CN110084094A - A kind of unmanned plane target identification classification method based on deep learning - Google Patents
A kind of unmanned plane target identification classification method based on deep learning Download PDFInfo
- Publication number
- CN110084094A CN110084094A CN201910168705.7A CN201910168705A CN110084094A CN 110084094 A CN110084094 A CN 110084094A CN 201910168705 A CN201910168705 A CN 201910168705A CN 110084094 A CN110084094 A CN 110084094A
- Authority
- CN
- China
- Prior art keywords
- signal
- unmanned plane
- deep learning
- data
- target
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/12—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation operating with electromagnetic waves
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Health & Medical Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Image Analysis (AREA)
Abstract
The unmanned plane target identification classification method based on deep learning that the invention discloses a kind of, specific processing step include: (1) by radio monitoring means, collect the link signal of Small Civil unmanned plane;(2) link signal is sent to data processing centre, carries out Signal Pretreatment, obtains the temporal signatures and frequency domain character of signal;(3) temporal signatures of signal and frequency domain character are input in deep learning network and are calculated, output obtains the target identification classification results of signal;(4) signal data collected and recognition result are saved to database, if newfound target category, signal identification library is added to after manual confirmation.Unmanned plane target identification classification method provided by the present invention, deep learning algorithm is applied to unmanned aerial vehicle radio signal monitoring field, to break through the application bottleneck of classical wireless telecommunication analysis method, unmanned plane signal monitoring and identification in low altitude safe protection are solved the problems, such as practically.
Description
Technical field
The present invention relates to a kind of unmanned aerial vehicle radio signal monitoring technical field more particularly to a kind of civilian unmanned planes
Signal identification and classification method.
Background technique
Currently, unmanned plane is used widely in various fields, such as agricultural plant protection, geographical mapping, forest fire protection, electricity
Power inspection etc..In addition, unmanned plane is also used for some unlawful activities, terrorist activity, airport " black to fly " thing of explosive are such as installed additional
Part etc..It is threatened as it can be seen that Small Civil unmanned plane brings serious low altitude safe, is all one in worldwide and urgently solves
Problem certainly.In order to realize the inspection identification and effective control of unmanned plane, domestic and international many scientific research institutions and enterprise are exerting always
Power explores the implementation of unmanned plane signal monitoring identification and management game, but there is presently no form mature effective solution side
Case.
UAV system is to be communicated by radio signal with ground system, including be remotely controlled, count biography and figure communication number,
Wherein remote control module is mainly used for the transmission of operational order, and digital transmission module is mainly used for sending unmanned plane during flying state, such as
The information such as position, electricity, figure transmission module are to image data captured by camera on terrestrial system passback unmanned plane.Cause
This, can find these communications link signals of unmanned plane by radio signal monitoring means, and carry out target identification and determine
Position, and then realize effective supervision of unmanned plane.
For urban area complex electromagnetic environment, Small Civil unmanned plane is carried out using traditional radio monitoring technology
Investigation identification, application effect are extremely limited.Current radio monitoring technology, typical processing step include: that signal inputs, is pre-
Processing, feature extraction, Classification and Identification, wherein extremely important step is feature extraction, by the feature for extracting radio signal
Then parameter, such as centre frequency, bandwidth further analyze the statistical natures such as the modulation type for obtaining signal by these parameters
Parameter finally carries out target identification according to the classification difference between these features.These traditional radio signal processing methods
When noise is relatively high, it can detecting finds a certain range of unmanned plane target.However, the electromagnetic environment in practice is very
Complexity, various radio signals and noise mix, and signal-to-noise ratio is relatively low, are difficult to correctly extract these in this case
Characteristic parameter, so that target identification effect is unstable.
Summary of the invention
Technical problem to be solved by the present invention lies in: the signal-to-noise ratio of traditional radio monitoring technology in practice is low
In the case of be difficult to correctly extract the characteristic parameter of Small Civil unmanned aerial vehicle radio signal, provide a kind of based on deep learning
Unmanned plane target identification classification method.
