CN109902712A - Transmission line of electricity bird repellent method based on unmanned plane inspection - Google Patents
Transmission line of electricity bird repellent method based on unmanned plane inspection Download PDFInfo
- Publication number
- CN109902712A CN109902712A CN201910045339.6A CN201910045339A CN109902712A CN 109902712 A CN109902712 A CN 109902712A CN 201910045339 A CN201910045339 A CN 201910045339A CN 109902712 A CN109902712 A CN 109902712A
- Authority
- CN
- China
- Prior art keywords
- birds
- transmission line
- electricity
- image
- classifier
- 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
- Catching Or Destruction (AREA)
Abstract
This application discloses the transmission line of electricity bird repellent methods based on unmanned plane inspection, and the sound collection frequency of microphone is arranged, and utilize the voice signal in the self-contained microphone acquisition transmission line of electricity setting range of unmanned plane；The image taking resolution ratio of camera is set, the image in the self-contained camera acquisition transmission line of electricity setting range of unmanned plane is utilized；It is input to after voice signal is handled in preparatory trained first classifier, exports the first classification results；It is input to after image is handled in preparatory trained second classifier, exports the second classification results；Whether consistent first and second classification results is judged, if unanimously, it is determined that transmission line of electricity nearby has bird, driving unmanned plane executes setting flare maneuver, drives to birds；If it is inconsistent, returning to the sound collection frequency step of setting microphone and the image taking resolution ratio step of setting camera simultaneously.
Description
Technical field
This disclosure relates to the transmission line of electricity bird repellent method based on unmanned plane inspection.
Background technique
The statement of this part is only to refer to background technique relevant to the disclosure, not necessarily constitutes the prior art.
Understand according to inventor, there are following technical problems for the prior art:
Existing transmission line of electricity bird repellent method does not account for benefit only from the viewpoint of scarecrow device, scarer
Bird repellent is realized with unmanned plane, moreover, it is limited not solve the self-contained service time of battery of unmanned plane, how to increase unmanned plane
Under the premise of battery capacity is constant, realizes and the birds detection of transmission line of electricity and birds is driven for a long time using unmanned plane,
And the existing image that is only used only carries out birds to know the case where method for distinguishing is in the presence of wrong report, comes to the work people of inspection center
Bring unnecessary puzzlement.
Summary of the invention
In order to solve the deficiencies in the prior art, present disclose provides a kind of transmission line of electricity bird repellent sides based on unmanned plane inspection
Method, it is accurate with testing result, unmanned plane single trip use energy is saved, the unmanned plane single trip use time is improved.
Present disclose provides the transmission line of electricity bird repellent methods based on unmanned plane inspection；
Transmission line of electricity bird repellent method based on unmanned plane inspection, comprising:
The sound collection frequency of microphone is set, sets model using the self-contained microphone acquisition transmission line of electricity of unmanned plane
Enclose interior voice signal；
The image taking resolution ratio of camera is set, is set using the self-contained camera acquisition transmission line of electricity of unmanned plane
Image in range；
It is input to after voice signal is handled in preparatory trained first classifier, exports the first classification results；
It is input to after image is handled in preparatory trained second classifier, exports the second classification results；
Whether consistent first and second classification results is judged, if unanimously, it is determined that transmission line of electricity nearby has bird, driven
Unmanned plane executes setting flare maneuver, drives to birds；If it is inconsistent, the sound for returning to setting microphone simultaneously is adopted
Collect frequency step and the image taking resolution ratio step of camera is set.
As a kind of possible implementation, the concrete mode of the sound collection frequency of the setting microphone are as follows:
When being arranged for the first time, it is set as low frequency acquisition；When being arranged for second, it is set as middle frequency acquisition, in third
When secondary setting, it is set as high frequency acquisition, wherein low frequency acquisition is less than middle frequency acquisition；Middle frequency acquisition is less than high acquisition
Frequency.
As a kind of possible implementation, the specific steps of the image taking resolution ratio of the setting camera are as follows:
When being arranged for the first time, it is set as low resolution；When being arranged for second, it is set as intermediate-resolution, is set for the third time
When setting, it is set as high-resolution, wherein low resolution is less than intermediate-resolution, and intermediate-resolution is less than high-resolution.
