CN109813276A - A kind of antenna for base station has a down dip angle measuring method and its system - Google Patents
A kind of antenna for base station has a down dip angle measuring method and its system Download PDFInfo
- Publication number
- CN109813276A CN109813276A CN201811555714.3A CN201811555714A CN109813276A CN 109813276 A CN109813276 A CN 109813276A CN 201811555714 A CN201811555714 A CN 201811555714A CN 109813276 A CN109813276 A CN 109813276A
- Authority
- CN
- China
- Prior art keywords
- antenna
- image
- base station
- candidate frame
- training
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses the measurement methods and its system of a kind of antenna for base station angle of declination, and wherein method flies or flies around base station to obtain the image of base station the following steps are included: unmanned plane is used to pinpoint, and data set is made according to image；Training is iterated to data set using convolutional neural networks, obtains antenna training model；The antenna part in image is identified using antenna training model；The antenna in antenna part is identified according to auto-thresholding algorithm；Calculate Downtilt.Compared to traditional technology, it is more intelligent using unmanned plane, and it can reduce the measurement problem and work load of staff, even if inspection can also be carried out in inclement weather, and image is handled with adaptive threshold fuzziness means based on convolutional network, Downtilt can be calculated after the treatment, and measurement efficiency is higher and measurement result is also more accurate, therefore the present invention has many advantages, such as that safety is good, environmental suitability is strong, measurement fast and stable.
Description
Technical field
The present invention relates to electronic surveying field, especially a kind of antenna for base station has a down dip angle measuring method and its system.
Background technique
Terminal of the antenna for base station as mobile communications network, carries electromagnetic radiation and received duplex capability, is
The carrier of mobile communication signal transmitting, the quality of application effect directly determine the superiority and inferiority of mobile communications network.With people
In the case that requirement to mobile communications network is higher and higher, each network communication operator increases the investment to base station one after another,
It carries out the enlarging of base station and increases to the investment of base station daily maintenance and inspection, therefore regularity carries out maintenance to base station and patrols
Inspection, is an indispensable link in common carrier commodity network maintenance work.But existing technological means is to antenna
There is also certain defects for the detection of progress angle of declination:
For operator, the existing means for obtaining Downtilt are mainly had a down dip by manually stepping on tower
Angular dimensions, and the antenna of base station is often placed on higher place, this undoubtedly brings great safety to technical work personnel
Hidden danger and operating difficulties, and this means only rely on manual measurement, and efficiency is also more low；Moreover, inspection base station
Working time, place be affected by weather, environment, in the case where bad environments, can not to base station carry out inspection.
Summary of the invention
To solve the above-mentioned problems, the purpose of the embodiment of the present invention is that providing that a kind of safety is good, environmental suitability is strong, surveys
The antenna for base station of amount fast and stable has a down dip angle measuring method and its system.
In order to make up for the deficiencies of the prior art, technical solution used in the embodiment of the present invention is:
A kind of measurement method of antenna for base station angle of declination, comprising the following steps:
It uses the flight of unmanned plane fixed point or flies around base station to obtain the image of base station, data set is made according to image；
Training is iterated to data set using convolutional neural networks, obtains antenna training model；
The antenna part in image is identified using antenna training model；
The antenna in antenna part is identified according to auto-thresholding algorithm；
Calculate Downtilt.
Further, data set is made according to image, comprising: image is divided to and is labeled as several antenna candidate frames, is made
Make the corresponding data collection of any antenna candidate frame, the total collection of the corresponding data collection is data set.
Further, training is iterated to data set using convolutional neural networks, obtains antenna training model, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first
Secondary network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of net
Network model；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
Further, the antenna part in image is identified using antenna training model, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates that any antenna is candidate
Whether the IOU between the region and either objective region of frame is greater than 0.7, if so, include antenna in the antenna candidate frame, thus
It is identified, otherwise the antenna candidate frame is background.