The present invention is to solve above-mentioned technical problem by the following technical programs, the method specifically includes following steps:
(1) by radio monitoring means, the link signal of Small Civil unmanned plane is collected;
(2) link signal is sent to data processing centre, carries out Signal Pretreatment, obtain signal when, frequency domain it is special
Sign;
(3) by signal when, frequency domain character be input in deep learning network and calculate, output obtain the target of signal
Identify classification results;
(4) signal data collected and recognition result are saved to database, if newfound target category, through people
It is added in signal identification library after work confirmation.
In the step (1), it is as follows that link signal obtains process: scanning unmanned plane work frequency by data-link detector
Section, when unmanned plane target is close to prevention and control region, data-link detector can be found that target, receives the radio signal of target,
It completes to acquire the data of unmanned plane link signal.
In the step (2), Signal Pretreatment includes resampling, spectrum analysis, obtains the time domain of unmanned plane signal
Characteristic and frequency domain character data.
In the step (3), the method for constructing deep learning network is as follows:
(31) deep learning algorithm of neural network framework is constructed;
(32) data in database are obtained into data set through manual sorting and mark, and is classified as training set and test
Collection;
(33) training of deep learning algorithm is carried out, by preceding to the continuous iteration derived with backpropagation, so that target is known
Other accuracy reaches desired value, terminates training, and retain each layer neural network parameter, obtains unmanned plane target identification deep learning
Algorithm model.
In the step (31), deep learning algorithm of neural network includes input layer, three layers of hidden layer, output layer;Wherein,
The temporal signatures data and frequency domain character data merge the input layer as deep learning algorithm later, i.e., the described input layer is total
There is n+N neuron, send the neural network to three layers of hidden layer to calculate input layer, be automatically performed comprehensive information
After excavation, send to the output layer;
In the step (31), deep learning algorithm of neural network include input layer, stack self-encoding encoder, full articulamentum,
Output layer;Wherein, the temporal signatures data and frequency domain character data respectively pass through two layers of the stack self-encoding encoder, respectively
It carries out the information excavating of time domain or frequency domain and then merges, be input to the full articulamentum, finally send to the output
Layer.
In the step (3), the target identification classification results are as follows: being exported according to output layer is the general of each target category
Rate, the maximum classification of select probability are the target identification result of the unmanned plane signal;The wherein each neuron of the output layer
Each known unmanned plane classification and " unknown " classification in target identification library are respectively corresponded, i.e., if in identification library altogether
There is T target category, then the neural network output layer neuron number is T+1.
In the step (4), signal data collected and processing result are saved to database, following son is specifically included
Step:
(41) signal data and result are stored in the database, carries out data accumulation, to expand algorithm training number
According to collection, the iteration that can be used for later algorithm model updates;
(42) if algorithm recognition result is " unknown ", by artificial nucleus couple, after confirming the true classification of unmanned plane target,
Database is modified, " unknown " is changed to true target category;
(43) it if unmanned plane target is newfound classification, also needs for corresponding signal data and target category to be added to
In signal identification library.
The present invention is had the advantage that compared with prior art the present invention is based on deep learning, designs and Implements unmanned chain
Road signal target identifies that sorting algorithm avoids characteristic parameter extraction and selection compared to classical wireless telecommunication Processing Algorithm
The step of, algorithm noiseproof feature is improved, the practicability in practical application is enhanced;Meanwhile in order to excavate signal spy comprehensively
Sign, using the deep learning target identification sorting algorithm based on time-domain and frequency-domain characteristic binding, time domain and frequency-region signal are contained
The information of reflection signal characteristic enough, frequency-region signal and time-domain signal are combined, then can further increase the target of algorithm
Recognition accuracy increases unmanned plane target monitoring range and investigation efficiency.
Detailed description of the invention
Fig. 1 is Whole Work Flow schematic diagram;
Fig. 2 is the deep learning algorithm architecture diagram of embodiment 1;
Fig. 3 is the deep learning algorithm architecture diagram of embodiment 2.
