CN109871788A - A kind of transmission of electricity corridor natural calamity image recognition method - Google Patents
A kind of transmission of electricity corridor natural calamity image recognition method Download PDFInfo
- Publication number
- CN109871788A CN109871788A CN201910089314.6A CN201910089314A CN109871788A CN 109871788 A CN109871788 A CN 109871788A CN 201910089314 A CN201910089314 A CN 201910089314A CN 109871788 A CN109871788 A CN 109871788A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- disaster
- noise reduction
- aerial images
- 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 25
- 230000005540 biological transmission Effects 0.000 title claims abstract description 14
- 230000005611 electricity Effects 0.000 title claims abstract description 14
- 230000001537 neural effect Effects 0.000 claims abstract description 5
- 238000011946 reduction process Methods 0.000 claims abstract description 4
- 238000001914 filtration Methods 0.000 claims description 7
- 230000002146 bilateral effect Effects 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 abstract description 8
- 238000000605 extraction Methods 0.000 abstract description 5
- 238000003384 imaging method Methods 0.000 abstract description 3
- 238000011410 subtraction method Methods 0.000 abstract description 3
- 239000010410 layer Substances 0.000 description 13
- 238000012544 monitoring process Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000004913 activation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Abstract
This application provides a kind of transmission of electricity corridor natural calamity image recognition method, this method includes obtaining unmanned plane image；Noise reduction process is carried out to the aerial images；The disaster characteristic in aerial images using preset algorithm, after extracting the noise reduction；According to the disaster characteristic, the first Disasters Type is determined；The aerial images after the noise reduction are identified using neural convolutional network, determine the second Disasters Type；First Disasters Type and the second Disasters Type are compared, recognition result is obtained.Feature extraction is carried out to foreground image using background subtraction method and inter-frame difference, convolutional neural networks can efficiently extract category feature automatically, doubtful disaster image is accurately identified, geological disaster intelligent imaging in electric system corridor is identified to realize, recognition efficiency is high, and recognition result is accurate.
Description
Technical field
This application involves field of image processing more particularly to a kind of transmission of electricity corridor natural calamity image recognition methods.
Background technique
The public utilities and basic energy resource that electric power is paid high attention to as the whole society, in national security, economic development and daily
Irreplaceable role is played in life, once occur large area blackout, can to national economy, public safety and
The daily life of the people makes an extremely bad impression.
In recent years, worldwide the physical and mechanicalness impact that power grid generates occurs again and again for natural calamity,
Because the electric power system fault that natural calamity directly or indirectly causes is in the situation of cumulative year after year.Such as: 2008 beginning of the years, China south
Square drug in some provinces is subject to serious low-temperature freezing rain and snow disaster, since wet down amplitude is big, ice and snow weather range is wide, the duration
It is long, power grid occur successively transmission line of electricity fall bar, fall tower, broken string situations such as, lead to electrical network facilities subject to severe risks of damage, cause big model
The power supply enclosed is interrupted, and some areas disastrous accidents such as power failure for a long time are resulted in.According to statistics, in this sleet and snow ice calamity
In evil, nationwide power grid is because of calamity stoppage in transit power circuit totally 36740, wherein and 119,500kV route, 220kV route 348
Totally 36273, item, 110kV and following route.Disaster-affected suspended substation totally 2018.110-500kV route is because to fall tower total for calamity
8381 bases cause 33,480,000 families to power off, and direct economic loss is up to 1516.5 hundred million yuan.In mid-February, 2009, Hunan Electric Grid is because of mountain fire
Occur 500kV line tripping 3 times, 220kV transmission line of electricity trips 14 times, accounts for as many as 64.3% of tripping sum.Huge economy
Loss and social influence warn us, carry out electric system and take precautions against natural calamities research, ensure the safety of electric system in a natural environment surely
Fixed operation is current very urgent challenge and difficult task.