Further, the antenna in antenna part is identified according to auto-thresholding algorithm, calculates Downtilt, packet
It includes:
Anisotropic according to the gray value differences between antenna and background image, being found out by adaptivenon-uniform sampling algorithm makes to be identified
Antenna candidate frame the maximum adaptive threshold of inter-class variance, be partitioned into the antenna candidate frame being identified based on this condition
Antenna and background image.
Further, Downtilt is calculated, comprising: using the pole of antenna as reference point, calculate the horizontal axis of antenna candidate frame
With the ratio between the longitudinal axis and when the ratio maximum, obtain antenna angle of declination be the ratio between horizontal axis and longitudinal axis of antenna candidate frame anyway
Cut value.
A kind of antenna for base station has a down dip angle measuring system, comprising:
Image capture module obtains the image of base station for controlling the flight of unmanned plane fixed point or around base station flight, according to
Data set is made in image；
Image training module obtains antenna training mould for being iterated training to data set using convolutional neural networks
Type；
Picture recognition module, for identifying the antenna part in image using antenna training model；
Image separation module, for identifying the antenna in antenna part according to auto-thresholding algorithm；
Image computing module, for calculating Downtilt.
Further, data set is made according to image in described image acquisition module, comprising:
Image is divided to and is labeled as several antenna candidate frames, makes the corresponding data collection of any antenna candidate frame, institute
The total collection for stating corresponding data collection is data set.
Further, described image training module is iterated training to data set using convolutional neural networks, obtains antenna
Training pattern, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first
Secondary network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of net
Network model；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
Further, picture recognition module, for identifying the antenna part in image using antenna training model, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates that any antenna is candidate
Whether the IOU between the region and either objective region of frame is greater than 0.7, if so, include antenna in the antenna candidate frame, thus
It is identified, otherwise the antenna candidate frame is background.
The one or more technical solutions provided in the embodiment of the present invention, at least have the following beneficial effects: using nobody
Machine obtains the external structure image of base station, and by convolutional neural networks come to data set corresponding to image be trained with
And identification, to identify the antenna part in the image, and further use adaptive threshold fuzziness means by this portion
Antenna in point individually splits out, and calculates so as to carry out final independent analysis to it；As it can be seen that compared to traditional technology,
It is more intelligent using unmanned plane, and can reduce the measurement problem and work load of staff, even if in inclement weather
Inspection can be carried out, and image is handled with adaptive threshold fuzziness means based on convolutional network, after the treatment
Downtilt can be calculated, compared to previous means, measurement efficiency is higher and measurement result is also more accurate.Therefore, originally
Invention has many advantages, such as that safety is good, environmental suitability is strong, measurement fast and stable.
Detailed description of the invention
Present pre-ferred embodiments are provided, with reference to the accompanying drawing with the embodiment that the present invention will be described in detail.
Fig. 1 is a kind of step flow chart of the measurement method of antenna for base station angle of declination of the embodiment of the present invention；
Fig. 2 is that the unmanned plane of the embodiment of the present invention obtains the schematic illustration of base station image；
Fig. 3 is the functional schematic of the convolutional neural networks of the embodiment of the present invention；
Fig. 4 is the schematic illustration of the convolutional neural networks of the embodiment of the present invention.
Specific embodiment
- Fig. 4 referring to Fig.1, a kind of measurement method of antenna for base station angle of declination of the embodiment of the present invention, comprising the following steps:
It uses the flight of unmanned plane fixed point or flies around base station to obtain the image of base station, data set is made according to image；
Training is iterated to data set using convolutional neural networks, obtains antenna training model；
The antenna part in image is identified using antenna training model；
The antenna in antenna part is identified according to auto-thresholding algorithm；
Calculate Downtilt.
Further, data set is made according to image, comprising: image is divided to and is labeled as several antenna candidate frames, is made
Make the corresponding data collection of any antenna candidate frame, the total collection of the corresponding data collection is data set.