Specific embodiment
It elaborates below to the embodiment of the present invention, the present embodiment carries out under the premise of the technical scheme of the present invention
Implement, the detailed implementation method and specific operation process are given, but protection scope of the present invention is not limited to following implementation
Example.
The present invention is the classification of Small Civil unmanned plane signal automatic detection and identification, the frequency hopping emitted with unmanned controller
Signal, unmanned plane figure communication number are used as target, can it be detected and be found rapidly, by data processing, realize unmanned plane
The intelligent recognition and classification of target.
Embodiment 1
As shown in Figure 1, the Whole Work Flow of unmanned plane signal identification classification is as follows:
The radio signal that unmanned plane is sent carries out data acquisition and is sent at data after detector detection discovery
Reason center obtains unmanned plane target identification classification results, analysis is tied after Signal Pretreatment and the analysis of deep learning algorithm
Fruit is stored in database together with acquisition data.
As shown in Fig. 2, a kind of deep learning algorithm architecture diagram for unmanned plane target identification classification is given, by inputting
Layer, three layers of hidden layer, output layer constitute deep neural network algorithm, and the unmanned plane signal detected is carried out resampling and Fu
In the pretreatment such as leaf transformation, respectively obtaining time domain waveform, (data length n) and spectral feature data (data length N) are closed
And later, while the input layer as deep learning algorithm, i.e. input layer share n+N neuron.When, frequency domain data is by each
The calculating of layer neural network, is automatically performed after comprehensive information excavating, and output layer is the probability of each target category, select probability
Maximum classification is the target identification result of the unmanned plane signal.Wherein, each neuron of output layer respectively corresponds target knowledge
Each known unmanned plane classification and " unknown " classification in other library, i.e., if sharing T target category in identification library,
Then neural network output layer neuron number is T+1.
Storage and data accumulation are carried out finally, saving signal data collected and processing result to database, to expand
Algorithm training dataset is filled, the iteration that can be used for later algorithm model updates.If algorithm recognition result is " unknown ", pass through people
Work verification after confirming the true classification of unmanned plane target, modifies database, " unknown " is changed to true target category.Nobody
When machine target is newfound classification, also need for corresponding signal data and target category to be added in signal identification library.
Embodiment 2
As shown in figure 3, this gives another deep learning algorithm framves for unmanned plane target identification classification
Composition constitutes deep neural network algorithm by input layer, stack self-encoding encoder, full articulamentum, output layer, and wherein stack is self-editing
Code device includes two layers of self-encoding encoder network.The unmanned plane signal detected is subjected to the pretreatment such as resampling and Fourier transformation,
Obtain time domain waveform (data length n) and spectral feature data (data length N).Time domain and frequency spectrum data respectively pass through two
The stack self-encoding encoder of layer carries out the information excavating of time domain or frequency domain respectively and then merges, is input to full articulamentum,
Most Zhongdao output layer.Output layer is the probability of each target category, and the maximum classification of select probability is the mesh of the unmanned plane signal
Mark recognition result.Wherein, each neuron of output layer respectively corresponds each known unmanned plane classification in target identification library, and
One " unknown " classification, i.e., if sharing T target category in identification library, neural network output layer neuron number is T+1
It is a.
Other embodiments and embodiment 1 are identical.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (8)
1. a kind of unmanned plane target identification classification method based on deep learning, which is characterized in that specific steps include:
(1) by radio monitoring means, the link signal of Small Civil unmanned plane is collected;
(2) link signal is sent to data processing centre, carries out Signal Pretreatment, obtain signal temporal signatures data and
Frequency domain character data;
(3) the temporal signatures data of signal and frequency domain character data are input in deep learning network and are calculated, exported
To the target identification classification results of signal;
(4) signal data collected and recognition result are saved to database, if newfound target category, through artificial true
It is added to after recognizing in signal identification library.
2. a kind of unmanned plane target identification classification method based on deep learning according to claim 1, which is characterized in that
In the step (1), it is as follows that link signal obtains process: unmanned plane working frequency range is scanned by data-link detector, when nobody
When machine target is close to prevention and control region, data-link detector can be found that target, receives the radio signal of target, completes to nobody
The data of machine link signal acquire.