As a kind of possible implementation, trained first classification in advance is input to after voice signal is handled
In device, the first classification results specific steps are exported are as follows:
Feature extraction is carried out to actual voice signal, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, is obtained fused
Actual sound feature；
Fused actual sound feature is input in preparatory trained first classifier, output the first classification knot
Fruit: present sound signals are the voice signal of birds, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are not
It is the voice signal of birds, i.e., transmission line of electricity does not have birds presence nearby.
As a kind of possible implementation, preparatory trained second classifier is input to after image is handled
In, export the second classification results specific steps are as follows:
Feature extraction, texture feature extraction, pixel characteristic and gray feature are carried out to actual image；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtained fused
Real image feature；
Fused real image feature is input in preparatory trained second classifier, output the second classification knot
Fruit: there are birds in present image, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, birds are not present in present image, i.e., it is defeated
Electric line does not have birds presence nearby.
As a kind of possible implementation, unmanned plane is driven to execute setting flare maneuver, birds drive specific
Flare maneuver, comprising: dive, spiral, along transmission line of electricity upper straight traveling one of or any number of combined mode.
As a kind of possible implementation, the training step of the first classifier are as follows:
Be arranged unmanned plane microphone voice signal use frequency for it is low, in or high sample frequency；
The voice signal of birds and the voice signal of non-birds are acquired using unmanned plane microphone；
The voice signal of acquisition is subjected to feature extraction, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, is obtained fused
Training sound characteristic；
By the fused trained sound characteristic of birds and the fused trained sound characteristic of non-birds, it is input to first
In classifier, the first classifier is convolutional neural networks, and export the first classification results: present sound signals are that the sound of birds is believed
Number, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are the voice signals of non-birds, i.e. transmission line of electricity is attached
Closely exist without birds；When nicety of grading reaches given threshold, deconditioning obtains trained first classifier.
As a kind of possible implementation, the training step of the second classifier are as follows:
Be arranged unmanned plane camera image taking resolution ratio be it is low, in or high-resolution；
Using unmanned plane camera to there are the transmission line of electricity image of birds and there is no the transmission line of electricity images of birds
It is acquired；
Acquisition there is no the transmission line of electricity image of birds is carried out there are the transmission line of electricity image of birds and by feature and mentioned
It takes, texture feature extraction, pixel characteristic and gray feature；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtained fused
Training image feature；
The fused training image feature of birds will be present and there is no the fused training image feature of birds is equal
It being input in the second classifier, the second classifier is neural network, export the second classification results: there are birds in present image,
I.e. transmission line of electricity is nearby with the presence of birds；Alternatively, birds are not present in present image, i.e., transmission line of electricity is nearby deposited without birds
?；When nicety of grading reaches given threshold, deconditioning obtains trained second classifier.
Compared with prior art, the beneficial effect of the disclosure is:
First, it merges, considers by way of to temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing
Multiple aspect ratios consider that single feature is more able to achieve the exact classification to voice signal；
Second, by being merged in such a way that feature is concatenated to textural characteristics, pixel characteristic and gray feature, consider
Multiple aspect ratios consider that single feature is more able to achieve the accurate identification to image；
Third learns whether transmission line of electricity nearby has birds by judging whether first and second classification results is consistent, can
Effectively to avoid judging by accident, erroneous judgement caused by voice signal inaccuracy or image procossing inaccuracy is avoided；
4th, by using low resolution for the first time, and intermediate-resolution and high-resolution are used for the second time, for the third time, a side
Face is the electricity in order to save unmanned plane, and still further aspect is realized more acurrate also for the acquisition precision improved below several times
Recognition result.
5th, by using low frequency acquisition when being arranged for the first time, and second, use for the third time in frequency acquisition and height
Frequency acquisition is on the one hand to save the electricity of unmanned plane, and second aspect is also for the essence for improving subsequent acquisition signal
Degree realizes preferably identification.
Testing result is accurate, saves unmanned plane single trip use energy, improves the unmanned plane single trip use time.