Further, training is iterated to data set using convolutional neural networks, obtains antenna training model, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first
Secondary network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of net
Network model；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
Further, the antenna part in image is identified using antenna training model, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates that any antenna is candidate
Whether the IOU between the region and either objective region of frame is greater than 0.7, if so, include antenna in the antenna candidate frame, thus
It is identified, otherwise the antenna candidate frame is background.
Further, the antenna in antenna part is identified according to auto-thresholding algorithm, calculates Downtilt, packet
It includes:
Anisotropic according to the gray value differences between antenna and background image, being found out by adaptivenon-uniform sampling algorithm makes to be identified
Antenna candidate frame the maximum adaptive threshold of inter-class variance, be partitioned into the antenna candidate frame being identified based on this condition
Antenna and background image.
Further, Downtilt is calculated, comprising: using the pole of antenna as reference point, calculate the horizontal axis of antenna candidate frame
With the ratio between the longitudinal axis and when the ratio maximum, obtain antenna angle of declination be the ratio between horizontal axis and longitudinal axis of antenna candidate frame anyway
Cut value.
Specifically, referring to Fig. 2, the flight shape of unmanned plane can be controlled by professional unmanned plane manipulator on the ground
State, allow unmanned plane in height with antenna is contour is advisable, and the position of vertical range antenna 7m to 10m is carried out around flight,
To which collected antenna condition image (being also possible to truncated picture in video) is real-time transmitted to artificial background server.
When being converted into data set, can also label simultaneously to different antenna candidate frames, i.e., as antenna
Frame is demarcated, in this way convenient for distinguishing, because usually depositing in the terminals such as computer when it is as storage data, utilizing label
It distinguishes and is similar to file name differentiation on PC, can be convenient and easily find out satisfactory antenna candidate in step below
Frame.
Referring to Fig. 3, the feature of antenna candidate frame (that is to say that the aforementioned antenna labeled demarcates frame) is extracted in convolutional layer
When attribute, multistage region convolutional neural networks can be extracted antenna using the convolution+activation+pond layer on one group of basis first and be waited
The feature for selecting frame, specifically can refer to Fig. 4, and a kind of preferred convolutional coding structure implementing procedure is: assuming that input picture size is 224*
221*3, then it successively passes through 7*7 convolutional layer, 3*3 maximum pond layer, 5*5 convolutional layer, 3*3 maximum pond layer and 4 3*3 volumes
Lamination (specific stepping can be arranged according to situation difference, not limit), is finally come by two different 1*1 convolutional layers respectively
It realizes respectively and distinguishes target antenna and output target frame coordinate value (x, y, w, h), since it is desired that identifying that antenna extracts mesh
Mark antenna characteristic attribute, extract this feature after will be used in subsequent antenna candidate frame network layer and full articulamentum, so as to
To carry out the selection of antenna candidate frame and accurately identifying for antenna；Then the selection for carrying out candidate frame region, due to only needing
Recognition detection goes out in the frame region whether have antenna, is the detection for belonging to two classification, so carrying out candidate frame regional choice
When, output is target and non-targeted probability；In general, region candidate frame network has two-lines, one passes through classification
Subregion separate background and target antenna, another can calculate the offset of frame recurrence, accurately be predicted with this
Frame, and antenna candidate frame network layer is then responsible for obtaining integration objective and boundary and returns the offset of frame, at the same weed out it is too small and
Antenna candidate frame beyond boundary.When IOU > 0.7 of candidate frame region and either objective region, then it is determined as thering is target；In addition,
If IOU < 0.3 of certain candidate frame region and either objective region, then be determined as background.So-called IOU is exactly prediction block (i.e. day
Line candidate frame) and true frame (target area frame) coverage rate, value be equal to two frames intersection divided by two frames union, i.e.,
Assuming that A, B two are different individuals, then registration IOU=(A ∩ B)/(A ∪ B).
