CN113627299A - Intelligent wire floater identification method and device based on deep learning - Google Patents
Intelligent wire floater identification method and device based on deep learning Download PDFInfo
- Publication number
- CN113627299A CN113627299A CN202110879169.9A CN202110879169A CN113627299A CN 113627299 A CN113627299 A CN 113627299A CN 202110879169 A CN202110879169 A CN 202110879169A CN 113627299 A CN113627299 A CN 113627299A
- Authority
- CN
- China
- Prior art keywords
- target
- floater
- wire
- data
- conductor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000013135 deep learning Methods 0.000 title claims abstract description 25
- 239000004020 conductor Substances 0.000 claims abstract description 60
- 238000012549 training Methods 0.000 claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 18
- 238000003708 edge detection Methods 0.000 claims abstract description 17
- 230000006798 recombination Effects 0.000 claims abstract description 11
- 238000005215 recombination Methods 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 230000004927 fusion Effects 0.000 claims description 4
- 230000000149 penetrating effect Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 16
- 208000034699 Vitreous floaters Diseases 0.000 description 31
- 238000010586 diagram Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 239000000725 suspension Substances 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001629 suppression Effects 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a device for intelligently identifying a wire floater based on deep learning, wherein the method comprises the following steps: performing data enhancement on the acquired satellite image data by adopting an image recombination mode to acquire a training set; processing the training set by adopting a Canny edge detection operator to obtain a target lead; fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage, and obtaining a target wire flotage; and if the target conductor is successfully compared with the target conductor floater, acquiring a final target conductor floater. According to the invention, data enhancement is carried out through image recombination, and a target conductor floater obtained by training in combination with an improved SSD model is compared with a conductor benchmark, so that the identification accuracy and the detection efficiency are improved.
Description
Technical Field
The invention relates to the technical field of target identification, in particular to a method and a device for intelligently identifying a wire floater based on deep learning.
Background
Modern society relies heavily on electric power service, and only reliable electric power service can maintain normal electricity demand in production and life, and in order to provide reliable electric power service for cities and rural areas, the maintenance of important power grid components such as power lines, power transmission towers and related accessories is very important, however, the facilities are exposed to outdoor environment and are easily damaged, and when floating garbage such as kites, balloons, plastic bags, advertising cloths and the like is hung on power transmission leads, the facilities are damaged, and huge safety accidents and economic losses are caused. Therefore, the power line system is required to be monitored in real time so that maintenance departments can clean the power line system in time when dangerous conditions such as suspension of wire floaters occur.
In recent years, with the development of deep learning and the popularization and application of satellite remote sensing images in civil use, the detection of the state of a power transmission line by processing satellite images through the deep learning becomes possible, however, because the satellite remote sensing images are large in resolution and complex in background, a sufficient number of sample images with wire suspenders are lacked to train a deep learning network, sufficient positive and negative samples are the key for ensuring the accuracy of a target detection algorithm, the lack of samples can cause the performance of the trained network to be insufficient, the wire suspenders are relatively small in area in the satellite remote sensing images, the small target is represented, and the existing network model is low in detection accuracy of the small target. The existing research mostly focuses on identifying the wire suspension object based on the aerial image of the unmanned aerial vehicle, for example, the image segmentation algorithm research based on the aerial image of the suspended foreign matter on the power transmission line and the aerial image of the unmanned aerial vehicle provide a method for segmenting the wire suspension object and the image background and a method for detecting the foreign matter on the power transmission line based on YOLOv 4. The target extracted by the method is only a target similar to the wire floater, the position of the target is uncertain relative to the wire, and whether the target is the wire floater cannot be distinguished.
Disclosure of Invention
The invention aims to provide a wire floater intelligent identification method based on deep learning, and aims to solve the problems that a few-sample training set is insufficient and whether a final floater is obtained cannot be distinguished.
In order to achieve the above object, the present invention provides a method for intelligently identifying a wire float based on deep learning, comprising:
performing data enhancement on the acquired satellite image data by adopting an image recombination mode to acquire a training set;
processing the training set by adopting a Canny edge detection operator to obtain a target lead;
fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage, and obtaining a target wire flotage;
and if the target conductor is successfully compared with the target conductor floater, acquiring a final target conductor floater.