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are not constituted an undue limitation on the present application for explaining the application.
Fig. 1 is the entire flow figure of one or more embodiments；
Fig. 2 is the first classifier training flow chart of one or more embodiments；
Fig. 3 is the second classifier training flow chart of one or more embodiments.
Specific embodiment
It is noted that described further below be all exemplary, it is intended to provide further instruction to the application.Unless another
It indicates, all technical and scientific terms used herein has usual with the application person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
MFCC (Mel Frequency Cepstral Coefficents, mel-frequency cepstrum coefficient)
As shown in Figure 1, the transmission line of electricity bird repellent method based on unmanned plane inspection, comprising:
The sound collection frequency of microphone is set, sets model using the self-contained microphone acquisition transmission line of electricity of unmanned plane
Enclose interior voice signal；
The image taking resolution ratio of camera is set, is set using the self-contained camera acquisition transmission line of electricity of unmanned plane
Image in range；
It is input to after voice signal is handled in preparatory trained first classifier, exports the first classification results；
It is input to after image is handled in preparatory trained second classifier, exports the second classification results；
Whether consistent first and second classification results is judged, if unanimously, it is determined that transmission line of electricity nearby has bird, driven
Unmanned plane executes setting flare maneuver, drives to birds；If it is inconsistent, the sound for returning to setting microphone simultaneously is adopted
Collect frequency step and the image taking resolution ratio step of camera is set.
Optionally, the concrete mode of the sound collection frequency of the setting microphone are as follows:
When being arranged for the first time, it is set as low frequency acquisition；When being arranged for second, it is set as middle frequency acquisition, in third
When secondary setting, it is set as high frequency acquisition, wherein low frequency acquisition is less than middle frequency acquisition；Middle frequency acquisition is less than high acquisition
Frequency.
Such as: low sample frequency is 1-2kHz；Middle sample frequency is 3-4kHz；High sample frequency is 5-6kHz；
Features described above the utility model has the advantages why use low frequency acquisition when being arranged for the first time, and second, adopt for the third time
With middle frequency acquisition and high frequency acquisition, it is on the one hand to save the electricity of unmanned plane, after second aspect is also for improve
The precision of the acquisition signal in face, realizes preferably identification.
Optionally, the specific steps of the image taking resolution ratio of the setting camera are as follows:
When being arranged for the first time, it is set as low resolution；When being arranged for second, it is set as intermediate-resolution, is set for the third time
When setting, it is set as high-resolution, wherein low resolution is less than intermediate-resolution, and intermediate-resolution is less than high-resolution.
For example, low resolution is 640 × 480；Intermediate-resolution is 960 × 720；High-resolution is 1024 × 768.
Why the beneficial effect of features described above uses low resolution when being arranged for the first time, and uses for the second time, for the third time
Intermediate-resolution and high-resolution are on the one hand to save the electricity of unmanned plane, and still further aspect is several below also for improving
Secondary acquisition precision realizes more accurate recognition result.
Optionally, it is input to after voice signal being handled in preparatory trained first classifier, exports first point
Class result specific steps are as follows:
Feature extraction is carried out to actual voice signal, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, is obtained fused
Actual sound feature；
Fused actual sound feature is input in preparatory trained first classifier, output the first classification knot
Fruit: present sound signals are the voice signal of birds, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are not
It is the voice signal of birds, i.e., transmission line of electricity does not have birds presence nearby.
Optionally, it is input to after image being handled in preparatory trained second classifier, output the second classification knot
Fruit specific steps are as follows:
Feature extraction, texture feature extraction, pixel characteristic and gray feature are carried out to actual image；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtained fused
Real image feature；
Fused real image feature is input in preparatory trained second classifier, output the second classification knot
Fruit: there are birds in present image, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, birds are not present in present image, i.e., it is defeated
Electric line does not have birds presence nearby.
Optionally, driving unmanned plane executes setting flare maneuver, carries out driving specific flare maneuver to birds, comprising:
It dives, spiral, along one of transmission line of electricity upper straight traveling or any number of combined mode.