Currently, a large amount of scholar have carried out research abundant for the contingency management of electric system both at home and abroad, although for electricity
The research of net natural calamity contingency management achieves faster progress, but since research is started late, with electric network emergency chain of command
The situation faced is compared with task, compared with natural calamity bring severe challenge, the research of power grid natural calamity contingency management
Content and depth are all inadequate.On the one hand, power grid disaster emergency management is laid particular emphasis in thing or subsequent reply is disposed, even if carrying out
Early warning in advance is also mainly focused on natural calamity oneself is formed and in that will cause urgent moment of power grid accident, leaves management for
The time of policymaker is too short, and often difficulty avoids for loss.On the other hand, traditional power grid disaster emergency management is more focused on power grid system
System itself, is limited to inside electric system, is more focused on grid equipment and fails and less consideration equipment failure reason, for risk
The type or degree in source consider that cannot not enough effectively realize the risk for easily leading to large-area power-cuts for natural calamity carries out in advance
It is alert.Need further thoroughgoing and painstaking carry out case study for power grid natural calamity forewarning management.Traditional Geological Hazards Monitoring
Based on mainly artificial field exploring, due to artificial field exploring inefficiency with a varied topography, predictablity rate is not high.
Summary of the invention
This application provides a kind of transmission of electricity corridor natural calamity image recognition methods, are mainly to solve Geological Hazards Monitoring
Based on artificial field exploring, due to artificial field exploring inefficiency with a varied topography, the not high problem of predictablity rate.
This application provides a kind of transmission of electricity corridor natural calamity image recognition methods, which comprises
Obtain unmanned plane image；
Noise reduction process is carried out to the aerial images using bilateral filtering algorithm；
The disaster characteristic in aerial images using preset algorithm, after extracting the noise reduction；
According to the disaster characteristic, the first Disasters Type is determined；
The aerial images after the noise reduction are identified using neural convolutional network, determine the second Disasters Type；
First Disasters Type and the second Disasters Type are compared, recognition result is obtained.
From the above technical scheme, this application provides a kind of transmission of electricity corridor natural calamity image recognition methods, utilize
Background subtraction method and inter-frame difference carry out feature extraction to foreground image, and convolutional neural networks can efficiently extract classification automatically
Feature accurately identifies doubtful disaster image, identifies to realize to electric system corridor geological disaster intelligent imaging,
Recognition efficiency is high, and recognition result is accurate.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow chart of transmission of electricity corridor natural calamity image recognition method of the application body；
Fig. 2 is convolutional neural networks characteristic extraction procedure schematic diagram.
Specific embodiment
Referring to Fig. 1, this application provides a kind of transmission of electricity corridor natural calamity image recognition methods, and the method includes as follows
Step:
Step 11: obtaining unmanned plane image.
Using unmanned plane routine inspection mode, acquires power grid transmission line Hidden hazrads and high-resolution boat in periphery region takes the photograph data.
Step 12: noise reduction process being carried out to the aerial images using bilateral filtering algorithm.
In large space is monitored, due to being easy the interference by natural phenomenas such as weather, monitoring environment is relatively multiple
It is miscellaneous, therefore the quality of video image is poor, there are much noises.Thus to video figure acquired in large space monitoring place
It is particularly important as carrying out image denoising pretreatment, Preprocessing Algorithm will can be improved picture quality, inhibit making an uproar in image
Sound and the operand for reducing subsequent motion region and feature extraction improve the accuracy and fast speed of video smoke detection.One
As for, noise present in video image usually by external disturbance, situations such as video camera is mobile, shake and generate, compare
At random, therefore common Denoising Algorithm is median filtering algorithm, Mean Filtering Algorithm etc..But when disaster occurs (such as fire
Calamity), in early days in a zonule the variation phenomenon (such as the smog of fire) of generation in monitoring range, and become
The phenomenon that change (such as smog), has fuzzy behaviour to background image, and edge is not obvious.And the bilateral filtering for retaining boundary is calculated
Method can not only retain the minutia of target image better, and have good suppression to the noise of logo image again
System.