After selected target frame, so that it may export target frame coordinate value (x, y, w, h), what wherein x and y was represented is target frame
Centre coordinate, w and h indicate the width and length of target frame, can use formula calculating:Make predicted value with true value t*=(tx,ty,tw,th)
Between gap it is minimum, obtain loss function are as follows:
φ5(P) be input antenna candidate frame feature vector, w* be need learn parameter (* indicate x, y, w and h,
I.e. each converts a corresponding objective function), the predicted value that d* (P) is；Further, function optimization target is obtained
Are as follows:
The two simultaneous, then the integral function taken in this algorithm antenna identification is:
Wherein, LclsAnd LregPredicted value and true value are respectively corresponded, λ is proportionality constant, and i indicates the rope of antenna candidate frame
Draw, pi indicates preceding to classification predicted value, and pi* indicates the predicted value probability of antenna candidate frame, comes to identify antenna accurately with this；
By continuing to optimize network model, modification and the repetitive exercise of optimal threshold are carried out, may ultimately reach an optimal training
Model.
Adaptive threshold fuzziness method, can be by image point using the having differences property of gray value between antenna and background image
At target and background, and the maximum segmentation of inter-class variance is set to mean that wrong point of probability is minimum, therefore pass through adaptive threshold
Segmentation can accurately find out an adaptive threshold, so that inter-class variance is maximum, thus accurately by target antenna and background
It is separated.For image I (m, n), the segmentation threshold of prospect (i.e. target) and background is remembered
Make T, the pixel number for belonging to prospect accounts for the ratio of entire image and is denoted as ω 0, and average gray is μ 0；Background pixel
The ratio that points account for entire image is ω 1, and average gray is μ 1；The overall average gray scale of image is denoted as μ, and inter-class variance is denoted as g；
Assuming that the background of image is darker, and the size of image is M × N, the gray value of the pixel pixel less than threshold value T in image
Number scale makees N0, and number of pixels of the grey scale pixel value greater than threshold value T is denoted as N1, then has:
ω0=N0/M·N
ω1=N1/M·N
N0+N1=MN
ω0+ω1=1
μ=ω0*μ0+ω1*μ1
G=ω0(μ0-μ)2+ω1(μ1-μ)2
Equivalence formula can be obtained in abbreviation:
G=ω0ω1(μ0-μ1)2
Then, it only need to can just obtain making the maximum threshold value T of inter-class variance using the method for traversal, it is as required, thus base
Keep the antenna separated with background image in the adaptive threshold fuzziness, the convenient calculating followed by angle of declination.
Final step is the calculating of angle of declination:
After outlining target antenna using algorithm, by calculating the transverse and longitudinal ratio for the antenna candidate frame chosen, work as transverse and longitudinal
I.e. antenna is in front position at this time when than reaching maximum value.
After determining the front of antenna, unmanned plane can be positioned to the front of height herein, desired vertical distance is 7m
To 10m, so that unmanned plane is rotated clockwise or counter-clockwise 90 ° of flight around base station and reach its side, at this time the pole of antenna with
The longitudinal axis of antenna candidate frame is parallel to each other, and antenna is exactly on the diagonal line of antenna candidate frame, then obtains the moment day
The image of line, and image is split, obtain the side view of the antenna.
The relational expression for finally applying trigonometric function, calculate the angle, θ between antenna holding pole and antenna, i.e. antenna has a down dip
Angle is the arc-tangent value of transverse and longitudinal ratio.
In fact, Downtilt=mechanical tilt angle+preset lower decline angle+electricity adjusts angle of declination.Mechanical tilt angle: can lead to
Adjustment mounting bracket is crossed, changes antenna physical position, to realize that mechanical tilt angle continuously adjusts；
Preset lower decline angle: by antenna figuration technology, adjusting antenna feeding network, changes each oscillator in aerial array
Phase, to realize the adjusting of preset lower decline angle under the premise of antenna physical position is constant.
Electricity adjusts angle of declination: by antenna Primary Component phase shifter, continuously adjusting antenna feeding network, continuously changes antenna array
The phase of each oscillator in column, to realize that electricity adjusts the tune of angle of declination continuously adjusted under the premise of antenna physical position is constant
Section mode.