As a kind of possible implementation, as shown in Fig. 2, the training step of the first classifier are as follows:
Be arranged unmanned plane microphone voice signal use frequency for it is low, in or high sample frequency；
The voice signal of birds and the voice signal of non-birds are acquired using unmanned plane microphone；
The voice signal of acquisition is subjected to feature extraction, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, is obtained fused
Training sound characteristic；
By the fused trained sound characteristic of birds and the fused trained sound characteristic of non-birds, it is input to first
In classifier, the first classifier is convolutional neural networks, and export the first classification results: present sound signals are that the sound of birds is believed
Number, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are the voice signals of non-birds, i.e. transmission line of electricity is attached
Closely exist without birds；When nicety of grading reaches given threshold, deconditioning obtains trained first classifier.
Optionally, as shown in figure 3, the training step of the second classifier are as follows:
Be arranged unmanned plane camera image taking resolution ratio be it is low, in or high-resolution；
Using unmanned plane camera to there are the transmission line of electricity image of birds and there is no the transmission line of electricity images of birds
It is acquired；
Acquisition there is no the transmission line of electricity image of birds is carried out there are the transmission line of electricity image of birds and by feature and mentioned
It takes, texture feature extraction, pixel characteristic and gray feature；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtained fused
Training image feature；
The fused training image feature of birds will be present and there is no the fused training image feature of birds is equal
It being input in the second classifier, the second classifier is neural network, export the second classification results: there are birds in present image,
I.e. transmission line of electricity is nearby with the presence of birds；Alternatively, birds are not present in present image, i.e., transmission line of electricity is nearby deposited without birds
?；When nicety of grading reaches given threshold, deconditioning obtains trained second classifier.
The foregoing is merely preferred embodiment of the present application, are not intended to limit this application, for the skill of this field
For art personnel, various changes and changes are possible in this application.Within the spirit and principles of this application, made any to repair
Change, equivalent replacement, improvement etc., should be included within the scope of protection of this application.
Claims (8)
1. the transmission line of electricity bird repellent method based on unmanned plane inspection, characterized in that include:
The sound collection frequency of microphone is set, using in the self-contained microphone acquisition transmission line of electricity setting range of unmanned plane
Voice signal；
The image taking resolution ratio of camera is set, acquires transmission line of electricity setting range using the self-contained camera of unmanned plane
Interior image；
It is input to after voice signal is handled in preparatory trained first classifier, exports the first classification results；
It is input to after image is handled in preparatory trained second classifier, exports the second classification results；
Whether consistent first and second classification results is judged, if unanimously, it is determined that transmission line of electricity nearby has bird, driving nobody
Machine executes setting flare maneuver, drives to birds；If it is inconsistent, returning to the sound collection frequency of setting microphone simultaneously
The image taking resolution ratio step of rate step and setting camera.
2. the method as described in claim 1, characterized in that the concrete mode of the sound collection frequency of the setting microphone
Are as follows:
When being arranged for the first time, it is set as low frequency acquisition；When being arranged for second, it is set as middle frequency acquisition, is set for the third time
When setting, it is set as high frequency acquisition, wherein low frequency acquisition is less than middle frequency acquisition；Middle frequency acquisition is less than high frequency acquisition.
3. the method as described in claim 1, characterized in that the specific steps of the image taking resolution ratio of the setting camera
Are as follows:
When being arranged for the first time, it is set as low resolution；When being arranged for second, it is set as intermediate-resolution, when third time is arranged,
It is set as high-resolution, wherein low resolution is less than intermediate-resolution, and intermediate-resolution is less than high-resolution.
4. the method as described in claim 1, characterized in that be input in advance trained the after being handled voice signal
In one classifier, the first classification results specific steps are exported are as follows:
Feature extraction is carried out to actual voice signal, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, fused reality is obtained
Sound characteristic；
Fused actual sound feature is input in preparatory trained first classifier, the first classification results are exported: when
Preceding voice signal is the voice signal of birds, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are not birds
Voice signal, i.e. transmission line of electricity nearby do not have a birds presence.