3. a kind of unmanned plane target identification classification method based on deep learning according to claim 1, which is characterized in that
In the step (2), Signal Pretreatment includes resampling, spectrum analysis, obtains the temporal signatures data of unmanned plane signal
With frequency domain character data.
4. a kind of unmanned plane target identification classification method based on deep learning according to claim 1, which is characterized in that
In the step (3), the method for constructing deep learning network is as follows:
(31) deep learning algorithm of neural network framework is constructed;
(32) data in database are obtained into data set through manual sorting and mark, and is classified as training set and test set;
(33) training of deep learning algorithm is carried out, by preceding to the continuous iteration derived with backpropagation, so that target identification is just
True rate reaches desired value, terminates training, and retain each layer neural network parameter, obtains unmanned plane target identification deep learning algorithm
Model.
5. a kind of unmanned plane target identification classification method based on deep learning according to claim 4, which is characterized in that
In the step (31), deep learning algorithm of neural network includes input layer, three layers of hidden layer, output layer;Wherein, the time domain
Characteristic and frequency domain character data merge the input layer as deep learning algorithm later, i.e., the described input layer shares n+N
Input layer is sent the neural network to three layers of hidden layer to calculate by neuron, be automatically performed comprehensive information excavating it
Afterwards, it send to the output layer.
6. a kind of unmanned plane target identification classification method based on deep learning according to claim 4, which is characterized in that
In the step (31), deep learning algorithm of neural network includes input layer, stack self-encoding encoder, full articulamentum, output layer;Its
In, the temporal signatures data and frequency domain character data respectively pass through two layers of the stack self-encoding encoder, carry out time domain respectively
Or frequency domain information excavating and then merge, be input to the full articulamentum, finally send to the output layer.
7. a kind of unmanned plane target identification classification method based on deep learning according to claim 1, which is characterized in that
In the step (3), the target identification classification results are as follows: be the probability of each target category according to output layer output, selection is general
The maximum classification of rate is the target identification result of the unmanned plane signal;Wherein each neuron of the output layer respectively corresponds mesh
Each known unmanned plane classification and " unknown " classification in other library are identified, i.e., if sharing T target class in identification library
Not, then the neural network output layer neuron number is T+1.
8. a kind of unmanned plane target identification classification method based on deep learning according to claim 1, which is characterized in that
In the steps (4), signal data collected and processing result are saved to database, following sub-step is specifically included:
(41) signal data and result are stored in the database, carries out data accumulation, to expand algorithm training dataset,
The iteration that can be used for later algorithm model updates;
(42) if algorithm recognition result is " unknown ", pass through artificial nucleus couple, after confirming the true classification of unmanned plane target, modification
" unknown " is changed to true target category by database;
(43) it if unmanned plane target is newfound classification, also needs corresponding signal data and target category being added to signal
It identifies in library.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910168705.7A CN110084094B (en) | 2019-03-06 | 2019-03-06 | Unmanned aerial vehicle target identification and classification method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910168705.7A CN110084094B (en) | 2019-03-06 | 2019-03-06 | Unmanned aerial vehicle target identification and classification method based on deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110084094A true CN110084094A (en) | 2019-08-02 |
CN110084094B CN110084094B (en) | 2021-07-23 |
Family
ID=67413092
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910168705.7A Active CN110084094B (en) | 2019-03-06 | 2019-03-06 | Unmanned aerial vehicle target identification and classification method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110084094B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111222430A (en) * | 2019-12-27 | 2020-06-02 | 湖南华诺星空电子技术有限公司 | Unmanned aerial vehicle identification method and system based on artificial intelligence |
CN111800216A (en) * | 2020-05-29 | 2020-10-20 | 中南民族大学 | System and method for generating anti-electromagnetic waveform of unmanned aerial vehicle |
CN111817794A (en) * | 2020-05-29 | 2020-10-23 | 中南民族大学 | Multi-domain cooperative unmanned aerial vehicle detection method and system based on deep learning |
CN112163569A (en) * | 2020-10-29 | 2021-01-01 | 上海特金无线技术有限公司 | Signal detection method and device, electronic equipment and storage medium |
CN112505620A (en) * | 2021-02-06 | 2021-03-16 | 陕西山利科技发展有限责任公司 | Rotary direction finding method for unmanned aerial vehicle detection |
CN113327461A (en) * | 2021-08-03 | 2021-08-31 | 杭州海康威视数字技术股份有限公司 | Cooperative unmanned aerial vehicle detection method, device and equipment |
CN114154545A (en) * | 2021-12-07 | 2022-03-08 | 中国人民解放军32802部队 | Intelligent unmanned aerial vehicle measurement and control signal identification method under strong mutual interference condition |
CN114760172A (en) * | 2022-04-13 | 2022-07-15 | 北京博识广联科技有限公司 | Method and device for identifying radio frequency baseband comprehensive characteristic signal |
CN115308813A (en) * | 2022-10-10 | 2022-11-08 | 成都本原聚能科技有限公司 | Dual-directional antenna aircraft detection system and method based on deep learning |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105550636A (en) * | 2015-12-04 | 2016-05-04 | 中国电子科技集团公司第三研究所 | Method and device for identifying target types |
CN108280395A (en) * | 2017-12-22 | 2018-07-13 | 中国电子科技集团公司第三十研究所 | A kind of efficient identification method flying control signal to low small slow unmanned plane |
CN109088695A (en) * | 2018-07-24 | 2018-12-25 | 广州海格亚华防务科技有限公司 | A kind of unmanned plane detecting system of defense and method |
CN109343046A (en) * | 2018-09-19 | 2019-02-15 | 成都理工大学 | Radar gait recognition method based on multifrequency multiple domain deep learning |
-
2019
- 2019-03-06 CN CN201910168705.7A patent/CN110084094B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105550636A (en) * | 2015-12-04 | 2016-05-04 | 中国电子科技集团公司第三研究所 | Method and device for identifying target types |
CN108280395A (en) * | 2017-12-22 | 2018-07-13 | 中国电子科技集团公司第三十研究所 | A kind of efficient identification method flying control signal to low small slow unmanned plane |
CN109088695A (en) * | 2018-07-24 | 2018-12-25 | 广州海格亚华防务科技有限公司 | A kind of unmanned plane detecting system of defense and method |
CN109343046A (en) * | 2018-09-19 | 2019-02-15 | 成都理工大学 | Radar gait recognition method based on multifrequency multiple domain deep learning |
Non-Patent Citations (1)
Title |
---|
MING ZHANG等: "Convolutional Neural Networks for Automatic Cognitive Radio Waveform Recognition", 《IEEE ACCESS》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111222430A (en) * | 2019-12-27 | 2020-06-02 | 湖南华诺星空电子技术有限公司 | Unmanned aerial vehicle identification method and system based on artificial intelligence |
CN111800216A (en) * | 2020-05-29 | 2020-10-20 | 中南民族大学 | System and method for generating anti-electromagnetic waveform of unmanned aerial vehicle |
CN111817794A (en) * | 2020-05-29 | 2020-10-23 | 中南民族大学 | Multi-domain cooperative unmanned aerial vehicle detection method and system based on deep learning |
CN111817794B (en) * | 2020-05-29 | 2021-04-13 | 中南民族大学 | Multi-domain cooperative unmanned aerial vehicle detection method and system based on deep learning |
CN111800216B (en) * | 2020-05-29 | 2021-07-20 | 中南民族大学 | System and method for generating anti-electromagnetic waveform of black-flying unmanned aerial vehicle |
CN112163569B (en) * | 2020-10-29 | 2022-11-29 | 上海特金无线技术有限公司 | Signal detection method and device, electronic equipment and storage medium |
CN112163569A (en) * | 2020-10-29 | 2021-01-01 | 上海特金无线技术有限公司 | Signal detection method and device, electronic equipment and storage medium |