Above-mentioned is the regulative mode that can change Downtilt, for reference.
A kind of antenna for base station has a down dip angle measuring system, comprising:
Image capture module obtains the image of base station for controlling the flight of unmanned plane fixed point or around base station flight, according to
Data set is made in image；
Image training module obtains antenna training mould for being iterated training to data set using convolutional neural networks
Type；
Picture recognition module, for identifying the antenna part in image using antenna training model；
Image separation module, for identifying the antenna in antenna part according to auto-thresholding algorithm；
Image computing module, for calculating Downtilt.
Further, data set is made according to image in described image acquisition module, comprising:
Image is divided to and is labeled as several antenna candidate frames, makes the corresponding data collection of any antenna candidate frame, institute
The total collection for stating corresponding data collection is data set.
Further, described image training module is iterated training to data set using convolutional neural networks, obtains antenna
Training pattern, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first
Secondary network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of net
Network model；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
Further, picture recognition module, for identifying the antenna part in image using antenna training model, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates that any antenna is candidate
Whether the IOU between the region and either objective region of frame is greater than 0.7, if so, include antenna in the antenna candidate frame, thus
It is identified, otherwise the antenna candidate frame is background.
Further, Downtilt is calculated, comprising: using the pole of antenna as reference point, calculate the horizontal axis of antenna candidate frame
With the ratio between the longitudinal axis and when the ratio maximum, obtain antenna angle of declination be the ratio between horizontal axis and longitudinal axis of antenna candidate frame anyway
Cut value.
Specifically, the external structure image of base station is obtained using unmanned plane, and by convolutional neural networks come to image
Corresponding data set is trained and identifies, to identify the antenna part in the image, and further using adaptive
Threshold segmentation means are answered individually to split out the antenna in this part, so as to carry out final independent analysis meter to it
It calculates；As it can be seen that compared to traditional technology, it is more intelligent using unmanned plane, and can reduce the measurement problem and work of staff
Burden, even if can also carry out inspection in inclement weather, and based on convolutional network with adaptive threshold fuzziness means come to figure
As being handled, Downtilt can be calculated after the treatment, compared to previous means, measurement efficiency is higher and measurement is tied
Fruit is also more accurate.
Presently preferred embodiments of the present invention and basic principle is discussed in detail in the above content, but the invention is not limited to
Above embodiment, those skilled in the art should be recognized that also have on the premise of without prejudice to spirit of the invention it is various
Equivalent variations and replacement, these equivalent variations and replacement all fall within the protetion scope of the claimed invention.
Claims (10)
1. a kind of measurement method of antenna for base station angle of declination, which comprises the following steps:
It uses the flight of unmanned plane fixed point or flies around base station to obtain the image of base station, data set is made according to image；
Training is iterated to data set using convolutional neural networks, obtains antenna training model；
The antenna part in image is identified using antenna training model；
The antenna in antenna part is identified according to auto-thresholding algorithm；
Calculate Downtilt.
2. a kind of measurement method of antenna for base station angle of declination according to claim 1, which is characterized in that be made according to image
Data set, comprising: image is divided to and is labeled as several antenna candidate frames, makes the corresponding data of any antenna candidate frame
Collection, the total collection of the corresponding data collection is data set.
3. a kind of measurement method of antenna for base station angle of declination according to claim 2, which is characterized in that utilize convolutional Neural
Network is iterated training to data set, obtains antenna training model, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first time net
Network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of network mould
Type；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
4. a kind of measurement method of antenna for base station angle of declination according to claim 3, which is characterized in that utilize antenna training
Model identifies the antenna part in image, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates any antenna candidate frame
Whether the I OU between region and either objective region is greater than 0.7, if so, include antenna in the antenna candidate frame, thus by
It identifies, otherwise the antenna candidate frame is background.