The bilateral filtering algorithm basic principle for retaining details is as follows:
G (x, y)=f (x, y)+n (x, y)
Wherein, g is noise-containing image, and f is rebuild without noise image, and n is noise, and two-sided filter is to pass through
To the pixel value of partial image pixel weighted average method operation and then acquisition restored image, it is shown below:
Wherein, SX, yIt is field of the central point (x, y) with (2N+1) * (2N+1) apart from size, g (i, j) is in the field
Pixel value:
W (i, j)=ws(i,j)wr(i,j)；
Wherein, wsIt is the spatial neighbor degree factor, wrIt is the brightness degree of approximation factor, filters half-breadth N, will affect two-sided filter
Performance.N is bigger, and expression smoothing effect is stronger, δsAnd δrThe spatial neighbor degree factor and the brightness degree of approximation factor are controlled respectively
Attenuation degree.
Step 13: utilizing preset algorithm, the disaster characteristic in aerial images after extracting the noise reduction.
Preset algorithm can be background subtraction algorithm or frame differential method.Background subtraction algorithm includes the following steps:
(1) following formula is utilized, pixel difference absolute value is calculated,
BDt(x, y)=| It(x, y)-Bt(x, y) |,
Wherein, It(x, y) indicates current frame image, Bt(x, y) indicates background frames image, BDt(x, y) indicates background difference
Image.
(2) judge whether the pixel difference absolute value is greater than or equal to first threshold, if the pixel difference is absolute
Value is greater than or equal to first threshold, then the pixel is foreground pixel；If the pixel difference absolute value is less than the first threshold
Value, then the pixel is background pixel；
(3) foreground pixel, the disaster feature in the aerial images after obtaining noise reduction are extracted.
Inter-frame difference algorithm includes the following steps:
(1) difference of the image of present frame and the image of former frame is calculated；
(2) judge whether the difference is greater than or equal to second threshold, if the difference is greater than or equal to first threshold,
Then the pixel is foreground pixel；If the difference is less than first threshold, the pixel is background pixel；
(3) foreground pixel, the disaster feature in the aerial images after obtaining noise reduction are extracted.
Step 14: according to the disaster characteristic, determining the first Disasters Type.
Step 15: the aerial images after the noise reduction being identified using neural convolutional network, determine the second disaster class
Type.
Specifically, referring to fig. 2, in convolutional layer, aggregate statistics is carried out to the pixel of the aerial images after the noise reduction, are mentioned
Take the local feature of image.Aerial Images can actually regard the matrix of pixel composition, after image gray processing, each picture as
The size of vegetarian refreshments is between 0~255.
Assuming that l layers are convolutional layers, then the feature vector that goes out of this layer is indicated with following formula are as follows:
In formula,Indicate i-th of layer output；Indicate that a convolution kernel of the convolutional layer, * indicate convolution operation,
Indicate that the convolutional layer biases, f is activation primitive.
In the layer of pond, dimensionality reduction is combined to the local feature of the image.The pixel of disaster of taking photo by plane image itself
It is more, it needs to carry out aggregate statistics to the feature of different location, while also achieving the purpose that Data Dimensionality Reduction.
Assuming that m layers are pond layers, then the icing feature vector exported is expressed as follows shown in formula
In formula,For connection weight, down (x) is that all pixels of the block different to icing image are summed,For the layer
Biasing, f is activation primitive.
Interlayer connection weight optimization based on back-propagation algorithm, the reversed biography (Back of convolutional neural networks
Propagation, BP) it is also referred to as the error propagation stage, this stage needs to calculate the parameter gradients of each layer.It is defeated firstly the need of calculating
It deviates, is shown below.
In formula, yiIt is the reality output of IBP-CNN, oiIt is data original tag.It is implied using BP back-propagation algorithm
After the connection weight gradient of layer, connection weight parameter correction is shown below.
In formula, η is learning rate, and β is connection weight.