Preferably, the obtaining of the final target conductor float specifically includes:
and if the position of the target conductor is overlapped with the position of the target conductor floater, determining that the target conductor floater is the final target conductor floater if the position of the target conductor is overlapped with the position of the target conductor floater, otherwise, determining that the training set is input into the improved SSD model for inaccurate conductor floater identification.
Preferably, the data enhancement of the acquired satellite image data by using an image reconstruction method specifically includes:
preprocessing the satellite image data to obtain preprocessed data, segmenting target data and background data in the preprocessed data, and fusing the segmented target data and preset background data.
Preferably, the satellite image data is preprocessed, specifically: and carrying out graying, median filtering and Gaussian filtering on the satellite image data.
Preferably, the segmenting the target data and the background data in the preprocessed data specifically includes:
and traversing the preprocessed data, selecting the target data by adopting a template matching frame, establishing a target frame of the preprocessed data, and segmenting the target frame by adopting a Grabcut algorithm.
Preferably, the segmented target data is fused with preset background data, specifically:
combining the pixel point P of the divided target datatThe pixel point P of the preset background databAnd corresponding weight coefficientsLine fusion, as follows:
wherein, PnewAnd representing pixel points of the training set, and alpha and beta represent weight coefficients.
Preferably, the processing the training set by using a Canny edge detection operator to obtain the target wire specifically includes:
after denoising the training set by adopting Gaussian filtering, calculating a gradient mode and a gradient direction and reserving a maximum gradient value in the gradient direction;
filling missing values of target edges broken due to noise interference in the training set by adopting a hysteresis threshold value, and obtaining a processing result of the Canny edge detection operator;
and extracting the conducting wire in the Canny edge detection operator processing result according to the conducting wire characteristics to obtain the target conducting wire, wherein the conducting wire characteristics comprise images penetrating through the training set and a plurality of parallel lines.
Preferably, the fusing deep features and shallow features in the preset SSD model according to the feature pyramid model specifically includes:
and acquiring semantic information of the deep features by adopting a bilinear interpolation method, and transmitting the semantic information to the shallow features containing position information for fusion.
The invention also provides a device for intelligently identifying the wire floater based on deep learning, which comprises:
the data processing module is used for performing data enhancement on the acquired satellite image data in an image recombination mode to acquire a training set;
the first extraction module is used for processing the training set by adopting a Canny edge detection operator to obtain a target lead;
the second extraction module is used for fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage and obtain a target wire flotage;
and the comparison module is used for obtaining the final target conductor floater if the target conductor is successfully compared with the target conductor floater.
Preferably, the alignment module is further configured to:
and if the position of the target conductor is overlapped with the position of the target conductor floater, determining that the target conductor floater is the final target conductor floater if the position of the target conductor is overlapped with the position of the target conductor floater, otherwise, determining that the training set is input into the improved SSD model for inaccurate conductor floater identification.
Compared with the prior art, the invention has the beneficial effects that:
the invention realizes the recombination of the target and the background according to the image recombination mode, thereby solving the problem of insufficient training set in the prior art, simultaneously providing a basis for a better training model, effectively fusing a deep characteristic diagram containing rich semantic information and a shallow characteristic diagram containing rich position information after up-sampling, increasing the shallow characteristic semantic information to improve the detection precision when detecting a small target, obtaining an improved SSD model, comparing an extracted lead with a target lead floater of the improved SSD model, further confirming that the output result of the improved SSD model is a final target lead floater, and improving the detection precision and the detection efficiency.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for intelligently identifying a wire float based on deep learning according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a Canny edge detector process according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of an improved SSD model according to yet another embodiment of the present invention;
FIG. 4 is a schematic diagram of a bilinear interpolation algorithm provided in accordance with an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a device for intelligently identifying a wire float based on deep learning according to an embodiment of the present 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.
Referring to fig. 1, an embodiment of the present invention provides a method for intelligently identifying a wire float based on deep learning, including the following steps:
s101: and performing data enhancement on the acquired satellite image data by adopting an image recombination mode to acquire a training set.