5. a kind of measurement method of antenna for base station angle of declination according to claim 4, which is characterized in that according to adaptive thresholding
Value partitioning algorithm identifies the antenna in antenna part, calculates Downtilt, comprising:
It is anisotropic according to the gray value differences between antenna and background image, the day for making to be identified is found out by adaptivenon-uniform sampling algorithm
The maximum adaptive threshold of the inter-class variance of line candidate frame is partitioned into the day in the antenna candidate frame being identified based on this condition
Line and background image.
6. according to a kind of measurement method of any antenna for base station angle of declination of claim 2-5, which is characterized in that calculate day
Line angle of declination, comprising: using the pole of antenna as reference point, calculate the ratio between horizontal axis and longitudinal axis of antenna candidate frame and work as the ratio most
When big, the angle of declination for obtaining antenna is the horizontal axis of antenna candidate frame and the arc-tangent value of the ratio between the longitudinal axis.
7. a kind of have a down dip angle measuring system using any antenna for base station of claim 1-6 characterized by comprising
Image capture module, for controlling the flight of unmanned plane fixed point or flying around base station to obtain the image of base station, according to image
Data set is made；
Image training module obtains antenna training model for being iterated training to data set using convolutional neural networks；
Picture recognition module, for identifying the antenna part in image using antenna training model；
Image separation module, for identifying the antenna in antenna part according to auto-thresholding algorithm；
Image computing module, for calculating Downtilt.
The angle measuring system 8. a kind of antenna for base station according to claim 7 has a down dip, which is characterized in that described image acquires mould
Data set is made according to image in block, comprising:
Image is divided to and is labeled as several antenna candidate frames, makes the corresponding data collection of any antenna candidate frame, it is described right
The total collection for answering data set is data set.
The angle measuring system 9. a kind of antenna for base station according to claim 8 has a down dip, which is characterized in that described image training mould
Block is iterated training to data set using convolutional neural networks, obtains antenna training model, comprising:
The characteristic attribute that antenna candidate frame is extracted according to data set carries out network training to antenna candidate frame, obtains first time net
Network model；
It when first time, network model reached threshold value, then takes it as a basis and executes previous step again, obtain second of network mould
Type；
Adjust the network parameter of antenna candidate frame and the extracting parameter of characteristic attribute, second of network model of training；
When second of network model reaches optimal threshold, using it as antenna training model.
The angle measuring system 10. a kind of antenna for base station according to claim 9 has a down dip, which is characterized in that picture recognition module,
For identifying the antenna part in image using antenna training model, comprising:
Corresponding target area is determined based on the extracted characteristic attribute of antenna training model, differentiates any antenna candidate frame
Whether the I OU between region and either objective region is greater than 0.7, if so, include antenna in the antenna candidate frame, thus by
It identifies, otherwise the antenna candidate frame is background.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811555714.3A CN109813276B (en) | 2018-12-19 | 2018-12-19 | Base station antenna downward inclination angle measuring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811555714.3A CN109813276B (en) | 2018-12-19 | 2018-12-19 | Base station antenna downward inclination angle measuring method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109813276A true CN109813276A (en) | 2019-05-28 |
| CN109813276B CN109813276B (en) | 2021-01-26 |
Family