Image treating monitors error wave in larger vision difference situation caused by the reasons such as shooting angle, light
Move larger, therefore the test validity of this method relies on corresponding specific icing image, and Generalization Capability is poor.CNN model exists
Model, which is not enough to that model can be expanded when expressing sample characteristics completely, introduces new parameter, compensate for original CNN model cross redundancy or
Simplified problem is crossed, test effect is more preferably.
According to the local feature of the image, the disaster feature of image is obtained.
According to the disaster feature, the second disaster type is determined.
Step 16: first Disasters Type and the second Disasters Type being compared, recognition result is obtained.
If the first Disasters Type is identical as the second Disasters Type, final recognition result can determine.
From the above technical scheme, this application provides a kind of transmission of electricity corridor natural calamity image recognition methods, utilize
Background subtraction method and inter-frame difference carry out feature extraction to foreground image, and convolutional neural networks can efficiently extract classification automatically
Feature accurately identifies doubtful disaster image, identifies to realize to electric system corridor geological disaster intelligent imaging,
Recognition efficiency is high, and recognition result is accurate.
Claims (4)
1. a kind of transmission of electricity corridor natural calamity image recognition method, which is characterized in that the described method includes:
Obtain unmanned plane image；
Noise reduction process is carried out to the aerial images using bilateral filtering algorithm；
The disaster characteristic in aerial images using preset algorithm, after extracting the noise reduction；
According to the disaster characteristic, the first Disasters Type is determined；
The aerial images after the noise reduction are identified using neural convolutional network, determine the second Disasters Type；
First Disasters Type and the second Disasters Type are compared, recognition result is obtained.
2. the method as described in claim 1, which is characterized in that the preset algorithm is background subtraction algorithm, described using pre-
Imputation method, the disaster characteristic in aerial images after extracting the noise reduction include:
Using following formula, pixel difference absolute value is calculated,
BDt(x, y)=| It(x, y)-Bt(x, y) |,
Wherein, It(x, y) indicates current frame image, Bt(x, y) indicates background frames image, BDt(x, y) indicates background difference image；
Judge whether the pixel difference absolute value is greater than or equal to first threshold, if the pixel difference absolute value be greater than or
Equal to first threshold, then the pixel is foreground pixel；If the pixel difference absolute value is less than first threshold, described
Pixel is background pixel；
The foreground pixel is extracted, the disaster feature in the aerial images after obtaining noise reduction.
3. the method as described in claim 1, which is characterized in that the preset algorithm is frame differential method, described using default
Algorithm, the disaster characteristic in aerial images after extracting the noise reduction include:
Calculate the difference of the image of present frame and the image of former frame；
Judge whether the difference is greater than or equal to second threshold, it is described if the difference is greater than or equal to first threshold
Pixel is foreground pixel；If the difference is less than first threshold, the pixel is background pixel；
The foreground pixel is extracted, the disaster feature in the aerial images after obtaining noise reduction.
4. the method as described in claim 1, which is characterized in that described to utilize neural convolutional network to taking photo by plane after the noise reduction
Image is identified, determines that the second Disasters Type includes:
In convolutional layer, aggregate statistics are carried out to the pixel of the aerial images after the noise reduction, extract the local feature of image；
In the layer of pond, dimensionality reduction is combined to the local feature of the image；
According to the local feature of the image, the disaster feature of image is obtained；
According to the disaster feature, the second disaster type is determined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910089314.6A CN109871788A (en) | 2019-01-30 | 2019-01-30 | A kind of transmission of electricity corridor natural calamity image recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910089314.6A CN109871788A (en) | 2019-01-30 | 2019-01-30 | A kind of transmission of electricity corridor natural calamity image recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109871788A true CN109871788A (en) | 2019-06-11 |
Family
ID=66918427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910089314.6A Pending CN109871788A (en) | 2019-01-30 | 2019-01-30 | A kind of transmission of electricity corridor natural calamity image recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871788A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027422A (en) * | 2019-11-27 | 2020-04-17 | 国网山东省电力公司电力科学研究院 | Emergency unmanned aerial vehicle inspection method and system applied to power transmission line corridor |
| CN111898861A (en) * | 2020-06-29 | 2020-11-06 | 中国矿业大学 | Grading evaluation method for geological disaster to geographic interest point dangerousness |
| CN113538524A (en) * | 2021-06-18 | 2021-10-22 | 云南电网有限责任公司 | 5G-fused intelligent early warning method and system for debris flow of power transmission line |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005050581A2 (en) * | 2003-11-17 | 2005-06-02 | Vidient Systems, Inc | Video surveillance system |
| CN105809679A (en) * | 2016-03-04 | 2016-07-27 | 李云栋 | Mountain railway side slope rockfall detection method based on visual analysis |
| CN106778902A (en) * | 2017-01-03 | 2017-05-31 | 河北工业大学 | Milk cow individual discrimination method based on depth convolutional neural networks |
| CN106778650A (en) * | 2016-12-26 | 2017-05-31 | 深圳极视角科技有限公司 | Scene adaptive pedestrian detection method and system based on polymorphic type information fusion |
| CN107229907A (en) * | 2017-05-09 | 2017-10-03 | 宁波大红鹰学院 | A kind of method using unmanned machine testing marine red tide occurring area |
| US20170286774A1 (en) * | 2016-04-04 | 2017-10-05 | Xerox Corporation | Deep data association for online multi-class multi-object tracking |
| CN108133188A (en) * | 2017-12-22 | 2018-06-08 | 武汉理工大学 | A kind of Activity recognition method based on motion history image and convolutional neural networks |
| CN108275114A (en) * | 2018-02-27 | 2018-07-13 | 苏州清研微视电子科技有限公司 | A kind of Security for fuel tank monitoring system |
| CN108447123A (en) * | 2018-03-27 | 2018-08-24 | 贵州电网有限责任公司输电运行检修分公司 | A kind of power transmission line corridor geological disaster investigation method and system |
| CN108599863A (en) * | 2018-04-26 | 2018-09-28 | 王晓楠 | Overhead transmission line wind based on shallow-layer CNN disaggregated models waves monitoring and pre-warning system and method |
| CN108680602A (en) * | 2018-05-18 | 2018-10-19 | 云南电网有限责任公司电力科学研究院 | A kind of detection device, the method and system of porcelain insulator internal flaw |
| CN108961675A (en) * | 2018-06-14 | 2018-12-07 | 江南大学 | Fall detection method based on convolutional neural networks |
| US20190025858A1 (en) * | 2016-10-09 | 2019-01-24 | Airspace Systems, Inc. | Flight control using computer vision |
-
2019
- 2019-01-30 CN CN201910089314.6A patent/CN109871788A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005050581A2 (en) * | 2003-11-17 | 2005-06-02 | Vidient Systems, Inc | Video surveillance system |
| CN105809679A (en) * | 2016-03-04 | 2016-07-27 | 李云栋 | Mountain railway side slope rockfall detection method based on visual analysis |
| US20170286774A1 (en) * | 2016-04-04 | 2017-10-05 | Xerox Corporation | Deep data association for online multi-class multi-object tracking |
| US20190025858A1 (en) * | 2016-10-09 | 2019-01-24 | Airspace Systems, Inc. | Flight control using computer vision |
| CN106778650A (en) * | 2016-12-26 | 2017-05-31 | 深圳极视角科技有限公司 | Scene adaptive pedestrian detection method and system based on polymorphic type information fusion |
| CN106778902A (en) * | 2017-01-03 | 2017-05-31 | 河北工业大学 | Milk cow individual discrimination method based on depth convolutional neural networks |
| CN107229907A (en) * | 2017-05-09 | 2017-10-03 | 宁波大红鹰学院 | A kind of method using unmanned machine testing marine red tide occurring area |
| CN108133188A (en) * | 2017-12-22 | 2018-06-08 | 武汉理工大学 | A kind of Activity recognition method based on motion history image and convolutional neural networks |
| CN108275114A (en) * | 2018-02-27 | 2018-07-13 | 苏州清研微视电子科技有限公司 | A kind of Security for fuel tank monitoring system |