Specifically, the satellite image data is preprocessed, segmented and fused, and the method specifically comprises the following steps:
preprocessing operations such as graying, median filtering and advanced filtering are carried out on input satellite image data, then the preprocessed data are traversed, template matching is adopted to frame the target data, a target frame of the preprocessed data is established, Grabcut algorithm is adopted to segment the target frame, and pixel points P of the segmented target data are combinedtPixel point P of preset background databAnd corresponding weight coefficients are fused as follows:
wherein, PnewAnd representing pixel points of the training set, and alpha and beta represent weight coefficients. And randomly rotating, reversing and scaling the fused new image sample.
S102: processing the training set with a Canny edge detection operator to obtain a target wire.
Referring to fig. 2, based on the above processing procedure, a Canny edge detection operator is used to extract a wire, which is used as a comparison reference for the following model output results, and the specific wire extraction procedure is as follows:
after denoising a training set by adopting Gaussian filtering, calculating the difference in the horizontal and vertical directions by adopting an edge difference operator Sobel to obtain a gradient mode and a gradient direction, inhibiting and retaining the maximum gradient value in the gradient direction by adopting a non-maximum value, deleting other pixels, filling missing values of target edges broken due to noise interference in the training set by adopting a hysteresis threshold value, obtaining a Canny edge detection operator processing result, extracting wires in the Canny edge detection operator processing result according to the characteristics of the wires, and obtaining a target wire, wherein the characteristics of the wires comprise approximate straight lines, images penetrating through the training set and a plurality of parallel lines.
S103: and fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage, and obtaining a target wire flotage.
Specifically, when a preset SSD model detects a target, firstly, a prior frame with different scales and aspect ratios is adopted to predict a boundary frame of the target, then VGG is adopted as a feature extraction network, an input image enters the network, feature graphs of 6 different layers are respectively adopted to predict the category and the coordinate of the target, and finally, a non-maximum suppression (NMS) method is adopted to carry out final detection. The preset SSD has the main idea that a large target is detected based on a deep low-resolution feature map, and a small target is detected by utilizing a shallow high-resolution feature map. However, the shallow feature map does not contain rich semantic information, so that the model has the problem of low detection accuracy when detecting a small target.
The sizes of objects in the images are various, objects in the data set cannot cover all scales, so that the image pyramid (downsampling with different resolutions) is used for assisting CNN learning, but the speed is too slow, so that the objects are usually predicted by using a single scale, an intermediate result is also predicted, a layer of transposed convolution is added behind a plurality of layers of residual modules to improve the resolution and obtain a segmentation result, or a classification result is obtained through convolution of 1x1 or GlobalPool, and the framework is largely used when auxiliary information and an auxiliary loss function exist.
The author of the feature pyramid model improves the above method by a very ingenious method, and adds top-down connection in addition to lateral connection, and fuses the top-down result and the lateral result together by adding method, the key point here is that the low-level feature map is not rich enough in semantic meaning and can not be directly used for classification, but the deep features are more reliable, and the combination of lateral connection and top-down connection can obtain feature maps with different resolutions, and the feature maps contain the semantic information of the original deepest feature map.
By using a characteristic pyramid model for reference, the small target detection is superior to a preset SSD target detection algorithm, the characteristic pyramid model is mainly used for solving the problem that pictures in object detection have different scales, the improved SSD model is constructed by combining the characteristic pyramid model and the preset SSD model, semantic information of deep features is obtained by a bilinear interpolation method, the semantic information is transmitted to shallow features containing position information to be fused, the deep features are sampled, and the deep feature information is transmitted to the shallow features, so that the shallow feature map contains more semantic features, and the small target can be accurately captured.
Referring to fig. 3, Block11 corresponding to Conv11_2 in the SSD model is fused with Conv10_2 by bilinear interpolation upsampling to become Block10, and Block9, Block8, Block7, and Block4 are sequentially obtained. By the method, the semantic information of the deep features and the high resolution of the shallow features are fused, and the small target detection precision can be improved by predicting the fused features.
Referring to fig. 4, when upsampling, an unknown point in an image is predicted by using a known point in the image, so as to enlarge the image size, the upsampling is implemented by using a two-line interpolation method in the present invention, and assuming that a point Q in fig. 4 is a known point and a point R is an insertion point in the x-axis direction, the calculation is as follows:
wherein P is the unknown point, x1,x2,y,y1,y2Respectively, horizontal and vertical coordinate points, and f represents the pixel value of each point.