5. the method as described in claim 1, characterized in that be input to trained second point in advance after being handled image
In class device, the second classification results specific steps are exported are as follows:
Feature extraction, texture feature extraction, pixel characteristic and gray feature are carried out to actual image；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtain fused reality
Characteristics of image；
Fused real image feature is input in preparatory trained second classifier, the second classification results are exported: when
There are birds in preceding image, i.e., transmission line of electricity is nearby with the presence of birds；Alternatively, birds, i.e. power transmission line are not present in present image
Road does not have birds presence nearby.
6. the method as described in claim 1, characterized in that driving unmanned plane executes setting flare maneuver, drives to birds
Catch up with specific flare maneuver, comprising: dive, spiral, along one of transmission line of electricity upper straight traveling or any number of combined mode.
7. the method as described in claim 1, characterized in that the training step of the first classifier are as follows:
Be arranged unmanned plane microphone voice signal use frequency for it is low, in or high sample frequency；
The voice signal of birds and the voice signal of non-birds are acquired using unmanned plane microphone；
The voice signal of acquisition is subjected to feature extraction, extracts temporal signatures, frequency domain character and MFCC feature；
The mode of temporal signatures, frequency domain character and MFCC characteristic use characteristic weighing is merged, fused training is obtained
Sound characteristic；
By the fused trained sound characteristic of birds and the fused trained sound characteristic of non-birds, it is input to the first classification
In device, the first classifier is convolutional neural networks, and export the first classification results: present sound signals are the voice signal of birds,
I.e. transmission line of electricity is nearby with the presence of birds；Alternatively, present sound signals are the voice signals of non-birds, i.e., transmission line of electricity does not have nearby
With the presence of birds；When nicety of grading reaches given threshold, deconditioning obtains trained first classifier.
8. the method as described in claim 1, characterized in that the training step of the second classifier are as follows:
Be arranged unmanned plane camera image taking resolution ratio be it is low, in or high-resolution；
Using unmanned plane camera to there are the transmission line of electricity image of birds and there is no the progress of the transmission line of electricity image of birds
Acquisition；
To acquisition is there are the transmission line of electricity image of birds and there is no the transmission line of electricity images of birds to carry out feature extraction,
Texture feature extraction, pixel characteristic and gray feature；
Textural characteristics, pixel characteristic and gray feature are merged in such a way that feature is concatenated, obtain fused training
Characteristics of image；
The fused training image feature of birds will be present and there is no the fused training image features of birds to input
Into the second classifier, the second classifier is neural network, and export the second classification results: there are birds in present image, i.e., defeated
Electric line is nearby with the presence of birds；Alternatively, birds are not present in present image, i.e., transmission line of electricity does not have birds presence nearby；When
When nicety of grading reaches given threshold, deconditioning obtains trained second classifier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910045339.6A CN109902712B (en) | 2019-01-17 | 2019-01-17 | Unmanned aerial vehicle inspection-based bird repelling method for power transmission line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910045339.6A CN109902712B (en) | 2019-01-17 | 2019-01-17 | Unmanned aerial vehicle inspection-based bird repelling method for power transmission line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109902712A true CN109902712A (en) | 2019-06-18 |
| CN109902712B CN109902712B (en) | 2021-04-16 |
Family
ID=66943912