ID=66602158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811555714.3A Active CN109813276B (en) | 2018-12-19 | 2018-12-19 | Base station antenna downward inclination angle measuring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109813276B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688904A (en) * | 2019-08-30 | 2020-01-14 | 中通服建设有限公司 | Base station antenna parameter surveying method and device based on 5G unmanned aerial vehicle |
| CN113128281A (en) * | 2019-12-31 | 2021-07-16 | 中国移动通信集团福建有限公司 | Automatic base station opening method and device |
| WO2021189353A1 (en) * | 2020-03-26 | 2021-09-30 | Wuyi University | Method and system of antenna measurement for mobile communication base station |
| CN116052003A (en) * | 2023-02-07 | 2023-05-02 | 中科星图数字地球合肥有限公司 | Method and device for measuring antenna angle information and related equipment |
| CN117097421A (en) * | 2023-10-18 | 2023-11-21 | 公诚管理咨询有限公司 | Base station antenna installation parameter detection method and system based on image recognition |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202077190U (en) * | 2011-05-09 | 2011-12-14 | 厦门特力通信息技术有限公司 | Downtilt angle device of remote real-time monitoring base station antenna |
| CN104978580A (en) * | 2015-06-15 | 2015-10-14 | 国网山东省电力公司电力科学研究院 | Insulator identification method for unmanned aerial vehicle polling electric transmission line |
| CN105528595A (en) * | 2016-02-01 | 2016-04-27 | 成都通甲优博科技有限责任公司 | Method for identifying and positioning power transmission line insulators in unmanned aerial vehicle aerial images |
| CN106469304A (en) * | 2016-09-22 | 2017-03-01 | 西安理工大学 | Handwritten signature location positioning method in bill based on depth convolutional neural networks |
| CN106683091A (en) * | 2017-01-06 | 2017-05-17 | 北京理工大学 | Target classification and attitude detection method based on depth convolution neural network |
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| CN107341488A (en) * | 2017-06-16 | 2017-11-10 | 电子科技大学 | A kind of SAR image target detection identifies integral method |
| CN107664491A (en) * | 2016-07-28 | 2018-02-06 | 中国电信股份有限公司 | Antenna for base station has a down dip angle measuring method, device and system |
| CN107833220A (en) * | 2017-11-28 | 2018-03-23 | 河海大学常州校区 | Fabric defect detection method based on depth convolutional neural networks and vision significance |
| WO2018155765A1 (en) * | 2017-02-22 | 2018-08-30 | 연세대학교 산학협력단 | Method and device analyzing plaque from computed tomography image |
| CN108550133A (en) * | 2018-03-02 | 2018-09-18 | 浙江工业大学 | A kind of cancer cell detection method based on Faster R-CNN |
| CN108648233A (en) * | 2018-03-24 | 2018-10-12 | 北京工业大学 | A kind of target identification based on deep learning and crawl localization method |
| CN108898145A (en) * | 2018-06-15 | 2018-11-27 | 西南交通大学 | A kind of image well-marked target detection method of combination deep learning |
-
2018
- 2018-12-19 CN CN201811555714.3A patent/CN109813276B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202077190U (en) * | 2011-05-09 | 2011-12-14 | 厦门特力通信息技术有限公司 | Downtilt angle device of remote real-time monitoring base station antenna |
| CN104978580A (en) * | 2015-06-15 | 2015-10-14 | 国网山东省电力公司电力科学研究院 | Insulator identification method for unmanned aerial vehicle polling electric transmission line |
| CN105528595A (en) * | 2016-02-01 | 2016-04-27 | 成都通甲优博科技有限责任公司 | Method for identifying and positioning power transmission line insulators in unmanned aerial vehicle aerial images |
| CN107664491A (en) * | 2016-07-28 | 2018-02-06 | 中国电信股份有限公司 | Antenna for base station has a down dip angle measuring method, device and system |
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| CN106469304A (en) * | 2016-09-22 | 2017-03-01 | 西安理工大学 | Handwritten signature location positioning method in bill based on depth convolutional neural networks |
| CN106683091A (en) * | 2017-01-06 | 2017-05-17 | 北京理工大学 | Target classification and attitude detection method based on depth convolution neural network |
| WO2018155765A1 (en) * | 2017-02-22 | 2018-08-30 | 연세대학교 산학협력단 | Method and device analyzing plaque from computed tomography image |
| CN107341488A (en) * | 2017-06-16 | 2017-11-10 | 电子科技大学 | A kind of SAR image target detection identifies integral method |
| CN107833220A (en) * | 2017-11-28 | 2018-03-23 | 河海大学常州校区 | Fabric defect detection method based on depth convolutional neural networks and vision significance |