S104: and if the target conductor is successfully compared with the target conductor floater, acquiring a final target conductor floater.
Specifically, since the target conductor float solved in the above step is a similar target, and it cannot be determined by using the network model whether the obtained target conductor float is a target that really floats on the conductor, therefore, based on the processing method for satellite image data in step S101, the conductor is extracted by using the Canny edge detection operator, which is used as a reference for screening a final target in this step, if the position of the target conductor overlaps with the position of the target conductor float, it is determined that the target conductor float is a final target conductor float, otherwise, it is determined that the training set is input into the improved SSD model for conductor float identification is inaccurate, and the target float obtained by the model is not a final target, so that irrelevant suspicious targets are removed, and the accuracy of detection is improved.
The method comprises the steps of obtaining a target image only containing a small amount of background information by adopting template matching and Grabcut algorithm segmentation, then carrying out pixel point fusion on the obtained target image and the new background image to obtain a new image so as to expand a training set, solving the problems of few satellite remote sensing image samples and insufficient training set, using a characteristic pyramid model for reference, fusing a deep layer characteristic image after sampling with a shallow layer characteristic, increasing semantic information of the shallow layer characteristic so as to improve the detection precision of the model on small targets such as wire floaters and the like in the satellite remote sensing image, and removing irrelevant targets extracted by deep learning by comparing the positions of wires with the position of a suspicious target as the condition for screening the final wire floaters so as to improve the detection efficiency.
Referring to fig. 5, another embodiment of the present invention provides an intelligent wire floater identification device based on deep learning, including:
and the data processing module 11 is configured to perform data enhancement on the acquired satellite image data in an image recombination manner to acquire a training set.
A first extraction module 12, configured to process the training set with a Canny edge detection operator to obtain a target wire.
And a second extraction module 13, configured to fuse deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, and input the training set into the improved SSD model to perform wire floater identification, so as to obtain a target wire floater.
A comparison module 14, configured to obtain a final target conductor float if the target conductor is successfully compared with the target conductor float.
For specific limitations of the intelligent identification device for wire floats based on deep learning, reference may be made to the above limitations, which are not described herein again. The modules in the intelligent identifying device for the wire floating objects based on deep learning can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. A wire floater intelligent identification method based on deep learning is characterized by comprising the following steps:
performing data enhancement on the acquired satellite image data by adopting an image recombination mode to acquire a training set;
processing the training set by adopting a Canny edge detection operator to obtain a target lead;
fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage, and obtaining a target wire flotage;
and if the target conductor is successfully compared with the target conductor floater, acquiring a final target conductor floater.
2. The intelligent wire float identification method based on deep learning according to claim 1, wherein the obtaining of the final target wire float specifically comprises:
and if the position of the target conductor is overlapped with the position of the target conductor floater, determining that the target conductor floater is the final target conductor floater, otherwise, determining that the training set is input into the improved SSD model for identifying the conductor floater is inaccurate.
3. The intelligent wire floater identification method based on deep learning as claimed in claim 2, wherein the obtained satellite image data is subjected to data enhancement by adopting an image recombination mode, and specifically comprises:
preprocessing the satellite image data to obtain preprocessed data, segmenting target data and background data in the preprocessed data, and fusing the segmented target data and preset background data.
4. The intelligent wire floater identification method based on deep learning as claimed in claim 3, wherein the satellite image data is preprocessed, specifically: and carrying out graying, median filtering and Gaussian filtering on the satellite image data.
5. The intelligent wire floater identification method based on deep learning as claimed in claim 4, wherein the target data and the background data in the preprocessed data are segmented, specifically:
and traversing the preprocessed data, selecting the target data by adopting a template matching frame, establishing a target frame of the preprocessed data, and segmenting the target frame by adopting a Grabcut algorithm.
6. The intelligent wire floater identification method based on deep learning as claimed in claim 5, wherein the segmented target data is fused with preset background data, specifically:
combining the pixel point P of the divided target datatThe pixel point P of the preset background databAnd corresponding weight coefficients are fused as follows:
wherein, PnewAnd representing pixel points of the training set, and alpha and beta represent weight coefficients.
