equipment identification method and device |
| CN111563495A (en) * | 2020-05-09 | 2020-08-21 | 北京奇艺世纪科技有限公司 | Method and device for recognizing characters in image and electronic equipment |
| CN111860641A (en) * | 2020-07-17 | 2020-10-30 | 东莞理工学院 | Power grid image identification method, electronic device and storage medium |
| CN112070134A (en) * | 2020-08-28 | 2020-12-11 | 广东电网有限责任公司 | Power equipment image classification method and device, power equipment and storage medium |
-
2021
- 2021-05-31 CN CN202110598038.3A patent/CN113392730A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104281843A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Image recognition method and system based on self-adaptive features and classification model selection |
| US20170011257A1 (en) * | 2015-07-08 | 2017-01-12 | University Of Central Florida Research Foundation, Inc. | Supervised facial recognition system and method |
| CN105678344A (en) * | 2016-02-29 | 2016-06-15 | 浙江群力电气有限公司 | Intelligent classification method for power instrument equipment |
| CN109389180A (en) * | 2018-10-30 | 2019-02-26 | 国网四川省电力公司广元供电公司 | A power equipment image-recognizing method and inspection robot based on deep learning |
| CN109685075A (en) * | 2018-11-27 | 2019-04-26 | 山东鲁能软件技术有限公司 | A kind of power equipment recognition methods based on image, apparatus and system |
| CN110110726A (en) * | 2019-05-15 | 2019-08-09 | 深圳供电局有限公司 | The recognition methods of power equipment nameplate, device, computer equipment and storage medium |
| CN110674684A (en) * | 2019-08-15 | 2020-01-10 | 深圳壹账通智能科技有限公司 | Micro-expression classification model generation method, micro-expression classification model generation device, micro-expression classification model image recognition method, micro-expression classification model image recognition device, micro-expression classification model image recognition equipment and micro-expression classification model image recognition medium |
| CN111079792A (en) * | 2019-11-19 | 2020-04-28 | 全球能源互联网研究院有限公司 | Power equipment identification method and device |
| CN111563495A (en) * | 2020-05-09 | 2020-08-21 | 北京奇艺世纪科技有限公司 | Method and device for recognizing characters in image and electronic equipment |
| CN111860641A (en) * | 2020-07-17 | 2020-10-30 | 东莞理工学院 | Power grid image identification method, electronic device and storage medium |
| CN112070134A (en) * | 2020-08-28 | 2020-12-11 | 广东电网有限责任公司 | Power equipment image classification method and device, power equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108805076B (en) | Method and system for extracting table characters of environmental impact evaluation report | |
| Khan et al. | An efficient contour based fine-grained algorithm for multi category object detection | |
| CN106203539B (en) | Method and device for identifying container number | |
| CN110807775A (en) | Traditional Chinese medicine tongue image segmentation device and method based on artificial intelligence and storage medium | |
| CN112257711B (en) | Method for detecting damage fault of railway wagon floor | |
| CN110852327A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN114863189B (en) | Intelligent image identification method based on big data | |
| CN114170418A (en) | Automobile wire harness connector multi-feature fusion image retrieval method by searching images through images | |
| CN111079543B (en) | Efficient vehicle color identification method based on deep learning | |
| CN104463091B (en) | A kind of facial image recognition method based on image LGBP feature subvectors | |
| CN110569848A (en) | feature extraction method and system for power equipment nameplate | |
| CN111401434B (en) | Image classification method based on unsupervised feature learning | |
| CN111127407B (en) | Fourier transform-based style migration forged image detection device and method | |
| CN110555406B (en) | Video moving target identification method based on Haar-like characteristics and CNN matching | |
| Özyurt et al. | A new method for classification of images using convolutional neural network based on Dwt-Svd perceptual hash function | |
| CN113392730A (en) | Power distribution network equipment image identification method and computer readable storage medium | |
| CN112766082B (en) | Chinese text handwriting identification method and device based on macro-micro characteristics and storage medium | |
| CN110084125B (en) | Agricultural insurance investigation technical method based on deep learning | |
| Rani et al. | Object Detection in Natural Scene Images Using Thresholding Techniques | |
| CN113628113A (en) | Image splicing method and related equipment thereof | |
| CN111696070A (en) | Multispectral image fusion power internet of things fault point detection method based on deep learning | |
| CN112115949B (en) | Optical character recognition method for tobacco certificate and order | |
| CN117422958B (en) | Financial data verification method and system based on deep learning | |
| CN117037019A (en) | Base station inspection video processing method and device, electronic equipment and storage medium | |
| Antony et al. | An Examination of Image Mining: Approaches, Techniques, and Application |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210914 |
|
| RJ01 | Rejection of invention patent application after publication |