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910045339.6A Active CN109902712B (en) | 2019-01-17 | 2019-01-17 | Unmanned aerial vehicle inspection-based bird repelling method for power transmission line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109902712B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110367239A (en) * | 2019-08-26 | 2019-10-25 | 华北电力大学（保定） | A kind of bird trouble on transmission line control method based on Internet of Things |
| CN110555460A (en) * | 2019-07-31 | 2019-12-10 | 国网江苏省电力有限公司 | Image slice-based bird detection method for power transmission line at mobile terminal |
| CN110679586A (en) * | 2019-09-30 | 2020-01-14 | 深圳供电局有限公司 | Bird repelling method and system for power transmission network and computer readable storage medium |
| CN113889124A (en) * | 2021-11-08 | 2022-01-04 | 国网山东省电力公司乳山市供电公司 | Power transmission line bird prevention platform based on voice recognition |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010023253A1 (en) * | 2008-08-28 | 2010-03-04 | Sa Speir Aviation Limited | A bird collision avoidance system |
| CN102288613A (en) * | 2011-05-11 | 2011-12-21 | 北京科技大学 | Surface defect detecting method for fusing grey and depth information |
| CN203692294U (en) * | 2014-02-12 | 2014-07-09 | 冯伟 | Ultrasonic bird repelling unmanned aerial vehicle |
| CN205196824U (en) * | 2015-12-28 | 2016-05-04 | 国家电网公司 | Intelligent electric power transmission line high tower drives bird system |
| CN106094867A (en) * | 2016-08-01 | 2016-11-09 | 深圳供电局有限公司 | Multifunctional power equipment inspection unmanned aerial vehicle |
| CN106297770A (en) * | 2016-08-04 | 2017-01-04 | 杭州电子科技大学 | The natural environment sound identification method extracted based on time-frequency domain statistical nature |
| CN106259288A (en) * | 2016-08-09 | 2017-01-04 | 广州大气候农业科技有限公司 | Drive the method for birds, server and information collecting device |
| CN205942444U (en) * | 2016-08-01 | 2017-02-08 | 深圳供电局有限公司 | Multifunctional power equipment inspection unmanned aerial vehicle |
| CN106826846A (en) * | 2017-01-06 | 2017-06-13 | 南京赫曼机器人自动化有限公司 | The intellect service robot and method driven based on abnormal sound and image event |
| CN107969417A (en) * | 2017-12-16 | 2018-05-01 | 佛山市神风航空科技有限公司 | A kind of bird-repeller system |
| CN108398689A (en) * | 2018-01-26 | 2018-08-14 | 广东容祺智能科技有限公司 | A kind of birds identification guide device and its bootstrap technique based on unmanned plane |
| CN108924483A (en) * | 2018-06-27 | 2018-11-30 | 南京朴厚生态科技有限公司 | A kind of automatic monitoring system and method for the field animal based on depth learning technology |
| CN108985377A (en) * | 2018-07-18 | 2018-12-11 | 太原理工大学 | A kind of image high-level semantics recognition methods of the multiple features fusion based on deep layer network |
| CN109117732A (en) * | 2018-07-16 | 2019-01-01 | 国网江西省电力有限公司电力科学研究院 | A kind of transmission line of electricity relates to the identification of bird failure bird kind figure sound and control method |
-
2019
- 2019-01-17 CN CN201910045339.6A patent/CN109902712B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010023253A1 (en) * | 2008-08-28 | 2010-03-04 | Sa Speir Aviation Limited | A bird collision avoidance system |
| CN102288613A (en) * | 2011-05-11 | 2011-12-21 | 北京科技大学 | Surface defect detecting method for fusing grey and depth information |
| CN203692294U (en) * | 2014-02-12 | 2014-07-09 | 冯伟 | Ultrasonic bird repelling unmanned aerial vehicle |
| CN205196824U (en) * | 2015-12-28 | 2016-05-04 | 国家电网公司 | Intelligent electric power transmission line high tower drives bird system |
| CN205942444U (en) * | 2016-08-01 | 2017-02-08 | 深圳供电局有限公司 | Multifunctional power equipment inspection unmanned aerial vehicle |
| CN106094867A (en) * | 2016-08-01 | 2016-11-09 | 深圳供电局有限公司 | Multifunctional power equipment inspection unmanned aerial vehicle |
| CN106297770A (en) * | 2016-08-04 | 2017-01-04 | 杭州电子科技大学 | The natural environment sound identification method extracted based on time-frequency domain statistical nature |
| CN106259288A (en) * | 2016-08-09 | 2017-01-04 | 广州大气候农业科技有限公司 | Drive the method for birds, server and information collecting device |
| CN106826846A (en) * | 2017-01-06 | 2017-06-13 | 南京赫曼机器人自动化有限公司 | The intellect service robot and method driven based on abnormal sound and image event |
| CN107969417A (en) * | 2017-12-16 | 2018-05-01 | 佛山市神风航空科技有限公司 | A kind of bird-repeller system |