CN112505620A (en) * | 2021-02-06 | 2021-03-16 | 陕西山利科技发展有限责任公司 | Rotary direction finding method for unmanned aerial vehicle detection |
CN112505620B (en) * | 2021-02-06 | 2021-04-27 | 陕西山利科技发展有限责任公司 | Rotary direction finding method for unmanned aerial vehicle detection |
CN113327461A (en) * | 2021-08-03 | 2021-08-31 | 杭州海康威视数字技术股份有限公司 | Cooperative unmanned aerial vehicle detection method, device and equipment |
CN113327461B (en) * | 2021-08-03 | 2021-11-23 | 杭州海康威视数字技术股份有限公司 | Cooperative unmanned aerial vehicle detection method, device and equipment |
CN114154545A (en) * | 2021-12-07 | 2022-03-08 | 中国人民解放军32802部队 | Intelligent unmanned aerial vehicle measurement and control signal identification method under strong mutual interference condition |
CN114760172A (en) * | 2022-04-13 | 2022-07-15 | 北京博识广联科技有限公司 | Method and device for identifying radio frequency baseband comprehensive characteristic signal |
CN114760172B (en) * | 2022-04-13 | 2024-04-19 | 北京博识广联科技有限公司 | Method and device for identifying radio frequency baseband comprehensive characteristic signals |
CN115308813A (en) * | 2022-10-10 | 2022-11-08 | 成都本原聚能科技有限公司 | Dual-directional antenna aircraft detection system and method based on deep learning |
CN115308813B (en) * | 2022-10-10 | 2023-08-22 | 成都本原聚能科技有限公司 | Double-directional antenna unmanned aerial vehicle detection system and method based on deep learning |
Also Published As
Publication number | Publication date |
---|---|
CN110084094B (en) | 2021-07-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110084094A (en) | A kind of unmanned plane target identification classification method based on deep learning | |
CN108932480A (en) | The study of distributing optical fiber sensing signal characteristic and classification method based on 1D-CNN | |
CN105471882A (en) | Behavior characteristics-based network attack detection method and device | |
CN108718310A (en) | Multi-level attack signatures generation based on deep learning and malicious act recognition methods | |
CN106915462A (en) | Forestry pests & diseases intelligent identifying system based on multi-source image information | |
CN110213222A (en) | Network inbreak detection method based on machine learning | |
CN108710126A (en) | Automation detection expulsion goal approach and its system | |
CN108055094B (en) | Unmanned aerial vehicle manipulator frequency spectrum feature identification and positioning method | |
Chen et al. | Collaborative spectrum sensing for illegal drone detection: A deep learning-based image classification perspective | |
CN109255286A (en) | A kind of quick detection recognition method of unmanned plane optics based on YOLO deep learning network frame | |
CN106226212A (en) | EO-1 hyperion haze monitoring method based on degree of depth residual error network | |
CN111709329B (en) | Unmanned aerial vehicle measurement and control signal high-speed recognition method based on deep learning | |
CN109981212B (en) | Low-slow small prevention and control system and method based on data chain detection and countermeasures | |
CN108769047A (en) | A kind of big data risk monitoring system | |
CN105468922A (en) | Site electromagnetic environment signal real time analysis and threat evaluation method | |
CN108256759A (en) | A kind of attack of terrorism risk Temporal Simulation and intelligent analysis method | |
CN114598551A (en) | Information network security early warning system for dealing with continuous threat attack | |
CN114124420A (en) | Network flow abnormity detection method based on deep neural network | |
Lofù et al. | URANUS: Radio Frequency Tracking, Classification and Identification of Unmanned Aircraft Vehicles | |
CN109740845A (en) | A kind of AMI reliability index system | |
CN102622345B (en) | High-precision land-utilization remote sensing updating technology with synergistic multisource spatio-temporal data | |
CN109614926B (en) | Distributed optical fiber sensing signal mode identification method and system based on prediction model | |
Wang et al. | [Retracted] Remote Sensing Satellite Image‐Based Monitoring of Agricultural Ecosystem | |
Yu et al. | Network security monitoring method based on deep learning | |
CN110380801A (en) | The method that collaborative sensing algorithm and more USRP based on LSTM are realized |
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 |