| CN108550133A (en) * | 2018-03-02 | 2018-09-18 | 浙江工业大学 | A kind of cancer cell detection method based on Faster R-CNN |
| CN108648233A (en) * | 2018-03-24 | 2018-10-12 | 北京工业大学 | A kind of target identification based on deep learning and crawl localization method |
| CN108898145A (en) * | 2018-06-15 | 2018-11-27 | 西南交通大学 | A kind of image well-marked target detection method of combination deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 王天雷等: "《基于OTSU算法的BP神经网络网球识别方法》", 《测控技术》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688904A (en) * | 2019-08-30 | 2020-01-14 | 中通服建设有限公司 | Base station antenna parameter surveying method and device based on 5G unmanned aerial vehicle |
| CN113128281A (en) * | 2019-12-31 | 2021-07-16 | 中国移动通信集团福建有限公司 | Automatic base station opening method and device |
| WO2021189353A1 (en) * | 2020-03-26 | 2021-09-30 | Wuyi University | Method and system of antenna measurement for mobile communication base station |
| CN116052003A (en) * | 2023-02-07 | 2023-05-02 | 中科星图数字地球合肥有限公司 | Method and device for measuring antenna angle information and related equipment |
| CN116052003B (en) * | 2023-02-07 | 2024-05-14 | 中科星图数字地球合肥有限公司 | Method and device for measuring antenna angle information and related equipment |
| CN117097421A (en) * | 2023-10-18 | 2023-11-21 | 公诚管理咨询有限公司 | Base station antenna installation parameter detection method and system based on image recognition |
| CN117097421B (en) * | 2023-10-18 | 2023-12-19 | 公诚管理咨询有限公司 | Base station antenna installation parameter detection method and system based on image recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109813276B (en) | 2021-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109813276A (en) | A kind of antenna for base station has a down dip angle measuring method and its system | |
| US20210073573A1 (en) | Ship identity recognition method based on fusion of ais data and video data | |
| CN108334844B (en) | Line automatic tracking method for power transmission line inspection | |
| WO2020093630A1 (en) | Antenna downward inclination angle measurement method based on multi-scale deep semantic segmentation network | |
| CN105023014B (en) | A kind of shaft tower target extraction method in unmanned plane inspection transmission line of electricity image | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN108229587A (en) | A kind of autonomous scan method of transmission tower based on aircraft floating state | |
| CN114240868A (en) | Unmanned aerial vehicle-based inspection analysis system and method | |
| CN111709994B (en) | Autonomous unmanned aerial vehicle visual detection and guidance system and method | |
| AU2020215794B2 (en) | Downscaling weather forecasts | |
| CN114034296B (en) | Navigation signal interference source detection and identification method and system | |
| CN106709529B (en) | Visual detection method for photovoltaic cell color difference classification | |
| CN109116298B (en) | Positioning method, storage medium and positioning system | |
| CN112561899A (en) | Electric power inspection image identification method | |
| CN110263390B (en) | Antenna downward inclination angle automatic adjustment method and system based on unmanned aerial vehicle vision measurement | |
| CN110245592A (en) | A method of for promoting pedestrian's weight discrimination of monitoring scene | |
| CN108509844A (en) | VPER cloud cluster recognition methods based on ground cloud atlas | |
| CN109458978A (en) | A kind of Downtilt measurement method based on multiple scale detecting algorithm | |
| CN114089786A (en) | Autonomous inspection system based on unmanned aerial vehicle vision and along mountain highway | |
| CN115880231A (en) | Power transmission line hidden danger detection method and system based on deep learning | |
| CN115809986A (en) | Multi-sensor fusion type intelligent external damage detection method for power transmission corridor | |
| CN109684914A (en) | Based on unmanned plane image intelligent identification Method | |
| CN109271904A (en) | A kind of black smoke vehicle detection method based on pixel adaptivenon-uniform sampling and Bayesian model | |
| Bao et al. | E-unet++: A semantic segmentation method for remote sensing images | |
| Zhong et al. | Identification and depth localization of clustered pod pepper based on improved Faster R-CNN |
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 |