| CN108398689A (en) * | 2018-01-26 | 2018-08-14 | 广东容祺智能科技有限公司 | A kind of birds identification guide device and its bootstrap technique based on unmanned plane |
| CN108924483A (en) * | 2018-06-27 | 2018-11-30 | 南京朴厚生态科技有限公司 | A kind of automatic monitoring system and method for the field animal based on depth learning technology |
| CN109117732A (en) * | 2018-07-16 | 2019-01-01 | 国网江西省电力有限公司电力科学研究院 | A kind of transmission line of electricity relates to the identification of bird failure bird kind figure sound and control method |
| CN108985377A (en) * | 2018-07-18 | 2018-12-11 | 太原理工大学 | A kind of image high-level semantics recognition methods of the multiple features fusion based on deep layer network |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555460A (en) * | 2019-07-31 | 2019-12-10 | 国网江苏省电力有限公司 | Image slice-based bird detection method for power transmission line at mobile terminal |
| CN110367239A (en) * | 2019-08-26 | 2019-10-25 | 华北电力大学（保定） | A kind of bird trouble on transmission line control method based on Internet of Things |
| CN110679586A (en) * | 2019-09-30 | 2020-01-14 | 深圳供电局有限公司 | Bird repelling method and system for power transmission network and computer readable storage medium |
| CN110679586B (en) * | 2019-09-30 | 2022-08-30 | 深圳供电局有限公司 | Bird repelling method and system for power transmission network and computer readable storage medium |
| CN113889124A (en) * | 2021-11-08 | 2022-01-04 | 国网山东省电力公司乳山市供电公司 | Power transmission line bird prevention platform based on voice recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109902712B (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902712A (en) | Transmission line of electricity bird repellent method based on unmanned plane inspection | |
| CN110261394A (en) | Online fan blade damages real-time diagnosis system and method | |
| CN105841961A (en) | Bearing fault diagnosis method based on Morlet wavelet transformation and convolutional neural network | |
| CN108898085A (en) | Intelligent road disease detection method based on mobile phone video | |
| CN111832398B (en) | Unmanned aerial vehicle image distribution line pole tower ground wire broken strand image detection method | |
| CN106846803A (en) | Traffic incidents detection device and method based on audio | |
| CN109472284A (en) | A kind of battery core defect classification method based on zero sample learning of unbiased insertion | |
| CN102201236A (en) | Speaker recognition method combining Gaussian mixture model and quantum neural network | |
| CN212787168U (en) | Novel transmission line unmanned aerial vehicle intelligence drive bird device | |
| CN103618638B (en) | The method of assessment power telecom network maintenance solution | |
| CN106992011A (en) | Engineering machinery sound identification method based on MF PLPCC features | |
| CN110503070A (en) | Traffic automation monitoring method based on Aerial Images object detection process technology | |
| CN112304614A (en) | End-to-end rolling bearing intelligent fault diagnosis method adopting multi-attention machine system | |
| CN109978847A (en) | Drag-line housing disease automatic identifying method based on transfer learning Yu drag-line robot | |
| CN106203437B (en) | Individual driving behavior recognition methods and device | |
| CN110490428A (en) | Job of air traffic control method for evaluating quality and relevant apparatus | |
| CN109473119A (en) | A kind of acoustic target event-monitoring method | |
| CN110533103A (en) | A kind of lightweight wisp object detection method and system | |
| CN113252701B (en) | Cloud edge cooperation-based power transmission line insulator self-explosion defect detection system and method | |
| CN110532925A (en) | Driver Fatigue Detection based on space-time diagram convolutional network | |
| CN115758258B (en) | Traction transformer fault diagnosis method based on vibration signals and domain knowledge | |
| CN101833843A (en) | Monitoring system based on voiceprint authentication | |
| CN110652726A (en) | Game auxiliary system based on image recognition and audio recognition | |
| CN114581799A (en) | Unmanned aerial vehicle small target detection method based on multi-scale feature fusion | |
| CN109684990A (en) | A kind of behavioral value method of making a phone call based on video |
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 |