7. The method according to claim 6, wherein the Canny edge detection operator is adopted to process the training set to obtain the target wire, specifically:
after denoising the training set by adopting Gaussian filtering, calculating a gradient mode and a gradient direction and reserving a maximum gradient value in the gradient direction;
filling missing values of target edges broken due to noise interference in the training set by adopting a hysteresis threshold value, and obtaining a processing result of the Canny edge detection operator;
and extracting the conducting wire in the Canny edge detection operator processing result according to the conducting wire characteristics to obtain the target conducting wire, wherein the conducting wire characteristics comprise images penetrating through the training set and a plurality of parallel lines.
8. The intelligent wire floater identification method based on deep learning as claimed in claim 6, wherein the deep layer features and the shallow layer features in the preset SSD model are fused according to the feature pyramid model, and specifically:
and acquiring semantic information of the deep features by adopting a bilinear interpolation method, and transmitting the semantic information to the shallow features containing position information for fusion.
9. The utility model provides a wire floater intelligent recognition device based on degree of depth study which characterized in that includes:
the data processing module is used for performing data enhancement on the acquired satellite image data in an image recombination mode to acquire a training set;
the first extraction module is used for processing the training set by adopting a Canny edge detection operator to obtain a target lead;
the second extraction module is used for fusing deep features and shallow features in a preset SSD model according to the feature pyramid model to obtain an improved SSD model, inputting the training set into the improved SSD model to identify the wire flotage and obtain a target wire flotage;
and the comparison module is used for obtaining the final target conductor floater if the target conductor is successfully compared with the target conductor floater.
10. The intelligent deep learning-based identification device for conductor floating objects according to claim 9, wherein the comparison module is further configured to:
and if the position of the target conductor is overlapped with the position of the target conductor floater, determining that the target conductor floater is the final target conductor floater if the position of the target conductor is overlapped with the position of the target conductor floater, otherwise, determining that the training set is input into the improved SSD model for inaccurate conductor floater identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110879169.9A CN113627299B (en) | 2021-07-30 | 2021-07-30 | Wire floater intelligent recognition method and device based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110879169.9A CN113627299B (en) | 2021-07-30 | 2021-07-30 | Wire floater intelligent recognition method and device based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113627299A true CN113627299A (en) | 2021-11-09 |
| CN113627299B CN113627299B (en) | 2024-04-09 |
Family
ID=78382122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110879169.9A Active CN113627299B (en) | 2021-07-30 | 2021-07-30 | Wire floater intelligent recognition method and device based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113627299B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114842006A (en) * | 2022-07-04 | 2022-08-02 | 南方电网科学研究院有限责任公司 | Digital twin station power line detection method and related device thereof |
| CN115432331A (en) * | 2022-10-10 | 2022-12-06 | 浙江绿达智能科技有限公司 | Intelligent classification dustbin |
| CN115861359A (en) * | 2022-12-16 | 2023-03-28 | 兰州交通大学 | Self-adaptive segmentation and extraction method for water surface floating garbage image |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105447530A (en) * | 2016-01-05 | 2016-03-30 | 国网四川省电力公司检修公司 | Power transmission line hidden risk and fault detection method based on image identification technology |
| CN108985170A (en) * | 2018-06-15 | 2018-12-11 | 山东信通电子股份有限公司 | Transmission line of electricity hanger recognition methods based on Three image difference and deep learning |
| CN111881937A (en) * | 2020-06-22 | 2020-11-03 | 深圳金三立视频科技股份有限公司 | Transmission line hardware target detection and defect identification method and terminal |
| CN111967335A (en) * | 2020-07-21 | 2020-11-20 | 广东工业大学 | Method for identifying foreign matters on power transmission line based on image processing |