| CN108447123A (en) * | 2018-03-27 | 2018-08-24 | 贵州电网有限责任公司输电运行检修分公司 | A kind of power transmission line corridor geological disaster investigation method and system |
| CN108599863A (en) * | 2018-04-26 | 2018-09-28 | 王晓楠 | Overhead transmission line wind based on shallow-layer CNN disaggregated models waves monitoring and pre-warning system and method |
| CN108680602A (en) * | 2018-05-18 | 2018-10-19 | 云南电网有限责任公司电力科学研究院 | A kind of detection device, the method and system of porcelain insulator internal flaw |
| CN108961675A (en) * | 2018-06-14 | 2018-12-07 | 江南大学 | Fall detection method based on convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| 张宁: "基于无人机平台的动态目标检测系统开发", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
| 张洁: "烟雾视频图像识别算法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027422A (en) * | 2019-11-27 | 2020-04-17 | 国网山东省电力公司电力科学研究院 | Emergency unmanned aerial vehicle inspection method and system applied to power transmission line corridor |
| CN111898861A (en) * | 2020-06-29 | 2020-11-06 | 中国矿业大学 | Grading evaluation method for geological disaster to geographic interest point dangerousness |
| CN111898861B (en) * | 2020-06-29 | 2023-09-26 | 中国矿业大学 | Grading evaluation method for geographical interest point dangers by geological disasters |
| CN113538524A (en) * | 2021-06-18 | 2021-10-22 | 云南电网有限责任公司 | 5G-fused intelligent early warning method and system for debris flow of power transmission line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107145846B (en) | A kind of insulator recognition methods based on deep learning | |
| CN109871788A (en) | A kind of transmission of electricity corridor natural calamity image recognition method | |
| CN106595551B (en) | Ice covering thickness detection method in powerline ice-covering image based on deep learning | |
| CN109741320A (en) | A kind of wind electricity blade fault detection method based on Aerial Images | |
| CN110334660A (en) | A kind of forest fire monitoring method based on machine vision under the conditions of greasy weather | |
| Zheng et al. | Component identification and defect detection in transmission lines based on deep learning | |
| CN112184628B (en) | Infrared duplex wave image and cloud early warning system and method for flood prevention and danger detection of dike | |
| CN115240075B (en) | Construction and training method of electric power vision multi-granularity pre-training large model | |
| CN115331384B (en) | Fire accident early warning system of operation platform based on edge calculation | |
| CN115690730A (en) | High-speed rail contact net foreign matter detection method and system based on single classification and abnormal generation | |
| CN115330718A (en) | Insulator defect detection method based on improved YOLOv4 network | |
| Liu et al. | Summary of insulator defect detection based on deep learning | |
| CN114037910A (en) | Unmanned aerial vehicle forest fire detecting system | |
| CN117058396B (en) | Fan blade defect area rapid segmentation method and system based on artificial intelligence | |
| Yuan et al. | Identification method of typical defects in transmission lines based on YOLOv5 object detection algorithm | |
| Liu et al. | Image dehazing method of transmission line for unmanned aerial vehicle inspection based on densely connection pyramid network | |
| CN112651337A (en) | Sample set construction method applied to training line foreign object target detection model | |
| Daogang et al. | Anomaly identification of critical power plant facilities based on YOLOX-CBAM | |
| Zhou et al. | Intelligent identification method for natural disasters along transmission lines based on inter-frame difference and regional convolution neural network | |
| CN114612855A (en) | Power line hazard source detection system and method fusing residual error and multi-scale network | |
| Wei et al. | An early warning model for damage of power grid under typhoon disaster | |
| Changfu et al. | Research of substation equipment abnormity identification based on image processing | |
| Pan | Intelligent image recognition research on status of power transmission lines | |
| CN106093729A (en) | The intelligent analysis method of the transmission line insulator flashover of feature based identification | |
| Su et al. | Research on Stereoscopic Monitoring and Positioning Technology of Mountain Fire in EHV Transmission Lines |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190611 |