| CN111985499A (en) * | 2020-07-23 | 2020-11-24 | 东南大学 | High-precision bridge apparent disease identification method based on computer vision |
| CN112613343A (en) * | 2020-12-01 | 2021-04-06 | 浙江大学 | Improved YOLOv 4-based river waste monitoring method |
| CN112800952A (en) * | 2021-01-27 | 2021-05-14 | 山东大学 | Marine organism identification method and system based on improved SSD algorithm |
-
2021
- 2021-07-30 CN CN202110879169.9A patent/CN113627299B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105447530A (en) * | 2016-01-05 | 2016-03-30 | 国网四川省电力公司检修公司 | Power transmission line hidden risk and fault detection method based on image identification technology |
| CN108985170A (en) * | 2018-06-15 | 2018-12-11 | 山东信通电子股份有限公司 | Transmission line of electricity hanger recognition methods based on Three image difference and deep learning |
| CN111881937A (en) * | 2020-06-22 | 2020-11-03 | 深圳金三立视频科技股份有限公司 | Transmission line hardware target detection and defect identification method and terminal |
| CN111967335A (en) * | 2020-07-21 | 2020-11-20 | 广东工业大学 | Method for identifying foreign matters on power transmission line based on image processing |
| CN111985499A (en) * | 2020-07-23 | 2020-11-24 | 东南大学 | High-precision bridge apparent disease identification method based on computer vision |
| CN112613343A (en) * | 2020-12-01 | 2021-04-06 | 浙江大学 | Improved YOLOv 4-based river waste monitoring method |
| CN112800952A (en) * | 2021-01-27 | 2021-05-14 | 山东大学 | Marine organism identification method and system based on improved SSD algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 金立军 等: "基于航拍图像的输电线路异物识别", 同济大学学报(自然科学版), vol. 41, no. 2, pages 277 - 281 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114842006A (en) * | 2022-07-04 | 2022-08-02 | 南方电网科学研究院有限责任公司 | Digital twin station power line detection method and related device thereof |
| CN115432331A (en) * | 2022-10-10 | 2022-12-06 | 浙江绿达智能科技有限公司 | Intelligent classification dustbin |
| CN115861359A (en) * | 2022-12-16 | 2023-03-28 | 兰州交通大学 | Self-adaptive segmentation and extraction method for water surface floating garbage image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113627299B (en) | 2024-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113627299B (en) | Wire floater intelligent recognition method and device based on deep learning | |
| CN110310264B (en) | DCNN-based large-scale target detection method and device | |
| CN109029381B (en) | Tunnel crack detection method and system and terminal equipment | |
| CN110598784B (en) | Machine learning-based construction waste classification method and device | |
| CN109631848B (en) | Transmission line foreign matter intrusion detection system and detection method | |
| CN110675415B (en) | Road ponding area detection method based on deep learning enhanced example segmentation | |
| CN112232391A (en) | Dam crack detection method based on U-net network and SC-SAM attention mechanism | |
| Zheng et al. | Vehicle detection based on morphology from highway aerial images | |
| CN112666553B (en) | Road ponding identification method and equipment based on millimeter wave radar | |
| CN111639530B (en) | Method and system for detecting and identifying power transmission tower and insulator of power transmission line | |
| CN111160065A (en) | Remote sensing image ship detection method, device, equipment and storage medium thereof | |
| CN114049356B (en) | Method, device and system for detecting structure apparent crack | |
| CN112733886A (en) | Sample image processing method, device, equipment and storage medium | |
| CN114187505A (en) | Detection method and device for falling-off of damper of power transmission line, medium and terminal equipment | |
| CN104598906B (en) | Vehicle outline detection method and its device | |
| CN115439319A (en) | Exposed detection method for electric slide wire protection device | |
| CN112116561B (en) | Power grid transmission line detection method and device based on image processing fusion network weight | |
| CN110472472B (en) | Airport detection method and device based on SAR remote sensing image | |
| Cetin et al. | Automated electric utility pole detection from aerial images | |
| CN113095224A (en) | Method and device for identifying floaters in unmanned aerial vehicle patrolling river channel | |
| CN111080562A (en) | Substation suspender identification method based on enhanced image contrast | |
| Shaw et al. | Automated image processing of historical maps | |
| CN117830882B (en) | Deep learning-based aerial image recognition method and related product | |
| CN117876362B (en) | Deep learning-based natural disaster damage assessment method and device | |
| CN116310880A (en) | Remote sensing image building extraction method based on impulse neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |